Walking Back Dissents: A Reply to Helfer and Voeten

We see no need to respond to Helfer and Voeten’s reply (Stone Sweet, Sandholtz, and Andenas, 2021), beyond clarifying one major point. The basis of our critique of the Helfer and Voeten article (2020) concerns our inability to confirm the classifications reported in their data. Our focus is on those cases that are most likely to fit Helfer and Voeten’s expectations, given their own theory.

We sought to verify Helfer and Voeten’s classifications for all 23 Grand Chamber judgments and dissents that (1) involved qualified rights (rights enumerated in Articles 8, 9, 10, and 11 ECHR and ECHR Protocol 1) and (2) triggered ‘consensus analysis’, on the part of the Grand Chamber, as one determinant of the size of ‘the margin of appreciation’ to be afforded the respondent state. Again, these are the cases most likely to fit Helfer and Voeten’s own theory, as they describe it. The vast bulk of these dissenting opinions and rulings do not fit Helfer and Voeten’s expectations; we found that the majority of these cases has been wrongly coded by the students to whom Helfer and Voeten had tasked the analysis. We believe that the fact that three independent scholars could not confirm Helfer and Voeten’s main findings, but instead found myriad errors, necessitates rejecting of Helfer and Voeten’s conclusions.

To repeat, we read and analyzed the 23 Grand Chamber judgments and dissents that addressed qualified rights (as detailed above), decisions that Helfer and Voeten claim to contain ‘walking-back dissents’. These 23 judgments concern considerations that are at the core of Helfer and Voeten’s thesis: evolutive interpretation, margin of appreciation, and consensus analysis. The simple rules that guided our recoding are available in the Appendices to our article. These rules simply operationalize Helfer and Voeten’s own arguments; they do not substitute for Helfer and Voeten’s priorities. To be counted as a ‘walking-back judgment,’ a Grand Chamber ruling must: (i) breach an existing precedent, (ii) restrict a qualified right, and/or (iii) expand the regulatory authority of state officials to restrict the scope of a qualified right. This coding protocol thus reflects Helfer and Voeten’s discussion of their own theory.

In our analysis, we found that only five of the 23 judgments contained – arguably – a ‘walking-back dissent’ (that is, at least two out of three of us deemed a dissent as having made a claim to the effect that that the Grand Chamber had ‘walked back’ rights). In the crucial period of 2012 – 2018 (Period 3), we found that only four of 17 cases included a ‘walking-back’ dissent. We also analyzed the 23 judgments of the Grand Chamber, given that these rulings would provide more direct evidence on ‘walking back’ rights. Of the 23 judgments we analyzed, we were only able to identify one as a ‘walking-back judgment’ (S.A.S. v. France, with two ‘yes’ votes against one ‘no’). The only other Grand Chamber judgment in the set of cases that came close was Scoppola v. Italy, which received two ‘no – tossup’ votes and one ‘no walking-back judgment’ vote. In short, we were unable to confirm Helfer and Voeten’s thesis; indeed, the data weaken their thesis, fatally in our view.

We concluded by inviting readers to evaluate our coding of the ECtHR judgments against that of Helfer and Voeten’s students, and we have summarized how and why we coded cases as we did in Appendices. We are confident that Helfer and Voeten’s findings are wrong, and that their conclusions must be rejected.