Electronic copy available at: http://ssrn.com/abstract=2070906
LEGAL RESEARCH PAPER SERIES
Paper No 23/2012 May 2012
Using Gender Equality Analysis to Improve the Wellbeing
of Prostitutes
BARBARA HAVELKOVÁ
(2011) 18 Cardozo Journal of Law and Gender 53-86
The full text of this paper can be downloaded without charge from the
Social Science Research Network electronic library at:
<http://ssrn.com/abstract=2070906>
An index to the working papers in the
University of Oxford Legal Research Paper Series is located at:
<http://www.ssrn.com/link/oxford-legal-studies.html>
Electronic copy available at: http://ssrn.com/abstract=2070906
101
USING GENDER EQUALITY ANALYSIS TO
IMPROVE THE WELLBEING OF PROSTITUTES
BARBARA HAVELKOVÁ
INTRODUCTION
The idea for this Article comes from my research on the legal treatment of
prostitution in the Czech Republic since the fall of communism.1 Right after the
Velvet Revolution of 1989, prostitution was decriminalized2 in Czechoslovakia.
The subsequent “boom” in prostitution resulted in the adoption of various public 
order provisions by individual towns, the violation of which was punishable by
fines for prostitutes only. The main proposals for change in the past twenty years
in the Czech Republic have been for the regulation of prostitution, which would
introduce obligatory registration and health checks for prostitutes, move
prostitution indoors3 into registered establishments, and, in effect, outlaw
“outdoor”4 prostitution. Neither the current legal response nor the proposals are
concerned with the wellbeing of prostitutes as a legislative goal; instead, public
order and morality, health, images of the town and fiscal considerations have been
primary concerns. Nor is prostitution addressed as a gender equality problem. The
fact that no feminist conceptualizationneither the sex-work nor sexualdomination
understanding of the phenomenonis present in the Czech policy
debates can be seen as a specific post-communist phenomenon.5 However, it is
contended that the lack of a gender equality analysis is a problem facing other legal
systems, especially those which have not yet adopted an explicitly genderconscious
policy.
 CSET Teaching Fellow in EU Law, Faculty of Law, University of Oxford. I would like to thank the
following for reading and discussing earlier drafts: Blanka  Hančilová,  Sandra  Fredman,  Hana 
Havelková, Tarunabh Khaitan, Laura Hilly, Farrah Ahmed, Jarrod Hepburn and the participants of the
Oxford Jurisprudence Discussion Group on November, 24 2010. Substantial portions of Parts II and III
draw on my thesis, Barbara Havelková, European Gender Equality Under and After State Socialism:
Legal Treatment of Prostitution in the Czech Republic (2010) (unpublished MSt. Thesis, Faculty of
Law, University of Oxford), http://ora.ouls.ox.ac.uk/objects/uuid%3Aad0b1fa1-28ca-4400-908b-
4b34e08ca064.
1 Id.
2 Prostitution was not explicitly criminalized under state socialism (1948-1989) in Czechoslovakia
but was in some circumstances prosecuted under the criminal offence of parasitism. Id. at 59-75.
3 Indoor prostitution refers to prostitution done in sex clubsoften dissimulating  as  “massage 
parlors,” night clubs, and cabaretshotels, and private flats or by means of escort services.
4 Outdoor prostitution refers to street, road and highway prostitution.
5 I argue that due to its intellectual isolation and lack of freedoms of speech and association, the
Czech Republic missed what in the West was the second wave of feminism, including the debates on
sexuality, gender and gender-based violence.
Electronic copy available at: http://ssrn.com/abstract=2070906
This Article proposes a framework for a gender equality analysis of
prostitution with the aim of offering national regulators and judges a tool for
improving the wellbeing of prostitutes. It compares the treatment of clients and
prostitutes, as they are the relevant parties to the transaction of exchange of sex for
money. While the Article acknowledges the existence of male, transsexual and
transgender prostitutes, its analysis is based on the observation that the distribution
of men and women in the positions of buyers and sellers makes prostitution an
extremely sex-segregated field where the demand is overwhelmingly created by
men and the supply by women.
Part I presents a typology of regulatory regimes, based mainly on secondary
literature about European Union member states and common law jurisdictions.
Then, Part II offers a synthesis of the two feminist positions on prostitution. It
observes that the two positions, sex-work and sexual-domination, conceptualize
prostitution in vastly different manners and disagree strongly on the best policy
solution: decriminalization with legalization and abolition, respectively. Most
feminist writing adheres to one of these two seemingly irreconcilable positions. As
a result, a legal gender equality argument that transcends this divide has yet to be
fully developed. Part III provides a framework for a gender equality analysis based
on a comparison of the treatment of the client and the prostitute. The Article
discusses the steps of a gender equality test, extrapolated from constitutional
review in European and common law jurisdictions, which often ask similar
questions: (i) Is there a difference in treatment or impact (ii) between persons that
are comparable? (iii) Is the ground for the distinction suspicious? (iv) Is the
distinction fair? (v) Is it pursuing a legitimate aim? (vi) Is the measure
proportionate to that aim?
There are three possible scenarios in terms of comparative treatment of the
prostitute and the client. First, there can be asymmetric treatment benefiting the
client, a still-existing practicewhether in statutory regulation or in
enforcementthat violates the principle of gender equality. Second, the treatment
can be symmetric, subjecting both parties to equal treatment. The principle of
equality requires at least this standard. Third, an asymmetric treatment for the
benefit of the prostitute can be adopted. This Article addresses these three types of
regimes. In particular, it challenges existing justifications for asymmetric treatment
benefiting the client, the arguments that the prostitute is a repeat offender and that
she is the merchant or dealerboth which were prominently discussed in the
decision of the South African Constitutional Court in State v. Jordan,6 and the
argument of enforcement practicality. It further points out that even when the aim
is legitimate, such as public health or public order, an asymmetric practice
benefiting the client will often fail the test of proportionality, as it often is not
suitable, necessary or properly tailored. The Article then suggests that areas in
which the prostitute is particularly disadvantaged and vulnerable are aspects of
prostitution which support symmetric measures and even asymmetric treatment
benefiting the prostitute: (i) the negative social meaning of prostitution; (ii) the risk
of harm to the prostitute; and (iii) the existing inequality between the prostitute and
the client in terms of class, age, race, gender, nationality, immigration status and
socio-economic status.
6 See State v. Jordan 2002 (6) SA 642 (CC) (S. Afr.). For a discussion of the case, see note 12.
I. A TYPOLOGY OF REGULATORY REGIMES
The term prostitution is in everyday parlance understood as commercial sex7
or “the exchange of sex or sexual services for money or other material benefits.”8
Rarely is prostitution defined in law,9 and states respond to it in different ways.
The literature denominates the various approaches as prohibition, criminalization,
abolition, decriminalization, regulation, legalization, and deregulation,10 but the
actual understanding of these terms varies. Upon closer examination, these terms
are not helpful, especially since they do not use a single reference framework; their
categorizing criteria vary. In order to classify regulatory regimes, or to create a
map or spectrum, it is imperative to consider at least the three interconnected
issues: (i) the regime’s overall attitude to prostitution: whether prostitution is seen 
as a negative phenomenon to be eliminated, whether the state is resigned to it and
controls or contains it, or whether it accepts prostitution and facilitates it; (ii) the
treatment of the actors involved in prostitution: the prostitute, the client, and the
procurer; and (iii) what legislative goals the regime primarily pursues and whether
it is concerned with the wellbeing of the prostitute.
A. Overall Attitudes towards Prostitution
On the one side of the spectrum are regimes that consider prostitution a
negative phenomenon and aim at its elimination. One way of suppressing
prostitution is to prohibit it in its entirety by criminalizing the procuring, buying
and selling of sex and related activities. Such a regime, often referred to as
prohibitionist or criminalization, exists in the United States.11 The second
approach, which criminalizes the procurement of brothel keeping and the selling of
7 See Laurie Shrage, Should Feminists Oppose Prostitution, 99 ETHICS 348 (1989).
8 JOYCE OUTSHOORN, Introduction: Prostitution, Women and Politics in THE POLITICS OF
PROSTITUTION: WOMEN’S MOVEMENTS, DEMOCRATIC STATES, AND THE GLOBALISATION OF SEX
COMMERCE 3 (Joyce Outshoorn ed., 2004).
9 A comparative study of seventeen European countries revealed that only two of themAustria
and the UKhad a legal definition. See Blanka  Hančilová  &  Camille Massey, Legislation and the
Situation Concerning Trafficking in Human Beings for the Purpose of Sexual Exploitation in EU
Member States, INT’L. CTR. FOR MIGRATION POLICY DEV. 107 (2009), available at
http://ec.europa.eu/anti-trafficking/download.action?nodeId=519395a8-fe4e-417f-bf65-
fb65d004779e&fileName=ICMPD+Report+on+EU+Legislation+on+trafficking_en.pdf&fileType=pdf.
10 Common is the distinction between three regulatory regimes: abolitionism (referring to Sweden),
prohibitionism (often referring to the U.S.) and regulation (referring to the Netherlands). See
OUTSHOORN, supra note 8, at 6; Purchasing Sexual Services in Sweden and the Netherlands: Legal
Regulation and Experiences, NORWEGIAN MINISTRY OF JUSTICE AND THE POLICE 47 (2004), available
at http://www.regjeringen.no/upload/kilde/jd/rap/2004/0034/ddd/pdfv/232216-purchasing_sexual_
services_in_sweden_and_the_nederlands.pdf. The term legalization is often used for the Dutch model
as well. See Hinke Beukema, Legalisation of Prostitution in the Netherlands: A Trial to Decriminalise
Prostitution and to Improve the Economical, Physical and Emotional Situation of Prostitutes, 9 FEMINA
BOREALIS (2004). Prostitutes themselves often call for decriminalization or deregulation. See
International  Committee  for  Prostitutes’  Rights,  World  Charter  for  Prostitutes’  Rights  (adopted  in 
Amsterdam in 1985) SOCIAL TEXT (1993) [hereinafter ICPR]; KAMALA KEMPADOO, Introduction, in
GLOBAL SEX WORKERS: RIGHTS, RESISTANCE, AND REDEFINITION (Kamala Kempadoo & Jo Doezema
eds., 1998). Other authors have created new terminology, similar to mine. For example Hančilová & 
Massey, supra note 9, distinguish between four approaches: a) legalization, b) criminalization of
connected activities, c) criminalization of the client, and d) complete criminalization. See also MARJAN
WIJERS, Women, Labor and Migration: The Position of Trafficked Women and Strategies for Support, in
GLOBAL SEX WORKERS: RIGHTS, RESISTANCE, AND REDEFINITION (Kamala Kempadoo & Jo Doezema
eds., 1998).
11 This is true for the entirety of the United States with the exception of parts of Nevada. See
CATHARINE A. MACKINNON, SEX EQUALITY 1236 (Foundation Press; Thomson/West 2nd ed. 2007); see
also RONALD JOHN WEITZER, The Politics of Prostitution in America, in SEX FOR SALE: PROSTITUTION,
PORNOGRAPHY, AND THE SEX INDUSTRY (Ronald John Weitzer ed., 2000).
sex without criminalizing the buying, targets procurers and prostitutes but not
clients. South Africa used to follow this approach, where the system was famously
subject to constitutional review in the case of State v. Jordan.12 Both of these
repressive approaches have now been abandoned by an overwhelming majority of
countries in the European Union.13
A third approach to the elimination of prostitution does not target the
prostitute’s behavior but instead focuses on stifling demand by criminalizing the
clients as well as the procurers. Often referred to as abolitionist or neoabolitionist,14
it  is  exemplified  today  by  the  “Swedish  model.”15 In Sweden,
prostitution itself is neither legal nor illegal; however, the use of the prostitute is
punishable.  Since 1999, “A person who [. . .] obtains a casual sexual relation in 
return for payment, shall be sentenced for purchase of sexual services to a fine or
imprisonment for at most six months.”16 In addition, procurement“promot[ing] 
or  improperly  financially  exploit[ing]  a  person’s  engagement  in  casual  sexual 
relations”is punishable by up to four years in prison.17 This policy is
accompanied by programs for reintegration into society or a range of social
provisions for prostitutes, including shelter and training.18 The wellbeing of the
12 State v. Jordan 2002 (6) SA 642 (CC) (S. Afr.). The case concerned a constitutional challenge to
the provisions of the Sexual Offences Act 23 of 1957 that criminalized brothel-keeping and providing
sex for reward (the offense  of  “living  on  earnings  of  prostitution” was defined in Art. 20(1)(A) as
applying to “any person who… has unlawful carnal intercourse, or commits an act of indecency, with 
any other person for reward”). The challenge was on several grounds, in particular dignity, privacy and
equality. As far as equality is concerned, the majority dismissed the application, arguing that, because a
client was criminally liable as an accessory or a conspirator in the act, there was no discrimination on
the basis of sex. Id. at para. 11. The minority disagreed, arguing that the fact that the  “prostitute [is] 
the primary offender of the actual offence… carr[ies] a difference in social stigma and impact.”  Id. at
para. 63. They considered the provision to be unfair indirect discrimination on the basis of gender that is
not justified. For a more detailed discussion of the case, see Denise Meyerson, Does the Constitutional
Court of South Africa Take Rights Seriously? The Case of S v Jordan, ACTA JURIDICA (2004); Rosaan
Kruger, Sex Work from a Feminist Perspective: A Visit to the Jordan Case, 20 S. AFR. J. HUM. RTS. 138
(2004). The purchase of sex was explicitly criminalized in 2007. South Africa is currently debating an
overhaul of its legal treatment of prostitution. See SOUTH AFRICAN LAW REFORM COMMISSION.
DISCUSSION PAPER ON SEXUAL OFFENCES AND ADULT PROSTITUTION. (2009) available at
www.justice.gov.za/salrc/dpapers/dp0001-2009_prf107_2009pdf.
13 Romania was the last country to decriminalize prostitution in 2009. E-mail from Daniela Ortner
to author (May 31, 2011) (on file with author).
14 Abolitionism has its origins in the nineteenth century.    The  “old abolitionism,” mainly
characterized by opposition to regulation regimes which normalize prostitution, was an inspiration for
many national regulatory regimes in the first half of the twentieth century as well as to the international
instruments of that period, such as the 1949 Convention for the Suppression of the Traffic in Persons
and of the Exploitation of the Prostitution of Others, Dec. 2, 1949, 96 U.N.T.S. 271, 282.   Today’s 
abolitionism,  also  referred  to  as  “neo-abolitionism,”  keeps  the  anti-legalization position and adds an
emphasis on fighting demand.
15 BROTTSBALKEN [BrB] [CRIMINAL CODE] 4:1a (Swed.). Trafficking is punishable with up to life
imprisonment.
16 Originally adopted by the Law of Sweden on the Prohibition of the Purchase of Sexual Services
of 1999, it is now part of the Swedish Criminal Code. Id. 6:11.
17 Id. 6:12.
18 For more on the “Swedish model,” see MACKINNON, supra note 11, at 1235–1236; YVONNE
SVANSTRÖM, Criminalizing the John - a Swedish Gender Model?, in THE POLITICS OF PROSTITUTION:
WOMEN’S MOVEMENTS, DEMOCRATIC STATES, AND THE GLOBALISATION OF SEX COMMERCE (Joyce
Outshoorn ed., 2004); NORWEGIAN MINISTRY OF JUSTICE AND THE POLICE, supra note 10; JULIE
BINDEL & LIZ KELLY, CHILD AND WOMAN ABUSE STUDIES UNIT, LONDRON METRO. UNIV., A
CRITICAL EXAMINATION OF RESPONSES TO PROSTITUTION IN FOUR COUNTRIES: VICTORIA, AUSTRALIA;
IRELAND; THE NETHERLANDS; AND SWEDEN (2003), available at http://www.glasgow.gov.
uk/NR/rdonlyres/C19E010B-1A4F-4918-97BD-F96AF7D7F150/0/mainreport.pdf; Gunilla Ekberg, The
Swedish Law that Prohibits the Purchase of Sexual Services: Best Practices for Prevention of
Prostitution and Trafficking in Human Beings, 10 VIOLENCE AGAINST WOMEN 1187 (2004); Max
Waltman, Sweden’s  Prohibition  of  Purchase  of  Sex:  The  Law’s  Reasons,  Impact,  and  Potential, 34
prostitute is a primary concern. Estimates of numbers of prostitutes in Sweden are
considerably low with about 0.29 prostitutes per 1,000 inhabitants.19 The Swedish
model has now been adopted in Iceland and Norway.20
Most of the other regulatory regimes are based on the decriminalization of
prostitution; neither the prostitute nor the client is criminally liable. The regimes
range greatly in their further responses to prostitution; however, they may be
broadly categorized into three main types: control, containment and legalization.
The control approach, referred to here as “decriminalization with state control,” is 
resigned to the existence of prostitution but does not accept it as a normal, morally
neutral activity. It decriminalizes the buying, selling and procuring of prostitution,
but regulates and controls it. A highly controlling regulation approach was
common in the nineteenth century,21 when prostitutes were policed through
registers,22 obligatory regular health checks, strict zoning of street prostitution, and
regulation of brothels.23 The Czech 2005 proposal24 falls under this category. I
distinguish it from the “decriminalization with legalization” approach, the example 
of which is the “Dutch model,”25 which views prostitution as morally neutral, treats
it as work and is explicitly concerned with the wellbeing of prostitutes.
The  “decriminalization  with  containment”  regime,  which  is  prevalent  in 
Europe,26 differs  from  the  “decriminalization  with  control”  regime  in  that 
procurement usually remains illegal, and thus brothels are prohibited. The state
addresses only some aspects of prostitution, such as health or prostitution-related
activities, such as soliciting in public places or curb-crawling.27 Prostitution itself
operates in a grey zone and is sometimes not even mentioned in statutes.28 Its
policing can occur under general public order or public health provisions. Fewer
regulations surrounding prostitution indicates a more permissive regime. A regime
which did not address prostitution specifically at all could theoretically be
considered a “pure decriminalization” regime.29
WOMEN’S STUDIES INTERNATIONAL FORUM 449 (2011).
19 Report by the Czech Parliamentary Institute cited by Deputy Karas (KDU-ČSL) speaking on 29 
March 2005, available at http://www.psp.cz/sqw/tisky.sqw?F=H&PT=U&dx=1&o=4&na=prostituce&
T=&ZA=&f_N=on&f_I=on&f_U=on&f_S=on&f_Z=on&f_P=on&f_R=on&f_V=on&f_O=on&ra=20.
20 See Sheila Jeffreys, Prostitution, Trafficking and Feminism: An Update on the Debate, 32
WOMEN’S STUDIES INT’L FORUM 316, 319 (2009); SHEILA JEFFREYS, THE INDUSTRIAL VAGINA: THE
POLITICAL ECONOMY OF THE GLOBAL SEX TRADE 203-204 (Routledge 2009).
21 See ALEXANDRE JEAN B. PARENT-DUCHATELET, DE LA PROSTITUTION DANS LA VILLE DE PARIS
(1836).
22 Prostitutes work on the basis of a license and have to register with local authorities. Practicing
prostitution without a license makes the prostitute liable in administrative or criminal law.
23 This approach existed for example in the Austro-Hungarian Empire. Havelková, supra note 1, at
59. Today, it still exists in some Länder in Austria. See BIRGIT SAUER, Discourses on Prostitution in
Austria, in THE POLITICS OF PROSTITUTION: WOMEN’S MOVEMENTS, DEMOCRATIC STATES, AND THE
GLOBALISATION OF SEX COMMERCE (Joyce Outshoorn ed., 2004).
24 Government Of The Czech Republic, Governmental Proposal Of An Act On The Regulation Of
Prostitution (Print No. 1073) (2005), available at http://www.psp.cz/sqw/text/tiskt.sqw?O=4&CT=
1073&CT1=0.
25 MACKINNON, supra note 11, at 1235-1236.
26 Hančilová, supra note 9, at 104.
27 Curb-crawling, or in the UK terminology  “kerb-crawling”, is the client activity of soliciting
street prostitutes by driving along the curbside.
28 This is for example the case in the Czech Republic. Havelková, supra note 1, at 76 and
following.
29 Under a pure decriminalization regime, prostitution would not be criminal and would be subject
to the same regulatory requirements as other occupations and businessesfor example, in relation to
occupational health and safety and municipal planning. The New Zealand Prostitution Reform Act 2003
Legalization of prostitution is usually the aim of regimes that openly accept
prostitution as work or service. This approach has been adopted in the Netherlands
and is thus often referred to as the “Dutch model.”30 The state of the Netherlands
lifted a ban on brothels in 2000 in a move from “passive tolerance”31 to “active 
tolerance”  of  prostitution.32 The state entitled municipalities to license brothels
and regulate other aspects of the sex industry, such as setting time and place
restrictions and regulating advertisements.33 While maintaining criminal sanctions
for trafficking and emphasizing the prosecution of exploitation of involuntary
prostitution,34 the state aims to normalize the status of voluntary prostitutes. The
wellbeing of the prostitute is an important concern of this approach. Prostitutes can
either be employed by brothel-keepers or be self-employed, and they are liable for
tax and social security contributions.35 While health checks are recommended,
there is no obligation that they be undergone.36 A failure to legalize their activities
does  not  lead  to  criminal  liability;  however,  “the  sale and purchase of sexual
services on the street outside the [tolerance] zone is a criminal offence [. . .]
punishable by fines.”37 The numbers of prostitutes are high under such a regime; it
is estimated that about 3.13 out of 1,000 inhabitants in the Netherlands are
prostitutes.38
B. Regulating the Actors Involved in Prostitution
Of the actorsthe prostitute, the client, and the procurer or brothelkeeperthe
prostitute is often the primary target of criminalization, control,
regulation or containment. Many  “decriminalization  with  containment”  regimes 
only address obligations and restrictions through the prostitutes. This was the case
in the United Kingdom until 198539 and is still the case with many municipal
ordinances governing public order in the Czech Republic.40 A recent legislative
is often considered an example. NEW ZEALAND GOVERNMENT, REPORT OF THE PROSTITUTION LAW
REVIEW COMMITTEE ON THE OPERATION OF THE PROSTITUTION REFORM ACT 2003 (2008). Because it
legalizes prostitution and procurement and requires registration for both, it would however fall under a
legalization regime in my typology.
30 MACKINNON, supra note 11, at 1235-1236.
31 The ban on brothels has not, according to Dutch government, been enforced for the fifty years
preceding the 2000 lift of the ban, and  the  “prostitution  business”  was  not  interfered  with  unless  it 
caused    “inadmissible  nuisance”  or  breached  other  provisions  of  law.  NORWEGIAN MINISTRY OF
JUSTICE AND THE POLICE, supra note 10, at 26.
32 Id. at 26.
33 NETHERLANDS MINISTRY OF FOREIGN AFFAIRS, DUTCH POLICY ON PROSTITUTION. QUESTIONS
AND ANSWERS 2004 (2004), available at http://www.mfa.nl/contents/pages/743/prost.pdf.
34 Id. at 8. For more on the “Dutch model,” see Beukema, supra note 10; A.L. Daalder, Lifting the
Ban on Brothels: Prostitution in 2000-2001 (English summary) (2004), available at http://www.wodc.
nl/onderzoeksdatabase/01.035g-lifting-the-ban-on-brothels.aspx; NORWEGIAN MINISTRY OF JUSTICE
AND THE POLICE, supra note 10; REPORT OF THE SPECIAL RAPPORTEUR ON VIOLENCE AGAINST
WOMEN, ITS CAUSES AND CONSEQUENCES: MISSION TO THE NETHERLANDS (2007).
35 See NETHERLANDS MINISTRY, supra note 33, at 6.
36 See id. at 4.
37 Id.
38 See Report by the Czech Parliamentary Institute cited by Deputy Karas (KDU-ČSL) speaking on 
29 March 2005, available at http://www.psp.cz/sqw/tisky.sqw?F=H&PT=U&dx=1&o=4&na=
prostituce&T=&ZA=&f_N=on&f_I=on&f_U=on&f_S=on&f_Z=on&f_P=on&f_R=on&f_V=on&f_O=
on&ra=20.
39 In the UK, the 1959 Street Offences Act made it illegal for a “common prostitute” to loiter or 
solicit for prostitution. Only in 1985 did curb-crawling—soliciting by the client—receive the same
treatment. See, e.g., Nikki Adams, Prostitute women, justice and the law, 4 A CULTURAL REVIEW 295
(1993).
40 See Havelková, supra note 1, at 79; Government of the Czech Republic, supra note 24.
governmental proposal in the Czech Republic, falling under the “decriminalization 
with  state  control”  category,  would  have  introduced  restrictions,  obligations  and 
sanctions diametrically different for the prostitute and the client. The proposal
foresaw eight different offenses for the prostitute, punishable by up to
approximately $2,900,41 and only three for the client, punishable by up to $870.
While selling sex without a license was an administrative offense for the prostitute,
buying sex from a prostitute without a license was not. Repeated exercise of
prostitution without a license by the prostitute was to be a crime, punishable with
up to one year’s imprisonment or a fine.  A client who repeatedly bought sex from
an unlicensed prostitute, on the other hand, would not be punished, notwithstanding
the fact that the lack of a license might have meant that the prostitute had been
trafficked.
Many regulatory regimes ignore the client completely or pay much less
regulatory attention to him. Restrictive regimes that aim to prohibit prostitution in
its entirety sometimes do criminalize the client but other times choose not to
criminalize his conduct.42 The exception is the “Swedish model,” under which the 
client is the sole target of criminalization.43 A remark should be made regarding
the implementation of the rules governing both the selling and the purchasing of
sex in practice—large discrepancies exist between statutory provisions and their
enforcement. Even where the legal treatment of client and prostitute is formally
equal, an enforcement bias can exist against the prostitute. In repressive
prohibitionist regimes that criminalize both buying and selling of sex, such as the
United States, prostitutes are more often targeted than clients.44 Even in regimes
where prostitution has been decriminalized, such as France and Canada,45
prostitutes are nonetheless disproportionately targeted and harassed in policing.
C. Legislative Goals
Joyce Outshoorn, an editor of a multi-country study of policy debates about
prostitution, distinguishes between four goals of state intervention: maintaining law
and order, preserving morals, preventing the spread of sexually transmitted diseases
(“STDs”), and protecting  women from exploitation.46 This Article is concerned
with the wellbeing of prostitutes as a legislative goal and with the principle of
gender equality as a standard and corrective for any legislative goal. Both the
“Dutch model” and the “Swedish model” are influenced by feminist discussions of
41 The average wage in the Czech Republic in the last quarter of 2010 was 25803 CZK ($1,500
USD), according to the Czech Statistical Office. http://www.czso.cz.
42 This has been the case in South Africa. See the discussion of State v. Jordan, supra note 6,
supra note 12 & discussion infra Part III.A.
43 See supra note 18.
44 See DOROTHY MCBRIDE STETSON, The Invisible Issue: Prostitution and Trafficking of Women
and Girls in the United States, in THE POLITICS OF PROSTITUTION: WOMEN’S MOVEMENTS,
DEMOCRATIC STATES, AND THE GLOBALISATION OF SEX COMMERCE (Joyce Outshoorn ed., 2004).
45 See LESLIE ANN JEFFREY, Prostitution as Public Nuisance: Prostitution Policy in Canada in
THE POLITICS OF PROSTITUTION: WOMEN’S MOVEMENTS, DEMOCRATIC STATES, AND THE
GLOBALISATION OF SEX COMMERCE (Joyce Outshoorn ed., 2004); see also AMY G. MAZUR, Prostitute
Movements Face Elite Apathy and Gender-biased Universalism in France, in THE POLITICS OF
PROSTITUTION: WOMEN’S MOVEMENTS, DEMOCRATIC STATES, AND THE GLOBALISATION OF SEX
COMMERCE (Joyce Outshoorn ed., 2004).
46 Most regimes have a mix of legislative goals. OUTSHOORN, supra note 8. The image of town
and fiscal considerations came up as prominent aims in the Czech Republic. See Havelková, supra note
1, at 102.
the late twentieth century47 and count the “wellbeing of the prostitute” prominently 
among their legislative goals. Their understanding of what is good for the
prostitute is, however, dramatically different, and it mirrors the two different
feminist positions. Thus, the Swedish government states that it criminalized the
purchase of sex because “[t]he ability of men to purchase sexual access to women 
in order to gratify their own sexual needs runs contrary to the conviction of
universal human equality and to the pursuit of full equality between women and
men.”48 The  Dutch  government’s  goals  in  legalizing  prostitution  included  the 
“amelioration  of  the  prostitutes’  position:  greater  personal  safety  and  more  legal 
protection of their rights. Furthermore it was hoped that the taboos and
stigmatization surrounding prostitution would disappear and that prostitution would
become socially acceptable.”49
Even today, however, some regimes or legislative proposals decline to
consider the wellbeing of the prostitute. For example, the 2005 Czech Proposal,
presented as liberal and permissive and inspired by the Dutch model,50 was actually
highly restrictive of prostitutes and was unconcerned with their wellbeing.51 This
Article considers how the wellbeing of prostitutes can be improved using the
principle of gender equality, especially in countries that have not yet chosen a
legislative framework explicitly based on concern for the prostitute. To enable the
discussion to go beyond the sex-work versus sexual-domination binary, it is
important to first present the landscape of the feminist debate.
II. FEMINIST APPROACHES
The issue of prostitution has been difficult for feminists. The two things
feminists agree on are that the act of prostitution itself should never be
criminalized52 and that the exploitation of women by using coercion, deceit, abuse
or violence to bring or keep them in prostitution always should be deemed illegal.
However, feminists are diametrically opposed about whether a person can choose
prostitution freely as a profession. Two distinct positions can be identified.53 The
first position conceptualizes prostitution as sex work and speaks about sex workers,
clients and procurers; the second sees it as sexual domination and the essence of
women’s  oppression  and  speaks  about  prostituted persons or persons in
prostitution, johns and pimps. These positions further disagree on the following: (i)
whether prostituting oneself is a free choice and an expression of agency or
whether people are coerced into prostitution; (ii) whether what is being sold is the
service or the self; (iii) whether the harm of prostitution is merely the repressive
laws or inherently prostitution itself; (iv) whether economic or sexual inequality is
47 Historically, in the nineteenth and first half of the twentieth century, many regimes had a strong
emphasis on public order, morality and health. The well-being of the prostitute was either a non-issue or
a marginal concern.
48 MACKINNON, supra note 11, at 1315.
49 Beukema, supra note 10, at 152.
50 A governmental proposal of an Act on the Regulation of Prostitution was presented to the
Parliament in 2005. See Government Of The Czech Republic, supra note 24.
51 Havelková, supra note 1, at 101.
52 See GRAHAM SCAMBLER & ANNETTE SCAMBLER, RETHINKING PROSTITUTION: PURCHASING SEX
IN THE 1990S 186 (Routledge 1997); Catharine A. MacKinnon, Prostitution and Civil Rights, 1 MICH. J.
OF GENDER & L. 13 (1993); see also ICPR, supra note 10.
53 See, e.g., Joyce Outshoorn, The Political Debates on Prostitution and Trafficking of Women, 12
SOCIAL POLITICS 1 (2005).
at the root of prostitution; (v) whether or not a distinction should be drawn between
voluntary and forced prostitution; (vi) whether policy should be informed by sexworkers
speaking for themselves or by experiences of the most marginalized
prostitutes extracted by qualitative research; and (vii) whether the appropriate legal
response to prostitution is the  “decriminalization with legalization” regime or the 
criminalization of the client under the “abolition” regime.
A. The Sex-Work Position
The sex-work position54 emphasizes the autonomy, agency, choice and selfdetermination
of prostitutes. Most of its proponents would argue that, like any
other worker, the prostitute sells alienable labor power.55 Than-Dam Troung56
conceptualizes prostitution as “sexual labor,” which should be considered “similar 
to other forms of labor that humankind performs to sustain itself.”57 According to
most sex-work proponents, the inherent problem is not in the nature of sex work
but rather with the conditions that such work exists in today. It is the laws
criminalizing sex workers and repressing their migration that need changing, not
prostitution itself. Many organizations of prostitutes58 are among the proponents of
the sex-work position. It also finds support among many feminists and has been
identified as the currently predominant position in academic writing.59 Moreover,
it resonates with a liberal emphasis on personal choice and agency as well as with
socialist feminism’s analysis of sex work as work that places prostitutes within the 
context of the international labor movement.60
The sex-work position appears to be more diverse than the sexual-domination
position, and is comprised of four principled sub-positions and one pragmatic subposition.
The first principled sex-work sub-position addresses sex and sexuality.
Its proponents, such as Gayle Rubin and Pat Califia61 who are often referred to as
“sex radicals,”62 stress positive aspects of prostitution. They celebrate consensual
sexual practices that can be read as subverting binaries of normal versus abnormal,
healthy versus unhealthy, and pleasurable versus dangerous sex. Subverting these
54 See, e.g., FRÉDÉRIQUE DELACOSTE & PRISCILLA ALEXANDER, SEX WORK: WRITINGS BY WOMEN
IN THE SEX INDUSTRY (1987); WENDY CHAPKIS, LIVE SEX ACTS: WOMEN PERFORMING EROTIC LABOR
76 (1997); SHANNON BELL, READING, WRITING, AND REWRITING THE PROSTITUTE BODY (1994); THAN
DAM TROUNG, SEX, MONEY AND MORALITY: THE POLITICAL ECONOMY OF PROSTITUTION AND
TOURISM IN SOUTH EAST ASIA (1990); KAMALA KEMPADOO & JO DOEZEMA, GLOBAL SEX WORKERS:
RIGHTS, RESISTANCE, AND REDEFINITION (1998).
55 The use of the, originally Marxist, term “alienable  labor  power”  equates  prostitution  to  other 
types of work, where “contracts transfer powers of command from seller to buyer (the extent of those
powers and the terms of the transfer being the subject of the contract), and so require the seller to
temporarily surrender or suspend aspects of her will.” Julia  O’Connell  Davidson,  The Rights and
Wrongs of Prostitution, 17 HYPATIA 84, 86 (2002).
56 TROUNG, supra note 54.
57 KEMPADOO & DOEZEMA, supra note 54, at 4 (summarizing Truong’s arguments).
58 See VALERIE JENESS, MAKING IT WORK: THE PROSTITUTES’ RIGHTS MOVEMENT IN
PERSPECTIVE (1993).
59 See Jeffreys, supra note 20.
60 See D. KELLY WEISBERG, APPLICATIONS OF FEMINIST LEGAL THEORY TO WOMEN’S LIVES: SEX,
VIOLENCE, WORK, AND REPRODUCTION (1996).
61 See, e.g., PAT CALIFIA-RICE, MACHO SLUTS: EROTIC FICTION (Alyson Publications 1988); PAT
CALIFIA, PUBLIC SEX: THE CULTURE OF RADICAL SEX (Cleis Press 1st ed. 1994); GAYLE RUBIN,
Thinking sex: Notes for a radical theory of the politics of sexuality, in CULTURE, SOCIETY AND
SEXUALITY: A READER (Richard Parker & Peter Aggleton eds., 1999).
62 Davidson, supra note 55, at 95.
binaries  transforms  prostitution  into  a  legitimate  feature  of  “erotic  diversity.”63
Any repressive approach is blamed for “society’s negative attitudes to all women’s 
sexuality,”64 and prostitution has  a  valuable  social  function  of  “facilitat[ing]  the 
gratification  of  erotic  needs  that  would  otherwise  go  unmet”.65 However, other
sex-work  proponents  have  criticized  these  “sex  radicals.”    O’Connell  Davidson 
challenges the notion of a “transcendental human need for prostitution” and stresses 
the need to recognize the social construction of desire. Secondly, she criticizes the
fiction of a sovereign sexual subject. Under patriarchy, an individual cannot
liberate him or herself “from [a] fixed relationship to the sexual community” by 
“exchanging money for commodified sex,”66 as sex radicals believe.
A second principled sub-position concerns the nature of work. To Chapkis,
sex  work  is  normal  “emotional  labor”  like  child  care,  massage  work, 
psychotherapy, acting or other service work.67 Similarly, Troung compares sex
work to any other mental or manual labor, “all of which involve specific parts of 
the body and particular types of energy and skill.”68 On the basis of its normalcy,
prostitution should be treated exactly like any other work. This is a relatively
common, but not universally held, sub-position among sex-work proponents.
A third sub-position stresses agency.69 It argues that respecting choice is the
basis  of  women’s  empowerment.    As  many  sex-worker organizations claim that
prostitution is their work of choice, these sex-work proponents insist that the law
should enable this choice. Moreover, recognition of agency is seen as a
precondition for any bottom-up attack on patriarchal structures. This understanding
is also quite common in sex-work writing, even though many socialist feminists70
argue that prostitution is caused by women’s poverty and recognize that, as a result, 
choice can be limited. A fourth sub-position is the socialist feminist one. Like
radical feminists, socialist feminists recognize external conditions to be the root of
prostitution. Where radical feminists identify gender as the organizing axis of
disadvantage,71 socialist feminists emphasize class and social conditions.
Whether they believe that prostitution is sexually liberating or not, normal
work or not, the result of free choice or of external conditions, sex-work proponents
are unified by a pragmatic argument: prostitution is a phenomenon that cannot be
eradicated. Legalizing it would enable guarantees of equal rights for sex workers.
The sex-work position develops the distinction between forced or involuntary
prostitution, known as trafficking, on one hand, and voluntary or consensual
prostitution, referred to as “sex work”, on the other. Because trafficked women are
coerced, involuntary prostitution must remain criminalized; however, voluntary
prostitution,  which  is  an  expression  of  a  woman’s  agency,  should be
63 Id. at 88-98.
64 HELEN VICQUA, Scarlet Alliance internal communication, May 19th
1995, in ALISON MURRAY,
Debt-Bondage and Trafficking. Don’t Believe the Hype, in KAMALA KEMPADOO & JO DOEZEMA supra
note 54, at 61.
65 Davidson, supra note 55, at 89.
66 Id.
67 See CHAPKIS, supra note 54, at 76.
68 KEMPADOO & DOEZEMA, supra note 54, at 4.
69 See, e.g., Davidson, supra note 55.
70 See, e.g., Stephanie A. Limoncelli, The  Trouble  with  Trafficking:  Conceptualizing  Women’s 
Sexual Labor and Economic Human Rights, 32 Women’s Studies International Forum 261 (2009).
71 In equality law, this is the protected ground or characteristic.
decriminalized. In terms of regulatory regimes, the sex-work position mostly calls
for decriminalization, often accompanied by legalization of voluntary prostitution.
Under it, sex work should be treated as any other work or service and sex migration
as any other labor migration. The proponents argue for equal human and labor
rights, protection for all sex workers and working visas for migrant sex workers.72
It is thus also sometimes referred to as the “pro-rights” approach,73 and the “Dutch 
model” is often seen as an example of good practice.
B. The Sexual-Domination Position
The sexual-domination approach argues that prostitution is a form of violence
against women, and it contests the claims of choice, consent and voluntariness,
citing prostitution survivors’ description of it as  “the choice  made by those  who 
have  no  choice.”74 Kathleen Barry argues that prostitution is a form of female
sexual slavery.75 In contrast to the sex-work approach, which concerns itself with
conditions of labor, the sexual-domination claims that the nature of the activity is
the problem and that it is inherently harmful. Unlike the sex-work proponents,
sexual-domination theorists argue that what is being sold is the person herself and
not just her services. Carole Pateman, describing what is wrong with prostitution,
states that for the client to buy mastery of an objectified female body, the prostitute
must sell herself in a different and much more real sense than that which is required
by any other occupation.76 This is inherently damaging to the prostitute.77 Sexualdomination
proponents do not dispute the fact that prostitution is often a result and
an indicator of economic inequality, but they see sexual domination as a more
fundamental basis and explanation of prostitution.78
Radical feminists are proponents of the sexual-domination position,79 which,
unlike the relatively diverse sex-work position, is quite unified in terms of
principled arguments. Sexual-domination proponents argue that patriarchy is at the
root of the subordination that is prostitution and see the fight against prostitution as
a necessary part of challenge to gender inequality in society. They try to attain an
ideal—they hope for the eradication of prostitution by curbing its demand. The
principles of the sexual-domination position cannot be compromised, and thus there
is no pragmatic position. As a result, a permissive policy, such as legalization,
becomes unacceptable even as a temporary solution, as it would create a culture
accepting of prostitution and be in direct clash with the principle of gender
equality.
In terms of evidence, sexual-domination proponents often use in-depth
interviews and biographical narratives. These sources are combined with statistical
data to show the violence inherent in prostitution. Citing extensive research
72 See KEMPADOO & DOEZEMA, supra note 54.
73 Outshoorn, supra note 53.
74 Melissa Farley, Prostitution, Trafficking, and Cultural Amnesia: What We Must Not Know in
Order to Keep the Business of Sexual Exploitation Running Smoothly, 18 YALE J.L. & FEMINISM 109,
110 (2006).
75 See KATHLEEN BARRY, FEMALE SEXUAL SLAVERY 40 (1984).
76 See CAROLE PATEMAN, THE SEXUAL CONTRACT 207 (1988).
77 See also Dorchen Leidholdt, Prostitution: A Violation of Women’s Human Rights, 1 CARDOZO J.
L. & GENDER 133, 135 (1993); MacKinnon, supra note 52.
78 See BARRY, supra note 75, at 9-10.
79 Andrea Dworkin, Catharine MacKinnon, Janice Raymond, Kathleen Barry, among others.
conducted in different countries, Melissa Farley points out that the overwhelming
majority of women in prostitution report repeated instances of verbal abuse,
physical assault, and rape by both procurers and buyers.80 Cross-culturally, the
prevalence of post-traumatic stress disorder (“PTSD”) is at 78 – 80 %, levels which
are higher than those of Vietnam veterans,81 and a Canadian statistic shows that
prostitutes are forty times more likely to become murder victims than the general
populace.82 Sheila Jeffreys conceptually distinguishes between this “unpaid for” 
violence and the “paid for violence of everyday penetration.”83 Sexual proponents
understand not only the excess brutality inflicted on prostitutes in some cases, but
rather all prostitution, to fall into the category of violence against women. Because
lack of choice and harm are considered intrinsic to prostitution, prostitution is not
seen as conceptually separate from trafficking; the distinction between involuntary
and voluntary prostitution is contested. As a result, international anti-trafficking
instruments84 are given the broadest possible meaning. The preferred legislative
response is the curbing of demand85 through the criminalization of procurers and
clients, accompanied by a support system for women to escape prostitution86the
“Swedish model.”87
C. Critique of Proposed Legal Responses
It is obvious from the above mentioned postulates that the two feminist
positions  will  be  highly  critical  of  each  other’s  preferred  legal  response  to 
prostitution. Sex-work proponents point out that the proposal for suppression of
demand, the “Swedish model,” is blinded to the realities of prostitution by its naïve 
idealism. The wellbeing of women in or about to enter prostitution should be
tantamount,  and  not  an  abstract  goal  of  abolition.    One  result  of  the  “Swedish 
model,”  asserted  by  the sex-work proponents, is that the actual situation of
prostitutes is not improved; they still cannot operate openly and claim equal rights.
It is argued that clients’ fear of criminalization leads to two further negative effects:
the negotiation time for transactions shortens, which makes it more difficult for the
prostitute to assess whether the client is dangerous;88 and clients, who are
sometimes the only possible link between a victim of trafficking and the police,
lose any incentive for reporting potential instances of trafficking.89 According to a
Norwegian government’s report, one Swedish prostitute informant stated, “There is 
great pressure on prices, demands for unprotected sex have increased and there is
more violence.”90
80 See generally Farley, supra note 74.
81 See MACKINNON, supra note 11, at 1259.
82 See Farley, supra note 74, at 115.
83 Jeffreys, supra note 20.
84 See Protocol to Prevent, Suppress and Punish Trafficking in Persons, Especially Women and
Children, Supplementing the U.N. Convention Against Transnational Organized Crime, G.A. Res.
55/25, 55 U.N. GAOR Supp. (No. 49), U.N. Doc. A/45/49 (Vol. I) (2001).
85 See, e.g., MacKinnon, supra note 52; BARRY, supra note 75; KATHLEEN BARRY, THE
PROSTITUTION OF SEXUALITY (1995); Farley, supra note 74.
86 See ANDREA DWORKIN, LETTERS FROM A WAR ZONE: WRITINGS 1976-1987 147 (1988).
87 The proponents of the sexual-domination approach also argue for the availability of civil
remedies. MACKINNON, supra note 11, at 1322, 1327. A civil claim for coercion into prostitution exists
in the state of Florida. FLA. STAT. § 796.09(1) (2011).
88 See NORWEGIAN MINISTRY OF JUSTICE AND THE POLICE, supra note 10.
89 From interviews conducted by Hančilová.  See Hančilová & Massey, supra note 9, at 103-11.
90 NORWEGIAN MINISTRY OF JUSTICE AND THE POLICE, supra note 10, at 19.
On the other side, sexual-domination  proponents’  critique  of  permissive 
approaches  concentrates  on  the  “prostitution  culture”an increase in both legal
and illegal prostitution91 and an acceptance of objectification of women. They
argue that legalization does not coincide with equal treatment, normalization and
destigmatization. They point out that the preference for indoor prostitution and the
fact that outdoor prostitution is permitted in very limited urban areas under the
“Dutch model” show that the aim is to keep prostitution out of sight, which implies
that prostitution is still not viewed and normal and is not destigmatized. They also
criticize health checks as a measure that aims to secure an STD-free service to
clients rather than protection to the prostitute. The target of legal control of
prostitution  is  thus  its  “outward  appearance  rather  than  the  conditions  in  which 
women find themselves. On the whole, governments are far more anxious about
public order and public health than about abuse and violence.”92 It is also often
pointed out that only a small percentage of prostitutes have availed themselves of
the registration option,93 which shows that not all prostitutes are in the position to
regularize their situation.
D. A Middle Ground?
It is difficult to locate a middle ground in the feminist debate, especially as
far as legal solutions are concerned. Several writers have extremely important
insights and avoid the binary. For example, many socialist feminists argue that
prostitution should not be attacked directly but that women’s poverty needs to be 
addressed.  For example, Stephanie Limoncelli states that “strategies for social and 
economic justice [are needed], which at the same time, will help to combat the
exploitation of women in prostitution.”94 Another solution is offered by Laurie
Shrage, who sees prostitution as gendered and oppressive only under the current
gender order95 and argues for a change in the cultural context.96 Shrage stated,
“By striving to overcome discriminatory structures in all aspects of society—in the
family, at work outside the home, and in our political institutions—feminists will
succeed in challenging some of the cultural presuppositions which sustain
prostitution.”97 Consequently,  “if  prostitution  were  sufficiently transformed to
91 For an example, see Mary Sullivan, What Happens When Prostitution Becomes Work? An
Update on Legalisation of Prostitution in Australia (2005), available at http://action.web.ca/home/
catw/attach/Sullivan_proof_01.pdf.
92 MARJAN WIJERS & LIN LAP-CHEW, TRAFFICKING IN WOMEN FORCED LABOUR AND SLAVERYLIKE
PRACTICES IN MARRIAGE, DOMESTIC LABOUR AND PROSTITUTION cited in Farley, supra note 74, at
137.
93 According to the Prostitution Information Center in Amsterdam, in the Netherlands only 5% to
10% of the nearly 20,000 prostitutes are registered and pay taxes. See Dan Bilefsky, Belgian
Experiment: Make Prostitution Legal to Fight Its Ills, WALL ST. J., May 26, 2005 at 2, available at
http://www.aegis.org/news/wsj/2005/WJ050507.html.
94 Limoncelli, supra note 70, at 267.
95 [A]lthough the commercial availability of sexuality is not in every existing or
conceivable society oppressive to women, in our society this practice depends upon the
general acceptance of principles which serve to marginalize women socially and
politically. Because of the cultural context in which prostitution operates, it epitomizes
and perpetuates pernicious patriarchal beliefs and values and, therefore, is both damaging
to the women who sell sex, and as an organized social practice, to all women in our
society.
Shrage, supra note 7, at 349.
96 See id.
97 Id. at 360.
make it completely non-oppressive to women, though commercial transactions
involving sex might still exist, prostitution as we now know it would not.”98
Both of these positions sidestep, rather than go beyond, the binary with
regard to legal solutions. They neither help governments that might be pondering
how to construct, amend, or change their legal response to prostitution nor judges
who are asked to review an existing policy or its application in light of their
constitutional, statutory or international requirements. This Article aims to offer a
gender  equality  test  that  transcends  the  two  positions,  is  agnostic  to  the  “best 
solution,” and can be practically implemented.
E. On Whom to Base a Legal Response?
The two feminist positions accuse each other of disconnect with the reality
and the homogenization of the phenomenon of prostitution. The sex-work position
argues that the sexual-domination position disregards the heterogeneity of
prostitution, ignores agency and imposes a victim narrative on everyone. The
sexual-domination position argues that sex-work proponents ignore the violent
reality of prostitution. Sex-work proponents who complain about homogenization
argue that the diversity in sex-workers’ lives and experiences must be recognized.
They are not always victims or objects.99 An example is sometimes given of a rich
university student who occasionally sells sex for luxuries.100 It is argued that there
is choice and possibly sexual enjoyment, and that these women need to be taken
into account when designing a legal solution.
The sexual-domination proponents dismiss this argument101 with four
counterarguments. First, conceptually, the possibility of exercising free choice
under patriarchy is questioned.102 Second, it is argued that whether the prostitute
accepts to be a victim or not, the context and social meaning of prostitution in
patriarchal society still victimize her.103 Third, the existence of prostitution harms
women’s equality because it creates a “culture of prostitution.”  In a society where 
prostitution is considered normal and the objectification of women is acceptable,
men can base their understanding of all women on prostitution, whether as
husbands and partners or as executives who decide about women’s advancement in
the workplace.104 Fourth, it is argued that the “creamy layer” in prostitution does 
harm  by  “false  advertising”105creating an impression about the realities of
prostitution which are not true for the majority of women involved.
As far as homogenization goes, I believe that both positions commit this error
to some extent. Let us, for the sake of argument, say that there are three
“segments”  of  prostitution— color-coded  as  traffic  lights.    The  first  is  a  “red 
segment,”  where  prostitutes  are  victims  of  coercion, and as such fall under the
98 Id. at 359.
99 See Davidson, supra note 55, at 91-92.
100 Internationally, especially third wave feminists stress the difference between the First and the
Third World and criticize the Western cultural imperialism in dominating the prostitution discourse, a
lack of historical and geo-political contextualization. See KEMPADOO & DOEZEMA, supra note 54, at
13.
101 Often referred to as “happy hooker.”
102 “[P]rostitution is about the absence of meaningful choices.”  Leidholdt, supra note 77, at 136.
103 See Shrage, supra note 7, at 358.
104 JEFFREYS, supra note 20.
105 I thank Michelle Madden Dempsey for this insight.
trafficking definition of the Palermo Protocol,106 or are underage. The second is a
“yellow segment,” where the prostitutes are vulnerable by reasons of gender, race, 
homelessness, drug or alcohol addiction, immigration status, or poverty, and are
often under the control of a pimp. The freedom of their choice to enter and remain
in  prostitution  could  be  disputed.    The  third  is  a  “green  segment,”  where  the 
prostitute has other options but chooses prostitution and is independent and in
control.
While  the  “red  segment”  is  considered  unacceptable  to  both  positions  and 
there is consensus for its criminalization, the proponents of the sex-work and
sexual-domination positions disagree about the existence and the size of the
“yellow” and the “green” segments. To sexual-domination proponents, the “green 
segment” is an illusion and the “yellow segment” is virtually indistinguishable from 
the “red.”  To sex-work proponents, the “green segment” definitely exists and is 
substantial in size, but their policies are based on the argument that the situation of
both  prostitutes  in  the  “green  segment”  and  the  “yellow  segment”  would  be 
improved if prostitution were normalized, accepted and treated equally with other
work or services.
It is very difficult  to  know  which  portion  of  the  “sex  industry”  falls  into 
which segment. While a proponent of the sex-work position might estimate the
segments as 10% red, 50% orange, and 40% green, a sexual-domination proponent
might believe it is 50% red and 50% orange without any green segment. Empirical
research into prostitution faces difficulties of practical nature, but more importantly
of theoretical nature, connected to the binary conceptualizations of prostitution.
Sexual-domination proponents doubt that declarations of agency and freedom by
prostitutes are creditable; as to them, these utterances are either made under
coercion or are at any rate expressions of false consciousness. It is thus difficult to
foresee a universally acceptable empirical answer about what happens in
prostitution. And it is not the aim of this Article to provide it. However, it is
important to briefly address what the division into “segments” means for a legal 
response to prostitution.
While I am sympathetic to the radical feminist analysis that under patriarchy
all choice is problematic and that violence is an omnipresent threat, it is hard to
entirely  dismiss  the  existence  of  the  “green  segment.”    It  is  difficult  to  tell 
prostitutes that they are victims of sexual slavery and of false consciousness,
especially when they claim that they know what they are doing, that prostitution is
the best they can do, and that they want to continue doing it. Rather, the questions
to be asked are different: 1) Can a legislative response be based on more segments
at once, or 2) should a legislative response be based on one segment only? The
first question is answered in the affirmative by the current difference in treatment
by most countries of trafficking on the one hand—criminalization of the  “red 
106 “Trafficking in persons” shall mean the recruitment, transportation, transfer, harboring or
receipt of persons, by means of the threat or use of force or other forms of coercion, of
abduction, of fraud, of deception, of the abuse of power or of a position of vulnerability
or of the giving or receiving of payments or benefits to achieve the consent of a person
having control over another person, for the purpose of exploitation. Exploitation shall
include, at a minimum, the exploitation of the prostitution of others or other forms of
sexual exploitation, forced labour or services, slavery or practices similar to slavery,
servitude or the removal of organs.
G.A. RES. 55/25, U.N. GAOR 55th
Sess.SUPP. No. 49, (Vol. I), U.N. DOC. A/45/49, at 2 (2001).
segment”—and the remaining “green” and “yellow” segments on the other.  While 
not common practice, further distinguishing between legal treatment of the “green” 
and “yellow” segments is feasible.  “Green segment,” or “voluntary” prostitutes, 
could be  allowed  to  normalize  their  situation,  but  any  identification  as  “green” 
would have to be made by the prostitute herself and be entirely voluntary. In other
words, the prostitutes in the “yellow” segment should never be forced to conform 
to the rules applicable to the “green” segment, nor should they be punished for not 
conforming.
The second question basically asks whether, even recognizing that there is
coercion and abuse, the overall legal response should be based on “green segment” 
prostitutes who would benefit from legalization, as sex-work proponents argue. Or
whether, on the other hand, even recognizing that a certain percentage of prostitutes
choose prostitution and want it legalized, the legislative response should be based
on another vulnerable and  endangered  group  of  prostitutes  in  the  “yellow”  and 
“red” segments, as sexual-domination proponents advise. Here, I should admit to
sympathy with the latter argument that the danger inherent in prostitution could
justify legislatively limiting some prostitutes’  freedom,  even  if  they  made  a 
voluntary decision to risk it. I accept the sexual-domination arguments of “false 
advertising”, as well as the argument that acceptance of prostitution leads to a
“culture of prostitution” which threatens the achievement of a gender equal society.
These observations show my preliminary thoughts on the ultimate solution and I
disclose them to admit to my position. However, in the following, I aim to
transcend the binary conceptualizations of prostitution and the opposed
understandings of what the requirements of gender equality mean for the regulation
of prostitution.
III. GENDER EQUALITY
To begin with, let us explore why prostitution is a matter of gender and also a
matter of equality. Two reasons could be mentioned as to why prostitution is a
gender issue. First, the distribution of men as buyers and women as sellers
suggests that prostitution is an extremely sex-segregated field where the demand is
overwhelmingly created by men and the supply by women.107 This is a fact on
which both feminist positions agree108 and that few people generally would
dispute.109 Secondly, the sexual-domination position would further argue that
prostitution is inherently gendered. Dorchen Leidholdt, an anti-pornography and
anti-prostitution feminist, commented, “What other job is so deeply gendered that 
one’s breasts, vagina, and rectum constitute the working equipment? Is so deeply
gendered that the workers are exclusively women and children and young men used
like women?”110
As far as equality is concerned, the legal treatment of prostitution is a matter
107 Rosaan Kruger, pointing to South African statistics, says that 95% of prostitutes are women. See
KRUGER, supra note 12.
108 See Outshoorn, The Political Debates on Prostitution and Trafficking of Women, supra note 53,
at 145-46.
109 See, e.g., State v. Jordan (6) SA 642 (CC), at 60 (S. Afr.) (O’Regan, J. & Sachs, J. dissenting)
(the agreement among the parties on this issue).
110 Leidholdt, supra note 77, at 138-39. While  I  agree  with  Leidholdt’s  analysis,  the  first 
observation about the distribution of the sexes in the respective roles of the prostitute and the client is
sufficient to support my argument that prostitution is a matter of gender.
of equality if two similarly situated parties to the same transaction are treated
differently without good reason. In terms of the standard test of equality,
national111 understandings of it vary,112 but they often ask similar questions: (i) Is
there a difference in treatment or impact113 (ii) between persons that are
comparable?114 (iii) Is the ground for the distinction suspicious?115 (iv) Is the
distinction fair?116 (v) Is it pursuing a legitimate aim?117 (v) Is the measure
proportionate to that aim118? Of these, difference impact,  or  “indirect 
discrimination” in European terminology, will be discussed in more detail below.
Particular attention will further be paid to the questions of comparability and
justification,119 as well as that of proportionality.
A formal and a substantive understanding of equality can be distinguished.120
Whereas a formal understanding of equality calls for consistency of treatment
111 I am drawing on comparisons of national standards, as these allow me to use existing national
case-law. However, the international standard definition of sex discrimination, as contained in
CEDAW, comprises many of the elements as well:
For  the  purposes  of  the  present  Convention, the  term “discrimination  against  women” 
shall mean any distinction, exclusion or restriction made on the basis of sex which has the
effect or purpose of impairing or nullifying the recognition, enjoyment or exercise by
women, irrespective of their marital status, on a basis of equality of men and women, of
human rights and fundamental freedoms in the political, economic, social, cultural, civil
or any other field.
Convention on the Elimination of All Forms of Discrimination Against Women, Dec. 18, 1979, 1249
U.N.T.S. 14.
112 A comparative analysis of the definitions of equality and the judicial tests goes beyond the aims
of this paper. My reasoning is based primarily on the case law of the Court of Justice of the European
Union (CJEU), the European Court of Human Rights (ECHR), the U.S. Supreme Court, the Supreme
Court of Canada and the Constitutional Court of South Africa. I am using the analysis available in
JANNEKE H. GERARDS, JUDICIAL REVIEW IN EQUAL TREATMENT CASES (2005); and also MACKINNON,
supra note 11, at 2-40.
113 A difference in treatment is termed “direct discrimination” in the European and South African 
context  and  “disparate  treatment”  in  the  U.S., and a difference in impact of an otherwise neutral
provision is termed “indirect discrimination” in the European and South African context and “disparate 
impact” in the US.
114 The emphasis on comparability varies. In the U.S., the question whether groups are “similarly 
situated,” is often crucial for deciding whether prima facie discrimination occurred or not. The CJEU
refers to comparability seldom, almost never in indirect discrimination (disparate impact) cases.
115 Most  jurisdictions  consider  some  distinctions  more  “invidious”  (to  use  the  U.S. term) than
others. Sex/gender is usually one of the grounds that raise suspicion. In the E.U., the specially
protected grounds can be found in the competence provision of Art. 19 Treaty on the Functioning of the
European Union; in the South African Bill of Rights, they are listed in Art. 9(3); in the U.S., the list is
case-law based, and sex is subject to intermediate scrutiny.
116 Among the jurisdictions that I am looking at, this part of the test exists separately only in South
Africa. See State v. Jordan 2002 (6) SA 642 (CC), discussed below in Part III.A.
117 This question is about “justification”—it inquires after the reasons for which this distinction has
been made by the government. While in most cases, this is a judge-made part of the test; some
countries have a constitutional legal basis for this. For example in South Africa, the general rule on
limitations of rights in used in Art. 36(1) of the South African Bill of Rights, which states that “[t]he
rights […] may be limited only in terms of law of general application to the extent that the limitation is 
reasonable and justifiable in an open and democratic society based on human dignity, equality and
freedom, taking into account all relevant factors”. S. AFR. CONST., 1996.
118 Proportionality is a test used by the ECHR and the CJEU. See GERARDS, supra note 112, at 209
& 223. In the U.S., the requirement on the relationship between the measure and the aim depends on the
standard of scrutiny.
119 Awarding great importance to comparability has been criticized by Gerards. She argued that too
much depends on the choice of criteria which measure comparability—non-identical situations, or
groups, are always in some ways the same and in some ways different. She proposed that such
questions should be addressed in a more nuanced way under the rubric of justification and
proportionality. See id. at 57.
120 See SANDRA FREDMAN, DISCRIMINATION LAW (2002); see also MACKINNON, supra note 11, at
20.
without regard to the possibly disparate situations in which the subjects may find
themselves, a substantive understanding of equality is context-conscious and reacts
to the actual realities in which the individuals or groups find themselves.121 This
Article is written from substantive equality perspective. The factual inequalities
between prostitutes and clients are thus considered salient for the equality analysis.
In the following, I examine three possible legal scenarios for the comparative
treatment of the prostitute and the client. First, there can be asymmetric treatment
benefiting the client, which is still an existing practice and also a violation of
gender equality. Second, the treatment can be symmetric, in which both parties are
subject to the same treatment. The principle of equality requires at least this
standard and this solution will resonate with the sex-work proponents, even though
their comparison is usually to other workers or providers of services and not to the
client. Third, an asymmetric treatment for the benefit of the prostitute can be
adopted. While this approach would be more popular with the sexual-domination
proponents, I argue that it is paradigmatically reconcilable with the sex work
position as well.
A. Asymmetric Treatment Benefiting the Client
An asymmetric rule benefiting the client can either be directly or indirectly
discriminatory. A case of direct discrimination would require an explicit reference
to the sex of the prostitute or the client, while a case of indirect discrimination
would entail different treatment of groups consisting mostly of persons of one
sex.122 Until recently, some jurisdictions criminalized prostitution only when a
woman was engaged in it. Catharine MacKinnon points out that the state of
Louisiana used to define prostitution as “the practice by a female of indiscriminate
sexual  intercourse  with  males  for  compensation.”123 At the time, the court
reviewing it for compatibility with the Equal Protection Clause ruled that
“[D]ifferences between the sexes does bear a rational relationship to the prohibition
of  prostitution  by  females.”124 Arguably, such directly discriminatory provision
would be found unconstitutional today, especially following a later United States
Supreme Court decision in Craig v. Boren, in which the Court applied intermediate
scrutiny rather than rationality review.125 It is difficult to imagine a government
policy on prostitution that would require a distinction to be made between the sexes
in regulating prostitutes. As such, the provision should fail any test of legitimate
121 Substantive equality, in recognizing the context—existing de facto inequality and
disadvantage—,allows for different treatment for the benefit of the individual or the group that is worse
off culturally and socio-economically. I argue that a substantive understanding of equality is implicit in
any jurisdiction’s acceptance of affirmative action—or positive action in European terminology. The
substantive equality approach has been explicitly adopted by various courts. See, e.g., Andrews v. Law
Society of British Columbia, [1989] S.C.R. 143 (Can.). In the area of prostitution, it was the approach
of the dissent in the State v. Jordan case 2002 (6) SA 642 (CC) (O’Regan, J. & Sachs, J. dissenting),
which is discussed below in IV.A and which this Article endorses.
122 Without explicit reference to the sexes of the groups, indirect discrimination would entail
different treatment of clients, mostly men and prostitutes, mostly women.
123 State of Louisiana v. Devall, 302 So. 2d 909, 910 (La. 1974) (quoting LA. REV. STAT. ANN. §
14:82).
124 Id. at 913. See also Julie Lefler, Shining the Spotlight on Johns: Moving Toward Equal
Treatment of Male Customers and Female Prostitutes, 10 HASTINGS WOMEN’S L. J. 11, 24 (1999).
125 Two years after DeVall, the United States Supreme Court established that different treatment or
impact based on sex is subject to intermediate scrutiny—a policy is only constitutional when it furthers
an important government interest in a way that is substantially related to that interest. See Craig v.
Boren, 429 U.S. 190, 197 (1976).
aim or justification.
Most statutes today, at least in the European and common law jurisdictions
referred to in this Article, use gender-neutral formulations. They do not distinguish
between male and female prostitutes, nor do they identify prostitutes as female and
clients as male. A more common problem would be that of indirect discrimination
consisting in the different treatment of prostitutes and clients.
Practices of indirect discrimination are still wide-spread. The legal
provisions can either foresee rules that only apply to prostitutes and not
clientsfor example, the sole criminalization of selling of sex as in State v.
Jordan126or the rules can apply differently to the prostitute and the client, being
stricter on the prostitute. More serious offenses and harsher sanctions for the
prostitute are still common in criminal law.127 For example, the Czech 2005
proposal foresaw criminal punishment for repeated prostitution without license for
the prostitute but none for repeated use of prostitutes without license for the client.
Areas of regulation other than criminal law need scrutiny as well, including
administrative, police, social, and health law.128 The Czech 2005 Proposal foresaw
obligatory registration and information duties vis-à-vis various public authorities on
the prostitute, obligatory health checks for the prostitutes and not clients, and
various activity restrictions for the prostitutes, such as time and space limits in
outdoor prostitution. The breach of these obligations would impose liability and
would be punishable by fines or even imprisonment. It should be mentioned that
even if different treatment is not explicit in the legal regulation, it often occurs in
its implementation. Application and enforcement should be subject to gender
equality scrutiny as well.
A question which needs addressing under an equality test is whether there are
differences between the groups of clients and prostitutes that warrant their
differential treatment. This question is in some jurisdictions framed as one of
comparability; in others, it is a question of justification.129 A good starting point is
the South African case State v. Jordan, which involved a constitutional challenge to
provisions of the Sexual Offenses Act which criminalized providing sex for reward
but did not explicitly criminalize paying the reward for sex.130 The argument in
the case is of general interest and applies to any legal response to prostitution, not
just to “criminalization” regimes.
The  court’s  majority  in  State v. Jordan did not find that provisions of the
Sexual Offenses Act discriminated unfairly against women because the client was
criminally liable as an accomplice and conspirator under the Riotous Assemblies
Act.131 The court found substantive differences between the prostitute and the
client, which warranted making the prostitute the primary offender. The majority
pointed out that the prostitute was more likely to be a repeat offender and was the
126 See State v. Jordan 2002 (6) SA 642 (CC).
127 Julie Lefler cites several examples of different criminal sanctions from the U.S. See Lefler,
supra note 124, at 17.
128 Julie Lefler, who provides a convincing case for equality within criminalization, is
overoptimistic  when  she  states  that  “[o]bviously  legalization  or  decriminalization  would  obviate  the
need for a discussion of differential treatment since neither male patrons nor female prostitutes would be
within the grasp of the justice system.” Id. at 12 n.1.
129 See supra notes 114 & 119.
130 See supra note 12.
131 See The Riotous Assemblies Act 17 of 1956 §18(2) (S. Afr.).
“merchant”  and  “dealer,”  whereas  the  client  was  the  “customer.”132 It likened
prostitution to other areas, such as the sale of dangerous weapons, medicines and
liquor.133 The majority considered both the potential higher stigma on the
prostitute,134 as well as disparate enforcement targeting the prostitute135 to be
irrelevant.  Before looking at the majority’s position in more detail, the dissenting 
opinion merits discussion.
The  dissent  disagreed  with  the  majority’s  representation  of  the  difference 
between prostitutes and clients and saw only three differences between the
prostitute and the client:
The first is that the one pays and the other is paid. The second is that in
general the one is female and the other is male.  The third is that the one’s actions
are  rendered  criminal  by  section  20(1)(A)  but  the  other’s  actions  are  not.  
Moreover, the effect of making the prostitute the primary offender directly
reinforces a pattern of sexual stereotyping which is itself in conflict with the
principle of gender equality. The differential impact between prostitute and client
is therefore directly linked to a pattern of gender disadvantage which our
Constitution is committed to eradicating.136
As far as the criminalization of the client as an accessory or conspirator is
concerned,  the  dissent  conceded  that  “[t]he  difference  between  being  a  principal 
offender and an accomplice or co-conspirator may have little impact in formal legal
terms;” however, it saw an important “difference in social stigma and impact.”137
Moreover,  the  dissenting  judges  argued  that,  “[i]n  imposing  a  direct  criminal 
liability for the prostitute, the law chooses to censure and castigate the conduct of
the prostitute directly. . .[T]he primary crime and the primary stigma lie in offering
sexual intercourse for reward, not in purchasing it.”138 Unlike the majority, the
dissent was very concerned about the context and the social reality of prostitution:
The female prostitute has been the social outcast, the male patron has been
accepted or ignored. She is visible and denounced, her existence tainted by her
activity. He is faceless, a mere ingredient in her offence rather than a criminal in
his own right, who returns to respectability after the encounter. In terms of the
sexual double standards prevalent in our society, he has often been regarded either
as having given in to temptation, or as having done the sort of thing that men do.
Thus, a man visiting a prostitute is not considered by many to have acted in a
morally reprehensible fashion. A woman who is a prostitute is considered by most
to be beyond the pale. The difference in social stigma tracks a pattern of applying
different standards to the sexuality of men and women.139
The primary criminalization of the prostitute, according to the dissenting
judges,  “reflects  and  reinforces”  “harmful  social  prejudices  against  women”  and 
132 State v. Jordan 2002 (6) SA 642 (CC) at 10.
133 See id.
134 The majority commented: “If the public sees the recipient of reward as being ‘more to blame’ 
than the ‘client’, and a conviction carries a greater stigma on the ‘prostitute’ for that reason, that is a 
social attitude and not the result of the law.”  Id. at 16.
135 The  majority  commented:  “What  happens  in  practice  may  therefore  point  to  a  flaw  in  the 
application of the law but it does not establish a constitutional defect in it.”  See id. at 19.
136 Id. at 60 (O’Regan, J. & Sachs, J., dissenting).
137 Id. at 63.
138 State v. Jordan 2002 (6) SA 642 (CC) at 63 (O’Regan, J., and Sachs, J., dissenting).
139 Id. at 64.
“stems  from  and  perpetuates  gender  stereotypes.”140 The dissenting judges
identify one very important element of the reality of prostitution that shows a
particular vulnerability and should be considered in a legal response – the social
meaning of prostitution. Under my analysis, the social meaning of prostitution,
which is in most societies stigmatizing to the prostitute, is one of justifications for
equal treatment of the client and the prostitute and even for asymmetric treatment
benefiting the prostitute.
B. Challenging Justifications of Asymmetric Treatment Benefiting the Client
The majority in State v. Jordan proposed two justifications: (1) the prostitute
as a repeat offender and (2) the prostitute as a dealer. This section will also analyze
the potential justification of (3) targeting prostitutes for enforcement practicality
and (4) will look at how to assess other possible justifications, especially
highlighting the requirement of proportionality and correct tailoring.
First, the argument that prostitutes can legitimately be the primary target of
criminal provisions because they are more likely to be repeat offenders is an
example of flawed circular reasoning. When prostitutes are the explicitly named
offenders in criminal or administrative law, they are, logically, the primary target of
enforcement and are more often prosecuted and become repeat offenders. That this
is then held against them as a justification for a different treatment in criminal law
is a serious logical fallacy in reasoning about legal regulation.
Second, courts, including the majority in State v. Jordan, have allowed a
harsher treatment of the prostitute based on their understanding of prostitutes as
“merchants,”  “dealers,”141 or  as  “profiteers  in  commercial  crimes,”142 and they
have used analogies to drugs, liquor and arms.143 There are, however, fundamental
differences between prostitution and the drug trade in the potential harms that they
pose to the “consumer.”  A drug dealer sells a product that is potentially deadly and 
addictive. The same can be said of alcohol and at least the former for arms. The
only potential physical harm to the client from prostitution is that resulting from
sexually transmitted diseases, and here, both sides of the transaction share the risk
and the blame at least equally. Actually, it is the prostitute and not the client who is
at greater risk of serious physical and psychological harm, as I discuss below; a fact
which justifies symmetrical treatment of both parties or special treatment for the
prostitute, but makes preferential treatment of the client perverse.
Another aspect of the “dealer” narrative is the underlying belief that it is the 
prostitute who controls the transaction. While it is the dealer in the drug industry
that can dictate the price and has the power over a drug addict customer, the
situation in prostitution is rather the reverse. It is the prostitute who typically needs
the money and has little power and control of negotiation over price and other
conditions. Moreover, when a pimp is involved, the prostitute has virtually no
control over the transaction or her earnings, and when a trafficker is involved, she
does not even have control over being a prostitute. In such situations, she is not in
the position of the dealer or the merchant. There are inequalities between clients
140 Id. at 65.
141 Id. at 19.
142 This narrative is not uncommon in court decisions in the U.S. See, e.g., People v. Superior
Court, 19 Cal. 3d 338 (1977).
143 See State v. Jordan 2002 (6) SA 642 (CC) at 10; Lefler, supra note 124, at 21.
and prostitutes that need to be taken into account in devising a legal response.
These are issues of class, age, race, gender, nationality, immigration status and
socio-economic status,144 as I argue below, not the  prostitutes’  recidivism  or 
profiteering as was argued by the majority in State v. Jordan.
Third, the true reason for targeting the prostitute in a legal response is often
that  of  enforcement  practicality.    In  the  “decriminalization  with  state  control” 
regimes, the prostitute is registered, and she is much more easily accessible. In any
other regime, she can be easily found at her “usual workplace.”  The same cannot 
be said of the client, and as a result, the prostitute is the more easily regulated side
of the transaction. I argue that enforcement practicality has to be discarded when
presented as a justification when the equality of the sexes, a suspect characteristic
that is given special constitutional protection, is at stake. The prohibition of
discrimination is in place precisely so that legal regulation does the fair thing and
not the easy thing. Laws should not be made to suit enforcement. Ideally, full and
effective enforcement would follow a legal rule. However, a legal rule, especially
when promoting gender equality, still has important educative,145 symbolic146 and
expressive147 value. It is surely better if the educational, symbolic or expressive
value of the law is morally correct rather than morally wrong, whether or not it can
be fully enforced. Many authors acknowledge that these values are particularly
important elements of equality and anti-discrimination law.148 The British legal
sociologist Cotterrell observes that several purposes of the United Kingdom Race
Relations Act of 1965 went “well beyond the use of law to redress grievances or to 
control behavior, and [the aim to reduce prejudice by discouraging the behavior in
which prejudice finds expression] point[ed] unequivocally to the educative function
of  law.”149 Similarly, he points to the importance of the rhetoric of the desegregation
decision of the United States Supreme Court in Brown v. Board of
Education.150
Catherine Fieschi cites the example of the United Kingdom Racial and
Religious Hatred Act of 2006 as a statute which sends “ . . . a signal . . . to the 
Muslim minority . . . that its concerns are taken seriously [like those of Sikhs and
Jews which were already protected through race legislation], and then the law is
drafted and enforced on the assumption that it will only rarely, if ever, be used.”151
She argues that:
[p]artly because policing and enforcement elude us in increasingly borderless
situations, the trend, and not just in law, is away from sanction and punishment
towards changing attitudes and modifying behaviour. Governments are warming to
144 See discussion infra Part III.C.3.
145 “‘Educative legislation’ aims to promote ideas through governmental action.” An example given
by Roger Cotterrell is the first British Race Relations Act of 1965 which stressed conciliation rather than
redress. ROGER COTTERRELL, The SOCIOLOGY OF LAW: AN INTRODUCTION 53 (Butterworths 2d ed.
1992).
146 See id. at 102.
147 See Elizabeth S. Anderson & Richard H. Pildes, Expressive Theories of Law: A General
Restatement, 148 U. PA. L. REV. 1504 (2000).
148 See generally, W.G. CARSON, Law Making: Symbolism and Instrumentality, in LAW AND
SOCIETY 236-246 (Colin Campbell & Paul Wiles eds., 1979).
149 COTTERRELL, supra note 145, at 53 and therein cited ANTHONY LESTER & GEOFFREY BINDMAN,
RACE AND LAW 85 (Longman. 1972).
150 See Brown v. Board of Education, 347 U.S. 483 (1957); COTTERRELL, supra note 145, at 106.
151 Catherine Fieschi, Symbolic Laws, PROSPECT MAGAZINE (2006).
the notion that it is in part through moral persuasion and debate created by
legislation that attitudes and behaviour will change.152
The legal response to prostitution is one area where the state has to be
particularly careful to espouse norms and values that are compatible with the
principle of gender equality and to set aside those which are not. As such, mere
enforcement practicality of provisions targeting the prostitute cannot stand as
justifications for such discriminatory legislation.
Fourth, it is possible that specific measures will have specific legislative aims
that are not discussed in this Article. For example, a government might try to
justify obligatory health checks with public health concerns and zoning rules with
public order concerns. An important distinction to be made from the perspective of
the equality test is between the legitimacy and justifiability of the aim on one hand,
and the proportionality of the measure, especially its appropriateness, necessity153
and its tailoring,154 on the other. While the aim might be legitimate, a careful
examination will often show that a measure is not suitable or necessary to achieve
it. It may also fail the test of narrow tailoring by being under-inclusive, such as
targeting only prostitutes where clients should logically also be the object of
regulation. For example, the one-sided  prostitute’s  obligation  to  undergo  health 
checks, which was included in the Czech 2005 proposal, was intended to prevent
the spread of STDs. This is a laudable, legitimate legislative aim. However, the
question is whether targeting only the prostitute is an appropriate way of achieving
that objective. Experience from various countries shows that compulsory health
checks make the client feel safer to practice unprotected sex,155 resulting in a
serious danger to the health of the prostitute.156 As a consequence, obligatory
health checks seem to be more of a consumer protection measure for the benefit of
the client than a genuine public health protecting measure that stops STDs from
spreading. As a result, it fails the suitability test. It can also be argued to be
underinclusivewhy are only prostitutes but not clients the target of health
provisions? Similarly, zoning rules that only target soliciting by the prostitute but
not curb-crawling or soliciting by the client are too narrowly tailored. This
rigorous examination of suitability, necessity and tailoring would uncover many
other measures as inappropriate for the proclaimed aims of public health and public
order.
C. A Gender Equal Approach
I argue that in order for the requirement of gender equality to be fulfilled, the
152 Fieschi particularly highlights the importance of symbolic law for prompting a re-examination of
the law, as a trigger and shaper of political debate and as the creator of constituencies. Id.
153 See EVELYN ELLIS, THE PRINCIPLE OF PROPORTIONALITY IN THE LAWS OF EUROPE (Hart Publ’g. 
1999).
154 The requirement of narrow tailoring is sometimes referred to as a test of degree of fit, or of overor
under-inclusiveness. In the European context, it is often taken to form part of a proportionality
assessment; in the U.S., it is often assessed as part of the strict or intermediate scrutiny. See, e.g., Kahn
v. Shevin, 416 U.S. 351 (1974) (Brennan, J., dissenting); Grutter v. Bollinger, 539 U.S. 306 (2003).
155 Nina Lopez-Jones, Legalising Brothels, 1 THE NEW L.J. 594 (1992). See also HANA MALINOVÁ,
Připomínky organizace R-R k Návrhu zákona o regulaci prostitutce Zastupitelstva hlavního města Prahy 
(“,mnSuggestions of Rozkoš bez rizika to the Proposal of an Act on Regulation of Prostitution by the 
Municipal Assembly  of  the  Capital  City  of  Prague”) (2008), available at
www.rozkosbezrizika.cz/.../35_zakon_regulace_prostituce_pripominky_rr.doc.
156 Farley, supra note 74, at 138-39.
prostitute and the client need to be treated at least symmetrically, but an
asymmetric regime benefiting the prostitute is also justified. There are at least
three aspects of the phenomenon of prostitution which are important to this equality
analysis: 1) the social meaning of prostitution; 2) the risk of harm to the prostitute;
and 3) the de facto class, age, race, gender, nationality, immigration status and
socio-economic status and other inequalities between the client and the prostitute.
1. The Social Meaning of Prostitution
The dissent in State v. Jordan correctly identified that the social stigma of
prostitution is connected with the prostitute, who is the “social outcast . . . visible 
and  denounced,  her  existence  tainted  by  her  activity.”157 They  noted  that  “the 
stigma [of prostitution] is prejudicial to women, and runs along the fault lines of
archetypal presuppositions about male and female behavior, thereby fostering
gender inequality.”158 The stigma is a result of the social meaning of prostitution
in the Western societies.159 The “social meaning” of prostitution, to apply Jon
Herring  and  Michelle  Madden  Dempsey’s  analysis  of  social meaning of sexual
penetration,160 is not based on an intention or purpose of the individual client or
sufficient majority of clients, nor on the experience or registering by the individual
prostitute or sufficient majority of prostitutes.161 It means that “one of the current
social meanings of [prostitution] under current social conditions can only credibly
be explained as devaluing women qua women  and  disrespecting  women’s 
humanity.”162 Herring and Madden Dempsey use examples of the language which
we use to describe the act of sexual penetration,163 and the depictions of it in
“literature, film, advertising, television, pornography and internet discourse”164 to
support their claim that the current social meaning of sexual penetration make it
prima facie wrongful. As sexual intercourse, often in the form of penile
penetration, is a core element of prostitution, this analysis applies to prostitution as
well.165 Prostitution, however, has further degrading, devaluing and disrespecting
social meanings as “whores” and “sluts” engage in sexual contact repeatedly and 
for money. While Herring and Madden Dempsey use their analysis to argue that
sexual penetration is a prima facie wrong which requires justification as a
preliminary analysis for the construction of criminal offences,166 I would use it to
157 State v. Jordan 2002 (6) SA 642 (CC) at 64 (O’Regan, J. & Sachs, J., dissenting).
158 Id. at 65.
159 I believe that the analysis of social meaning of prostitution is valid in any patriarchal society, and
is thus almost universal.
160 See Michelle Madden Dempsey & Jonathan Herring, Why Sexual Penetration Requires
Justification, 27 OXFORD JOURNAL OF LEGAL STUDIES 467, 482 (2007).
161 The  point  that  an  individual’s  perspective  cannot  alter  the  social  meaning  of  prostitution  is 
shared by many authors. Laurie Shrage writes, “Although the prostitute may want the meaning of her
actions assessed relative to her own idiosyncratic beliefs and values, the political and social meaning of
her actions must be assessed in the political and social context in which they occur.”  Shrage, supra note
7, at 358. This argument is highly relevant for the debate about who—green, orange, or red segments—
to base the legal regulation of prostitution on.
162 Dempsey & Herring, supra note 160, at 485 (emphasis added;  the  term  “penile sexual
penetration” appearing in the original text substituted for “prostitution”).
163 Id.
164 Id.
165 It can possibly be a matter of discussion whether the social meaning of prostitution disrespects
“women qua women” or “women qua the gender associated with prostitutes” (emphasis added).  See id.
166 They  conclude  that  “a  man  who  penetrates  the  vagina  or  anus of a woman with his penis
commits a prima facie wrong in virtue of the negative social meaning of the conduct.”  Id. at 488.
argue, together with the dissenting judges in State v. Jordan, that it is wrong for
any law to be complicit or “partly constitutive of invidious social standards” and 
“sexual stereotypes which degrade the prostitute.”167
Admittedly, the argument of social meaning would be more acceptable to
sexual-domination proponents. Most sex-work proponents would argue that the
current poor image of prostitution and prostitutes is not inherent in its social
meanings but is rather caused by unequal legal provisions and that if prostitution
was normalized by a permissive legal regime, the stigma would disappear.
2. The Risk of Harm to the Prostitute
As noted earlier, some governments’ arguments for maintaining asymmetric 
provisions against the prostitute are often based on seeing her as a dealer and a
potential source of harm. However, the reverse is true. Empirical research has
shown that there is a serious risk of harm to the prostitute that is not faced by the
client and is, moreover, often inflicted by him.
There is little dispute between the sex-work and sexual-domination
proponents that some prostitutes experience violence.168 However, there is
disagreement about its extent. The sexual-domination proponents draw a
harrowing  picture.    Melissa  Farley  states  that  “in  order  to  keep  the  business  of 
sexual exploitation running smoothly . . . we cannot know that prostitution is
extremely  violent”  and  “that  prostitution,  pornography  and  trafficking  meet  or 
exceed legal definitions of torture.”169 She argues that there is a very high risk of
both physical harm in the form of beating, whipping, rape, other forms of sexual
assault and even murder and psychological harm resulting from the physical abuse
or frequent verbal abuse.170 Many sex-work proponents, however, criticize both
the methodologies and the results of this research. Furthermore, the sex-work
proponents disagree with the sexual-domination proponents’ conceptualization of 
prostitution as inherently constituting violence against womenthey accept the
existence of some “unpaid  for”  violence  but disagree with understanding of
prostitution as a “paid for violence of everyday penetration.”171 For my analysis, it
is sufficient to observe that there is some risk of harm to the prostitute and that it is
greater than that to the client.
3. De Facto Inequalities Between Clients and Prostitutes
Class, age, race, gender, nationality, immigration status and socio-economic
status inequalities exist between the prostitutes and the clients. The socialist
feminist Christine Overall summarizes:
The inherent asymmetrical exchange in sex work, in which some persons sell
sexual services to others, provides the context for other forms of asymmetry, all of
them with important implications for its moral assessment. Prostitution is a classist,
167 State v. Jordan 2002 (6) SA 642 (CC) at 72 (O’Regan, J. & Sachs, J., dissenting).
168 See Bridget Anderson  & Julia O’Connell Davidson, Is Trafficking in Human Beings Demand
Driven? A Multi-Country Pilot Study, IOM MIGRATION RESEARCH SERIES (Dec. 2003)
http://www.compas.ox.ac.uk/fileadmin/files/pdfs/Bridget_Anderson/BA1_Anderson%20IOM%20report
.pdf.
169 Farley, supra note 74, at 122.
170 See id.
171 Jeffreys, supra note 20.
ageist, racist, and sexist industry, in which the disadvantaged sell services to those
who are more privileged. It is classist because for the most part it uses the sex labor
of poor and disadvantaged persons largely for the service of those with disposable
income to spend on sexual gratification. It is ageist because it recruits and preys
upon very young people, often people who are still children, and discards them
when they are past the artificially created stage at which they are considered
sexually attractive. It is racist because it often victimizes black and Asian women
and thrives on race stereotypes of sexually insatiable yet subservient women of
color who exist only to serve the sex needs of whites. Last, and most important, it is
sexist because it is an industry in which, for the most part, women are exploited for
the purpose of serving men’s desires.172
The insight into the class differences in prostitution is also present in the
dissent in the State v. Jordan case:  “We  see  no  reason  why  the  plier  of  sex  for 
money should be treated as more blameworthy than the client. If anything, the fact
that the male customers will generally come from a class that is more economically
powerful might suggest the reverse.”173
These pre-existing structural inequalities and disadvantages can constitute
important external pressures to enter or stay in prostitution. They interact with and
compound individual problems. Former sex worker Amber Hollibaugh writes:
The bottom line for any woman in the sex trades is economics. However a
woman feels when she finally gets into the life [of prostitution], it always begins as
survival - the rent, the kids, the drugs, pregnancy, financing an abortion, running
away from home, being undocumented, having a ‘bad’ reputation, incest - it always
starts at trying to get by.174
Different feminist positions would disagree on defining the primary axis of
disadvantage. Sexual-domination proponents would consider it to be gender.
Marxist and socialist feminists would highlight class and socio-economic status.
Those sex-work proponents who emphasize choice would contest the importance of
these inequalities, but most would not dispute that some vulnerabilities exist.
D. Symmetric Approach
The three points about the context of prostitution need to be taken into
account for an equality analysis. As the dissenting judges argued in State v.
Jordan, they need to be balanced against any suggested justifications brought for
the maintenance of an asymmetric regime benefiting the client. They are also in
their own right justifications for symmetric treatment of client and prostitute.
When making a previously asymmetric regime symmetric, two options are
available: leveling down and leveling up. Leveling down, for example in the area
of public health protection from STDs, would involve not creating an obligation for
individuals to undergo health checks. This is currently the case in the Czech
Republic’s “decriminalization with containment regime” but also under the “Dutch 
model” of legalization where health checks are not required.  Leveling down could 
be considered preferable, as it means less state control and a lower likelihood that
172 Christine Overall, What’s Wrong with Prostitution? Evaluating Sex Work, 17 SIGNS 705, 717
(1992).
173 State v. Jordan 2002 (6) SA 642 (CC) at 68 (O’Regan, J. & Sachs, J. dissenting).
174 Amber Hollibaugh, On the Street Where We Live, 5 WOMEN’S REV. OF BOOKS 4 (1988) (citing
Overall, supra note 172).
the prostitute will face prosecution. Leveling up would involve creating an
equivalent burden for the client, such as a requirement of health checks or condom
use. Creating such an obligation might seem impractical. Should clients also have
health cards and regularly visit the doctor? How could public authorities ever
check for the use of condoms? As argued above, regardless of an enforcement
practicality argument, even an apparently impractical obligation might be worth
creating for its symbolic value. An obligation like condom use could even have
practical empowerment value for the prostitute.
The question of whether to level up or down, in order to achieve equal
treatment, rises in most areas of regulation of prostitution. For example the Czech
2005 contained an asymmetric provision benefiting the client in that it foresaw
administrative and criminal sanctions for the prostitute when selling sex without a
license without a corresponding liability for the client for buying sex from a
prostitute without a license. Leveling up would mean creating symmetric
obligations and sanctions for the client. We have seen, however, that any
criminalization of the prostitute, while unobjectionable from a formal equality
perspective, is criticized by both sex-work and sexual-domination proponents.
Thus, in this case, leveling down, for example by eliminating any administrative
and criminal sanctions, would be the preferable solution.175
Both given examples show that in order to arrive at a solution, the formal
requirement of symmetry might need to be complemented with a substantive
analysis. This Article, in an attempt to present a schema for assessing the legal
treatment of prostitution from a gender equality perspective that transcends the
binary between sex-work and sexual-domination conceptualizations, mostly
avoided substantive questions about the nature of prostitution and the best
legislative solution. When choosing between leveling up and leveling down,
however, this analysis sometimes cannot be avoided. Let us take the example of
obligatory registration of prostitutes that occurs in “decriminalization with control” 
and  “legalization”  regimes.    The  licensing  of  prostitutes  in  the  “Dutch  model” 
without a corresponding registration obligation for the client would be considered
entirely adequate by most sex-work proponents because it would mean that
prostitution is treated like any other trade. A sexual-domination proponent would
abhor it. Under my analysis, which considers the asymmetric regimes benefiting
the client as contrary to the principle of gender equality, the obligatory registration
by prostitutes is unacceptable. A measure creating an obligation with possible
sanctions for the prostitute that benefits the client as a consumer cannot be justified.
This does not mean that my framework is fundamentally incompatible with the sexwork
position. My analysis would be compatible with options for the
prostituteoptional as opposed to obligatory registration and health checks. A
regime that gave the  prostitute  the  full  choice  of  whether  to  enter  the  “green 
segment” or stay in the “yellow segment,” and that offered legalization merely as 
one of possible options, would be considered gender equal.
The requirement of symmetry must not end with statutory provisions but has
to extend to enforcement.  Julie Lefler cites an example of a judge “who refused to 
hear cases involving prostitutes unless the police arrested the customer of the
175 While not entirely achieved, the elimination of such sanctions is the aim of the legalization
“Dutch model.”
prostitute as well.”176 While this example comes from a prohibitionist regime of
the United States and thus concerns criminal law, it could be adopted as practice for
any authority dealing with prostitution in other regimes as well, from local police
enforcing public order provisions to health authorities enforcing public health
provisions.
E. Asymmetric Approach Benefiting the Prostitute
The three points made about prostitutionits social meaning, risk of harm to
prostitutes and de facto inequality between the partiesalso serve as justifications
for an asymmetric regime. An asymmetric measure would be acceptable if it
benefited the prostitute, especially if it decreased her disadvantage and the existing
inequality with the clients, lowered the risk of harm to her, or positively acted
against the existing negative social meaning of prostitution.
While it is foreseeable that the requirement of symmetric treatment of the
prostitute and the client would be tolerable to most feminists, the asymmetric
approach benefiting the prostitute prima facie resonates with the sexual-domination
proponents. A difference can be made between measures benefiting the prostitute
and measures targeting the client. Measures benefiting the client are the
cornerstone of sexual-domination  proponents’  agenda,  and  the  “Swedish  model” 
that criminalizes only the client is a prime example of it; they are, however, not
truly reconcilable with the sex-work position. Measures benefiting the prostitute,
on the other hand, while supported by sexual domination proponents, can be
reconciled with the sex-work position as well. In theory, measures benefiting the
prostitute could be conceived of as protective legislation or affirmative action.
Protective provisions for women, disabled or young workers in labor law reflect
special vulnerabilities of these groups. Affirmative action remedies inequality and
disadvantage based on race and sex. As sex-work proponents conceptualize
prostitution as work, the same protective measures and affirmative action that are
available to workers could be created for prostitutes as well. Such measures would
be paradigmatically reconcilable with the sex-work position. Applied to
prostitution, these provisions specifically benefiting the prostitute could include the
establishment of shelters and other specific social services and welfare benefits, as
well as counseling, legal aid, and reintegration programs such as training and work
placement.177
As far as measures targeting the client are concerned, the following have
been tried or proposed in the context of the United States: car forfeiture or
revocation of driver’s licenses when clients are caught curb-crawling, publication
of names of clients in various media, educational programs for clients, and the
institution of civil causes of action for prostitutes against clients and procurers.178
In “decriminalization” regimes, the asymmetric approach might imply addressing 
all legislative goals through the client and not the prostitute. To continue using the
example of public health and the prevention of STDs, creating an obligation
176 Lefler, supra note 124, at 21.
177 The ideological basis for these measures would remain different. Sexual-domination
abolitionists would support programs that enable the prostitute to temporarily or permanently leave
prostitution. For sex-work, they would be meant to improve conditions of prostitution and empower
prostitutes.
178 For an analysis of these measures or proposals, see Lefler, supra note 124, at 26-34.
specifically and only for the client could, for example, take the form of obligatory
use of condoms by clients with liability for non-compliance. Similarly, public
order issues could be addressed with anti-curb-crawling measures without an
equivalent liability for soliciting by the prostitutes.
CONCLUSION
This Article presented a framework for testing the compatibility of legal
responses to prostitution with the principle of gender equality with the aim to offer
a tool for improvement of the wellbeing of prostitutes. The proposed analysis was
based on the comparison of the treatment of the prostitute and the client. It argued
that an asymmetric preferential treatment of the client, which is still common in
many countries in criminal but also administrative, police and health law, as well as
in enforcement practice, is in breach of the principle of gender equality. It
proposed that the principle of gender equality requires at a minimum symmetric
equal treatment of both parties. This can either be achieved by leveling
upcreating a restriction or obligation for bothor leveling downabolishing it
for both. It also submitted that an asymmetric approach benefiting the prostitute or
targeting the client can be justified.
The Article identified three aspects of the phenomenon of prostitution which
are important to the gender equality analysis: 1) the negative social meaning of
prostitution, 2) the risk of harm for the prostitute, and 3) the de facto class, age,
race, gender, nationality, immigration status and socio-economic status and other
inequalities between the client and the prostitute. These describe the context of
prostitution and need to be balanced against any justification brought for the
maintenance of an asymmetric regime benefiting the client. They are also in their
own right justifications for symmetric treatment of client and prostitute and even
for an asymmetric treatment benefiting the prostitute.
There is much feminist writing on the issue of prostitution. The literature,
however, has been divided between those who see prostitution as sex work that
should be normalized, and those who see it as an inherently violent and exploitative
practice that should be abolished. This Article transcends the binary. First, it is
primarily concerned with what should not be the legal response to prostitution—an
asymmetric treatment harsher on the prostitute—rather than with proposing an
ultimate best legal regime. This critique of regimes repressive toward the prostitute
resonates with both feminist positions. A legal gender equality analysis based on
the comparison with the client has rarely been used, however. Second, when the
Article explores the proposed scenarios—symmetric treatment of prostitutes and
clients and asymmetric treatment benefiting the prostitute—, it discusses the
perspectives of the two feminist positions and incorporates both of their insights:
agency, as argued by the sex-work proponents, as well as vulnerabilities and
violence, as argued by radical feminists. The analysis presented, it is hoped, could
improve gender equality and the wellbeing of prostitutes in applying existing legal
frameworks, in adopting their changes and amendments, and in their judicial
review.