A. The Case for an Empirical Analysis

Accounts of the use of scholarship by international courts tend to be anecdotal and limited to specific fora. Although the usefulness of in-depth qualitative studies cannot be doubted, such an approach generally provides no useful insights into overall patterns or trends. Accordingly, the authors of this study chose to rely on a large-scale empirical analysis of all citations to scholarly sources in the judgments adopted by the Grand Chamber of the ECtHR.

There is, of course, an obvious difficulty, which should be addressed at the outset and may be enunciated as follows: scholarship can be used in many ways, but not all of them are overt and visible. It is acknowledged that it is entirely possible—and, in some cases, even likely—that Judges may rely on a scholarly source without ever acknowledging having done so. But, it is also possible—and, as the authors found, relatively common—for some Judges to decide to rely on scholarly authorities and cite the source, thereby acknowledging their importance for their reasoning and embedding them in the flow of their arguments in order to provide support, interpret or persuade. For this to occur, the authority needs to be mentioned overtly, and the citations themselves can generate rich data.

Inquiries based on this type of data, known as citation analyses, have a long history. Citation analysis has been employed successfully in a variety of fields, including law, with some pioneering studies dating back to the 1950s. In one such study on the sources and authorities cited by the California Supreme Court, Merryman highlighted the promises and perils of the approach, recognising that while the citation of authority ‘is often an uncritical unreflective process’,Footnote ²³ ‘[p]resumably a citation means something to the person citing, and presumably he anticipates that it will have some meaning to a reader’.Footnote ²⁴ Yet, as Posner has suggested, while it is possible for a citation to be meaningless in principle, it is unlikely in practice, ‘if only because citing is not costless—there is the bother of finding the citation, and the possibility of criticism for misciting or failing to cite’.Footnote ²⁵ To be sure, absence of evidence does not amount to evidence of absence, for international adjudicators tend to err on the side of caution when it comes to including specific references to scholarship.Footnote ²⁶ Sometimes a connection may be drawn between a particular argument or turn of phrase and a prominent piece of scholarly writing. Yet, an approach of this kind would only be suited to the identification of the influence of works which have been known to be influential, either because they have been acknowledged by relevant actors to be so, or because they provided an unmistakable approach to a given issue. In all other cases, an approach of this kind would require a significant expenditure of resources, and would not be guaranteed to yield better outcomes in light of the several other challenges that might be faced in tracing arguments back to their unsourced antecedents.

This study embraces citation analysis as the most robust approach, operating on the assumption that the inclusion of a citation represents an overt acknowledgement of the relevance of scholarship by the Court or by the Judge or Judges authoring a separate opinion. The authors acknowledged and mitigated the possible shortcomings of this technique by building as large and feature-rich a dataset of citations of Grand Chamber judgments as possible, thereby avoiding reliance merely on the fact of a citation being included, but also incorporating context and additional metadata, as discussed in the following sections.

B. The Sample

C. The Data Collection

The methodology for coding the judgments of the ECtHR involved the identification of all the citations to scholarship (broadly construed) in all of the 486 judgments adopted by the Grand Chamber up to September 2021. Four research assistantsFootnote ⁴³ were provided with a list of variables to include in the data, including the name of the case as it appears on the ECtHR's HUDOC database,Footnote ⁴⁴ the year and date of the judgment, the application number, the date and year of the authority being cited, the age of the citation as determined by the difference between the year of the judgment and the year of publication of the source, the simplified name of the authority being cited, the number of the paragraph where the citation appears, the text of the citing paragraph, and various topics and issues related to the case. The research assistants were asked to indicate whether the citation appeared in the majority judgment or a separate opinion, the respondent State, the type of authority being cited, and whether the paragraph where the citation appears also cited Strasbourg case law or domestic law. Finally, the research assistants were asked to provide the names and nationalities of the Judges authoring the minority opinion, if applicable. This coding was complemented by creating a Boolean variable for each Judge involved in any decision,Footnote ⁴⁵ and for each ECHR Article that could potentially be dealt with by the judgment so as to be able to map the data more accurately. As discussed in Section IV, by leveraging the text of the paragraphs including citations extracted by the research assistants, a topic model was also created to classify the citations further.

The work of the research assistants was checked and complemented by the authors who moderated samples of the judgments, and a few missing citations were added to the database through this process. Some judgments were analysed by two different research assistants to ensure consistency in coding. Despite this approach, the possibility of human error cannot be ruled out. It is conceivable that some citations were missed or misrepresented. Moreover, it was not possible to check the accuracy via automated searches given the lack of a consistent format for quoting scholarship, as discussed in Section IV. These potential issues do not, however, undermine the findings as the number of possible missed citations is minimal and statistically insignificant. In total, 1,427 instances of citation of scholarship were counted across 592 paragraphs in 122 cases.

A. Overall Use of Scholarship

The data shows that the ECtHR has regularly cited scholarship in its majority judgments and separate opinions. It is possible to appreciate an overall increase across time, with some fluctuation in the number of judgments citing scholarship each year. In general, citations are far more frequent in separate opinions than in majority judgments. For example, in 2014, there were 99 scholarship citations in separate opinions compared to just 8 instances in majority judgments, while in 2017, there were 362 citations in separate opinions, but none in majority judgments.Footnote ⁴⁶ Overall, the rate at which the ECtHR cites scholarship in its judgments and opinions seems to vary from year to year, but with a clear preference for references to scholarship in separate opinions, as is shown clearly in Figure 1.

Figure 1: Increase in the number of citations (logarithmic scale)

In terms of distribution, the average number per year of unique scholarly works referenced in a judgment citing scholarship (irrespective of whether the citation is included by the majority or in a separate opinion) is 10.4, though the median number is only 3. This pattern reveals that some judgments have been clear outliers in terms of including a large number of references. For example, G.I.E.M. S.r.l. and others v Italy cites 148 unique scholarly authorities, with 139 being cited by Judge Pinto de Albuquerque and 11 by Judge Motoc (with two authorities cited by both).Footnote ⁴⁷ A similar pattern may be observed in Ilnseher v Germany, which includes references to 100 unique scholarly authorities in the opinion of Judge Pinto de Albuquerque, joined by Judge Dedov.Footnote ⁴⁸ These anomalies reveal that the practice of including abundant references to scholarly authorities is far from widespread, and indeed these two instances represent almost 20 per cent of the authorities included in our dataset.Footnote ⁴⁹ The presence of these exceptions also explains why there appears to be little correlation between the number of citations included in a judgment and the number (rather than the presence) of separate opinions.

Only 30 cases contain citations of scholarship by the majority, with a grand total of 80 unique authorities cited. Here, the distribution of citations is not as uneven, but it remains difficult to interpret in light of the type of works cited. For example, the majority judgment in Jalloh v Germany contains the highest number of citations, but only if it is accepted that ‘case notes discussing domestic case law’ should be counted as scholarship.Footnote ⁵⁰ Other judgments, however, cite sources that may be unequivocally qualified as scholarly. This was the case, for example, in Georgia v Russia (II),Footnote ⁵¹ where several works concerning the international law of military occupation were cited, and in Naït-Liman v Switzerland, which cited several works on torture and universal jurisdiction.Footnote ⁵²

Despite some examples, the citation of scholarship by the majority remains a comparatively rare practice, with less than 7 per cent of all majority judgments containing any references to it. Thus, beyond a slight increase over time in the number of works cited annually, it is difficult to discern any other trends.

B. Judges

As noted in the previous section, there is a significant variation between Judges in the number of citations that they tend to include in their separate opinions. An accurate calculation of these citations was not entirely devoid of difficulties, as the attribution of individual opinions to a specific Judge was complicated by the conventions applying to the titling of the opinion itself in the Court's practice—contrasting, for example, the citations made in sole-authored opinions with those in joint ones, or those ostensibly authored by just one Judge to which others joined. Thus, to calculate the number of references attributable to a Judge accurately, two different metrics were considered. First, all of the references made in an opinion where the Judge was named in the title of the opinion were counted, irrespective of whether they were leading or simply joining the minority. Second, all of the instances were counted in the opinions in which a Judge was leading, on the basis of the first name listed in the opinion.Footnote ⁵³

Overall, the average number of references to scholarship per individual Judge is just over 20, but the median number is in fact 4. The inconsistency between the figures may be explained by the fact that some Judges appear to have cited far more than others, with four in particular accounting for over half of the total number of passages containing references. Thus, as Figure 2 shows, Judge Pinto de Albuquerque leads with 704 citations, whereas Judges Serghides, Dedov and Vehabović made 158, 145 and 138 references, respectively. This is particularly interesting as Judge Vehabović was not listed as first drafter in any opinion citing scholarship.Footnote ⁵⁴ Ranking by leading Judges, there is again evidence that some Judges are far more inclined to cite scholarship than others. Thus, as Figure 3 shows, Judge Pinto de Albuquerque accounts for over half of the citations (with 704 authorities cited in 25 opinions) and is followed, at some distance, by Serghides (158 in 10), Yudkivska (95 in 6), Tsotsoria (43 in 1), Motoc (31 in 5) and Zupančič (20 in 9).

Figure 2: Number of paragraphs containing citations to scholarship by individual Judge, however listed in the opinion

Figure 3: Number of paragraphs containing citations to scholarship by ‘lead’ individual Judge

C. Age of Cited Scholarship

Another important question concerns the average age of the cited sources. The question is significant insofar as it may provide evidence of the level of engagement of the Court with scholarly authorities. In other words, it shows how up to date the Court and its Judges and lawyers might be at any given time. Figure 4 shows the average and median age of scholarship throughout the years.

Figure 4: Average and median age of cited scholarship over the years

The results reveal that, despite some exceptions, the Court has tended to cite comparatively recent scholarship, with an average age of 18.8 years and a median age of 7 years—the latter measure being more indicative, considering that more dated works such as those of Grotius and Shakespeare feature among the cited sources, thereby skewing the averages significantly. The findings appear to support the idea that the Court will generally seek to consult and cite up-to-date writings, as opposed to following the informal rule whereby the authority of deceased authors is to be preferred to that of the Judges’ own contemporaries.Footnote ⁵⁵ There is no difference between the majority and the minority as to the median age of the cited works, but, unsurprisingly, the average age of the authorities cited by the majority is lower (12.5 years, against 19.44 years for the minority), pointing towards a tendency to cite more recent works.

D. Highly Cited Sources

When analysing the jurisprudence of the Grand Chamber of the ECtHR, it is useful to look at the most cited scholarly authorities. Most international courts have certain most cited works,Footnote ⁵⁶ and the ECtHR is no exception. However, there is less consistency in the most cited works in the ECtHR than in other international courts.

Table 1 lists the authorities that have been cited the most by the majority. An interesting pattern that emerges is that the most cited works are on international humanitarian law, specifically Henckaerts and Doswald-Beck's Customary International Humanitarian Law (three citations), Roberts’ ‘Transformative Military Occupation: Applying the Laws of War and Human Rights’ (two citations) and Dinstein's The International Law of Belligerent Occupation (two citations). Other sources on occupation are also cited in two cases.Footnote ⁵⁷ Other commonly cited authorities belong to the field of public international law.

Table 1: Most cited authorities by the majority

It is also worth noting that there are sources cited in the ECtHR's case law concerning topics not specifically related to law, such as Zhou, Hofman, Gooren and Swaab's landmark study on transsexuality, which was ‘the first to show … a female brain structure in genetically male transsexuals and [to support] the hypothesis that gender identity develops as a result of an interaction between the developing brain and sex hormones’.Footnote ⁵⁸

Thus, it appears reasonable to conclude that highly cited authorities tend to target matters that fall outside the area of expertise of the ECtHR, and on which the Court might feel uncomfortable to contribute without buttressing its arguments.

The patterns emerging from analysis of the minority opinions are rather different (Table 2). Here, too, public international law scholarship features heavily, including some classicsFootnote ⁵⁹ and well-known reference works.Footnote ⁶⁰ In minority opinions, however, citation of scholarly authorities discussing the ECHR itself are frequent. Thus, among the most heavily cited sources are Harris, O'Boyle and Warbrick's textbook on the ECHR,Footnote ⁶¹ and Rietiker's study on the principle of effectiveness.Footnote ⁶² The most cited items also reveal other dynamics. For example, Judge Serghides's article ‘The Principle of Effectiveness as Used in Interpreting, Applying and Implementing the European Convention on Human Rights’ is cited in four cases, and by Judge Serghides himself in three of them.Footnote ⁶³ Similarly, Zupančič's study on the privilege against self-incrimination and adjudication and on the rule of law are also cited, in two cases and one case, respectively, by Judge Zupančič himself.Footnote ⁶⁴ More generally, it is important to note that most highly cited sources do not appear to have enjoyed broad support. On the contrary, more often than not, such sources have been favoured by one or two Judges, who have repeatedly referred to them.

Table 2: Most cited authorities by the minority

A. (Lack of) Scholarship in the Majority Judgments

References to scholarship in majority judgments tend to follow certain patterns. First, references to academic writings sometimes appear in a judgment because they were referred to in the parties’ submissions, and the Court simply summarises their arguments, occasionally including the references used by the applicantsFootnote ⁶⁵ or respondents.Footnote ⁶⁶ These references, arguably, are less relevant for the present analysis as they are largely incorporated by the Court in the text of the judgment, rather than carefully researched and selected. Second, when the majority cites scholarship, it almost invariably does so in the descriptive parts of its judgment, where the Court sets out and explains the relevant international, comparative and domestic legal frameworks, rather than in the analytical part, where the Court interprets and applies the law of the ECHR. This method of referencing scholarship is preferable if the Court wishes to avoid the impression that its majority decision-making is unduly influenced by scholarship.

While it is true that the majority of the Court almost never deploys academic writing in its reasoning for the purposes of persuasion, academic sources are used for other purposes. First, the Court can establish a fact by referring to scholarship. These sources are usually unrelated to law and provide some clarification of the context of the case. An example is offered by Gorzelik and others v Poland, which dealt with protection of national minorities. In this judgment, the ECtHR stated: ‘According to some linguists, although the Polish language is relatively unaffected by regional variations, it is possible to identify at least two regional varieties: Kashubian and Silesian.’Footnote ⁶⁷ The Court then referred to John A. Dunn's ‘The Slavonic Languages in the Post-Modern Era’. A reference like this explains the factual background of the case but adds little to the Court's reasoning directly.

Second, the ECtHR can clarify and elaborate on the rules of international law by reference to relevant scholarship, a practice that is, as already explained, entirely absent when ECHR provisions are at stake. Examples may be found in cases such as Sargsyan v Azerbaijan, where the Court considered the definition of occupationFootnote ⁶⁸ and referred to a number of academic works, including monographs by Benvenisti,Footnote ⁶⁹ Arai-TakahashiFootnote ⁷⁰ and Dinstein,Footnote ⁷¹ as well as an article by Roberts.Footnote ⁷² Similar references were made in Chiragov v Armenia,Footnote ⁷³ and later repeated in Georgia v Russia (II).Footnote ⁷⁴ These sources were not used in the Court's reasoning for the purpose of persuasion. Having said that, in Sargsyan v Azerbaijan, dissenting Judge Pinto de Albuquerque relied on Hersch Lauterpacht's works for the purpose of persuasion in relation to arguments similar to the ones made by the majority in that case.Footnote ⁷⁵

Third, and finally, the Court uses legal scholarship to explain and evaluate the domestic law of the respondent State. In the case of Reinhardt and Slimane-Kaïd v France, the ECtHR referred to scholarship in many instances in order to identify the duties of the State Counsel at the Court of Cassation in France.Footnote ⁷⁶ In Jalloh v Germany, which concerned administration of emetics to suspected drug dealers, the Court stated the following: ‘A considerable number of legal writers, however, take the view that the Code of Criminal Procedure, Article 81a in particular, does not permit the administration of emetics’,Footnote ⁷⁷ and proceeded to list some academic authorities to substantiate this reading. In Bladet Tromsø and Stensaas v Norway, the Court referenced scholarship to explain the legal debate that existed at the national level in relation to the burden of proof in defamation cases.Footnote ⁷⁸

The analysis of the use of scholarship by the majority clearly reveals that it avoids using academic commentaries to interpret the ECHR. Moreover, the approach of the majority of the Grand Chamber to the selection and representation of citations is also inconsistent. Although the Court is not always very consistent in referring to other sources such as soft-law instruments, international treaties and its own precedents, it does so more frequently and explicitly relies on them in its reasoning, and has an accepted method of citation and other formal criteria. The same cannot be said in relation to academic writing: there is no uniform citation style and the approach of the Court to this type of citation is very difficult to discern.

B. Patterns in Separate Opinions

ECtHR Judges commonly follow the terminology and approach of the majority in their separate opinions. This is true both in terms of formal structureFootnote ⁷⁹ and substantive analysis.Footnote ⁸⁰ Minority Judges apply similar tests used by the majority, but will ordinarily emphasise different aspects. In other words, the majority and minority of the Court speak the same legal language and usually follow the same formal rules of citing and referring to sources. However, as established in the previous section, the majority of the Court rarely cites scholarship. Perhaps, as a result of this, there are no clearly established and accepted drafting conventions for citing scholarship in the ECtHR judgments.Footnote ⁸¹ This may partially explain why Judges, in their separate opinions, demonstrate very little consistency in the deployment of academic sources. In other words, as the majority has not conclusively established conventions concerning the correct way to employ scholarship, the work of individual Judges remains without guidance.

In this section two related points are addressed. First, it is demonstrated that dissenting Judges do not restrict themselves to using scholarship in the three ways in which the majority does. While, like the majority, individual Judges also use scholarship for the purposes of establishing the factsFootnote ⁸² or the interpretation of internationalFootnote ⁸³ and domesticFootnote ⁸⁴ law, they also employ it for the purpose of persuasion and, accordingly rely on academic writing for arguments directly related to the interpretation of the ECHR. Second, the degree of variety and lack of consistency in the Judges’ use of scholarship in separate opinions are noted. Both formal aspects, such as formatting, language, and source selection, and substantive aspects, such as the role of these sources in the arguments of the Judges and the supposed audiences that they try to persuade, are considered.

2. Lack of a single overarching pattern

This section demonstrates the diversity of styles in which minority Judges refer to scholarship, and maps out the role that scholarship plays in their separate opinions. It will consider both formal aspects of referencing, such as types of sources, language and the formatting style, and substantive questions, such as the role of citations in the reasoning and possible audiences targeted with the inclusion of scholarly citations.

a) Types of sources

The analysis demonstrates that minority Judges refer to various and sometimes quite unusual sources in their separate opinions. The spread of scholarship is much wider than the scholarship referred to by the majority.

In their separate opinions, Judges refer to academic writing authored by experts in the law of the ECHR. They have referred to works by Gerards,Footnote ⁹⁵ Besson,Footnote ⁹⁶ Tsakyrakis,Footnote ⁹⁷ Greer,Footnote ⁹⁸ de Londras,Footnote ⁹⁹ Benoît-Rohmer, KlebesFootnote ¹⁰⁰ and many others. Minority Judges have also referred to various textbooks and commentaries on the ECHR, such as those by Warbrick, Harris and O'Boyle,Footnote ¹⁰¹ Van Dijk, Van Hoof, Van Rijn and Zwaak,Footnote ¹⁰² Jacobs, White and Ovey,Footnote ¹⁰³ SchabasFootnote ¹⁰⁴ and Reid.Footnote ¹⁰⁵ In addition to articles, books and textbooks written by academics, the minority Judges also refer to blogposts related to the subject matter in the case at hand. For example, blogposts by Leach and DonaldFootnote ¹⁰⁶ and by Føllesdal and UlfsteinFootnote ¹⁰⁷ have been referred to. The appearance of these sources in separate opinions seems logical and appropriate as these sources discuss the issues arising in the cases at hand.

However, Judges also make far more unconventional references. For example, references to legal and general philosophers are not uncommon, and the diversity among them is quite striking. Thus, while references to classics such as CiceroFootnote ¹⁰⁸ and Kant do appear,Footnote ¹⁰⁹ references have also been made to authors such as AlexyFootnote ¹¹⁰ and Dworkin.Footnote ¹¹¹ These references are generally made to illustrate the general context of the case or for rhetorical purposes. Consider, for example, the reference made to Cicero by Judges Pinto de Albuquerque and Vučinić in De Souza Ribeiro v France, where it was stated that:

In times of unemployment and fiscal constraints, States refrain from granting migrants equal access to civil and social rights and give preference to nationals over migrants. This policy not only challenges social cohesion in European countries, but goes to the heart of the principle of equality. By failing to acknowledge the civil and social rights of undocumented migrants, States become morally responsible for the commodification of those persons who live at the very bottom of the social ladder. This responsibility is not only moral, but also legal. As Cicero once put it, Meminerimus autem etiam adversus infimos iustitiam esse servandam [Let us not forget that justice should also be done to the most humble].Footnote ¹¹²

Here, the reference to Cicero is not used for the purpose of persuasion, explanation or to analyse the factual background of the case, but to provide a rhetorical introduction to the core matter at issue. The same can be said about the references frequently made by certain Judges to works of fiction. Again, these sources are very diverse, including references to novels, plays and sonnets ranging from Shakespeare,Footnote ¹¹³ DostoyevskyFootnote ¹¹⁴ and Kafka,Footnote ¹¹⁵ to RushdieFootnote ¹¹⁶ and McCullough.Footnote ¹¹⁷ They might add rhetorical value, but they are ordinarily quite detached from the subject matter of the judgments. Only a few Judges refer to literature and the presence of these references in separate opinions is very much dependent on the style of writing of a particular Judge.

In the face of such variety, it is difficult to establish a clear approach of the minority Judges in the ECtHR to citing scholarship. The selection of sources appears to remain a highly personal choice and depends on the familiarity of a particular Judge with certain literature, as well as on their writing style. However, there are two trends which may be discerned. First, as already mentioned above, the Judges prefer quoting academic sources that originate from within the Court. There are plenty of examples of minority Judges referring to the writings of their colleaguesFootnote ¹¹⁸ and lawyers of the Registry.Footnote ¹¹⁹ This may be due to familiarity and the accessibility of this scholarship, or to the fact that the ECtHR Judges consider it more authoritative than scholarship produced externally. Moreover, Judges sometimes refer to their own extra-judicial writing in their separate opinions. In many cases, Judges provide ‘further reading’ on the issue by referring to their own work,Footnote ¹²⁰ but the purpose of these self-references might not always be clearly explained in the separate opinions. For instance, Judge Elósegui in her concurring opinion stated:

In different publications I have referred to the main current positions that have been developed in academia, especially in Europe, on the principle of proportionality, concluding that it is a tool which assists judges in structuring an orderly reasoning for the resolution of a case. I have affirmed that Robert Alexy's theory is useful when studying the manner in which the courts really argue. I also agree with Carlos Bernal's thesis when he states that in the weighting exercise it is not possible to exclude the subjective assessments of the judge. But this is compatible with rationality if the judge applies and justifies each step of the proportionality test. Undoubtedly this idea can be inserted into a theoretical framework that starts from the theses of Robert Alexy and his disciples … among whom I have the honour of finding myself.Footnote ¹²¹

This discussion does have some connection to the subject matter of the case, but the connection was never explicitly unpacked in the opinion. Moreover, this self-reference appears to be a statement of general relevance, which could be included in many judgments.

The second discernible trend is that Judges often use a particular source to establish a certain point and then continue to refer to that source from one separate opinion to the next. For instance, publications by Rietiker, a senior lawyer of the ECtHR, are routinely cited by Judge Serghides in his separate opinions, usually on a similar point. In the majority of cases, the Judge refers to the issue of interpretation of the ECHR from the perspective of the principle of effectiveness in the jurisprudence of the ECtHR.Footnote ¹²²

b) Languages of sources

Table 3 provides a breakdown of the languages of the cited authority against the number of respondents, showing that English and French sources are the most likely to be cited, irrespective of the national language of the respondent State, but are by no means the only languages of cited sources. The dominance of references to scholarship in English or French is likely because these are the Court's official languages, and most Judges can read them both. By all metrics, sources in English (either originally or in translation) tend to be cited the most. However, as Table 3 shows, French is second by number of cases citing sources in that language, but it is only fourth according to the absolute number of sources referred to, behind Italian and German.

Table 3: Language of cited authorities

The main factor in the citation of scholarship in languages other than English and French, however, appears to be the identity of the respondents and their official language. Thus, for example, Italian sources are cited against respondents other than Italy only twice. Judge Pinto de Albuquerque referred to a high number of sources in Italian, German and Portuguese in cases against Italy,Footnote ¹²³ GermanyFootnote ¹²⁴ and Portugal,Footnote ¹²⁵ respectively.

However, not all rationales for the language of scholarship are so easily explained. Sometimes, sources in languages other than the languages of the Court or the languages of the respondent State appear in separate opinions. Thus, sources in German were used in cases against RomaniaFootnote ¹²⁶ and the Czech Republic.Footnote ¹²⁷ In Bedat v Switzerland, Judge Yudkivska referred to a book published in Russian, though the relevant quotation was translated into English in the text of the separate opinion.Footnote ¹²⁸ In Károly Nagy v Hungary, Judge Pinto de Albuquerque referred to sources in Italian and German.Footnote ¹²⁹

To briefly conclude, the approach of minority Judges to the language of the source is not always obvious. Although the majority of sources referred to are published in English, there are some notable exceptions.

C. Role of References

Although the issue of the role of references to scholarship may well warrant its own study, this section will briefly show how references to scholarship are used for a variety of purposes by the minority Judges. It has already been established that, not unlike the majority, minority Judges deploy scholarship to establish facts and interpret national and international law, and that minority Judges uniquely also refer to scholarship for persuasive purposes. Three other purposes may be discerned which illustrate the utility of academic references.

First, minority Judges use academic sources to establish the context of a particular issue and provide further reading on the topic. Judge Pinto de Albuquerque has referred to academic sources for this purpose in many of his separate opinions.Footnote ¹³⁵ Although this contextualisation might be helpful in some cases, a detailed list of possible sources might not be the most efficient use of separate opinions. Of course, there may be disagreements as to the purpose of separate opinions, but very few Judges use them to contextualise the matter at issue and to provide an extensive list of further reading.

Second, references to scholarship can be used as illustrative examples of a particular point. For instance, in Fernández Martínez v Spain, Judge Dedov simply referred to literary examples illustrating potential negative consequences of celibacy. Judge Dedov pointed out that:

Third, some references are used for purely rhetorical purposes. It seems that there is no purpose for some references other than to improve the ‘sound’ of these opinions. For instance, in Jaloud v the Netherlands, Judge Motoc compared an essay by Camus dealing with self-estrangement with the lack of jurisprudential consistency by the Court.Footnote ¹³⁷ Although interesting and creative, this reference's pragmatic utility might not be immediately obvious. The reference to Emma Lazarus's sonnet in De Souza Ribeiro v France is also used for rhetorical purposes.Footnote ¹³⁸

It is thus clear that Judges use references to scholarship in their minority opinions for purposes which vary significantly and often depend on the personality of the Judge and their approach to separate opinions. Undeniably, these purposes can overlap; moreover, some Judges use references to scholarship for a plethora of reasons while others do not use them at all. It is therefore not possible to discern a coherent approach of minority Judges to the role and purpose of their references to scholarship.

D. The Rationales for Citing Scholarship

This subsection examines the rationales for citing scholarship of the majority and minority of the ECtHR. Certain assumptions are necessary to discern the typical motivation for engaging with scholarship. The aim of this subsection is twofold. First, it seeks to explain the reluctance of the majority of the Court to cite academic sources for the purposes of persuasion, by reference to the practice of other courts and tribunals. This brief comparison is not exhaustive, as such an examination would warrant a separate study. Second, the related question of why some minority Judges are more willing to use legal scholarship in their judgments will be addressed. Aside from obvious practical reasons, such as that minority Judges need not collaborate or coordinate with other Judges in selecting sources and are able to follow their own style, it is hypothesised that when losing the argument in the Court the minority Judges are keen to persuade as broad an audience as possible, supported by authority such as scholarship, that their view, rather than that of the majority Judges, should have prevailed.

As noted above, the Court does not ordinarily cite scholarship in its majority opinions. If it does, it is by inclusion of a few references, and almost invariably in areas where the Court itself cannot claim unchallenged expertise. There are several reasons why the choice not to cite may be made—including, of course, the fact that citing is not ‘costless’, and that agreement may need to be reached not only on the proposition, but also on the authority itself.Footnote ¹³⁹ Academic sources are thus deliberately avoided. Yet, the current approach does little to enhance the transparency of decision-making at the Court.

One of the reasons why academic sources are not cited by the majority is that the ECHR does not provide a similar standing directive to that found in Article 38 of the ICJ Statute expressly allowing reference to specific authorities. That said, this does not prevent the Court from casting a wide net in its selection of sources of persuasive authority. Indeed, the Court has referred to its jurisprudence, rules of international law, soft law and reports of various organisations as means of interpretation of the ECHR.Footnote ¹⁴⁰ While the letter of the ECHR does not expressly allow reference to these materials, it does not forbid them either and, by the same logic, reference to scholarship should be subject to the same treatment. And yet, it is not. Materials of the type mentioned above are used by the majority much more frequently and extensively than legal scholarship. It follows that a lack of express regulation allowing reference to such external sources cannot alone explain the very limited number of such references in the judgments of majority.

Further explanations for legal scholarship rarely being used by the majority can be inferred from the cases where the majority has actually made reference to it. One of the problems here is the meaning and weight of a particular authority. To use the terminology pioneered by Henry SmallFootnote ¹⁴¹ and successfully applied to international law scholarship by Lianne Boer,Footnote ¹⁴² the ECtHR's reference to scholarship amounts to converting a cited work into a ‘concept symbol’ representing a majoritarian ‘consensus’ on a particular point. Moreover, the citation by the majority means that a particular source is elevated from an academic source to effectively a source of law. The Court perhaps is reluctant to grant such ‘power’ to academic commentators. This might also explain why the Court does not refer to specific ECHR scholarship but prefers academic sources discussing broader international or domestic law. As the ultimate authority on the ECHR, a citation by the Court would significantly boost the value of a particular source or scholar working in the area of ECHR law, much more than in broader international law.

Other adjudicators have been less hesitant to rely on scholarly authority. Thus, for example, a study by Helmersen confirms that the Appellate Body of the World Trade Organization has cited writings regularly, although with decreasing frequency.Footnote ¹⁴³ Similarly, looking at international investment arbitration, Ridi and Schultz found that references to scholarship are as common as citations to previous awards and other decisions, although the latter increasingly outpace the former.Footnote ¹⁴⁴

Such studies generally show that citations tend to favour old works by established writers.Footnote ¹⁴⁵ It was noted above that in contrast, the academic sources used by the ECtHR are not particularly old, with the vast majority of them having been published in the twenty-first century. The Court itself has never articulated its approach to citing legal scholarship, but a plausible explanation might be found in the distinctiveness of both human rights adjudication and the Court's own approach to legal interpretation. The ECtHR rarely uses an ‘originalist’ approach to interpretation of the ECHR to establish the historical pedigree of a particular provision. Harris, O'Boyle and Warbrick explained the rare use of travaux preparatoires by the Court in the following terms:

Moreover, the hard cases that the ECtHR is dealing with are often generated by contemporary challenges that are impossible to resolve by reference to older and more established academic sources. This perhaps makes citing older scholarship less appealing for the Court due to its limited value, while the Court's unwillingness to turn scholarship into a ‘concept symbol’ can explain its reluctance to cite more recent and relevant scholarship.

There is thus a very narrow window of opportunity for the Court to refer to legal scholarship. The use of scholarship by minority Judges may potentially make an impact on the approach of the majority over time and it is possible that their use of academic sources might increase. The capacity of the Court to control the content of separate opinions is limited and it seems that at least some Judges are (were) eager to refer to a wide array of scholarship.

What makes an individual Judge keen to cite academic sources in their judgments? One hypothesis would suggest a link between the frequency of scholarly citations of individual Judges with the approach of their national legal system of origin to such sources. This argument can be complemented by reference to the specific educational and professional backgrounds of the Judges with high ‘scholarship citation count’.

The argument that the country of origin can determine the Judge's inclination to cite scholarship is a challenging one. Some national courts accept no use of scholarship in judicial reasoning, while others use it more or less frequently. The weight of references, too, varies significantly. In Germany, for instance, references to domestic legal scholarship have heavy persuasive weight,Footnote ¹⁴⁷ although no ECtHR Judges elected in respect of Germany are known for high number of references to academic scholarship. In the courts of many Eastern European States, academic commentary is almost never cited;Footnote ¹⁴⁸ however, at least three Judges from the top-ten list (in Figure 3) are from Eastern Europe.

Moreover, the very suggestion that the inclination to cite depends on the Judge's country of origin is problematic because traditions can vary even within the judicial system in one given State. Within the same domestic system, there may of course be variations between lower and constitutional courts, for reasons such as the more academic composition of the bench.Footnote ¹⁴⁹ Other studies claim that ‘if ordinary courts do not refer to literature, constitutional ones will do the same’.Footnote ¹⁵⁰ With reference to the United States,Footnote ¹⁵¹ it has been argued that ‘[t]he role of legal scholarship and its potential utility are greater for courts of last resort’.Footnote ¹⁵² Yet, the increase in referring to scholarship might be skewed by certain Judges who are more inclined to cite than others. Thus, Schwartz and Petherbridge report an increase to citations to scholarship in US Federal Circuit Courts over time,Footnote ¹⁵³ but attribute the rise to just a small cohort of judges,Footnote ¹⁵⁴ and point to how generous referencing correlates negatively with a busy docketFootnote ¹⁵⁵ and political conservatism of the bench.Footnote ¹⁵⁶ It can also be seen that only a few particular ECtHR Judges are responsible for three-quarters of all minority Judge citations. Although the data this study is based on is somewhat limited, it seems that the inclination to cite scholarship cannot be explained by the origins, education or professional background of particular Judges alone.Footnote ¹⁵⁷

Another rationale for citing scholarship might be the minority Judges’ perception that it will make their arguments more persuasive to the various audiences to which the Court speaks. This explanation has been considered in legal scholarship. For example, building on the critical tradition of viewing international law as an argumentative practice,Footnote ¹⁵⁸ Venzke discusses the notion of semantic authority as ‘a specific form of power … generally sustained by a social expectation – an expectation that an actor at least refers to and deals with a specific claim in the international legal discourse’.Footnote ¹⁵⁹ For most international courts, a claim to semantic authority is generally sustained by self-referentiality (for example, reliance on an adjudicator's own precedent), but other uses of authority may be important: as Venzke continues, ‘an actor's semantic authority can be most effective if he takes himself out of the game and instead presents himself as a handmaiden of other authorities’.Footnote ¹⁶⁰ In this regard, it is clear that minority Judges have the potential to shape not just the development of the law, but, more broadly, the discourse on it. Applied to the use of scholarship, this may result in different deployments of authority when speaking to different audiences, so a minority opinion is likely to be written with one or more target audiences in mind.Footnote ¹⁶¹

The present empirical study provides further substantiation that at least some minority Judges might have specific audiences in mind when using academic sources in their separate opinions. First, this assumption can be supported by the fact that minority judges often use sources published in the language of the respondent State. Of course, scholarship written in the language of the respondent State may be most relevant in many cases, but it can also symbolise the familiarity of the minority Judge or Judges with the matter at issue. Second, the multiplicity of types of scholarship, from fiction to encyclopaedias, perhaps aims to show that minority Judges, or at least some of them, wish to speak to various groups and show their understanding of the local culture.

The idea of audiences of judicial decisions is not new and has been discussed in various contexts. It was argued in relation to the US Supreme Court that ‘Justices work with (or around) their audiences instrumentally to achieve their broader goals’,Footnote ¹⁶² meaning that the specific form of judicial reasoning can change depending on the audience that Judges are addressing.Footnote ¹⁶³ Since minority Judges were unable to persuade their colleagues on the bench, they may resort to using scholarly references to persuade other audiences which might support their point of view in future.

This broader use of scholarship may be intended to influence several key audiences, which may overlap. These audiences include, first and foremost, other Judges of the ECtHR (who were not on the bench in a particular case) and the lawyers arguing before it. Separate opinions, at the ECtHR and elsewhere, exist because one Judge or group of Judges failed to persuade their colleagues on the preference of their interpretation of the ECHR and they want to provide further scrutiny and discussion.Footnote ¹⁶⁴ The importance of airing these different interpretations lies in potentially influencing the Court in future cases, and references to scholarly writing can provide support for their argument.

The second key audience is national judges. It has been argued that national judges are the closest allies of the ECtHR at the national level and thus their acceptance of a particular position of the Court ensures that the ECHR is fully embedded at the national level.Footnote ¹⁶⁵ This is one of the reasons why some Judges use an abundance of sources which would be familiar to the judges and lawyers in the respondent State. It might also demonstrate the minority Judge's understanding of the legal context in that State. It might be argued that national governments also form part of the audience, and the inclusion of references to scholarship—especially domestic—is a powerful tool for making the opinion more persuasive.

The third audience is the legal community as a whole, including academic lawyers and commentators on the Court. To this audience, not only is reference to scholarship a tenet of scholarly writing, with this familiar format inviting them to support the minority position, but it is also a way for Judges to engage with the work of academics, and perhaps call for them to exercise their function of ‘vigilance’ over the Court's activity.Footnote ¹⁶⁶ This is particularly relevant in the context of those individual opinions which denounce a shift in a regressive direction, which have been fittingly termed by Helfer and Voeten as ‘walking back dissents’, which may well seek to rely on commentary which supports their assertion that the judgment represents a regression.Footnote ¹⁶⁷

The final key audience is the general public. The reference to literature, particularly for rhetorical purposes, may suggest that Judges are seeking to demonstrate their familiarity with the general cultural context of European society. Some Judges may include literary references to show that those people who make the judgments are not total strangers: they read similar books and sonnets, and can be trusted with the most complex matters.

Whilst these particular audiences are not unimportant for the majority, they are of greater significance for minority Judges, for at least two reasons. First, the judgments of the Court are legally binding in accordance with Article 46 of the ECHR and thus the majority Judges perhaps do not feel it necessary to use every persuasive tool to convince audiences as their writing becomes law regardless. Second, practically speaking, it is difficult to change the conservative drafting conventions of the judgments of the majority. Although there are some potential benefits in persuasive drafting, the Court perhaps does not see them as sufficient to warrant changing the established approach.

In conclusion, the majority of the Court only cites scholarship when it explains facts or law outside its primary domain, likely to avoid effectively elevating legal scholarship to the status of a source of ECHR law, or difficulties in agreeing on specific authorities being cited. For the majority the potential pitfalls of citing scholarship overweigh the potential benefits. The minority Judges have fewer such barriers to citing scholarship and are perhaps more concerned with persuading broader audiences, resulting in much wider use of legal scholarship in separate opinions.