PRIVATE PRISONS, PUBLIC FUNCTIONS, AND THE
MEANING OF PUNISHMENT
MARY SIGLER∗

I.

INTRODUCTION

For more than three decades, the prison population in the United
States has steadily increased.1 “For the first time in American history,
more than one in every 100 adults is now confined in an American jail or
prison.”2 During the course of this rapid expansion, states and the federal
government have come to rely increasingly on private prisons. In 2007,
private detention facilities housed more than seven percent of incarcerated adults in federal and state prisons, up from three percent in 1999.3
At least thirty-five states and the District of Columbia now have private
prisons; many more send inmates to private facilities.4 The Federal Bureau of Prisons pays private providers to house approximately eleven
percent of federal inmates,5 and Immigration and Customs Enforcement
houses about thirty-eight percent of its detainees in privately managed
facilities.6
The primary impetus for the private-prison boom that began in the
1980s was the belief that for-profit corporations, subject to the rigors of
∗
Professor of Law, Arizona State University College of Law. I am grateful
to Jeffrie Murphy, Antony Duff, and Richard Dagger for their thoughtful comments, and to participants in the 2009 Markets and Responsibilities workshop at
Oxford University.
1
Pew Center on the States, One in 100: Behind bars in America 2008, at 5.
2
Id. at 3; In 1985, the figure was one in every 450 people. Id.
3
Id. at 6. The United Kingdom has also experienced a burgeoning prison
population and increased reliance on private prisons. In England and Wales,
private prisons account for eleven percent of the prison population. Robert Verkaik, Private Prisons Performing Worse than State Run Jails, INDEPENDENT, June
29, 2009.
4
Leonard C. Gilroy et al., Reason Foundation Annual Privatization Report
2008, at 106.
5
Federal
Bureau
of
Prisons,
at
http://www.bop.gov/locations/weekly_report.jsp.
6
Gilroy, supra note _ at 106.. Immigration detainees are not “prisoners” in
the legal sense, and their incarceration does not constitute punishment. See 28
U.S.C. 1915(h) (defining “prisoner” as “any person incarcerated or detained in
any facility who is accused of, convicted of, sentenced for, or adjudicated delinquent for, violations of criminal law or the terms and conditions of parole probation, pretrial release, or diversionary program”). Although many of the same
practical concerns about privatization apply in the context of immigration detention, see infra Part II, the argument about the meaning of punishment does not.

Electronic copy available at: http://ssrn.com/abstract=1650872

2010

PRIVATE PRISONS

2

market competition, could deliver correctional services more efficiently
than could the state. According to the private providers, cost savings
come through lower payroll costs, consolidation of prisoner populations,
and the siting of facilities in low-cost markets.7 Indeed, one recent study
found that private prisons may reduce the cost of housing inmates by as
much as fifteen percent; another found that between 1999 and 2004 states
were able to save up to fifteen million dollars on their yearly corrections
budgets by using privately managed prisons to house at least some of
their inmate population.8 During the present economic crisis, many
states are poised to increase their reliance on private prisons. In Oklahoma, for example, where approximately one quarter of the state’s inmates are already housed in private facilities, legislators recently voted to
expand contracts with private prison providers.9 Similar proposals are
pending in California, Florida, and Arizona.10
Despite this enthusiasm for privatization, the cost-saving claim remains controversial. Some researchers have observed that private prison
contractors typically siphon off the least costly inmates—those who are
healthier and less violent than the incarcerated population as a whole.11
More generally, simple cost comparisons that appear to favor private
facilities are based on per diem rates that may not reflect the full cost of

7

Stephanie Chen, Larger Inmate Population Is Boon to Private Prisons,
WALL ST. J., Nov. 19, 2008.
8
James Blumenstein, et al., Do Government Agencies Respond to Market
Pressures? Evidence from Private Prisons, Health Policy Center, Vanderbilt
Institute for Public Policy Studies (2007).
9
Michael McNutt, Oklahoma House Fights Prospect of Terrorism Detainees, OKLAHOMAN, May 23, 2009. The legislation also extracts various concessions from private prison providers, including reduced costs, cancelation options,
and an increase in inmate services. Id.
10
Steven Harmon, Schwarzenegger Hopes Budget Debacle Leads to Lasting
Reform, MERCURY NEWS, June 2, 2009; Steve Bousquet, Florida Legislature
Gives OK to Ship Inmates Out of State, MIAMI HERALD, June 7, 2009; Luige del
Puerto, Arizona Senate Panel Passes FY10 Budget after 9-Hour Debate, ARIZ.
CAP. TIMES, May 22, 2009. In the United Kingdom, the “Government is committed to building five more private prisons to accommodate the growing prison
population, which is predicted to rise to 96,000 by 2014.” Verkaik, supra note 7.
11
Kevin Pranis, Cost-Saving or Cost-Shifting: The Fiscal Impact of Prison
Privatization in Arizona (2005), Private Corrections Institute Inc., at
http://www.nicic.org/Library/020388 (“Prisoners housed in private facilities were
far less likely to be convicted of serious or violent offenses, or to have high
medical and mental health needs, than prisoners housed in public facilities used
to generate cost comparisons.”); see also Pew, supra note 1 at 12 (noting that
medical care is one of the most expensive items in corrections budgets; JOHN D.
DONAHUE, THE PRIVATIZATION DECISION (1989) 158 (reporting on the tendency
to send “the best in the bunch” to private facilities).

Electronic copy available at: http://ssrn.com/abstract=1650872

2010

PRIVATE PRISONS

3

incarceration.12 Others contend that whatever cost savings private prisons achieve, they come at the expense of inmate well-being—that private
prison operators save money by skimping on personnel training and staffing, offering only minimal educational programming and vocational
training, and housing inmates in cramped and unsafe quarters. Moreover,
the profit motive is thought to create perverse incentives to extend inmate
sentences and promote criminal justice policies that yield more and
longer prison sentences regardless of whether they are in the public interest. Finally, critics decry the delegation of governmental functions to
private actors and the threat it poses to democratic accountability and the
rule of law.
While these important policy considerations may be reason enough
to worry about the proliferation of private prisons, I argue that an even
more fundamental consideration concerns the nature and justification of
punishment in a liberal democratic polity. Punishment under law is a
profound exercise of state power the meaning and justification of which
depend on the social and political institutions that authorize it. In a liberal state—in the United States—punishment is inflicted for public
wrongs in the name of the people themselves. Although it may be justified with reference to a plurality of public values, it is a predominantly
retributive practice that constitutes and expresses society’s moral condemnation of criminal conduct. Central to this conception of punishment
is the relationship between punisher and punished, for it transforms otherwise socially objectionable conduct, such as the deprivation of liberty,
into a just social practice. Punishment is thus meaningful not primarily
as a means to an end; rather, punishment instantiates justice. The delegation of punishment through prison privatization attenuates the meaning of
punishment in a liberal state and undermines the institution of criminal
justice.
Or so I shall argue. I begin by examining the phenomenon of privatization in the liberal democratic context generally, considering a range
of activities that, while publicly financed, are performed in whole or in
part by private entities. This inquiry suggests a number of criteria for
evaluating the wisdom of privatization in practical terms and the suitability of privatization more generally. In view of these considerations, I
outline some of the practical issues raised by prison privatization, ranging from the potentially distorting effects of the profit motive to the challenge of effective oversight of contractor discretion and the implications
for due process and the rule of law. This analysis highlights serious concerns about the trend toward penal privatization that ultimately rest on a
set of empirical claims about its tendency to yield undesirable social and
political consequences.
12

Gerry Gaes, Cost, Performance Studies Look at Prison Privatization
(2008),
National
Institute
of
Justice,
at
http://www.ojp.usdoj.gov/nij/journals/259/prison-privatization.htm.

2010

PRIVATE PRISONS

4

The case that I construct against prison privatization is, by contrast,
normative and conceptual. According to this approach, a judgment about
prison privatization depends not (only) on its potential to produce bad
consequences but on the meaning and justification of punishment itself. I
undertake this analysis, first, by canvassing the traditional purposes of
punishment, then developing a justificatory account that is predominantly
retributive and broadly consonant with liberal-democratic values. This
sets the stage for a discussion of the meaning of liberal-democratic punishment in the United States based on a conception of criminal justice as
a type of moral dialogue between individuals and their community. On
this view, the state is the legal embodiment of the political community,
calling offenders to account for their wrongful conduct. Imprisonment,
the primary mode of serious punishment in the United States, represents
a grave form of censure that constitutes the normative community’s
moral condemnation of serious wrongdoing. Filtered through the medium of privatization, this communication is necessarily garbled. In our
rush to privatization, we risk compromising the meaning and value of our
punitive institutions and practices. If and when the latest economic crisis
passes, we may find that it is too late to recover them.
II. PRIVATIZATION
As a general matter, privatization is “the use of the private sector in
the provision of a good or service, the components of which include financing, operations (supplying, production, delivery), and quality control.”13 Private actors involved in the provision of public goods or services may include corporations, interest groups, and nonprofit organizations. Their level of involvement varies widely, ranging from consulting
and standard-setting to financing, constructing, and operating facilities
designed to fulfill governmental responsibilities. For example, professional associations, such as the American Bar Association, often develop
and enforce professional standards in a variety of settings; religious organizations offer alternatives to welfare and public education funded
through voucher programs or tax credits; and for-profit corporations supply catering, medical, and waste management services to governmental
entities on a contract basis, as well as building and managing hospitals,
prisons, and military facilities to meet public-sector demand. At the far
reaches of public-private partnership are such quasi-governmental entities as “Fannie Mae” and Freddie Mac,”14 the United States Postal Ser-

13

Kevin R. Kosar, Privatization and the Federal Government: An Introduction, Congressional Research Service Report for Congress, 3 (Dec. 28, 2006).
14
Fannie Mae and Freddie Mac are government-sponsored enterprises chartered by Congress.
See http://www.fanniemae.com/about/index.html;
http://www.freddiemac.com/corporate/company_profile/faqs/.

2010

PRIVATE PRISONS

5

vice,15 and the American Red Cross.16 Although my current focus is
primarily “contracting out”—agreements between governments and private providers to supply public goods and services17—I first briefly review the historical context out of which this particular form of privatization developed.
A. Context and Case for Privatization
As many commentators have observed, the traditional distinction between public and private realms is both powerfully intuitive and somewhat misleading.18 Historically, a number of what we have come to regard as public functions were performed through private initiative, including police and fire protection, tax collection, and education. In this
setting, government relied on a variety of regulatory mechanisms, including tax policy and corporate law, to encourage private actors to pursue
the public interest along with their own.19 With the expansion of the administrative state in the middle part of the twentieth century, people’s
expectations of government, and thus the opportunity for contracting out,
greatly increased.20 In the modern era, public-private partnerships are
more likely to involve direct financing, joint ventures, and market-style
competition.21 Today, “[v]irtually any example of service provision or
regulation reveals a deep interdependence among public and private actors in accomplishing the business of governance.”22
Despite this interdependence, the distinction between public and private remains meaningful insofar as “[p]rivate firms and public agencies
tend to have different capacities, cultures, and priorities…and respond to
different incentives.”23 Indeed, the fact that public and private providers
may be animated by a different set of norms and goals gives rise to a
range of concerns about the privatization of governmental responsibilities. Before turning to these issues, I first briefly outline the case for privatization and identify the major grounds for criticism. I then examine
the phenomenon of prison privatization in particular, highlighting the
specific challenges it presents.
15
The United States Postal service is a semi- independent federal agency.
See http://usgovinfo.about.com/od/consumerawareness/a/uspsabout.htm.
16
The American Red Cross operates under a congressionally issued corporate charter. See http://www.redcross.org/museum/history/charter.asp.
17
Jody Freeman, The Private Role in Public Governance, 75 N.YU.L. REV.
543, 552 (2000).
18
See, e.g., Martha Minow, Public and Private Partnerships: Accounting
for the New Religion, 116 HARV. L. REV. 1229 (2003); Freeman, supra note 19 at
547.
19
Minow, supra note 20, at 1237.
20
Id. at 1240, Donahue, supra note 13, at 4.
21
Minow at 1240-41.
22
Freeman, supra note 19, at 547.
23
Id. at 550.

2010

PRIVATE PRISONS

6

The basic case for privatization, particularly contracting out, turns on
the greater efficiencies available through the operation of market mechanisms. Because governments function more or less as monopolies, they
lack adequate incentives to pursue cost-saving innovations. Through the
power of competition, however, private firms are motivated to deliver
goods and services more cost-effectively—lest competitors underbid
them—through stream-lined management and operations. Whereas governments must contend with entrenched bureaucracies and public employee unions, private entities have the flexibility to hire, fire, and adjust
staffing and wage levels to respond to prevailing market conditions. In
addition, the availability of private capital facilitates the timely design
and implementation of new ventures.
Beyond the economic advantages associated with privatization, the
competition it generates may lead to experimentation in the provision of
social services and novel responses to persistent social problems.24 At
least some proponents of privatization also view it as a means to reduce
the size and power of government and thereby promote greater individual
liberty and choice.25 The availability of vouchers or tax credits, for example, allows parents to choose educational options that reflect their values and traditions, while at the same time promoting pluralism and community.26 Such “[g]roup affiliations can encourage virtues of participation, self-governance, and mutual aid…while allowing freedom from the
controlling force of a powerful government.”27
B. Potential Problems
Despite the considerable potential and expectations for privatization
in the United States, the results to date reveal a mixed record.28 Although
one leading proponent of privatization contends that “the empirical question has long since been answered in its favor,” even he acknowledges
that a number of practices have weakened the case for privatization and
threaten its viability.29 In particular, he notes the troubling use of campaign contributions to influence the awarding of contracts, the lack of
transparent decision-making processes, and the use of anti-competitive
tactics by private providers.30 These concerns, and many others, high24

Minow, supra note 20, at 1245.
Id.
26
Id. at 1244. Minow endorses “nontoxic pluralism,” which requires that
individuals be permitted “to exit and participate in multiple groups or even none
at all.” Id. at 1245.
27
Id.
28
For a discussion of the mixed results of privatization in various contexts,
see John J. DiIulio, Jr., Government by Proxy: A Faithful Overview, 116 HARV.
L. REV. 1271, 1273, 1282 (2003); Minow, supra note 20, at 1248-49.
29
Robert W. Poole, Jr., Privatization, Library of Economics and Liberty, at
http://www.econlib.org/library/Enc/Privatization.html (visited Aug. 16, 2009).
30
Id.
25

2010

PRIVATE PRISONS

7

light the challenges associated with privatization. In what follows, I take
up these issues under four general headings—market failure, public accountability, legitimacy, and nonpublic motives.31
A market failure, as I use the term here, occurs when the ordinary
operation of market mechanisms cannot be counted on to yield optimal
outcomes. Thus, for example, if a government contracts with a private
firm for the provision of an essential service that requires significant initial capital expenditures and expertise, the government is in a poor position to negotiate—or deny—contract extensions if it has become dependent on the private provider’s service. In a variety of contexts, including
prison construction and management, the firm may be able to raise rates
dramatically over the initial contract bid because the government cannot
forgo the service—say, housing dangerous criminals—but lacks readily
available alternatives. Additionally, private firms face the risk of business failure. A corporation may mismanage its operation to the point of
bankruptcy, leaving government either to bail out the operation financially or scramble to identify alternative service providers, which may
themselves extract a premium based on the government’s desperation for
immediate supply. Other sources of market failure include the use of
campaign contributions to influence the public officials who award government contracts32 and the inherent challenges of drafting suitable contracts that specify with adequate precision the terms and expectations of
performance. In the absence of “solid and measurable performance standards,” it will be difficult to determine whether government is “getting
the full measure of services it expects at the promised lower cost.”33
The challenge of drafting sufficiently detailed contracts points to a
further set of concerns relating to privatization—democratic accountability. Effective public oversight and control requires transparency in the
contracting process as well as detailed public disclosure regarding contract terms and performance. Where the privatization process lacks
mechanisms for specifying public goals and evaluating the quality of
privately provided services, however, citizens cannot make informed
judgments about the performance of the contract—or of their elected
officials.34 As one commentator notes, “Self-government will not retain
meaning if major decisions about public resources and the shape of collective experiences occur without the knowledge and participation of the
nation’s citizens.”35 Finally, to the extent that the delegation of government functions to private actors diminishes legal liability, it weakens a

31

As will become clear, these are not so much discrete categories as they are
convenient terms for overlapping clusters of value.
32
Contracts awarded on the basis of aggressive lobbying, rather than the
competitiveness of the bid, are less likely to reflect prevailing market values.
33
Poole, supra note 31.
34
Minow, supra note _ at 1260.
35
Id.

2010

PRIVATE PRISONS

8

powerful mechanism for ensuring accountability in the exercise of public
power.36
The delegation of public functions to private actors also gives rise to
concerns about political legitimacy. In the liberal-democratic context,
legitimacy derives from the will of the people and the rule of law. Specifically, self-government entails a significant role for popular participation in the making and implementation of the rules and policies that bind
us. In the context of privatization, however, the imprecision of drafting
ensures that contracts will under-specify the terms and expectations of
service, leaving extensive discretion to private actors facing unanticipated contingencies. Under these circumstances, the quality and character of public services will depend on the ad hoc judgments of private actors, who may or may not be motivated by public concern. Although the
exercise of contractor discretion in some contexts—say, garbage collection—is likely to be unproblematic, in other cases, individual citizens—
welfare recipients, school children, inmates—may be subject to arbitrary
decision making and denied the protection of the rule of law. In a political environment where even inter-governmental delegation raises concerns, delegation to private actors is even more worrisome. For
“[p]rivate actors exacerbate all of the concerns that make the exercise of
[delegated] discretion so problematic.”37
A final set of concerns stems from the nonpublic motivations characteristic of private actors. Although private firms and public entities alike
rely on individual workers earning paychecks to carry out their activities,
firms and their employees operate within the domain of competitive
profit-seeking.38 In this environment, “most private organizations may
not develop the institutional norms of professionalism and public service
that characterize many public bureaucracies.”39 To say this is not, of
36
The state action doctrine imposes constitutional obligations on private actors under certain conditions. See, e.g., Edmonson v. Leesville Concrete Co., 500
U.S. 64 (1991); West v. Atkins, 487 U.S. 42 (1988). Despite its potential to hold
private actors accountable for the exercise of public power, the applicability of
the state action doctrine is quite narrow. Indeed, “[a]s a mechanism for disciplining private actors, the doctrine proves inept.” Freeman, supra note 19, at 579.
37
Freeman, supra note 19, at 574. As Freeman notes, private actors are
“one step further removed from direct accountability to the electorate” and “remain relatively insulated from the legislative, executive, and judicial oversight to
which agencies must submit.” Id. For an argument that delegation of prison
functions to private providers constitutes an unconstitutional delegation, see Joseph E. Field, Making Prisons Private: An Improper Delegation of a Governmental Power, 15 HOFSTRA L. REV. 649 (1987).
38
To the extent that they are not motivated by profit, it is unclear how the
market discipline that makes privatization appealing could yield the desired consequences.
39
Freeman, supra note 19, at 574. Freeman notes that private organizations
may be motivated by ideology or group allegiance rather than profit. Id. For
convenience, I will refer to these collectively as nonpublic motives. Also, private

2010

PRIVATE PRISONS

9

course, to denigrate the profit motive—or to glorify public service—only
to recognize that it is likely to generate a different set of workplace
norms and values. In particular, because public employees are generally
insulated from strict market discipline, their loyalty is to the government
and its purposes; private employees’ incentives are likely to be more directly linked to their firm’s bottom line.40 Moreover, just “the appearance of private motives in a public domain can undermine respect for
government and even generate doubt about whether the government is
sincerely pursuing public purposes.”41
C. Private Prisons
The private prison boom of the 1980s marked the beginning of only
the latest chapter in a long history of private sector involvement in public
corrections in the United States. Before the advent of punitive incarceration in the late eighteenth century, jails were run by for-profit providers
paid by local governments to house debtors and suspects awaiting trial or
capital punishment.42 The move to a government-run penitentiary system
around 1790 was based on the idea that a term of imprisonment might
facilitate the reform of wrongdoers, especially if they were occupied in
productive labor.43 Before long, private firms began contracting with
prisons for the use of inmate labor, transforming prison workshops into
sites of industrial production.44 In this way, government was able to defray the cost of incarceration and perhaps turn a profit, while private industry gained access to cheap labor and greater profit margins. During
the nineteenth century, a mix of leasing and contract arrangements proliferated, sending inmates to work for mines, railroads, and construction
contractors; in some cases, private contractors assumed responsibility for
all aspects of prison management in exchange for access to inmate labor.45
The contemporary practice of contracting out to private corporations
for total prison management is linked to significant public policy shifts in
the 1970s. First, the rehabilitative ideal, which had dominated penal
practices during much of the twentieth century, fell out of favor when it
produced disparate and indeterminate criminal sentences while failing to
prison contractors, unlike private schools and private rehabilitation providers, are
almost exclusively for-profit corporations.
40
Sharon Dolovich, State Punishment and Private Prisons, 55 DUKE L.J.
437, 518-19 (2005).
41
Minow, supra note 20, at 1234.
42
Dolovich, supra note 42, at 451.
43
DAVID SHICHOR, PUNISHMENT FOR PROFIT 26 (1995). The move to public
penitentiaries also reflected the developing Enlightenment view that criminal
conduct constituted a transgression against the community. See, e.g., CESARE
BECCARIA, ON CRIMES AND PUNISHMENTS (1764); see also infra, Part IV.
44
SHICHOR, supra note 45, at 28-29.
45
Id. at 34-35.

2010

PRIVATE PRISONS

10

reduce recidivism or deter crime. Soaring crime rates led to demands for
tougher criminal justice policies to protect the public and provide criminal offenders their just deserts. At the same time, laissez faire economics, characterized by faith in the productive capacity of private property,
suspicion of public regulation, and an aversion to “big government,” was
on the rise in both the United Kingdom and the United States, championed by both Margaret Thatcher and Ronald Reagan. In the United
States, the resulting criminal justice policies included a dramatic expansion in the number and type of offenses carrying a term of imprisonment
and the imposition of significantly longer prison sentences on serious
offenders. Facing burgeoning demand for prison beds, government
turned to the private sector, which promised to supply comparable corrections services at greatly reduced costs.
Despite the ideological appeal, the primary impetus for the move to
private prisons in the United States was and remains financial. As prison
costs continued to rise throughout the last several decades, private corrections companies offered to house and manage inmates at substantially
lower rates than the states were able to achieve themselves. In 2008, for
example, California’s contract with GEO Group, Inc. cost the state $40
per inmate per day, compared to $118 per day—the average cost for the
state to house inmates in its own facilities.46 Although savings rates in
other jurisdictions are less dramatic—ranging from two to fifteen percent—it still amounts to millions of dollars in savings annually.47 According to the private providers, the key to private sector cost savings is
cheaper private-sector labor. In California, for example, state corrections
officers, covered by collective bargaining agreements, earn up to 35.00
per hour, while GEO employees, who are not unionized, earn between
10.00 and 16.00 per hour. In addition, private operators can shift and
consolidate geographically disparate prisoner populations, concentrating
inmates in facilities located in areas with low real estate, wage, and construction costs. Finally, because private firms must compete against industry rivals, as well as government itself, they have the necessary incentives to develop innovative corrections strategies and streamline their
operations in order to win and retain government contracts.
Unfortunately, this upbeat picture of the public-private comparison
obscures more than it clarifies. As an initial matter, the studies (and contracts) that reflect dramatically reduced per diem rates in private facilities
are misleading even on their own terms.48 Inmates with significant men46

Andy Furillo, Schwarzenegger Seeks $67 Million Boost for Private Prison
Operator, SACBEE, Mar. 9, 2008. In 2008, GEO Group sought an increase to a
per diem rate of $60 per inmate, citing the need to raise the salaries of correctional officers. Id.
47
Gaes, supra note 14 (citing studies reflecting cost savings of between two
and fifteen percent).
48
One leading researcher points out the methodological complexities associated with cost comparisons. See Gaes, supra note 14. In particular, while some

2010

PRIVATE PRISONS

11

tal or physical health needs cost more to incarcerate than inmates without
such problems. Similarly, both violent inmates and particularly vulnerable inmates require more restrictive—and more expensive—security
measures to ensure the safety of inmates and prison staff.49 Private contractors routinely decline to accept such inmates, an option unavailable to
state-run facilities.50 Moreover, contract per diem rates typically do not
include the costs of programming and medical services that governments
must pay for separately. More generally, the promised innovation and
dramatically lower recidivism rates never materialized.51
Meanwhile, focusing on cost comparisons to the exclusion of other
considerations means neglecting a range of important values at stake in
the corrections context, implicitly accepting efficiency as the prime value
of penal policy.52 In fact, several of the practical concerns raised by privatization generally—market failure, public accountability, democratic
legitimacy, and nonpublic motives—apply with special force in the context of prison privatization.
The traditional market mechanisms for disciplining poor performance may not operate effectively in the private prison setting. As an initial matter, the “beneficiaries” of the contract—inmates—are not the purchasers of prison services. Thus, unlike the market for private education,
for example, where families can research alternatives, make informed
selections, and withdraw from unsatisfactory arrangements, inmates do
not have a say in the decision whether to enter or terminate a private
prison contract. Although the same is true when governments contract
out for garbage collection—the beneficiaries of the contract are not a
party to the contract—dissatisfied citizens are in a strong political position to demand improved service. Inmates, by contrast, are virtually
powerless to effect change in the face of unsatisfactory prison conditions.
Most lack the basic right to vote; and in any case, they constitute an unpopular minority without political influence or efficacy.
studies calculate costs based on actual governmental outlays for public facilities,
others estimate the costs that would have been incurred for comparable inmate
populations. These divergent approaches can yield dramatically different results.
I. For example, a pair of contemporaneous studies—one privately funded, one
done by the Board of Prisons—assessed the per diem costs of the same four facilities. The industry study found cost-savings of nearly 15%, while the BOP
found savings of only 2%. Id.
49
See Pew, supra note 1.
50
Gaes, supra note 14; Donahue, supra note 13, at 158.
51
DONAHUE, supra note 13, at 162 (“[I]n general, incarcerating people is an
enterprise with relatively little scope for resource-sparing technical progress.”);
id. at 154 (noting boast of private corrections official regarding recidivism);
Dolovich, supra note 42, at 476 (noting the lack of evidence of cost-saving innovation in private-sector prisons)..
52
Cf. LEO KATZ, BAD ACTS AND GUILTY MINDS 25 (1987) (characterizing s
as “price worship” the tendency to assign greater importance to quantitative over
qualitative assessments of value).

2010

PRIVATE PRISONS

12

Even governments may not be well positioned to respond to noncompliance by private prison contractors. Public officials dissatisfied
with a contractor’s performance—or rate increases—cannot realistically
cancel the contract before finding alternative placements for hundreds of
inmates. The high start-up costs for prison operations ensure that a relatively small number of players will (and do) dominate the market, giving
them considerable leverage when negotiating with governments desperate to place inmates.53 Although a handful of states have canceled contracts for noncompliance, they appear reluctant to rescind promptly even
in cases of extreme inmate abuse.54
A further source of concern arising from the private prison market is
the role of lobbyists. Apart from the usual worries about the use of campaign contributions to curry favor with elected officials or the potential
for self-dealing, prison industry lobbyists may play an even more pernicious role—developing and promoting criminal justice policies solely to
advance their financial interests. The most widely reported example is
the American Legislative Council, a Washington-based policy organization heavily funded by the two leading private prison firms, which successfully promoted such get-tough sentencing laws as “three strikes” and
“truth in sentencing.”55 These and similar policies contributed substantially to increased demand for private prison beds—and to the need for
contracting out. Indeed, private firms, as rational actors subject to market pressures, have every incentive to pursue such a strategy. As one
commentator cautions, “we should…be wary that private-corrections
corporations may initiate advertising campaigns to make the public feel
more fearful of crime than it already is, in order to fill the prisons and
jails.”56
The challenges of contract drafting also create special problems in
the private prison context. As many commentators have noted, the “incarceration function…proves difficult to specify.”57 As a result, contract
terms are likely to be imprecise, providing an insufficient basis for gauging contractor performance. This problem is exacerbated in the prison
53

DONAHUE, supra note 13, at 165 (noting the factors contributing to the
small number of private corrections firms and the difficulty for states of attempting to switch companies); see also Pranis, supra note 13 (describing the situation
in Louisiana during the mid-1990s when the state was prevented from canceling a
private prison contract for noncompliance when it learned that doing so would
adversely affect its bond rating).
54
Dolovich, supra note 42, at 498-99 (describing instances of inmate abuse
and the time lag before contract cancellation).
55
Id. at 526-27 (citing various sources).
56
Ira P. Robbins, Privatization of Corrections, 40 VAND. L. REV. 813, 82728 (1987).
57
Freeman, supra note 19 at 632; see also DONAHUE, supra note 13, at 166
(“Attempts to exhaustively spell out contingencies and assign rights and duties
for each conceivable case will be awkward and burdensome, and will almost
surely fail to cover everything.”.

2010

PRIVATE PRISONS

13

setting, where the quality of performance—from the provision of medical
care to the use of force—can mean for inmates the difference between
life and death. Moreover, these activities take place behind closed doors
in service of beneficiaries who lack meaningful recourse in cases of poor
performance. In these circumstances, officials can be confident about
neither the value of the contract nor the well being of inmates. These
obstacles to public accountability suggest the challenges to effective
oversight in precisely those circumstances that call for special vigilance.
A related set of issues is the threat to legitimacy resulting from inevitably vague contract terms. To the extent that the parameters for the use
of force, inmate discipline, and administrative classification are underspecified, private corrections employees exercise considerable discretion
on a daily basis. While these decisions can have a profound effect on the
length and conditions of confinement, they result from “uncontracted-for
contingencies” that cannot be settled in advance.58 Instead, private corrections employees—less well paid, less well trained, and less experienced than their public sector counterparts59—will be left to make critical
decisions, without reference to standards of due process or the rule of
law.60
Finally, the profit motive that fuels prison privatization exerts a constant pull in the direction of cost cutting. As various commentators have
observed, contractors can attempt to save costs by reducing the amount
spent on meeting inmates’ needs—food, housing, security, medical
care—and by keeping wages low.61 Because the delivery of these services is hidden from meaningful scrutiny, the temptation to cut corners is
likely to be overwhelming.62 At the same time, private prisons offer sub58

Dolovich, supra note 42, at 478-79.
See, e.g., Furillo, supra note 48 (noting inferior training and generally
lower levels of education of private corrections employees); DONAHUE, supra
note 13, at 164 (presenting “suggestive” data to illustrate that private corrections
employees are likely to be generally younger, less experienced, less well educated, and less well trained).
60
Some critics note that these decision makers also have a financial interest
in longer prison terms. See, e.g., Dolovich, supra note 42, at 518-19 (noting the
“possibility that private prison operators, whose profitability depends on maintaining a high occupancy rate, could encourage their employees in subtle and notso-subtle ways to make judgments regarding individual inmates’ behavior so as
to prolong the amount of prison time that inmates serve”); Ahmed A. White, Rule
of Law and the Limits of Sovereignty: The Private Prison in Jurisprudential
Perspective, 38 AM. CRIM. L. REV. 111 (2001) (noting the potential for private
prisons to “sustain their occupancy rates and therefore their revenues…by manipulating inmates’ terms of incarceration”).
61
See, e.g., DONAHUE, supra note 13, at 170; Dolovich, supra note 42, at
474-75.
62
This is not to suggest that all—or any—private prison contractors are indifferent to public values or inmate well-being. As Donahue notes, most firms
probably enter the industry with good intentions. But the structure of the enterprise—the quest for profits and the pressure of competition—create strong incen59

2010

PRIVATE PRISONS

14

stantially lower wages than public facilities and spend considerably less
on training and retention. Predictably, they draw younger workers with
less education and experience and have significantly higher turnover
rates.63 In this environment, employees are less likely to develop the
commitment to public values and shared norms of professionalism that
contribute to rule compliance and promotion of the common good.64 As
one observer lamented, “the private sector is more interested in doing
well than doing good.”65
III. JUSTIFYING PUNISHMENT
Criminal punishment, in general terms, is the authorized imposition
of deprivations—liberty, property, or other goods to which one has a
right—or the imposition of special burdens because one has been found
guilty of a criminal violation.66 In the absence of criminal wrongdoing,
the sort of treatment that we call punishment—taking life, liberty, or
property—would itself represent a grave injustice. For this reason, we
must be able to provide a justification, or some combination of justifications, that makes such otherwise prohibited treatment permissible or even
obligatory.
The range of acceptable justifications—even the need for such a justification—depends on the social and political institutions that authorize
criminal punishment. “Thus the color and texture of any possible justification for punishment will depend upon more general political and moral
theory, consistent with the responsibilities for legal protection afforded
by a just society.”67 In the Anglo-American tradition, legitimate punishment reflects such basic liberal-democratic values as liberty, equality,
and rule of law. Historically, the justification for punishment has shifted
among the traditional accounts according to changes in the prevailing
social and political norms. The ascendancy of the liberal commitment to
autonomy and individual rights seems to have secured the place of such
retributive values as culpability and desert, while discrediting utilitarian

tives to cut costs come what may. “And without robust measures to guarantee
conditions of confinement”—about which he is skeptical—“the businesspeople
least constrained by scruples are likely to enjoy a competitive advantage in the
imprisonment industry.” DONAHUE, supra note 13, at 170.
63
DONAHUE, supra note 13, at 164.
64
See generally LAURA DICKINSON, OUTSOURCING WAR AND PEACE (forthcoming 2010) (comparing operational norms among public and private military
personnel in Iraq).
65
Robbins, supra note 58, at 816.
66
See Hugo Adam Bedau, Punishment, in STANFORD ENCYCLOPEDIA OF
PHILOSOPHY, at http://plato.stanford.edu/ (last visited June 5, 2009).
67
Id.

2010

PRIVATE PRISONS

15

approaches to the extent that they lack side constraints that would confine
punishment to the blameworthy.68

A. Traditional Justifications
The traditional scheme of classification divides the justifications for
punishment in Anglo-American criminal law into two broad categories,
utilitarian and retributive. Utilitarian69 justifications—principally incapacitation,70 deterrence,71 and rehabilitation72—are defended in terms of
the positive consequences they are believed to bring about. In the case of
deterrence, for example, its viability as a justification for punishment is
measured in terms of its efficacy in achieving the goal of crime prevention by means of the threat of punishment. From the utilitarian perspective, punishment is justified in terms of its effectiveness in preventing
crime while at the same time generating the least possible amount of human suffering. Proportionality is thus defined in terms of the relevant
utilitarian goals, prescribing exactly that amount of punishment necessary
to achieve those goals; any suffering above that amount is excessive and
unjustifiable.
Retributivism, by contrast, is centrally concerned with the imposition
of punishment in proportion to an offender’s moral desert.73 On this
68

See, e.g., John Rawls, Two Concepts of Rules, 64 PHIL. REV. 3 (1955);
H.L.A. HART, PUNISHMENT AND RESPONSIBILITY 11-12 (1968).
69
Utilitarianism is the most prominent form of consequentialism, the view
that actions should be evaluated in terms of their consequences. For the utilitarian, the consequence to be maximized is happiness—or utility.
70
Incapacitation involves disabling an offender from engaging in further
criminal conduct. The most obvious forms of incapacitation are imprisonment
and execution; in both cases, offenders are physically prevented from offending
again. See, e.g., Jeremy Bentham, Panopticon Versus New South Wales, in 4 THE
WORKS OF JEREMY BENTHAM 183 (John Browning ed., Russell & Russell, Inc.
1962) (1838) (“This contrivance [incapacitation] was as firmly laid down in
school-logic as could be wished. Mischievously or otherwise, for a body to act in
a place, it must be there.”).
71
Deterrence may be either “general” or “specific.” General deterrence is
the “prevention of similar offenses on the part of individuals at large, viz. by the
repulsive influence exercised on the minds of bystanders by the apprehension of
similar suffering in case of similar delinquency.” Specific deterrence is “prevention of similar offenses on the part of the particular individual punished in each
instance, viz. by curing him of the will to do like in the future.” See id. at 174.
72
Rehabilitation involves the attempt to reform a wrongdoer, either in Bentham’s sense—by “curing” the offender of the impulse to engage in wrongdoing—or by otherwise reforming an “offender’s character, habits, or behavior
patterns so as to diminish his criminal propensities.” ANDREW VON HIRSCH,
DOING JUSTICE 11 n. (1976).
73
See MICHAEL S. MOORE, PLACING BLAME 105 (1997).

2010

PRIVATE PRISONS

16

view, punishment of the deserving is intrinsically good; its justification
does not depend on any further positive consequences that punishment
might be expected to produce. In Kant’s classic formulation: “The law
concerning punishment is a categorical imperative, and woe to him who
rummages around in the winding paths of a theory of happiness looking
for some advantage to be gained by releasing the criminal from punishment or by reducing the amount of it….”74 Although a retributivist will
welcome the positive consequences that punishment may incidentally
yield—crime prevention or character reformation, for example—such
consequences are not part of the justification for punishment.75 Thus, a
“retributivist punishes because, and only because, the offender deserves
it.”76
A further set of approaches to the justification of punishment—
expressive or communicative accounts—do not fit neatly into either the
utilitarian or retributive categories, for they typically reflect elements of
both.77 In general terms, they conceptualize punishment as a form of
communication that expresses society’s moral condemnation of criminal
wrongdoing. In Joel Feinberg’s influential formulation, “punishment is a
conventional device for the expression of attitudes of resentment and
indignation, and of judgments of disapproval and reprobation, on the part
of the punishing authority himself or of those in whose name the punishment is inflicted.”78 While Feinberg’s account highlights community
condemnation as an “essential ingredient” in legal punishment, it suggests that some alternative mechanism could also serve as an appropriate

74

IMMANUEL KANT, THE METAPHYSICAL ELEMENTS OF JUSTICE *331 (J.
Ladd trans. 1965) (1780).
75
See ROBERT NOZICK, PHILOSOPHICAL EXPLANATIONS 374 (1996) (“These
further consequences are not to be dismissed simply; but we shall see them as an
especially desirable and valuable bonus, not as part of a necessary condition for
justly imposed punishment.”).
76
Michael S. Moore, The Moral Worth of Retribution, [hereinafter Moral
Worth] in PUNISHMENT AND REHABILITATION 94 (Jeffrie G. Murphy ed., 1995).
Some conceptions of retributivism do not regard desert as the controlling value.
See generally Herbert Morris, Persons and Punishment, 4 MONIST 52 (1968)
(defending a conception of retributive punishment based on the mutual assumption of benefits and burdens); Jean Hampton, The Retributive Idea, in JEFFRIE G.
MURPHY & JEAN HAMPTON, FORGIVENESS AND MERCY 111 (1988). Although
these conceptions represent an important contribution to the literature, they do
not reflect the prevailing view. See Bedau, supra note 68.
77
For variations of the expressive theory, see NOZICK, supra note 77, at 370;
ANTONY DUFF, PUNISHMENT, COMMUNICATION, AND COMMUNITY (2001); Hampton, , supra note 78, 130; Joel Feinberg, The Expressive Function of Punishment,
in DOING & DESERVING 95 (1970).
78
Feinberg, supra note 79, at 98.

2010

PRIVATE PRISONS

17

vehicle for expression of this message.79 Feinberg’s expressivism thus
has a decidedly utilitarian cast.80
Contemporary communicative81 accounts are more explicitly retributive, reflecting the prevailing orientation of Anglo-American criminal
law.82 Indeed, “[p]unishment in its very conception is now acknowledged to be an inherently retributive practice, whatever may be the further role of retribution as a (or the) justification or goal of punishment.”83
For communicative accounts this means, first, that punishment must be
understood “not as a contingently efficient means towards a further and
independently identifiable end.”84 Rather, punishment itself constitutes
the condemnatory communication; censure is internal to the practice.85
Second, the retributive orientation requires that “the relationship between
past crime and present punishment [is] central to the meaning and justification of punishment.”86 In these ways, the communicative account reflects the plural values that underlie Anglo-American criminal punishment, including its essentially retributive character.87
79
Id. at 105 (concluding that a state might have other ways of expressing
various messages, “but when it speaks by punishing, its message is loud and sure
to get across”).
80
Michael Moore has emphasized the distinction between utilitarian and retributive conceptions of punishment, designating traditional expressivist accounts
utilitarian. See Moore, Moral Worth, supra note 78, at 96. See also DUFF, supra
note 79, at 27, 206 n.29 (distinguishing retributivist from utilitarian versions of
expressive purposes in punishment).
81
Following Antony Duff, I will use the term communicative to describe
approaches that view punishment as a kind of dialogue between the community
and some or all of its members. Although expressivist accounts also typically
contemplate a recipient for the messages expressed, they do not require it. DUFF,
supra note 79, at 79; see also Dan Markel, Retributive Damages: A Theory of
Punitive Damages and Intermediate Sanction, 94 CORNELL L. REV. 239, 261-62
(2009).
82
See The American Law Institute, Model Penal Code: Sentencing, Tentative Draft No. 1, Part I. General Provisions (April 9, 2007) (proposing revisions
to MPC’s punitive purposes to emphasize blameworthiness and retribution); Bedau, supra note 68.
83
Bedau, supra note 68.
84
Antony Duff, Expression, Penance and Reform, in PUNISHMENT AND
REHABILITATION, supra note 78, at 169, 170.
85
NOZICK, supra note 77, at 374 (“The wrongdoer has become disconnected
from correct values, and the purpose of punishment is to (re)connect him. It is
not that this connection is a desired further effect of punishment: the act of retributive punishment itself effects this connection.”); Markel, supra note 83, at
260 (noting that “the good achieved by punishment is bound up in the faithful
practice of retributive punishment itself, so that the practice of punishment has an
intrinsic value that makes the practice and its limits both internally intelligible
and attractive”).
86
DUFF, supra note 86, at 170;see also NOZICK, supra note 77, at 369;
Markel, supra note 83, at 260.
87
Bedau, supra note 68.

2010

PRIVATE PRISONS

18

B. Justifying Private Prisons
With the basic approaches to justification in hand, we can begin to
see how they bear on the question of prison privatization. Before developing the retributive account of communicative punishment that I believe
militates against prison privatization, I briefly sketch the relationship
between the traditional justifications in general terms and the suitability
of prison privatization. This suggests that private prisons are most compatible with utilitarian approaches to punishment, especially rehabilitation and incapacitation.88 Notably, the rehabilitative ideal that took hold
in the late eighteenth century facilitated the move to prison industrial
operations involving the private sector. Similarly, the abandonment of
that ideal, and the shift toward incapacitation, coincides with the rise of
the modern private corrections industry. At the same time, the retributive
turn in penal philosophy may itself have been instrumental in the privatization boom, though based on a misapprehension of retributive values.
1.

Rehabilitation

Despite the abandonment of the rehabilitative ideal in the 1970s,
there was never a complete rejection of the idea that punishment might
effect a change of heart, mind, or habit in criminal wrongdoers. To be
sure, the most ambitious models of rehabilitation—from the Quaker vision of reforming souls to the psychiatric goal of curing the disease of
crime—seem neither appealing nor realistic in contemporary liberal
terms. Whereas the eighteenth-century penitentiary model presupposed a
set of shared religious commitments that can neither be taken for granted
nor coercively enforced in a liberal society, the psychiatric conception of
crime denies wrongdoers their status as responsible moral agents worthy
of the rights that liberalism secures.89 More modest rehabilitative goals,
however, constrained by liberal-democratic principles, reflect our best
understanding of the correlation between drug addiction, poverty, and
mental illness on the one hand, and criminal misconduct on the other.
Effective rehabilitation programs that target these correlates of crime are
thus good for offenders, who are exposed to constructive options, and
good for the society that experiences a reduction in crime.
The value of rehabilitative services, then, consists primarily in their
effectiveness. The point is to provide programming—drug counseling,
vocational training, or therapeutic techniques—that yields positive con88
Indeed, Bentham’s utilitarian conception of punishment was based on the
idea of contracting out. See DONAHUE, supra note 13, at 171.
89
Morris, supra note 78, at 83 (noting that “we display a lack of respect for
the moral status of individuals, that is, a lack of respect for the reasoning and
choices of individuals” if we treat acts of intentional wrongdoing as symptoms of
disease).

2010

PRIVATE PRISONS

19

sequences in the lives of offenders and their community. Rehabilitation
is thus an instrumental good, valuable to the extent that it produces the
desired results.
The case for privatization in the development and delivery of rehabilitative services is easy to make out. A combination of religious organizations, other non-profits, and for-profit firms competing to provide
rehabilitation services is likely to offer a range of cost-effective alternatives. Whether governments adopt the programs offered by a particular
group or whether offenders are permitted to choose their own programs,
the instrumental nature of the services provides the means to gauge their
success. That is, we can determine the effectiveness of a drug program
by assessing the rate at which it succeeds, relative to other approaches, in
helping offenders end their drug dependency; vocational programs by
their ability to impart the knowledge and skills for a trade or profession;
counseling services to the extent that they equip offenders to cope with
the stresses and temptations of everyday life. Moreover, unlike the incarceration function itself, which provides little room for innovation or
alternative philosophies,90 rehabilitative programming is ideally suited to
creativity and experimentation.91 Finally, to the extent that privatized
rehabilitation services engage offenders in religious or other civic groups,
they promote the values of pluralism and community.92
2.

“Retribution” and Incapacitation

When the psychiatric conception of rehabilitation gave way to retribution and incapacitation as the dominant penal values in the 1970s, the
stage was set for the emergence of the private prison. But whereas the
instrumentalist goal of incapacitation may be well suited to privatization,93 the retributive—or “just deserts” philosophy—is not. Unfortunately, the form of retribution that took root during this period was not
always true to such fundamental retributive values as proportionality and
humanity.94 Instead, perhaps as a reaction to the perceived laxity of the
era that preceded it, the just deserts philosophy too often amounted to
nothing more than a get-tough approach to criminal justice, producing
mandatory minimum sentences, repeat-offender provisions, and generally
longer prison terms across the board. Even more troubling was the cultivation of a social and political environment in which officials who ex90

DONAHUE, supra note 13, at 162-63; Dolovich, supra note 42, at 501.
See, e.g., Minow, supra note 20, at 1236, 1245.
92
Id. at 1244.
93
That is, if incapacitation is the goal of punishment, then private prisons
may be just as successful in disabling dangerous offenders as public facilities are.
94
See KANT, supra note 76, at*332 (arguing for proportionality between
crime and punishment); id. at *333 (insisting that punishment “must be kept free
from any maltreatment that would make an abomination of the humanity residing
in the person suffering it”).
91

2010

PRIVATE PRISONS

20

pressed skepticism about these policies were branded “soft on crime” and
turned out of office.95
To the extent that retribution degenerates into a form of vengeance,
indifferent to considerations of culpability and desert, it is compatible
with penal privatization. Under these circumstances, private prisons may
have a role to play in delivering cost-effective punishment that provides a
more or less humane environment for housing criminal offenders. But
absent a concern for proportionality and humanity, punishment ceases to
be recognizably retributive.
Despite this bleak scenario, there is of course no necessary connection between various utilitarian justifications for punishment and the
worst excesses of prison privatization.96 But by focusing on instrumental
goals in evaluating prison privatization, we neglect the essentially retributive character of our punitive institutions and practices. Fleshing out
the meaning of punishment in the liberal-democratic context provides a
firmer foundation for assessing private prisons.
IV. THE MEANING OF PUNISHMENT
To flesh something out is to add detail to an existing structure—to
put meat on the bones.97 Fleshing out the meaning of punishment, as I
undertake it here, involves identifying the basic framework of liberaldemocratic punishment in the Anglo-American tradition, then filling it
out with some conceptual detail and defining its normative contours. The
resulting account of punishment will be recognizably our own, though it
will not reflect the prevailing approach to punishment in the United
States, the United Kingdom, or anywhere else. My aim is thus to suggest
“an ideal conception of what punishment ought to be, in whose light we

95
MICHAEL TONRY, THINKING ABOUT CRIME: SENSE AND SENSIBILITY IN
AMERICAN PENAL CULTURE 15 (2004).
96
For an especially thoughtful articulation of an instrumentalist conception
designed to ensure humanity in punishment, see Dolovich, supra note 42.. Although Dolovich’s argument against prison privatization is ultimately contingent
on empirical claims about the relative performance of public and private actors,
she makes a compelling normative case for resisting privatization. See also
DONAHUE, supra note 13, at 156 (“If a private prison treats inmates humanely,
protects them from indignity and assault, endeavors to aid their rehabilitation, and
charges the community a fair price, would the fact that its shareholders anticipated a return on their investment make that prison inferior to one in which public employees neglect, humiliate, and abuse prisoners while needlessly straining
the public purse?”).
97
See CAMBRIDGE DICTIONARY OF AMERICAN IDIOMS (last visited July 5,
2009). The origin of the expression is “based on the idea of adding flesh to a
picture that shows only the bones of a creature.” Id.

2010

PRIVATE PRISONS

21

can evaluate (and no doubt find seriously wanting) our existing practices.”98
I begin by outlining the familiar principles of the liberal-democratic
tradition that structure the institutions of criminal punishment, then draw
on various retributive and communicative theories to sketch a conception
of punishment that reflects the most compelling features of that tradition.99
Finally, I consider the implications of the retributivecommunicative account for prison privatization, concluding that private
prisons are largely inconsistent with the meaning of punishment in a liberal-democratic polity—that the proliferation of private prisons moves us
further away from the highest ideals of the Anglo-American tradition.
A. The Liberal-Democratic Political Tradition
The commitment to liberal-democracy sets the parameters for legitimate punishment in the Anglo-American tradition. In its classic formulation, the liberal-democratic polity arises from a state of nature into
which individuals are born free and equal.100 Endowed with rationality
and a bundle of natural rights, individuals come to recognize the advantages of mutual cooperation and consent to form themselves into political
communities that secure their rights and coordinate their activities
through the mechanisms of self-government and the rule of law.
One need not—should not—accept the state of nature as a historical
phenomenon to appreciate the liberal-democratic values it showcases. In
particular, because individuals are free and equal rights-bearers, a status
inherent in their humanity, they can neither be legitimately deprived of
their rights without their consent nor compelled to sacrifice their own
interests for the good of others. The commitment to self-government
provides individuals a say in establishing and enforcing the laws that
bind them, while the rule of law constrains arbitrary and unreasonable
manifestations of collective power. Contemporary conceptions of liberalism introduce autonomy and pluralism that provide individuals the authority and resources for determining the course of their lives according
to their own conceptions of meaning and value. Finally, part and parcel
of Anglo-American liberalism is a set of commitments—citizenship,
community, and civic responsibility—traditionally denominated republican.101
98
DUFF, supra note 79, at xv (describing the interplay between theory and
practice in developing a normative account of criminal punishment).
99
See especially DUFF, supra note 79; NOZICK, supra note 77.
100
See, e.g., THOMAS HOBBES, LEVIATHAN 117 (Richard Tuck ed., Cambridge Univ. Press 1991) (1651); JOHN LOCKE, TWO TREATISES OF GOVERNMENT
323-25 (Peter Laslett ed., Cambridge Univ. Press 1988) (1690).
101
This way of putting things is actually somewhat controversial. Historians
of political thought are divided on the precise sources of American constitutional
values. On one view, republican values predominated among the founders and

2010

PRIVATE PRISONS

22

This distinctive blend of liberal and republican values yields a social
and political environment that reflects neither extreme individualism nor
radical communitarianism, but a more or less stable balance between
individual and community interests that is constantly being negotiated
and renewed. At our worst, the obsession with individual rights vitiates
any sense of common purpose; at our best, political participation is deliberative, reshaping individual preferences in light of community norms
and values. Indeed, “[c]riminal law is … one area in which Americans
have conceded to the state an almost unqualified right to act in the name
of the polity, and hence one of the few places in which one can discern
an American conception of political community that is not a mere collage
of individual preferences.”102
Nothing in this brief account of liberal-democratic values obviously
disqualifies any of the traditional justifications for punishment. Utilitarian purposes—incapacitation, deterrence, and rehabilitation—honor the
basic liberal commitment to public order by securing the rights of individuals against criminal transgression. Retributivism respects the human
capacity for choice that the commitment to individual rights presupposes.
Yet the familiar weaknesses of these approaches quickly surface. Because utilitarianism conceives of the public good in the aggregate, it fails
to take seriously the distinction between persons and is formally indifferent regarding the allocation of benefits and burdens.103 Absent sideconstraints, it countenances the deliberate infliction of punishment on the
innocent104 and accommodates modes and methods of treatment that fail
to accord with our basic sense of justice and proportionality. Moreover,
because utilitarianism operates primarily through fear and manipulation
provide the raw material for a hoped-for “republican revival.” See, e.g., JGA
POCOCK, THE MACHIAVELLIAN MOMENT (1975); BERNARD BAILYN, THE
IDEOLOGICAL ORIGINS OF THE AMERICAN REVOLUTION (1967); GORDON S. WOOD,
THE CONFEDERATION AND THE CONSTITUTION (1979). Richard Sinopoli, among
others, argues, however, that the values identified as republican are in fact classically liberal and traceable to Locke himself. RICHARD C. SINOPOLI, THE
FOUNDATIONS OF AMERICAN CITIZENSHIP (1992). On this view, Lockean liberalism is less individualistic than popularly supposed, and American republicanism
is more rhetorical than actual. For my purposes, the exact source of the values is
less significant than their central place in the Anglo-American political tradition.
102
John J. DiIulio, Jr., What’s Wrong with Private Prisons 92 PUB. INT. 66,
81 (1988).
103
See JOHN RAWLS, A THEORY OF JUSTICE 26 (1971) (“[T]here is no reason
in principle why the greater gains of some should not compensate for the lesser
losses of others; or more importantly, why the violation of liberty of a few might
not be made right by the greater good shared by many.”); see also Jeremy Bentham, Panopticon, in THE PANOPTICON WRITINGS 48-50 (Miran Bozovic ed.
1995) (1791) (advocating that contracts “would make him [the private provider]
pay so much for every one that died, without troubling whether any care of his
could have kept the man alive”).
104
And the converse as well—forgoing punishment of the guilty in cases
where utilitarian values are not implicated.

2010

PRIVATE PRISONS

23

rather than appeals to shared values, it fails to address individuals as citizens or as members of a normative community.105
For its part, retributivism, without more, seems less like a justification for punishment than an article of faith. Despite the powerful intuitions that underwrite it, its historical and conceptual affinity with revenge
should give us pause. The concept of desert at the heart of retributivism
is similarly intuitive but also deeply mysterious, while the commitment
to proportionality cannot provide or even suggest a scale of deserved
punishment.106
Although neither the utilitarian nor retributive justifications provide
a complete, or completely satisfying, account of criminal punishment,
only the retributive approach is ultimately consistent with the liberaldemocratic values of the Anglo-American tradition. For retributive punishment is premised on the liberal individual with the distinctive set of
attributes and capacities that determine our moral status. As free and
rational agents, we are held accountable for our choices, including acts of
criminal wrongdoing; to refrain from punishing for such acts would be to
fail to treat wrongdoers as responsible moral agents.107 Moreover, because our rights are inviolable and cannot be subordinated to the interests
of others, the deliberate punishment of the innocent is ruled out of
bounds regardless of whatever social benefit it might produce. Finally,
the social condemnation that inheres in retributive punishment presupposes a community of value as well as a responsible moral agent. In the
absence of either, punishment lacks moral authority and retributive
meaning.
B. Retributive Communication
In the modern liberal-democracy, acts of criminal wrongdoing are
not only offenses against particular victims, but offenses against the
community as well.108 While the most serious forms of law violation—
105
JEREMY BENTHAM, AN INTRODUCTION TO THE PRINCIPLES OF MORALS AND
LEGISLATION 28 (1823) (characterizing criminal punishment as a form of “terror”).
106
As Michael Moore notes, a commitment to retributivism does not entail a
commitment to any particular metric for gauging desert. Moore, supra note 78,
at 94-95. For despite Kant’s commitment to lex talionis, see KANT, supra note
76, at *332, retributivism entails a commitment only to proportionality in punishment.
107
Samuel H. Pillsbury, The Meaning of Deserved Punishment: An Essay
on Choice, Character, and Responsibility, 67 IND. L.J. 719, 741 (“Moral responsibility signifies our belief that human worth comes from our choices.”).
108
This is among the developments associated with the shift from a state of
nature, in which individuals possess the executive authority to punish wrongdoing, to civil society, where the executive power transfers to the state. See LOCKE,
supra note 105, at 325 “Where-ever therefore any number of Men are so united
into one Society, as to quit every one of his Executive Power of the Law of Na-

2010

PRIVATE PRISONS

24

assault, robbery, and murder, for example—most dramatically affect direct and identifiable victims, law violation in all its forms constitutes a
transgression against the political community as a whole, a subversion of
its norms and values. Because it is a liberal-democratic polity, moreover,
its laws will reflect the self-determination of its members. For “[t]he
voice of the law is (or aspires to be) the voice of the community addressing itself, the voice of all the citizens addressing one another and themselves.”109 Legal punishment represents the community’s formal response to criminal attacks, a “special social convention that signifies
moral condemnation.”110
On the communicative conception, criminal justice represents a kind
of “moral dialogue”111 between citizens and the state as the legal embodiment of the political community.112 “The distinctive meaning of
criminal wrongdoing is its denial of some important value, such as the
victim’s moral worth.”113 Against the backdrop of the community’s
norms and conventions, the social meaning of criminal conduct is objective, conveying disrespect for victims and contempt for community values regardless of the offender’s subjective motive or intent.114 Likewise,
criminal punishment draws its meaning from the values of the community and its conventional forms of condemnatory expression. These reflect “deeply rooted public understandings” of particular modes of punishment that signify the gravity of criminal misconduct.115

C. The Meaning of Prison
Punishment, then, is and effects a form of community censure that
takes its meaning from the community’s values and conventions. The
Anglo-American criminal law contemplates a wide array of punitive
ture, and to resign is to the publick, there and there only is a Political, or Civil
Society.”); see also BECCARIA, supra note 45, at 15.
109
DUFF, supra note 79, at 60; see also Michael Walzer, Hold the Justice,
NEW REP., Apr. 8, 1985, at 11 (“Criminals are fellow citizens; when we punish
them we presume upon the fellowship.”).
110
Dan M. Kahan, What Do Alternative Sanctions Mean?, 63 Chi. L. Rev.
591, 593 (1996).
111
Pillsbury, supra note 112, at 744.
112
The doctrines of excuse in the criminal law tend to reflect the presupposition of a responsible moral agent as the addressee of the criminal law. See, e.g.,
MOORE, supra note 75, at 404 (“The presupposition is that any being who is held
responsible must be sufficiently rational and autonomous to be a rational agent.”).
113
Kahan, supra note 116, at 597-98; see also Hampton, supra note 78, at
124; Pillsbury, supra note 112, at 721.
114
See Kahan, supra note 116, at 597-98.
115
Id. at 593.

2010

PRIVATE PRISONS

25

practices, ranging from fines to capital punishment, each with a more or
less distinctive social meaning. Because “certain forms of hard treatment
have become the conventional symbols of public reprobation,”116 it is not
enough to attend to the severity of punishment; we must also consider the
mode of punishment as well. This accounts for why punishing a brutal
rapist with a monetary fine would offend our sense of justice. The problem is not (only) that the punishment is too lenient, it is rather the wrong
kind of punishment;117 it is insufficiently expressive of public condemnation, trivializing the seriousness of the offense and denigrating the worth
of the victim.118
In the Anglo-American tradition, “[i]t is…imprisonment in a penitentiary, which now renders a crime infamous.”119 Because it entails the
extreme curtailment of individual liberty and physical exclusion from the
political community, it expresses condemnation in the clearest possible
terms. In the liberal-democratic context, the loss of freedom and community “is our society’s most potent symbol of moral condemnation.”120
It signifies “that the offender has, by his crime, made the maintenance of
normal community with him impossible.”121
At its best, a term of imprisonment represents an extreme form of
censure that “dramatically and unequivocally” expresses social condemnation for acts and agents of serious wrongdoing.122 In the conventional
parlance of the Anglo-American criminal law, it addresses offenders as
responsible moral agents whose wrongful choices provoke the community’s punitive response. However, what is heard “depends not just on
the content of what is said, but on the context in which it is said, and the
accent in which it is spoken.”123 Effective communication thus depends
on the identity of the speaker as well as the identity of the listener, lest
“some offenders hear … its voice, not as the voice of a community to
which they belong and are treated as belonging, but as the voice of an
alien and oppressive power.”124 It must be “us against us” rather than “us
against them.”125
But perhaps, one might argue, the moral dialogue ends at the moment of conviction and sentencing. At that point, the community has
articulated its values through the legislative process, affirmed its commitments through the mechanisms of enforcement and prosecution, and
116

FEINBERG, supra note 79, at 100.
DUFF, supra note 79, at 146.
118
Kahan, supra note 116, at 620-22.
119
FEINBERG, supra note 79, at 111 (quoting Brandeis).
120
Kahan, supra note 116, at 621.
121
DUFF, supra note 79, at 149.
122
Kahan, supra note 116, at 592.
123
DUFF, supra note 79, at 192.
124
Id. at 193; see also Pillsbury, supra note 112, at 752 (“We punish offenders not because they stand outside of society, not because they are alien enemies,
but because they are fundamentally like the rest of us.”).
125
Pillsbury, supra note 112, at 752.
117

2010

PRIVATE PRISONS

26

communicated censure to the offender directly through the trial process
by pronouncing guilt and imposing a fitting sentence. What, if anything,
remains to be said? What possible significance could attach to the identity of the jailer?126
The first problem with this way of putting things is that it misplaces
the burden of justification. The enterprise of criminal justice, according
to the retributive-communicative account, is not a series of discrete processes that can be neatly distinguished and parceled out for delivery.
Rather, criminal justice encompasses the full range of decisions and actions that define, enforce, and affirm the community’s standards of
criminal behavior through a process of ongoing dialogue. In light of this,
we should expect the state, as the legal embodiment of the political
community, to assume responsibility for all aspects of criminal justice—
to take our part in the dialogue. Carving out one or more of these activities for private delivery thus requires justification in terms of the relevant
legal, moral, and political values. So instead of asking opponents of privatization why the enterprise of community censure extends beyond the
moment of conviction and sentencing, we should ask proponents of privatization why they believe that is the critical moment when the dialogue
ends. What is it about punishment, imprisonment in particular, that distinguishes it from the other aspects of criminal justice? Why is the identity of the jailer insignificant?
One way to make the case that it does not matter who owns and operates a prison—so long as inmates are treated fairly and humanely—
would be to draw a sharp distinction between the responsibilities of
prison personnel and those of legislators, prosecutors, and judges. On
this view, prison employees, whether public or private, are charged with
implementing the decisions of various public officials—housing inmates
for more or less determinate periods of time while maintaining a generally humane environment calculated to protect inmates and respect their
rights. As such, punishment is akin to a ministerial function, involving
the execution of policies and decisions made elsewhere by others. Although legislators, prosecutors, and judges (or juries) exercise considerable discretion in reaching their judgments, prison personnel, on this
conception, do not. Thus, a prison employee acts “in a prescribed manner, in obedience to the mandate of legal authority, without regard to, or
the exercise of, his or own judgment upon the propriety of acts being
done.”127
This argument is unavailing because it is based on a misconception
of prison operations. Prison personnel, ranging from top administrators
to line officers, in fact exercise considerable discretion in virtually every
126

For purposes of this inquiry, I assume sufficient oversight to ensure fair
and humane treatment. That is, I want to rule out as an answer to this question
the set of practical concerns about monitoring the conditions of confinement and
ensuring due process in private prisons.
127
636 AM. JUR. 2d Public Officers and Employees § 234 (2009).

2010

PRIVATE PRISONS

27

aspect of their work. Although legislative and judicial mandates set the
parameters of fair and humane treatment, it is not possible to anticipate
every situation that is likely to arise or to specify appropriate responses in
advance.128 As a result, prison personnel are necessarily afforded substantial discretion to establish and implement prison policy and to address the day-to-day contingencies that they encounter in the prison environment. For their part, administrators must develop policies regarding
the provision of medical care, standards for administrative classification,
and the procedures for inmate discipline.129 Corrections officers who
interact directly with inmates must not only implement these policies in a
variety of settings, they must also make on-the-spot judgments about
inmates and their behavior—determining whether they require medical
attention, represent a danger to themselves or others, or merit disciplinary
action, administrative segregation, or even the use of force.130 Moreover,
these decisions are not confined to the margins of the prison experience;
they arise on a daily basis and will dramatically affect the length and
character of a criminal sentence.
Since prison personnel exercise considerable discretion, their role is
not relevantly distinguishable from other actors in the criminal justice
process whose decisions we recognize as our own.131 By privatizing
punishment, however, we terminate the dialogue between offenders and
their community in just the same way as if we privatized prosecutors and
criminal courts. “Although some historical traditions permitted prosecutions initiated by private parties, contemporary U.S. practice consolidates
prosecutorial power in the government, with the symbolic message that
the government stands in for the community and private victims.”132
Indeed, even proponents of prison privatization balk at the idea of privatizing criminal courts.133 Our reasons for rejecting privatization of these
aspects of criminal justice should lead us to resist prison privatization as
well.
Moreover, despite the conventional meaning of prison in the AngloAmerican tradition, the message of punishment it constitutes can easily
be scrambled. Prison privatization interposes a filter between the community and the offenders whom it calls to account. In particular, by
transforming the institutions of punishment into commodities—fungible
objects of economic exchange—privatization alters the character of punishment, reducing the punitive enterprise to a question of price point and
logistics. It becomes a puzzle to be solved rather than a dialogue to be
128

Dolovich, supra note 42, at 478-79.
Id.
130
Id.
131
Thomas W. Pogge, How Should Human Rights Be Conceived?, in THE
PHILOSOPHY OF HUMAN RIGHTS 187, 194 (Patrick Hayden ed., 2001).
132
Minow, supra note 20, at 1234.
133
See, e.g., Clifford J. Rosky, Force, Inc.: The Privatization of Punishment, Policing, and Military Force, 36 CONN. L. REV. 879, 977 n.334 (2004).
129

2010

PRIVATE PRISONS

28

opened or renewed. For in the same way that the “law and the courts
speak and act in the name of the political community,”134 our conventions establish that our prisons do so as well.135 “That message ought to
be conveyed by the offended community of law-abiding citizens, through
its public agents, to the incarcerated individual.”136 As we distance ourselves from the condemnatory practice, however, we attenuate its message of censure, alienating offenders and ourselves from the meaning and
value that constitute the liberal-democratic community.137
Skeptics of the “social meaning” argument against prison privatization observe that the cultural context that confers meaning is by no means
fixed. Indeed, perhaps “there are already some legislators, judges, administrators, and entrepreneurs”—we might add citizens and criminal
offenders—“who actually and honestly do not believe that ‘private’ imprisonment is significantly different from ‘public’ imprisonment in cultural terms.”138 To the extent that this is the case, it suggests how far we
have strayed from the normative path of liberal-democratic meaning. In
fact, we can recall or envision changes in meaning regarding a number of
culturally significant phenomena, such as marriage, parenthood, and
rape. But presumably it is not a matter of indifference to us what course
these changes take—whether rape is or is not regarded as a serious violation of the self, whether marriage and family are limited to heterosexual
couples or extended to homosexuals, polygamists, or other nontraditional arrangements. In each instance, the challenge is to make a
case for meaning in terms of our liberal-democratic values and to promote or resist cultural change on that basis.
In the case of criminal punishment, the contemporary focus on incapacitation, combined with an “us v. them” mentality toward criminal
offenders, represents an impoverished conception of the liberaldemocratic community and charts a course in the wrong direction. It
fails to take seriously both the capacity of persons to make and remake
themselves and the number and variety of obstacles, affecting some more
than others, in the way of making socially responsible choices. By contrast, the communicative conception of punishment is predicated on precisely those features of the human condition—on our potential and our
limitations—that ground our liberal-democratic commitments. There is
thus nothing “mysterious” about the idea that it matters who inflicts punishment.139 For punishment engages fellow citizens in one of the most
serious and definitive enterprises of a liberal-democratic community—
134

DUFF, supra note 79, at 186.
See FEINBERG, supra note 79, at 111; Kahan, supra note 116, at 591.
136
DiIulio,, supra note 107, at 79.
137
This conceptual claim is distinct from, but related to, the empirical claim
that prison privatization causes us to be less concerned about the fate of inmates.
See, e.g., Walzer, supra note 114, at 12.
138
Rosky, supra note 137, at 968.
139
Id.
135

2010

PRIVATE PRISONS

29

holding ourselves and one another responsible for our actions—and the
voice of the community is clearest when it speaks for itself.
V. CONCLUSION
The case against private prisons may be overdetermined. In the prevailing liberal-democratic context, their mixed record of performance,
along with the seemingly intractable practical problems they present,
casts serious doubt on the value of private prisons. But the budgetary
realities that hit home during the present economic crisis heighten the
appeal of prison privatization. The burgeoning prison population in the
United States, resulting from a variety of dubious criminal justice policies, does not seem likely to decrease any time soon. In this environment, private prisons may act as a kind of escape valve that relieves the
pressure we might otherwise feel to critically examine our policies and
practices.
According to at least one commentator, however, advancing the
claim that the management and operation of prisons is an inherently public function means implicitly accepting public prisons in their current
form as the baseline for evaluating the private alternative.140 In this way,
one risks being “coldhearted and blind” about the fate of actual inmates
consigned to the deplorable conditions that prevail in our public prisons
and jails.141 This sort of complacency would indeed suggest a kind of
moral obtuseness that we are right to be on our guard against. My own
hope is that by focusing on the meaning of punishment in the AngloAmerican tradition, it may be possible to put our practices into fresh perspective, forcing us to confront the chain of events—the criminal justice
policies, the millions imprisoned, the overcrowded and indecent conditions—that led us astray. Punishment, especially imprisonment, is a serious matter, and we almost certainly punish too much—not in the utilitarian sense, but in terms of what our values are and what wrongdoers
deserve. They deserve to be taken seriously as moral agents in the way
that retributive punishment entails, and they deserve the full force of our
censure when their choices flout the values that constitute the liberaldemocratic community of which we take them to be a part.

140
141

Dolovich, supra note 42, at 443.
Id.