¹ Permanent Court of International Justice, Advisory Committee of Jurists, Procès-Verbaux of the Proceedings of the Committee, 16 June – 24 July 1920, The Hague: 1920, Annex I to Meeting 14, 2 July 1920, at 323. The Procès-Verbaux are available at www.icj-cij.org/pcij/other-documents.php?p1=9&p2=8 (accessed 8 December 2015).

² H. Harrison Dinniss, Cyber Warfare and the Laws of War (2014), 1. This is the title of her first chapter.

³ Ibid., at 8.

⁴ On the difficulty of applying Art. 2(4) to cyber-attacks, see Boer, L.J.M., ‘“Echoes of Times Past”: On the Paradoxical Nature of Article 2(4)’, (2014) 20 Journal of Conflict and Security Law 5 CrossRefGoogle Scholar.

⁵ Harrison Dinniss, supra note 2, at 43.

⁶ For references to all of these, see the appendix to this article. The possibility of invoking ‘“experts” “consensus”’ on interpretive claims (as persuasive, rather than valid) is briefly mentioned by Corten, O., ‘Formalization and Deformalization as Narratives of the Law of War’, in Beneyto, J.M. and Kennedy, D. (eds.), New Approaches to International Law: The European and the American Experiences (2012), 251–72CrossRefGoogle Scholar, at 253–5.

⁷ Harrison Dinniss, supra note 2, at 41.

⁸ See Boer, supra note 4.

⁹ This number was limited by practical considerations. Given the extensive nature of the trackback table, it was only possible to examine a limited number of claims. Furthermore, given the inductive approach to these claims – they were taken from an existing collection – it would be difficult to make any firm statements about how representative they are for the debate as a whole. One could only claim to know the frequency of consensus claims if one had a database containing all articles written on the subject of Art. 2(4) and cyber, in order to subsequently assess the percentage of those works containing consensus claims – which is far beyond the scope of this article. An indication of their frequency, however, might be that they are made, not just with regard to interpreting ‘force’, but also with regard to other interpretive issues (see Section 2).

¹⁰ To view supplementary material for this article, please visit http://dx.doi.org/10.1017/S0922156516000480. A similar ‘trackback’ is done by Urška Šadl. See Šadl, U., ‘Case – Case-Law – Law: Ruiz Zambrano as an Illustration of How the Court of Justice of the European Union Constructs Its Legal Arguments’, (2013) 9 European Constitutional Law Review 205.CrossRefGoogle Scholar

¹¹ In other words, it seems that we do not cite randomly. See Gilbert, G.N., ‘Referencing as Persuasion’, (1977) 7 Social Studies of Science 113 CrossRefGoogle Scholar. On the history of the role and function of ‘bibliographic citations’, see McInnis, R.G. and Symes, D., ‘David Riesman and the Concept of Bibliographic Citation’, (1988) 49 College & Research Libraries 387 CrossRefGoogle Scholar.

¹² Small, H.G., ‘Cited Documents as Concept Symbols’, (1978) 8 Social Studies of Science 327;CrossRefGoogle Scholar on repetition, see, inter alia, 329.

¹³ This term is also used by Šadl, supra note 10, at 208; see also Skouteris, T., ‘The New Tribunalism: Strategies of (De)Legitimation in the Era of International Adjudication’, (2006) 17 Finnish Yearbook of International Law 307 Google Scholar, at 334.

¹⁴ See also K. Hyland, Disciplinary Discourses: Social Interactions in Academic Writing (2004). The author makes a similar point about references in his second chapter: ‘The importance of citation as a constitutive element of the modern academic paper can be seen in its increasingly prominent role in the ways writers seek to construct facts through their communicative practices.’ (at 21, emphasis added).

¹⁵ White, following Swales, argues this crossover, between what he also refers to as an analysis of ‘quantitative studies’ and ‘rhetorical choices’. See White, H.D., ‘Citation Analysis and Discourse Analysis Revisited’, (2004) 25 Applied Linguistics 89 CrossRefGoogle Scholar, at 89. From the viewpoint of international law, and Art. 38(1)(d) of the ICJ Statute, Jörg Kammerhofer talks of ‘scholarship about scholarship’. See J. Kammerhofer, ‘Law-Making by Scholars’, available at papers.ssrn.com/sol3/papers.cfm?abstract_id=2182547 (accessed 23 May 2016), at 1.

¹⁶ On the history of cyberwar and international security, see M. Dunn Cavelty, Cyber-Security and Threat Politics: US Efforts to Secure the Information Age (2008), Chapter 4.

¹⁷ For examples of cyber-attacks, see Ziolkowski, K.C., ‘Computer Network Operations and the Law of Armed Conflict’, (2010) 49 Military Law and the Law of War Review 47 Google Scholar, at 57–61. The number of hostile, violent cyber-attacks is very small; oft-mentioned cases are the attacks on Estonia (2007), Georgia (2008) and the deployment of the Stuxnet worm.

¹⁸ The term ‘law-by-analogy approach’ is taken from Hollis, D.B., ‘New Tools, New Rules: Why States Need an International Law for Information Operations’, (2007) 11 Lewis and Clark Law Review 1023 Google Scholar, at 1039; the term ‘fifth domain of warfare’ is a familiar one in the cyberwar debate; see, for example, The Economist, ‘War in the Fifth Domain’, 1 July 2010, available at www.economist.com/node/16478792 (accessed 8 December 2015).

¹⁹ For an overview of these arguments, see Boer, supra note 4.

²⁰ M. Koskenniemi, From Apology to Utopia: The Structure of International Legal Argument (2005), 25.

²¹ On interpretations as reconstructions, see ibid., Chapter 8; see also I. Venzke, How Interpretation Makes International Law: On Semantic Change and Normative Twists (2012), Chapter 1.

²² I use the word ‘materials’ here, to avoid confusion with the sources of international law that are arguably found in Art. 38 of the ICJ Statute. The reason for using the word ‘material’ instead is that scholars draw from many different works to come to their conclusions. A comprehensive discussion on whether Art. 38(1)(d) is a complete account of international law's sources, or whether it represents sources in the first place, is beyond the scope of this article – I have, however, clarified my own view on this on the final pages of this article.

²³ For examples of the use of these arguments, see Boer, supra note 4.

²⁴ Foltz, A.C., ‘Stuxnet, Schmitt Analysis, and the Cyber “Use-of-Force” Debate’, (2012) 67 Joint Force Quarterly 40 Google Scholar, at 40, (emphasis added for ‘general consensus’; emphasis deleted for ‘some’).

²⁵ Dunlap, C.J. Jr., ‘Perspectives for Cyber Strategists on Law for Cyberwar’, (2011) 5 Strategic Studies Quarterly 81 Google Scholar, at 85, (emphasis added for ‘leading view’; emphasis deleted for ‘effects-based’).

²⁶ See the appendix for further examples.

²⁷ As the questions raised in this article proceed from an initial, previous inquiry into how scholars deal with the prohibition, the claims under scrutiny here have been taken from my own, existing database of academic works dealing with this question. To this list, I have furthermore added a few books that contain a chapter dealing with the same interpretive question, as well as some articles that have been found by means of a Google Scholar search, entering as search term ‘cyber & “article 2(4)”’. The works assembled in these different ways have all been screened for the use of consensus claims on the meaning of the word ‘force’. For a more extensive treatment of the methodology used here, see the cover page of the appendix to this article.

²⁸ I have taken two consensus claims from the same work in three cases: Buchan, R., ‘Cyber Attacks: Unlawful Uses of Force or Prohibited Interventions?’, (2012) 17 Journal of Conflict and Security Law 211 CrossRefGoogle Scholar; Harrison Dinniss, supra note 2; G. Kerschischnig, Cyberthreats and International Law (2012). It was not always clear to me whose ‘ruling opinion’ is argued: that of states (which happens occasionally), but most of the time, the ruling opinion of scholars seemed to be claimed; rarely, both that of scholars as well as states. For examples of all of these, see the appendix. This question into ‘whose consensus’ (Young, 148) is reminiscent of what others describe in different contexts; see Young, K.G., ‘The Minimum Core of Economic and Social Rights: A Concept in Search of Content’, (2008) 33 Yale Journal of International Law 113 Google Scholar, at 148 – though she describes a problem of sources: where we should be looking for consensus (see the introduction to her article); Helfer, L.R., ‘Consensus, Coherence and the European Convention on Human Rights’, (1993) 26 Cornell International Law Journal 133 Google Scholar, at 139; for a somewhat similar observation, see Horst, M. and Irwin, A., ‘Nations at Ease with Radical Knowledge: On Consensus, Consensusing and False Consensusness’, (2010) 40 Social Studies of Science 105 CrossRefGoogle Scholar, at 117 (‘the crucial issue might be how a shared “we” is constructed’).

²⁹ Consensus has been studied in different contexts. For its use as a decision-making tool in the UN, see Pronto, A.N., ‘Some Thoughts on the Making of International Law’, (2008) 19 European Journal of International Law 601 CrossRefGoogle Scholar (book review); Sperduti, G., ‘Consensus in International Law’, (1976) 2 Italian Yearbook of International Law 33.Google Scholar On the relation between states’ consensus and international law, see d'Amato, A., ‘On Consensus’, (1970) 8 Canadian Yearbook of International Law 104;Google Scholar for its use in the search for the ‘minimum core’ of social and economic rights, see Young, supra note 28. On the search for consensus by the US Supreme Court, see Markovits, R.S., ‘Learning from the Foreigners: A Response to Justice Scalia's and Professor Levinson's Professional Moral Parochialism’, (2003) 39 Texas International Law Journal 367;Google Scholar Lee, Y., ‘International Consensus as Persuasive Authority in the Eighth Amendment’, (2007) 156 University of Pennsylvania Law Review 63.Google Scholar On the use of consensus claims by the European Court of Human Rights, see Benvenisti, E., ‘Margin of Appreciation, Consensus, and Universal Standards’, (1999) 31 New York University Journal of International Law and Politics 843 Google Scholar; on the identification of consensus in amicus curiae briefs submitted to the ECtHR, and their use of sources, see L. Van den Eynde (2011), ‘NGOs’ Contribution to the European Court of Human Rights through Amicus Curiae Briefs', Paper for the Annual Meeting of the Law and Society Association, Westin St. Francis Hotel, San Francisco, CA, citation.allacademic.com/meta/p_mla_apa_research_citation/4/9/5/8/7/p495879_index.html (conference paper abstract). For a comparative analysis of the use of consensus-based arguments by the US Supreme Court, the Supreme Court of Canada, and the ECtHR, see K. Henrard and E. Mak, ‘The Use of Consensus Arguments in Transnational Judicial Decision-Making: Confirming or Jeopardising Universal Human Rights’, available at papers.ssrn.com/sol3/papers.cfm?abstract_id=2444682 (accessed 8 December 2015). On consensus as a source of law in Islam, see Hallaq, W.B., ‘On the Authoritativeness of Sunni Consensus’, (1986) 18 International Journal of Middle East Studies 427.CrossRefGoogle Scholar For ‘consensusing’ as a societal practice, see Horst and Irwin, supra note 28, at 107; for the function of public consensus in judicial decision-making, see J.N. Shklar, Legalism: An Essay on Law, Morals and Politics (1964), 12–3.

³⁰ Kessler and Werner likewise refer to the ‘fixation of meaning’. See Kessler, O. and Werner, W., ‘Expertise, Uncertainty, and International Law: A Study of the Tallinn Manual on Cyberwarfare’, (2013) 26 Leiden Journal of International Law 793 CrossRefGoogle Scholar, at 794, fn4.

³¹ The works used here deal with Art. 2(4) and cyber-attacks in a variety of ways. See the cover page of the trackback table for an elaboration.

³² Hathaway, O.A. et al., ‘The Law of Cyber-Attack’, (2012) 100 California Law Review 817 Google Scholar, at 842 (emphases added).

³³ On the different views held by ‘weaker states’ and the more powerful ones, see Waxman, M.C., ‘Cyber-Attacks and the Use of Force: Back to the Future of Article 2(4)’, (2011) 36 Yale Journal of International Law 421 Google Scholar, at 449.

³⁴ The phrase ‘reference point’ is taken from Venzke, supra note 21, at Chapter 2. However, he also uses it to indicate those points scholars have to refer to in legal argument (see, 27, 48; but see also 53, 70). This is not the case for consensus claims; I therefore use it here somewhat differently than Venzke does.

³⁵ See also Tsagourias's consensus claim in his discussion of the Tallinn Manual's treatment of force in Tsagourias, N., ‘The Tallinn Manual on the International Law Applicable to Cyber Warfare: A Commentary on Chapter II - The Use of Force’, in Gill, T.D. et al. (eds.), Yearbook of International Humanitarian Law 2012 (2014), 19–43 CrossRefGoogle Scholar, at 22. Crawford Camiciottoli describes this as ‘alignment with other experts in the field in order to reinforce their own expertise’, see Crawford Camiciottoli, B., ‘Collective and Individual Identities in Business Studies Lectures’, in Gotti, M. (ed.), Commonality and Individuality in Academic Discourse (2009), 141–62Google Scholar, at 153. She refers here to the work of Martin and White who use the terms ‘alignment’ and ‘disalignment’ to signal the speaker's position with respect to ‘actual or potential respondents’. See J.R. Martin and P.R. White, The Language of Evaluation (2005), 1. As elaborated below, a consensus claim may be a claim to authority: in any case, it functions (as stated in this section) at least as a reference point in a scholar's argument. For a critique on the use of consensus, and a reflection on its possibilities, see Koskenniemi, supra note 20, at 532 et seq., and 596–9.

³⁶ See the pieces by Moore, Antolin-Jenkins (who adds some more disclaimers to her interpretation, at 152–5), Benatar (who claims a ‘majority opinion’ and a contra opinion in the footnote), Kerschischnig (his consensus claim at 110), Silver (with immediate application to cyber-attacks), Barkham, Woltag and Harrison Dinniss (at a different point in her chapter than the one cited earlier). Something similar happens in case of the claim made by Ducheine (at 116). He points out the absence of a definition of force in international law, subsequently to claim a consensus on force as armed force. For full references of these works, see the appendix.

³⁷ In the articles written by Remus, Li and Sklerov. For full references of these works, see the appendix.

³⁸ Morth, T.A., ‘Considering Our Position: Viewing Information Warfare as a Use of Force Prohibited by Article 2(4) of the UN Charter’, (1998) 30 Case Western Reserve Journal of International Law 567 Google Scholar, at 593. He elaborates somewhat on this position, but only slightly.

³⁹ Buchan, supra note 28, at 216, footnotes omitted (emphasis added for ‘generally accepted’). A few pages prior to this claim, Buchan also states that ‘[t]he generally accepted interpretation of Article 2(4) is that only those interventions that produce physical damage will be regarded as an unlawful use of force. Consequently, according to positive international law those cyber attacks that do not produce physical damage will not constitute a violation of Article 2(4)’ (at 212). Though it is clear that the argument is contingent on an analysis yet to follow further on in the article, the jump from general acceptance to positive law is striking. Nor do we know the origins of this majority claimed by Buchan.

⁴⁰ See Waxman, supra note 33; Kerschischnig, supra note 28 (the other consensus claim in his work, at 106); Schmitt, M.N., ‘Computer Network Attack and the Use of Force in International Law: Thoughts on a Normative Framework’, (1999) 37 Columbia Journal of Transnational Law 885.Google Scholar

⁴¹ W.G. Sharp Sr., CyberSpace and the Use of Force (1999), 90. He refers to the ‘prevailing view’ at 88. This is described by Crawford Camiciottoli as ‘disalignment’; see Crawford Camiciottoli, supra note 35, at 154.

⁴² Sharp, supra note 41, at 102. Like Walter Sharp, Richard Aldrich expresses a ‘prevailing view’ (that force equals physical, or armed force) but he is unsure how to qualify economic coercion. Though this is not precisely the same as arguing against majority opinion – he would only be doing so, if he were to argue that the majority claims that economic coercion is not armed force, and that he disagrees with that qualification – most claims under review here do precisely that: the interpretation of force as armed force is mentioned in the same breath as the exclusion of economic coercion. This makes Aldrich's position unusual. See Aldrich, R.W., ‘How Do You Know You Are at War in the Information Age?’, (1999) 22 Houston Journal of International Law 223 Google Scholar, at 238, 242.

⁴³ See also Koskenniemi, supra note 20, at 597: ‘Consensus is . . . the end-point of a hegemonic process in which some agent or institution has succeeded in making its position seem the universal or “neutral” position.’ For a similar dynamic with expert opinions, see Kessler and Werner, supra note 30, at 805–6.

⁴⁴ See for an example of this argument Delanis, J.A., ‘Force under Article 2(4) of the United Nations Charter: The Question of Economic and Political Coercion’, (1979) 12 Vanderbilt Journal of Transnational Law 101 Google Scholar, for example at 109, 116.

⁴⁵ Schauer, F., ‘Authority and Authorities’, (2008) 94 Virginia Law Review 1931 Google Scholar, at 1944 (emphases added). Though I found this to be the case in some instances of invoking consensus, it does not apply to all. For example, Harrison Dinniss's chapter contains an elaborate interpretation of ‘force’.

⁴⁶ This is similar to Ingo Venzke's discussion of authority; see Venzke, I., ‘Understanding the Authority of International Courts and Tribunals’, (2013) 14 Theoretical Inquiries in Law 381.CrossRefGoogle Scholar In reference to Hart, Venzke writes how ‘[a]uthority implies at least relative content-independence’, at 395, more extensively 394–6. In his discussion, Schauer also refers to Hart's content-independent reasons. See Schauer, supra note 45, at 1935, fn13.

⁴⁷ Which they relate to the societal practice of ‘consensusing’ (at 107). The dynamic, however, is the same.

⁴⁸ Horst and Irwin, supra note 28, at 110. It should be noted that they talk of a specific strand of Science and Technology Studies, ‘which is not primarily focused on consensus as such, but on the mediation, negotiation and management of boundaries or entities [author citation omitted] . . . Habermasian ideals and accusations of false consensusing are replaced with the more neutral and empirically oriented language of “closure”’.

⁴⁹ Lee refers to the ‘Wisdom of Crowds’ argument. See Lee, supra note 29, at 101–2.

⁵⁰ Young cited in Lee, supra note 29, at 77, fn47.

⁵¹ The European Court of Human Rights uses consensus-based arguments to identify shared values in the area of human rights by the member states. The smaller this consensus, the larger the margin of appreciation for the states (see Helfer, supra note 28). In this context, Benvenisti refers to critique on the ‘vague process of identifying consensus’ by the Court (Benvenisti, supra note 29, at 851). His own critique of the consensus mechanism used by the Court can be found at 852. See for a similar critique on the identification of consensus, Yourow cited in Helfer, supra note 28, at 138.

⁵² An article in Social Forces in 1952 attempted to do just that – to measure consensus, the scale of which aimed to ‘measure the degree to which orientation of [a group] permeates all the individual members of the group, with respect to a given object’ (at 97, emphasis in original). The authors distinguish different degrees of ‘intensity’ of consensus. See Riley, M.W. et al., ‘The Measurement of Consensus’, (1952) 31 Social Forces 97.CrossRefGoogle Scholar The interest here, however, lies in what an argument does in terms of persuasion – and in that sense, the meaning of a norm is fixed by the word ‘majority’ as much as it is by the word ‘consensus’. A similar issue – measuring consensus – is addressed by Markovits in the context of the US debate on consensus, see Markovits, supra note 29, at 368. For reflections on the demands on identifying consensus by the US Supreme Court, see Young, E.A., ‘Foreign Law and the Denominator Problem’, (2005) 119 Harvard Law Review 148.Google Scholar

⁵³ Randelzhofer, A., ‘Article 2(4)’, in Simma, B. (ed.), The Charter of the United Nations: A Commentary (2002), 112–36, at 117 Google Scholar (emphasis added).

⁵⁴ See also Horst and Irwin, supra note 28, at 109–10.

⁵⁵ See the trackback table for an example. Nimis likewise refers to the ‘see also’ phenomenon, see Nimis, S., ‘Fussnoten: Das Fundament der Wissenschaft’, (1984) 17 Arethusa 105 Google Scholar, at 125.

⁵⁶ Similarly, Pareto states that ‘oftentimes the consensus is not based upon direct observation, but is merely taken for granted on the basis of certain sentiments held by the person asserting it’ (at 355). I do not consider the consensus claims studied here as based on ‘sentiments’ only; I cite Pareto here to point out their ‘empirical’ appearance. See V. Pareto, The Mind and Society, Vol. I (1935), 355–9. This is also cited by C. Perelman and L. Olbrechts-Tyteca, The New Rhetoric: A Treatise on Argumentation (1969), 33. I will briefly address those instances where consensus claims remain unsubstantiated, below.

⁵⁷ Thanks to one of the anonymous reviewers for pointing out the need for clarification on this point. On performativity, see J. Butler, Bodies that Matter: On the Discursive Limits of “Sex” (2011), xii. See, for a similar dynamic, Weber, C., ‘Performative States’, (1998) 27 Millennium 77 CrossRefGoogle Scholar, at 82.

⁵⁸ Buchan, supra note 28, at 217, (emphasis added in the last sentence; removed in the first for ‘effects’).

⁵⁹ Joyner, C.C. and Lotrionte, C., ‘Information Warfare as International Coercion: Elements of a Legal Framework’, (2001) 12 European Journal of International Law 825 CrossRefGoogle Scholar, at 845 (emphasis added).

⁶⁰ I. Brownlie, International Law and the Use of Force by States (1963), 361.

⁶¹ For the trackback of Brownlie's page 362, see the appendix.

⁶² D.W. Bowett, Self-Defence in International Law (1958), 148 (emphasis added). Bowett, in the same footnote, also refers to the Brazilian amendment to the Charter.

⁶³ Waldock, C.H.M., ‘The Regulation of the Use of Force by Individual States in International Law’, (1952) 81 RCADI at 492 Google Scholar (emphasis added). Bowett refers to Chapter 3 of Waldock's work; the only page relevant to Bowett's interpretation, in the sense that it is itself also footnoted, is 492.

⁶⁴ L. Goodrich and E. Hambro, Charter of the United Nations (1949), 104.

⁶⁵ Horst and Irwin similarly talk of ‘the performativity of consensus’, but their interest lies in what the process of ‘consensusing’ as a societal practice is productive of: they refer to ‘how [consensusing is] productive (but also reflective) of social and institutional arrangements’ (at 107), and of consensusing as ‘generating shared identities and understandings’ (at 118). See Horst and Irwin, supra note 28.

⁶⁶ Morth, supra note 38, at 593 and 593, fn169 (emphases added).

⁶⁷ For an inquiry into the ‘life of the concept’ of the ‘failed state’ in the discipline of international relations, using actor-network theory, see Bueger, C. and Bethke, F., ‘Actor-Networking the “Failed State”: An Enquiry into the Life of Concepts’, (2014) 17 Journal of International Relations and Development 30.CrossRefGoogle Scholar For a slightly different example, see Casey, D.L. and McMillan, G.S., ‘Identifying the “Invisible Colleges” of the Industrial & Labor Relations Review: A Bibliometric Approach’, (2008) 62 Industrial & Labor Relations Review 126.CrossRefGoogle Scholar

⁶⁸ For an example of a study into the function of footnotes in a particular case, see Paul, D., ‘In Citing Chaos: A Study of the Rhetorical Use of Citations’, (2000) 14 Journal of Business and Technical Communication 185 CrossRefGoogle Scholar, particularly at 201. Thomas and Hawes define citation analysis as ‘counts of references to other researchers, or reporting statements, [which] are employed as a way of assessing the contribution of researchers to the field’. See Thomas, S. and Hawes, T.P., ‘Reporting Verbs in Medical Journal Articles’, (1994) 13 English for Specific Purposes 129 CrossRefGoogle Scholar, at 130. The ‘hunch’ underlying this article was the question of how often the same authors are used for the same ‘knowledge claim’ (see Hyland, supra note 14, at 8), bringing me to use Small's work, further on in this article.

⁶⁹ This of course, is also a majority claim. It is based on the trackback table, and is therefore based on a discrete number of cases. I do not claim that this practice of self-referentiality extends beyond what I have found here, though, given findings in sociology of science and citation analysis, it may be expected that it does.

⁷⁰ As stated above, three authors appear twice in the trackback table. These counts are of the specific page or paragraph in which Randelzhofer makes his own consensus claim; see the appendix to this article.

⁷¹ Remus, supra note 37, at 181.

⁷² Hathaway et al., supra note 32, at 842.

⁷³ Sharp, supra note 41, at 88.

⁷⁴ This also includes different editions of the same work; e.g., Cot and Pellet's book was published in 1985; a new edition appeared in 1991. In bold are authors whose work is cited two or more times; there are variations in the number of times they are cited; e.g., Dinstein is cited more frequently than Singh.

⁷⁵ See the body text below for those writers who leave their claims unsubstantiated.

⁷⁶ Note that though Randelzhofer and Dörr in the 2012 edition of Simma's commentary still refer to Dinstein (in this case the fifth edition of his War, Aggression and Self-Defence of 2012), Dinstein himself ceases referring to Randelzhofer's chapter in his fifth edition of 2012. This could be related to his changed definition of ‘force’ compared to his fourth edition. Whereas in 2005 (4th ed.) he argues that ‘the term “force” in Article 2(4) must denote armed – or military – force’ (at 86) in 2012 (5th ed.) he states that ‘the term “force” in Article 2(4) must denote violence’ (at 88).

⁷⁷ Austin, A., ‘Footnote Skulduggery and Other Bad Habits’, (1990) 44 University of Miami Law Review 1009 Google Scholar, at 1022, fn87, citing Thorne (note 80).

⁷⁸ Watson, J.S., ‘Legal Theory, Efficacy and Validity in the Development of Human Rights Norms in International Law’, (1979) 1979 University of Illinois Law Forum 609 Google Scholar, at 636–7. Watson states that ‘it is much easier to cite authors who have already reached [a similar] conclusion . . . the writer does not need to concern himself with a full explanation of his argument’ (at 636). Watson is referred to by G.J.H. Van Hoof, Rethinking the Sources of International Law (1983), 178.

⁷⁹ Watson, supra note 78, at 610, 641, 637.

⁸⁰ Thorne, F.C., ‘The Citation Index: Another Case of Spurious Validity’, (1977) 33 Journal of Clinical Psychology 1157 3.0.CO;2-B>CrossRefGoogle Scholar, at 1160 (also cited by Austin, supra note 77).

⁸¹ White, supra note 15, at 110.

⁸² Douglas Kell at blogs.bbsrc.ac.uk/index.php/2009/03/the-matthew-effect-in-science/ (accessed 8 December 2015).

⁸³ Small, supra note 12, at 328.

⁸⁴ Ibid., at 329, 334. Referring to Small's work, Paul describes how ‘[citations] act as a discipline's shorthand for certain ideas’ (Paul, supra note 68, at 196); see also Gilbert, supra note 11, at 117.

⁸⁵ Small, supra note 12, at 329, 338.

⁸⁶ Ibid., at 328, 329. On repetition, see notes 87 and 88, below. Something similar is described by Gilbert when he discusses the use of citations: ‘In citing certain papers, the author can be seen to be making an assertion about his own opinion concerning the validity of the findings of the cited papers, and is thus contributing, albeit only in small measure, to the overall consensus of his research area, a consensus which is also continually being re-established through the choices of references in his fellow participants’ own publications’ (Gilbert, supra note 11, at 117). The kind of consensus Gilbert refers to here is similar to the process of an emerging ‘meta-consensus’ identified in this article: what he seeks to point out, is the emergence of a shared idea on the ‘validity of the findings of the cited papers’.

⁸⁷ Or, in Small's words, ‘[r]ecurring patterns of terminology used by citing authors when referring to these documents show that they have become standardized in their usage and meaning’. See Small, supra note 12, at 337. The ‘recurring pattern of terminology’, in this case, is the word ‘consensus’, ‘majority’, etc.

⁸⁸ As stated by Small, see ibid., at 329, 338. Small denies any claims to knowing ‘how’ a document turns into a concept symbol. However, he does refer to, for example, ‘recurring patterns of terminology’, ‘uniformity of usage’ and ‘standardized terminology’, which means that he understands this as simply a matter of repetition (though we still wouldn't know what started this process, see 338.). The notion of repetition is the point where we might link the Matthew effect – those who are cited often, will be cited more often – with Small's idea of ‘concept symbols’. Phrased differently, a concept symbol comes into being through its repeated use; and the more it is used, the more it will be used. Mahbuba and Rousseau likewise link the Matthew effect to Small's ‘concept symbols’. They state that ‘articles . . . considered as concept symbols become the hard-to-ignore default citation value. This is exactly the Matthew effect at work: the highly visible and highly cited become even more visible and cited, while the less known disappear completely from the “citation” radar’. See Mahbuba, D. and Rousseau, R., ‘The Matthew Effect and a Relation with Concept Symbols and Defaults’, (2011) 58 Annals of Library and Information Studies 335 Google Scholar, at 337.

⁸⁹ Small, supra note 12, at 338.

⁹⁰ As this section shows, as well as the appendix, this practice expands with Randelzhofer's references.

⁹¹ A. Grafton, The Footnote: A Curious History (1997), 33.

⁹² Ibid., at 16, 33.

⁹³ Ibid., at 33. It is necessary to add the context of Grafton's statement here: he refers to the practice of ‘theologians and lawyers of the Middle Ages and the Renaissance’ (at 33). Rodgers makes a similar point: he distinguishes between a footnote's evidentiary, and its authorizing function. ‘Through their authorising functions’, Rodgers writes, ‘citations legitimise the text and establish the author as trustworthy in the discourse community. Evidentiary functions . . . may justify claims of causality, explain terms or experimental operations, or provide evidence in an argument’. Rodgers adds a third, mapping function, which I have not included here, as it is not directly to the point. See Rodgers, J., ‘Whose Citations Are They?’, (2010) 9 Journal of Medical English Education 94 Google Scholar, at 95.

⁹⁴ Paul argues something similar – the idea of recognition – within science citations: ‘[N]aming indicates a sense of community. It indicates that the author believes the readers will recognize the name and its value.’ See Paul, supra note 68, at 199.

⁹⁵ Paul describes something similar in natural sciences, but in the context of the ‘obliteration phenomenon’. Paul, supra note 68, at 202–3, and note 109, below; Nimis discusses how a reference to a single author can start to lead a life of its own, and become an ‘institutional function’. Nimis, supra note 55, at 131 (emphasis omitted); Small likewise refers to the fact that an author cannot control the way in which his or her work is taken up by the discipline. See Small, supra note 12, at 338; Skouteris is more explicit. See Skouteris, supra note 13, at 334.

⁹⁶ He engages with Gilbert's work on citer motivations, and claims ‘citation as a symbol for an idea’, as ‘more nearly universal’; but he does consider ‘the concept symbol interpretation of citation practice’ as ‘complementary’ to other citer motivations. See Small, supra note 12, at 337, 338.

⁹⁷ The literature on citer motivations is enormous, and the basic question is whether it is possible to assess a scholar's reasons for citing in the first place; see B. Cronin, The Citation Process: The Role and Significance of Citations in Scientific Communication (1984), 57. For an example, see Brooks, T.A., ‘Private Acts and Public Objects: An Investigation of Citer Motivations’, (1985) 37 Journal of the American Society for Information Science 34.CrossRefGoogle Scholar

⁹⁸ Cozzens cited in Paul, supra note 68, at 187; see also ibid., at 196: ‘[T]he same citations that reward . . . also function as part of [a] community's rhetorical practices.’

⁹⁹ Cozzens cited in ibid., at 187.

¹⁰⁰ See also Kessler and Werner, supra note 30, at 805–6, about attempts to silence dissent.

¹⁰¹ As Gilbert states: ‘Referencing is essentially a device for persuasion . . . certain references will vary in their power of persuasion according to the opinions and concerns of those who are to be persuaded.’ See Gilbert, supra note 11, at 119; Bavelas, J.B., ‘The Social Psychology of Citations’, (1978) 19 Canadian Psychological Review/Psychologie canadienne 158 CrossRefGoogle Scholar, at 160.

¹⁰² Paul, supra note 68, at 199; Gilbert, supra note 11, at 116, 119; more explicitly, Skouteris, supra note 13, at 334. Thanks to one of the anonymous reviewers for pointing out the need for clarification on this point.

¹⁰³ Gilbert, supra note 11, at 117. My claim that these footnotes serve a persuasive purpose only extends to the use of footnotes to support consensus claims – I do not argue that all footnotes in an academic piece are there to persuade a reader. For a critique of Gilbert's work, see White, supra note 15, at 108–11.

¹⁰⁴ Budd, J.M., ‘Citations and Knowledge Claims: Sociology of Knowledge as a Case in Point’, (1999) 25 Journal of Information Science 265 Google Scholar, at 270. He later adds the caveat that ‘an author cannot possibly provide textual evidence of epistemic links to every potentially relevant piece of previous work’ (ibid., at 271), but this comment does not specifically pertain to a ‘list’ of references.

¹⁰⁵ Nimis, supra note 55, at 118, 119 (emphasis in original).

¹⁰⁶ The only change between the 1994/1995 and the 2002 edition is an update to the 2001 (third) edition of Dinstein's War, Aggression and Self-Defence. The same has been done for the 2012 Commentary which refers to the 2012 (fifth) edition of Dinstein's book. Moreover, the 2012 edition of Simma's Commentary omits Virally's chapter in Cot and Pellet as well as the 1962 article written by Dahm.

¹⁰⁷ Nimis, supra note 55, at 119.

¹⁰⁸ Ibid., at 122. Similarly, Bensman, J., ‘The Aesthetics and Politics of Footnoting’, (1988) 1 International Journal of Politics, Culture, and Society 443 CrossRefGoogle Scholar, at 445–6.

¹⁰⁹ This is reminiscent of the ‘obliteration phenomenon’, where the author of an idea is no longer cited, as the idea itself has become common knowledge. See Garfield in Paul, supra note 68, at 203; White, supra note 15, at 104; Rodgers, supra note 93, at 95. This does not apply here, however, as Randelzhofer or any other of the frequently cited authors are not the ‘creators’ of the idea of consensus.

¹¹⁰ See also Grafton, supra note 91, at 104.

¹¹¹ Hyland, supra note 14, at 168.

¹¹² Thanks to one of the anonymous reviewers for pointing out the need for clarification on this point. It is interesting to note that, in the discussions on the original proposal for the PCIJ Statute, the President of the Committee of Jurists suggested, inter alia, the work of scholars to ensure the application of ‘objective justice’ by the judges. See Procès-Verbaux of the Proceedings of the Committee, supra note 1, at 324.

¹¹³ Watson, supra note 78, at 637.

¹¹⁴ M. Koskenniemi (ed.), Sources of International Law (2000), xvi (emphasis added).

¹¹⁵ This is precisely Watson's complaint; see Watson, supra note 78, at 635–7.

¹¹⁶ This is similar to what Ken Hyland describes in his first chapter; see particularly Hyland, supra note 14, at 1.

¹¹⁷ M. Koskenniemi, ‘Between Commitment and Cynicism: Outline for a Theory of International Law as Practice', in, Collection of Essays by Legal Advisers of States, Legal Advisers of International Organizations and Practitioners in the Field of International Law (1999), 495–523, at 512–23. See also Cohen, H.G., ‘Lawyers and Precedent’, (2013) 46 Vanderbilt Journal of Transnational Law 1025 Google Scholar, at 1038–9.

¹¹⁸ See Hyland, supra note 14, at Chapter 2, on the construction of knowledge through citation.

¹¹⁹ Hyland also states that ‘discourses can reproduce unequal power relations through the ways in which things and people are represented and positioned’. See Hyland, supra note 14, at 156.