There are many reasons for putting rights in constitutions. Rights protections may serve important symbolic purposes, like expressing national identity and articulating the nation’s highest ideals. Relatedly, recognizing rights in a country’s founding document may be important for moral reasons. In one prominent articulation of this argument, Alon Harel contends that, even if they do not have instrumental value, including rights in constitutions is morally significant because it recognizes that governments have a duty to respect them.[1]
Rights might also be included in constitutions for instrumental reasons. For instance, including constitutional rights can help parties move forward in constitutional negotiations. In the U.S., for example, the addition of the Bill of Rights has been compared to “throwing a tub to a whale” because just as sailors would sometimes throw an old tub (or barrel) to whales to keep them entertained and distracted from harming ships, the federalists agreed to the inclusion of a bill of rights to distract antifederalists from pursuing amendments that would harm the Constitution.[2]
But one would imagine that the primary reason why countries enshrine rights in their constitution is that they expect that they will help improve the protection of those rights in practice. For the reasons we described in an earlier post, however, it’s an open question whether that actually happens. And, unfortunately, figuring out the answer to that question is not a straightforward task.
This is because the question is not simply whether countries with constitutional rights do better than countries without them. Instead, the question is whether countries with the right are better off than they would have been without these rights (and whether countries without them would have been better off with them). Answering such counterfactual questions empirically is vexing for a number of reasons: constitutional rights are not randomly assigned to countries, the protection of human rights is hard to measure, and there are many country-level traits—like wealth or age—that make pinpointing cause and effect difficult.
In fact, some scholars have said that answering these questions empirically is impossible, and as a result, it’s a mistake to try to offer an answer. For example, Holger Spamann has argued that “Comparative evidence alone will hardly ever be sufficient to establish a causal claim and that statistical methods that purport to do so are likely to do more harm than good in comparative settings.”[3]
We disagree. Although sorting out this kind of causal claim is particularly vexing, we think the question is too important to leave unanswered, simply because there is no single method available that can answer it with certainty. In our new book on the effectives of constitutional rights, instead of letting the perfect be the enemy of good, our approach is to use a range of methods to study this topic. We specifically conducted large-N analysis using our dataset that includes detailed information on a large number of rights from 1946 to 2016 for all 194 widely recognized countries in the international state system. We conducted case studies in countries where constitutional rights recently changed or where constitutional rights recently came under stress. And we conducted survey experiments that directly tested mechanisms that have been hypothesized as improving compliance with human rights obligations.
For our large-N analysis, our goal is to be as transparent as possible. As a result, we always start by simply graphing the raw data, and then walking through as we add additional elements to the analysis. We’ll say more about what exactly we did in our next post, but our concern is that some of the empirical literature on human rights can feel like a black box. The researchers make arguments about the “best” way to test the relationship between human rights and rights outcomes, plug their data into a complicated set of models, and then present numbers consistent with their theories. In contrast, our goal is to focus on showing the data, and estimating a wide range of different models that are plausible ways to test these relationships. And, by doing so, allow readers to see the relationships between rights protections and actual rights protections themselves.
For these case studies, we conducted a case study on union rights in Tunisia, on political parties in Myanmar, on religious freedom in Russia, on free speech in Poland, and on the right to healthcare in Colombia. We travelled to each of these countries and conducted semi-structured interviews with people like judges, political leaders, religious leaders, civil society groups, unionists, ombudsmen, practicing lawyers, and others with knowledge of constitutional rights enforcement. In total, we interviewed over a hundred people about constitutional rights enforcement.
The logic behind conducting the case studies is that they help probe the mechanisms we theorized as driving rights effectiveness. For example, a large-N research design can tell us that countries with the right to unionize in their constitution have higher respect for worker’s rights on average, but it does not tell us why.
For the survey experiments, we conducted two different experiments in Turkey and the United States. For example, in Turkey, we employed a survey firm that conducted 1,335 face-to-face interviews with a representative sample of the population. In this survey, we asked people about their views on the constitution as it related to the Turkish government’s Wikipedia ban (we’ll write more about this next week). In the United States, we asked people about a hypothetical terror threat, and whether people would be okay with torture even when doing so violates the Constitution.
The logic behind conducting survey experiments with the general public is that they make it possible to test specific theories on why constitutional rights may make a difference: those that focus on the general public as a compliance mechanisms. This is why survey experiments are popular in the literature on the effectiveness of international human rights treaties, as much of it focuses on democratic accountability as the mechanism treaties may improve rights protections.
Our hope is that using a range of methods will allow us to triangulate onto a more accurate understanding of the effect of constitutional rights than any single method could provide. In doing so, our approach is best understood as being part of the growing trend of Mixed Methods Research (MMR) in the social sciences. The advantage of MMR is that, when “[d]one well, multi-method research combines the strength of large-N designs for identifying empirical regularities and patterns, and the strength of case studies for revealing the causal mechanisms that give rise to political outcomes of interest.”[4] Our approach, then, reflects Ran Hirschl’s statement that “there is no magic bullet or one-size-fits-all research design ‘formula’ for a field as rich and diverse as comparative constitutional studies.”[5]
We know that this discussion of methods is a little abstract, but in the next posts, we’ll explain what we learned using these different methods to study if constitutional rights work.
Note: Our book, How Constitutional Rights Matter, was recently published by Oxford University Press. This is the third in a series of blog posts on parts of the argument and evidence from the book.
-----
[1] ALON HAREL, WHY LAW MATTERS 7 (2014).
[2] Kenneth R. Bowling, ‘A Tub to the Whale’: The Founding Fathers and Adoption of the Federal Bill of Rights, 8 J. Early Republic 233 (1988).
[3] Holger Spamann, Empirical Comparative Law, 11 Ann. Rev. L. & Soc. Sci. 131, 138 (2015).
[4] James D. Fearon & David D. Laitin, Integrating Qualitative and Quantitative Methods, in THE OXFORD HANDBOOK OF POLITICAL METHODOLOGY 758 (Janet M. Box-Steffensmeier et al. eds., 2008).
[5] RAN HIRSCHL, COMPARATIVE MATTERS: THE RENAISSANCE OF COMPARATIVE CONSTITUTIONAL LAW (2014).