Political Gender Equality and State Human Rights Abuse
| Published date | 01 March 2005 |
| Author | Erik Melander |
| Date | 01 March 2005 |
| DOI | 10.1177/0022343305050688 |
| Subject Matter | Articles |
149
Introduction
Feminist theorists argue that more equal
societies that are not based on gender hierar-
chies (or other hierarchies such as race, caste,
or ethnicity, for that matter) ought to be less
plagued by collective violence internally as
well as in relation to other societies. So far,
these theories have been subjected to no
more than a handful of empirical tests. Most
significantly, three recent studies indicate
that states with more gender equality are less
bellicose in their relations with other states,
as theoretically predicted (Caprioli, 2000;
Caprioli & Boyer, 2001; Regan & Paskevi-
ciute, 2003). To my knowledge, no quanti-
tative study has been published on the
relationship between gender equality and
how violently the state behaves within its
own borders.
The purpose of this study is to test
whether political gender equality is associ-
ated with lower levels of personal integrity
rights abuse carried out by state agents, such
as fewer political imprisonments, torture,
killings, and disappearances. My two indi-
cators of gender equality are (1) a dummy
indicating that the chief executive is a
woman; and (2) the percentage of women in
parliament. The foremost contribution of
this study is to subject a central tenet of
gender theory to a systematic empirical test.
© 2005 Journal of Peace Research,
vol. 42, no. 2, 2005, pp. 149–166
Sage Publications (London, Thousand Oaks, CA
and New Delhi) www.sagepublications.com
DOI 10.1177/0022343305050688
Political Gender Equality and State Human Rights
Abuse*
ERIK MELANDER
Department of Peace and Conflict Research, Uppsala University
Feminist theorists argue that more equal societies that are not based on gender hierarchies ought to be
less plagued by collective violence. This study tests whether political gender equality is associated with
lower levels of personal integrity rights abuse carried out by state agents, such as fewer political impris-
onments, torture, killings, and disappearances. Two indicators of political gender equality are used: (1)
a dummy indicating that the chief executive of a state is a woman; and (2) the percentage of women
in parliament. The impact of political gender equality on personal integrity rights abuse is tested using
multiple regression techniques and a dataset spanning most countries of the world during the period
1977–96. Female chief executives are rare, and their tenures are not significantly associated with the
level of abuse. The percentage of women in parliament is associated with lower levels of personal
integrity rights abuse. Results show both a direct effect of female representation in parliament and an
effect in interaction with the level of institutional democracy. These results hold when controlling for
the most important factors known or suspected to influence human rights behavior: democracy, leftist
regime, military regime, British colonial experience, civil war, international war, wealth, population,
ethnic heterogeneity, and regime transition and collapse.
* The data used in this article can be found at
http://www.pcr.uu.se/personal/anstallda/melander.htm.
All statistical results were generated using Stata 8.2. I would
like to thank Mats Hammarström, Magnus Öberg, Håvard
Hegre, and three reviewers for helpful comments, and
Linda Camp Keith for providing data. Please address ques-
tions and comments to erik.melander@pcr.uu.se.
02 melander (ds) 1/2/05 1:47 pm Page 149
At the same time, this appraisal of the
impact of political gender equality adds to
our understanding of state human rights
behavior.
The present study builds upon a series of
earlier quantitative studies of state human
rights behavior that use similar dependent
variables and have identified a set of
explanatory factors that consistently influ-
ence the level of personal integrity rights
abuse (Keith, 1999, 2002; Poe & Tate,
1994; Poe, Tate & Keith, 1999; Poe, Carey
& Vasquez, 2001; Zanger, 2000). I con-
struct a dataset based on the latest and most
comprehensive data available on state viola-
tions of personal integrity rights (Keith,
2002).
Female chief executives are rare, and their
tenures are not significantly associated with
the level of personal integrity rights abuse. In
contrast, the percentage of women in parlia-
ment is, as expected, associated with lower
levels of personal integrity rights abuse.
Results show both a direct effect of female
representation in parliament and an effect in
interaction with the level of institutional
democracy. These results hold when control-
ling for the most important factors known
or suspected to influence human rights
behavior: democracy, leftist regime, military
regime, British colonial experience, civil war,
international war, wealth, population, ethnic
heterogeneity, and regime transition and
collapse.
I begin the theoretical section by clarify-
ing what human rights are dealt with in this
study. Next, I discuss mechanisms through
which political gender equality conceivably
could affect state human rights behavior. In
this connection, I formulate my hypotheses.
I then review previous empirical findings on
the determinants of state human rights
behavior and introduce my control variables.
The subsequent section deals with research
design issues. Finally, I present the results
and my conclusions.
How Gender Equality Influences
State Human Rights Behavior
I will examine the variation in a subset of
human rights dealing with the integrity of
the person and ask what explains the extent
to which state agents violate these personal
integrity rights as a way of inducing compli-
ance among the population. Violations con-
sidered include politically motivated torture,
imprisonment, killings, and forced dis-
appearances. It has been argued that this
approach considers the core human rights
and that the violation of these rights consti-
tutes ‘the most egregious and severe crime
against humanity’ at the same time as these
violations ‘are of a sort that usually can be
avoided’ (Poe & Tate, 1994: 854).
The dependent variable, then, is the
extent to which states violate their own
citizens’ personal integrity rights through
political imprisonment, torture, killings, and
disappearances. Following Keith (2002),
who coded this variable and generously let
me use her data, I will refer to the dependent
variable as the level of personal integrity
rights abuse. In turn, Keith builds on an
expanding dataset on personal integrity
rights abuse that has been widely used and
gradually refined throughout several studies
(Keith, 1999, 2002; Poe & Tate, 1994; Poe,
Tate & Keith, 1999; Walker & Poe, 2002;
Zanger, 2000). The level of personal
integrity rights abuse is measured on a scale
of 0 to 4, with 0 representing states with no
or insignificant levels of abuse and 4 repre-
senting states with the worst abuses.1The
journal of PEACE RESEARCH volume 42 / number 2 / march 2005
150
1The original scale ranges from 1 to 5. The following rules
laid down by Gastil (1980, quoted in Keith, 2002: 117f.,
fn5) guide how a country is rated:
(1) Countries [are] under a secure rule of law, people are not
imprisoned for their views, and torture is rare or exceptional. . . .
Political murders are extremely rare.
(2) There is a limited amount of imprisonment for nonviolent
activity. However, few persons are affected, torture and beating
are exceptional. . .. Political murder is rare.
(3) There is extensive political imprisonment, or a recent history
of such imprisonment. Execution or other political murders and
02 melander (ds) 1/2/05 1:47 pm Page 150
To continue reading
Request your trial