![Sample our Law journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5942/TOC-Subject-Sample-2021-Law.jpg"})
Abstract
An expanding body of literature maintains that gender inequality heightens the probability of intrastate conflict by creating a structure of violence. The paper proposes the legal system as the missing link between social norms and conflict occurrence. Gender neutrality of the penal code coheres with norms of equality and, further, embodies egalitarian, progressive values associated with pacific norms of conflict resolution. The criminalization of rape enhances female empowerment by offering a legal commitment on the part of the state to safeguard women’s physical security. More broadly, legal prohibitions against rape protect women and other vulnerable individuals from sexual aggression. The statistical analysis uses novel data on rape legislation for 194 states over the 1965 to 2005 time period. The length of punitive sentence proxies for the stringency of rape legislation. The empirical findings demonstrate that longer punitive sentences against rape crimes are associated with a significantly lower probability of intrastate conflict. Sanctions against both female and male perpetrators of rape are analyzed separately. The results show that gender neutrality of law whereby the penal code establishes similar sentences for female and male offenders alike also significantly decreases conflict propensity.
Notes
1. One caveat is that this phenomenon may be tied predominantly to Western values and reserved for advanced democracies. Nevertheless, our data shows considerable variation in rape law both within advanced democracies and across regions.
2. A cynical perspective would be to state that states simply adopt measures to curry favor with the international community. However, as Fearon (Citation2010) points out, if governments institute measures increasing gender equality as a result of international pressure or inducements, this would not necessarily reduce the likelihood of civil conflict. Put differently, if in fact countries enact egalitarian penal code changes as a mere chimera, our coefficients would be biased upwards, making it more difficult to uncover the hypothesized reductive effect.
3. We do not argue that punitive measures eradicate rape; rather we claim that formal law has an impact on conflict regardless of its actual deterrent impact on violence against women.
4. Rape prevalence data is spotty and unreliable; McDermott’s rape conviction data on rapes per 100,000 population is only available for 2007. Related rape indicators in WomanStats are coded for 2011.
5. We utilize the Uppsala/PRIO country level data file conflict onset. The country-year file includes several predefined cutoff points of one, two, five and eight years of peace before recording a new dispute. We utilize the first specification although the results are robust to the latter two specifications.
6. In supplementary models, we found that the criminalization of rape operationalized as a binary measure fails to significantly reduce the probability of domestic conflict. This strengthens our argument that longer sentences meant to deter sexual violence are substantively important whereas mere criminalization, when not backed by significant penalties, does not exert the same impact.
7. In the source project, information prior to 1965 is also included. However, the quantitative indicators pertain to the post-1965 period.
8. As sensitivity checks, we examine all our main models with the –10 to 10 revised Polity score; results remained unaltered.
9. Our raw data does not distinguish punishments according to victim’s sex. Nonetheless, we do not regard this as a shortcoming of our analysis; we would expect societies that embrace gender neutrality in rape legislation to prohibit rape regardless of the victim’s sex.
10. More formally, the likelihood ratio test of rho returns a significant test statistic, indicating that the panel estimator is significantly different from the pooled estimator.
11. As sensitivity analysis, we also used the raw numbers for fertility rate and percentage measure for labor participation; the coefficients were in line with the results in but remained insignificant. Results are available upon request.
12. In fact, in separate models we look at ongoing civil war as the dependent variable. Consistent with prior literature, we find that women’s participation in the parliament dampens the probability of civil war.
13. Rape sanctions against males and females are correlated at the 0.45 level, which is not prohibitively high to introduce multicollinearity. Similarly the gender equality indicators showed a correlation coefficient of maximum –0.28.
14. Additionally, we separately compared model fit in models that incorporate rape sanctions against males and females, with models including one of the macro indicators of gender equality. In all these cases, the same pattern held whereby the AIC and BIC values were lower in the case of models that only include rape laws.