National Human Rights Institutions: Domestic implementation of international human rights law

ABSTRACT

Does adopting a National Human Rights Institution (NHRI) make states’ international commitments to not torture more constraining? Many researchers have explored international human rights treaties’ abilities to constrain leaders from violating human rights, some focusing exclusively on the United Nations Convention Against Torture (CAT). Thus far, findings are not promising unless certain domestic conditions apply such as sufficient democratic space to air grievances or independent judiciaries. This article continues to explore domestic conditions by focusing on another liberal institution—National Human Rights Institutions (NHRIs). Torture is usually a secretive practice, and NHRIs act as information providers to potential mobilizers and domestic legal systems assuring international legal commitments are not empty promises. Using statistical analysis on 153 countries over the years 1981–2007, I find that when a country has ratified the CAT, the presence of an NHRI substantively decreases the chances the state will be an egregious offender.

•  

Acknowledgments

I would like to thank Will H. Moore, William Berry, Courtenay Conrad, Sydney Gann, Brad Gomez, Danny Hill, Scott Meachum, Gina Miller, Baekkwan Park, Megan Shannon, Teresa Stanquist, Saadet Ulasoglu, and Rachel Wayne for helpful comments.

Notes on contributor

Ryan M. Welch is a PhD candidate in Political Science at Florida State University. His current research interests include human rights, international law, domestic institutions, political violence, and protest.

Notes

1. I define torture as does the CAT: “Any act by which severe pain or suffering, whether physical or mental, is intentionally inflicted on a person for such purposes as obtaining from him or a third person information or a confession, punishing him for an act he or a third person has committed or is suspected of having committed, or intimidating or coercing him or a third person, or for any reason based on discrimination of any kind, when such pain or suffering is inflicted by or at the instigation of or with the consent or acquiescence of a public official or other person acting in an official capacity” (United Nations 1984, Art. 1 Sec. 1).

2. Cole and Ramirez (2013), the only other quantitative piece on NHRI effectiveness known to this author, describe NHRIs as a consequence of the global polity’s concern with certain rights (physical integrity) over time.

3. Although Rejali (2009) reminds us that, despite having the reputation for innovative sadism, Nazi Germany borrowed many of its torture techniques from other cultures (e.g., electrotorture from the French).

4. However, see Fariss (2014) for the argument that these depressing results are due to the changing standard of accountability influencing the way in which we measure torture.

5. Including the CAT, American Torture Convention, and the African Charter.

6. Some specifications in Hathaway (2002) show no statistical relationship between ratification and practice.

7. However, see Conrad and Moore (2012) for an argument that elections may not lead to a decrease in torture because marginalized groups (not a priority of the majority) are often targeted.

8. Monists treat international and domestic law as one legal system with international law occupying a hierarchically superior position. Dualists treat domestic and international law as two separate spheres whereby international law is applied domestically to the extent that it is incorporated into the domestic law (von Glahn and Taulbee 2013: 118). For a more comprehensive review of monism and dualism, see Marian (2007).

9. The Paris Principles were adopted in by the United Nations General Assembly in 1993.

10. Studies, such as Cole and Ramirez (2013), suggest there is reason to study them as one homogenous group: “All types of human rights institutions improve long–term physical integrity outcomes but not civil and political rights practices” (1, emphasis added). As noted in the conclusion, future study should explore further what aspects of NHRIs are most effective at curbing rights violations and when.

11. Approximately 8,000 USD.

12. See Simmons (1998) for a helpful review that is but a starting point.

13. Although the Paris Principles call for independent NHRIs, a number of case studies show independence is not a binary concept and some NHRIs are more or less independent than others (e.g., Mertus 2009). Skepticism about NHRI independence (e.g., Rosenblum 2012) actually provides for a conservative test of my hypothesis, as I am looking for the average effect and those NHRIs that are less independent would dampen the overall effect.

14. Note the tests in this study do not directly address Optional Protocol to the United Nations Convention against Torture (OPCAT). The temporal range of the data go to 2007, and OPCAT did not go into force until June 22, 2006. Future studies should look further into OPCATs relationship with NHRIs.

15. This could be through a continuum of mechanisms from democratic elections to revolution. The most important point is domestic mobilization can threaten the executive’s office.

16. The global human rights polity includes international organizations, INGOs, states concerned with human rights, and companies concerned with human rights.

17. See Murdie (2014) for an interesting argument suggesting NGOs should not be treated as monolithic organizations but rather can be grouped according to their motivations.

18. The Polity IV researchers suggest democracies defined at Polity between 6 and 10, anocracies between −5 and 5, and autocracies at −10 to −6 (Marshall, Gurr, and Jaggers 2012). I thank an anonymous reviewer for the suggestion of dummies for democracy and anocracy. I have also used the Polity scale for democracy and a dummy variable for transitioning states with almost identical results, but the current specification makes for a more logical model fit.

19. Recall this would result in an increase in the dependent variable as it measures protection of rights.

20. Autocracy (Polity −10 to −6) is the referent category and is not included in the model.

21. Energy consumption is a more reliable and available measure than GDP, and the two are highly correlated, thus capturing the same concept (Jackman 1973).

22. Clague, Keefer, Knack, and Olson (1999), Feld and Voigt (2003), Howard and Carey (2004), Tate and Keith (2009), Cingranelli and Richards (2010a), Marshall and Jaggers (2010), Rìos-Figueroa and Staton (2010).

23. Three of the variables do not meet the assumption of parallel slopes (CAT ratification, International War, and Lagged Torture), but the overall model meets the assumption. Fitting a partial proportional odds model does not change the beta coefficient or standard error on the explanatory variable or its constituents in a substantial way; therefore, the results are the same and not model dependent. The results for the partial proportional odds model can be found in the online appendix.

24. Another way to deal with autocorrelation is the use of the Prais-Winston transformation (Plumper, Troeger, and Manow 2005). Estimating the Prais-Winston does not require a lagged dependent variable (which absorbs much of the time-series variance), so statistical results are “easier” to find, making my model choice a more conservative test. Results are similar and presented in the online appendix.

25. I used STATA 12 for statistical analyses (StataCorp 2011).

26. I set dichotomous variables to their medians and continuous variables to their means. The result is an average country that is slightly democratic (polity score of 1.68), is not at war (civil or international), has an economic standing on par with Brazil in 2006, has a population of around 11 million and has a judiciary with an independence score of 0.452.

27. This is the appropriate first difference to test my hypothesis: however, in full disclosure, the first difference that takes a country with an established NHRI and changes the state from a CAT nonratifier to ratifier yields opposite results. Not only does the first difference presented answer my question but it is more appropriate given that most states ratify the CAT before they establish an NHRI (modal CAT ratification year is 1987, modal NHRI adoption year is 1996). The disparate first difference does present a puzzle about timing of institutional adoption worth exploring in future work.

28. Throughout I assume movement from adjacent categories. It is possible that countries go from 0 to 2, and vice versa, but unlikely.

29. Judicial independence has the highest substantive effect on torture practices in this model. Despite that fact, knowing about NHRI effectiveness for states that commit to international agreements is worthwile.

30. Some may suggest a matching algorithm to preprocess the data in order to simulate experimental research design (e.g., Simmons and Hopkins 2005; D. W. Hill 2010). However, this method is unusually challenging given the explanatory variable of interest is an interaction term of two dichotomous constituent terms in the model. I leave this challenge to future research.

31. I present the results in the online appendix.

32. Cole and Ramirez (2013) begin to disaggregate by broadly grouping NHRIs based on their organizational structure and mandate (classic ombudsman, human rights commissions, and human rights ombudsman).

33. If not after investing in steps to make the judiciary independent given its large substantive effects in the model.