The Precedential Value Case Law of the WTO Dispute Settlement Mechanism

by Zihan Liu*



The Dispute Settlement Mechanism (DSM) of World Trade Organization (WTO), which is one of the important pillars of the WTO, improves the security and predictability of multilateral trade. However, the US has raised the issue of the status of precedent within the WTO DSM. WTO dispute settlement reports carry high precedential weight in judicial practice, and subsequent practice can be departed from only if there is a cogent reason;[1] the US regards such departures as precedent. Therefore, they constitute a systemic issue in WTO reform. This paper contends that the nature of WTO dispute settlement reports should be regarded as jurisprudence constante, in lieu of precedent, providing strong guidance to the Appellate Body (AB) and the panel compared to case law, but have no binding force for the subsequent cases; this is from the French legal system and is quite different from common law system. At the same time, this paper contends that the incorporation of jurisprudence consistante helps to address the concern of the US while continuing to preserve the judicial independence of the AB.



The Statement by the US on the Precedential Value of Panel or AB Reports was submitted at the meeting of WTO Dispute Settlement Body Geneva on December 18th in 2018. It asserted that the AB broke the rules of the Dispute Settlement Understanding and the WTO Agreement, and gave the authority in precedent to its dispute settlement reports, especially the interpretation of the treaties involved.[2] The US criticised the practice of the AB of considering the interpretation of the treaties as a precedent that is binding on subsequent cases. It used as an example the case the US – Stainless Steel (Mexico) (2008), one of twelve cases from 2004 to 2016. In this case, the panel disagreed with the interpretation of the provision in the well-established case law of the AB, but it agreed with the interpretation set out by the US. In contrast, the AB believed that the panel should take into account the legal interpretation embodied in an adopted panel and the AB reports, and is not free to disregard the legal interpretation.

Therefore, the US contends that the practical effect of this case is to attach significance to the role of precedent. It mentioned that the AB should not treat its adjudication as a precedent, which would go beyond the authority granted to it by WTO members.



The panel in the case the US – Stainless Steel (Mexico) (2008) said that it had “no option but to respectfully disagree with the line of reasoning developed by the AB regarding the WTO consistency of simple zeroing in periodic reviews”[3]. The AB mentioned in its report that it was ‘deeply concerned’ about the decision of the panel, which differed from the well-established case law of the AB, noting that the panel’s decision had “serious implications for the proper functioning of the WTO DSM”.[4] The AB stated:

AB reports are not binding…. Ensuring ‘security and predictability’ in the DSM implies that, absent ‘cogent reasons’, an adjudicatory body will resolve the same legal question in the same way in a subsequent case.[5]

According to the report of the AB, the adopted reports control subsequent practice[6], unless there are cogent reasons to depart from this assumption. Thus, if there is no cogent reason to do so, the panel should not depart from the legal interpretation embodied in the adopted reports of the panels and the AB, and should have consistency when resolving the same legal question.

This prompts the question: Is there any way to flesh out the precise elements of “cogent reason”?

The AB has proposed the condition of “cogent reason” in US – Stainless Steel (Mexico) (2008), but has not specified what constitutes a “cogent reason”. The panel in the China – Rare Earths (2014) stated, however, that it “may be understood as referring generally to a high threshold”.[7] Later that year, in the US – Countervailing and Anti-Dumping Measures (China) (2014), the panel defined the concept and the scope of “cogent reason” in the following terms:

To our minds, ‘cogent’ reasons, i.e. reasons that could in appropriate cases justify a panel in adopting a different interpretation, would encompass, inter alia: (i) a multilateral interpretation of a provision of the covered agreements under Article IX:2 of the WTO Agreement that departs from a prior AB interpretation; (ii) a demonstration that a prior AB interpretation proved to be unworkable in a particular set of circumstances falling within the scope of the relevant obligation at issue; (iii) a demonstration that the AB’s prior interpretation leads to a conflict with another provision of a covered agreement that was not raised before the AB; or (iv) a demonstration that the AB’s interpretation was based on a factually incorrect premise.[8]

Although the definition of “cogent reasons” provides the possibility of not following a prior AB’s decision, the panel did not provide other useful information or discussion except these reasons; therefore, the scope of cogent reason remains unclear. Moreover, the AB has not previously defined  what constitutes a ‘cogent reasons’ to depart from prior interpretations , [9] and it remains unclear whether future panels will do so.



In my opinion, the statement of the US in which it criticised the WTO’s AB’s regarding its rulings as precedent lacks merit. The arguments of the US are refuted as follows.

First, some WTO members such as China and Brazil disagree with the US on this issue. After the US presented its critique, many WTO members took the floor to comment that there was no allowance under the DSU or the WTO Agreement giving the AB authority to issue rulings that set a binding precedent, and questioned whether the AB was asserting such an authority.[10] They also said that the “cogent reasons” pronouncement that the US claimed the AB was asserting in Stainless Steel was not correct, and that in fact, the AB had said in the same ruling that it “is well settled that the AB reports are not binding, except with respect to resolving the particular dispute between the parties”.[11] There is a difference between previous rulings setting binding precedent and previous rulings serving as guidance for the AB and panels.

Second, the US selectively criticised the precedent role of the AB.[12] If the US thinks that the AB carries precedential weight in US – Stainless Steel (Mexico) (2008), then it carries precedential weight at the beginning of the WTO’s dispute settlement process. For example, in US – Gasoline (Venezuela) (1996), the AB sided with the US claim and overturned the panel’s interpretation of treaties in which the panel found that the measures involved a violation of of GATT Article 20(g) by the US; it then created a new “two-step” analysis method in interpreting the Article.[13] This interpretation was affirmed and applied in US – Shrimp (1998),[14] which subsequent panels and the AB continued to cite frequently over the next two decades, and about which the US said nothing. Thus, it is fair to say that the critique of the US is selective.

Finally, there are many de facto precedents in WTO dispute settlement procedures when a judicial decision provides authority or prescriptive principles for subsequent cases with the same or very relevant legal issues. Almost every panel or AB report in the DSB published reports cites the prior reports of the panels and the AB. For instance, Japan – Taxes on Alcoholic Beverages (1996), US – Gasoline (1996), EC – Bananas III (1997), and US – Softwood Lumber (2001), among others, were cited in more than 40 reports concerning the WTO DSM.



(1) Benefits in the Role of Precedent

First, carrying precedential weight in a WTO dispute settlement report can fill in gaps left by WTO rules and ensure the continuity, stability, and predictability [15]of WTO rules, as an adjudicatory body is much more likely to resolve the same legal question in the same way in a subsequent case.[16] The rules of the WTO are more abstract, general, and even vague, as when a deal is badly needed for political reasons, but a substantive agreement cannot be reached, negotiators sometimes opt for a vague arrangement that permits conflicting interpretations. The implicit decision in such instances is that the matter can be addressed further in future disputes while providing the immediate appearance of the accord that is required for political reasons.[17] But it still needs specific interpretation when applying the rules of the WTO in a specific dispute, so that it can be consistent and exercisable. Regarding reports of the panels and the AB as precedents is to apply and make a reasonable interpretation of WTO rules, which can help the panels and the AB find guidance from prior cases.[18] This then allows them to correctly apply abstract and generalised WTO rules to the settlement of specific disputes and ensures the consistent interpretation and application of WTO rules.

Second, the role of precedent can strengthen the persuasiveness of the WTO dispute settlement reports’ interpretation and reasoning, making them more convincing to the parties and thus enhancing the enforcement of the WTO dispute settlement reports. As Edgar Bodenheimer writes in Jurisprudence: The Philosophy and Method of the Law, “The authority and power of the jurisprudence does not come from the jurisprudence itself, but what is hidden behind the jurisprudence.[19] The power that makes the jurisprudence effective is not the judge’s will or order, but the inherent advantages of the principle or the reality of a habit embodied in a jurisprudence.[20]

Either the principle or the reality of a habit can enhance members’ confidence in the process of solving disputes under the WTO DSM. WTO members can predict legal effects before taking actions and confirm their rights, obligations, and duties, which enhances the enforcement of the WTO dispute settlement reports.

(2) Disadvantages in the Role of Precedent

Although the use of precedent has some benefits, it also has a number of disadvantages that could erode the legitimacy of the WTO DSM.

First, regarding the WTO dispute settlement reports as binding precedents is unfair to developing countries.[21] Developed countries, especially the US, are actively participating in the WTO dispute settlement, and can thus apply the rules of WTO through precedents in a way that serves their interests. This mechanism has the potential to put smaller and poorer countries, which do not have the legal capacity to participate intensively in the system, at a disadvantage in the legal process. [22] Their participation is limited because numerous interpretations already made by the panels and the AB are treated as binding precedents that benefit the interests of developed countries. [23]

Second, giving AB reports strong precedential weight would propagate any judicial mistake that might occur in one case over time. This has been portrayed by reference to by Maggi and Staiger’s (2011) cynical views on the DSB’s mandate to maximise the joint welfare of the disputing parties. [24] Professors Beshkar and Chilton also agree with it. [25] If there are mistakes in precedents, it is possible for subsequent cases to follow them, which hinders the development of the DSM.

Third, regarding the report of the AB as a binding precedent will increase both the frequency and the cost of disputes. Although an emphasis on precedent could reduce the uncertainty of disputes, reducing their frequency, it could, paradoxically, also lead more-frequent disputes, as to make the precedent established in the dispute settlement report more favorable to a given party, that party will repeatedly interact. As disputes emerge that conflict with the precedent established in dispute settlement reports, the likelihood of litigation and cost of lawsuits, as well as of instituting any changes to prior agreements, may increase. Because judges have the power to create binding precedents for the development of law, it is a matter of course for them to change the relevant law to adapt to changing circumstances. In fact, judges can invite legislators to pass new legislation if they disagree with a certain decision in some cases.[26] In general, however, overturning the agreements they made before requires consensus. The result is that the cost of treating WTO dispute settlement reports as precedents may be higher than that of similar actions in domestic settings.[27]

To sum up, the use of precedent in the WTO dispute settlement mechanism contains more disadvantages. Thus, this paper proposes that the AB’s decisions could be regarded as jurisprudence constante, as explained below.



The notion of iurisprudence constante derives from civil legal systems, while the notion of ‘precedent’ connotes the case law formed by common law systems. [28]In the view of this paper, WTO dispute settlement reports by their nature are better considered as jurisprudence constante. The book Legal Theory, Sources of Law and the Semantic Web explains jurisprudence constante as “basically a weakened formulation of stare decisis: the longer a series of identical decisions in identical cases based on the same legal rule by the courts becomes, the more determinative the decision found in the series becomes for future cases”[29].

The WTO Analytical Index: Guide to WTO Law and Practices, which is regularly updated by the WTO secretariat, uses the title of jurisprudence and treats the interpretation involved as jurisprudence constante to guide subsequent cases.[30] In addition, the panel of Brazil – Desiccated Coconut (1995) said that panel reports are useful and persuasive guidance in relevant cases. The panel of Japan –Taxes on Alcoholic Beverages (1995) also said that although unadopted panel reports have no validity in GATT or WTO, panels can find useful references in the analysis of unadopted panel reports. Therefore, my opinion is that it is better to consider the nature of WTO dispute settlement reports as jurisprudence constante than as precedent.

Treating the interpretation of treaties as jurisprudence constante is not unique to WTO dispute settlement. In fact, some international adjudication bodies, such as ICJ and ICSID, do so. For example, ICJ emphasised in the judgment of Temple of Preah Vihear (Cambodia v. Thailand) in 1961 that “[i]n so doing, the Court must apply its normal canons of interpretation, the first of which, according to the established jurisprudence of the Court, is that words are to be interpreted according to their natural and ordinary meaning in the context in which they occur”[31]. ICJ used the word “jurisprudence” as the basic principle of treaty interpretation. Another example is the ICSID case Daimler Financial Services AG v. Argentine. ICSID mentioned in this case that

[w]hile the Tribunal is not bound to follow the example of prior tribunals, it can find no justification either in the text of the German-Argentine BIT or in general international law to depart from the overwhelming jurisprudence constante that has emerged around this particular legal question[32].

Ultimately, WTO dispute settlement reports are not binding precedents, nor can they increase or decrease the rights and obligations under agreements. Although panels and the AB can cite prior reports, they can do so only for guidance. Although prior reports are not precedents and the panels and the AB have no requirement to follow them in subsequent cases, they should be taken into full account in subsequent practice, and the panels and the AB should follow the consistency principle, which is the key point of jurisprudence constante. Although jurisprudence constante may have a precedent effect, it is it is, in fact, merely strong guidance for subsequent judgments.



Reasons for the US’ objection about the nature of WTO dispute settlement reports are insufficient, which could confirmed from the following three grounds: the comments from other members, the historic case in the US, and cases in the practice of the WTO. It is unnecessary to treat the AB’s decisions as binding precedents, as the use of precedent in the WTO dispute settlement mechanism contains more disadvantages.. Considering it as jurisprudence constante is a better course.


*Attorney-at-law (PRC Bar). LL.M, University of Edinburgh. ORCID: 0000-0002-2845-6013

[1] Appellate Body Report, US – Stainless Steel (Mexico) (WT/DS344/R, 2008) para.160.

[2] Statement by the United States on the Precedential Value of Panel or Appellate Body Reports (Geneva 2018) para 3.

[3] Panel Report, US – Stainless Steel (Mexico) (WT/DS344/R, 2008) para 7.106.

[4] Ibid. para. 162.

[5] Ibid., para. 158-160.

[6] D Palmeter and P C Mavroidis, “The WTO Legal System: Sources of Law” (1998) 92 AM.J. INT’LL 398.

[7] Panel Report, China – Rare Earths (United States) (WT/DS433/R, 2014) paras 7.59–7.61.

[8] Panel Report, US [FIGURE DASH] Countervailing and Anti-Dumping Measures (China) (WT/DS449/R, 2014) paras 7.317

[9] B Mostafa and C. Adam, “Revisiting Procedure and Precedent in the WTO: An Analysis of US – Countervailing and Anti-Dumping Measures (China)” (2015) 728 Coase-Sandor Working Paper Series in Law and Economics 11.

[10] Supra note 3.

[11] Ibid.

[12] J Bacchus and S Lester, “Of Precedent and Persuasion: The Crucial Role of an Appeals Court in WTO Disputes” (2019)74 CATO Institute 2.

[13] US v Venezuela 1996 DSR 22.

[14] US v India 1998 DSR 118-110.

[15] G Gilbert, “The Use of Precedent by International Judges ad Arbitrators” (2012) 2 JIDS 6.

[16] Supra note 9, at 15.

[17] B M Hoekman and M M Kostecki, The Political Economy of The World Trading System


[18] L Simon, “What Is the Precedential Value of Appellate Body Reports?” (2018) 3 International Economic Law and Policy Blog, available at

[19] E Bodenheimer, Jurisprudence: The Philosophy and Method of the Law (1962) 390.

[20] Ibid.

[21] Supra note 9, at 17.

[22] Ibid.

[23] Ibid.

[24] Panel Report, US-CVD and AD Measures (China) (WT/DS379/12/Add.7) para. 7.317.

[25] Supra note 9, at 17.

[26] D Palmeter and P C Mavroidis, “The WTO Legal System: Sources of Law” (1998) 92 AM.J. INT’LL 398.

[27] Ibid.

[28] Z Naigen, “The Precedential Value or Treaty Interpretation of Appellate Body” (2019) 2 International Economic Review 45.

[29] A Roer, Legal Theory, Sources of Law and the Semantic Web (2009), 246.

[30] WTO Analytical, WTO Analytical Index (Guide to WTO Law and Practice), 1995-2011 (June 2012).

[31] International Court of Justice Reports, Temple of Preah Vihear (Cambodia v Thailand): Preliminary Objections para 32.

[32] Daimler Financial Services AG v. Argentine, ICSID Case No. ARB/05/1, Award, 22 August 2012, para 34.













Dancing into Conflict: TikTok, National Security and the WTO

by Alexander R. Kerr Alvarez* 

[1] Introduction

In 2018 the world was rocked by the dramatic rise of the video streaming platform TikTok. Its simple format of easily digestible sub-60 second videos has driven it to be the most downloaded app on the Apple App Store in late 2020. It has become a pop cultural tour ’de force with over 800 Million active users and 2.6 billion global downloads as of January 2021.[1] Its popularity has given rise to a new swath of celebrities, viral trends, and methods of political activism. Central to TikTok’s success is its algorithm: a tightly kept secret, that uses bulk meta-data collection to tailor users personal feed to content that they would find entertaining; this drives engagement and watch time- as well as personalizes advertisements for its users. Its virality has also brough scrutiny from tech communities, privacy activists, and governments, who have expressed concern over the app’s extraction of user’s personal data. Data miners and researchers have found that the app collects information beyond that which is allowed by the Android and Google terms of service- hiding some of its data collection behind layers of encryption. This includes, but is not limited to, geolocation tracking (even when the app is not in use), period copying of the users ‘clipboard’, internet search histories, type of phone the app is on, and cookies.[2] These concerns, paired with TikTok’s parent company ByteDance being based in the Peoples Republic of China (hereafter PRC or China) has caused significant worry across both the private and public sectors as to its seemingly benign nature. [3] In late 2019 the United States Army published a research paper “consider[ing TikTok] a cyber threat”.[4]

In May 2020 India banned TikTok and 59 other apps from doing business in India in a purge of Chinese apps from its domestic markets, following skirmishes between it and China on the Line of Actual Control (hereafter LAC) between their borders. The US followed suit in August, ordering that TikTok be added to the International Economic Powers Act (hereafter IEEPA) entities list, banning ‘any transaction’ between ByteDance and its American subsidiary, but has run into complications.[5] In response to this China has asserted their intent to seek redress at the World Trade Organization (hereafter WTO) over the banning of these apps, stating that these were ‘clearly inconsistent with WTO rules…[and] violate the basic principles of the multilateral trading system’.[6] In response both India and the US cited national security concerns as a principal basis for their banning of TikTok.

This paper will explore the likelihoods of both the United States’ and India’s success in a potential WTO dispute via an examination of the national security exception found in Article XIVbis(b) of the General Agreement on the Trade of Services (hereafter the GATS). To do this it will utilize the “objectivity” and “subjectivity” tests established by the Panel in Russian- Measures Concerning Traffic in Transit and apply them to Article XIVbis of the GATS.[7] Although Russia-Traffic in Transit relates to Article XXI of the General Agreement on Trade and Tariffs (hereafter the GATT), it is generally considered that Article XXI and Article XIVbis of the GATS are analogous, as they are nearly identical, and it is likely that the Panel would follow a similar route for Article XIVbis.[8] This paper will show that India is likely to succeed in its invocation of national security, however the US is unlikely to be successful. It would likely not pass the objective or subjective tests established in Russia-Traffic in Transit; as it is unlikely that an economic conflict would meet the standard of an “emergency in international relations”. Further this article will be working under the assumption that the banning of TikTok complies with US and India domestic law and any preferential trade agreements between them and China.

To make this argument, this paper will be divided into four major sections. The first will explore the trade of data flows and digital services under the WTO framework, discussing why the GATS is the relevant treaty for discussing the banning of TikTok. Then it will explore whether India and the United States would be successful in claiming the national security exception found in Article XIVbis of the GATS and per the tests established in Russian- Traffic in Transit.

[2] The Trade of Data Flows and the WTO

Two issues must first be explored prior to a discussion of whether the national security exception was lawfully invoked: 1) what, if any, treaty does TikTok fall under, and 2) if under the protection of a treaty, does the WTO have jurisdiction to intervene in the dispute. Broadly the WTO outlines that all tradable things can be categorized into one of three vertical silos: 1) a goods 2) a service or 3) intellectual property or IP, each with their own respective treaty governing it: The GATT for Goods, the GATS for Services, and the The Agreement on Trade-Related Aspects of Intellectual Property Rights for IP. Apps, data flows, and digital services have not been formally placed into one of the three silos -like traditional industries like steel or banking-, as the WTO texts make no explicit reference to the Internet, digital communication, and other relevant technologies, which creates ambiguity as to the role that the treaties have over the trade of data. To rectify this, the 1999 Council on Services found that much of the activities associated with E-Commerce and Digital Services would fall under the purview of Services and thus would thereafter be governed under the GATS framework.[9] Subsequent decisions pertaining to trade of electronic services, such as e-commerce, have worked under the assumption that digital services fell under the GATS.[10] In the US-Gambling report the Appellate Body concluded that the cross-border supply of services covered under the GATS encompassed all possible means of supplying a service between WTO members. This implies, though not outright says, digital services as well.[11] This was followed up in China-Publications and Audiovisual Products and China-Electronic Payment Services which explicitly state that the distribution of services, as covered in the GATS, includes online delivery and electronic payment services.[12] Further the 1999 Progress Report on the Work Programme on Electronic Commerce states that the “GATS is technology neutral” meaning it does not distinguish between the “different technological means” that “services may be supplied”.[13] However, a final confirmation on what treaty should govern digital services is still ongoing as of 2020.

For an exception to be invoked, it is required that the in-question measure be in violation of the GATS.[14] Although the GATS and the WTO texts in general do not make explicit reference to barriers to the trade of data flows as restrictions of trade, many scholars have argued that because the trade of data is governed by the GATS it can, therefore, be inferred that the WTO has jurisdiction over disputes arising from measures which restrict data flows.[15] Measures restricting cross-border data flows are analogous to restrictions of the cross-border supply of traditional services as they would unduly restrict market access for foreign services and be in violation of National Treatment obligations.[16] With these having knock on effects into other areas of trade.[17] Restrictions of this nature include: Data-localization, conditional flow regimes, local processing requirements, and bans.[18] Thus showing that the measures taken by Indian and the US governments would fall under the GATS framework and be in violation of it; creating a dispute where an exception may be invoked.

[3] The National Security Exemption

Central to the WTO’s mission is the reduction of -and eventual abolition of- tariffs, trade barriers, and other protectionist measures between its member states in order to rationalize and liberalize the global trading regime; facilitating the free flow of capital, goods, and services. However, member states can deviate from these goals in the pursuit of non-economic political or policy objectives such as public health, prevention of fraud, public morals, and data security. This is enshrined into Article XIV and Article XIVbis of the GATS. Because this dispute has arisen, chiefly, from national security concerns it is likely that India and the US would invoke Article XIVbis of the GATS which creates an exception to a state’s regular GATS obligations if that state feels that its national security may be threatened. India has already indicated that they are looking into invoking it if a WTO dispute were to arise. [19] Article XIVbis states:

  1. Nothing in this Agreement shall be construed:

a) to require any Member to furnish any information, the disclosure of which it considers contrary to its essential security interests; or

b) to prevent any Member from taking any action which it considers necessary for the protection of its essential security interests:

i) relating to the supply of services as carried out directly or indirectly for the purpose of provisioning a military establishment;

ii) relating to fissionable and fissionable materials or the materials from which they are derived;

iii) taken in time of war or other emergency in international relations;

c) to prevent any Member from taking any action in pursuance of its obligations under the United Nations Charter for the maintenance of international peace and security.

2. The Council for Trade in Services shall be informed to the fullest extent possible of measures taken under paragraphs 1 b) and c) and of their termination.

Its intent, like that of its near identical twin in Article XXI of the GATT, is to preserve the right of states to determine their “essential security interests” and create a mechanism in which states can established economic barriers of any kind against those that challenge its interests. It has a lower threshold for invocation, requiring only that a measure taken be “considered necessary for its protection”. While the general exceptions found in Article XIV require a strict examination which ‘weighs and balances’ the policy objective, impact, and whether an alternative measure is better suited.[20] Article XIVbis has yet to be invoked in any proceedings but has been rhetorically invoked by states when imposing cybersecurity regulatory measures, most recently with the Vietnam’s 2018 Cybersecurity Law,[21] and in disputes that did not reach the Dispute Settlement Body.[22]

Scholars often argue that the national security exception provides a carte blanche for protectionism and is ripe for abuse.[23] This is because unlike the other exceptions in Article XIV, Article XIVbis does not explicitly prohibit arbitrary or unjustifiable discrimination; requiring only that member states “consider” that their security interests are endangered. This creates a self-judging standard wherein only that state can determine its essential interest and thus can plausibly label anything as contrary to its interests. This was compounded by a general reluctance on the part of the WTO to intervene in situations where the national security exception had been invoked- though they have previously indicated an interest in limiting the exception.[24]

Though there has been a reluctance by member states to abuse these; Article XXI of the GATT has only been invoked sparsely since its original implementation in the 1947 GATT and Article XIVbis has never been invoked. The only recorded abuse of the national security exception, to date, was in 1975 when Sweden attempted to establish a quota system for footwear, arguing that it was done in the “spirit of Article XXI”, though it chose not to formally invoke it.[25] Yoo argues that this reluctance was due to the contracting parties of the 1947 GATT being primarily members of the western non-communist bloc and thus the majority of trade restrictions during the Cold War were externally focused towards Communist, non-GATT member, states. He argues that a spirit of internal cooperation and “economic and political stability” overrode economic concerns amongst the western alliance.[26] Of the seven disputes arising from Article XXI that occurred during the Cold War: six of them were directed against Communist states,[27] and one was against Argentina- a retaliatory measure by Commonwealth states for Argentina’s invasion of the Falkland Islands in 1982.[28] However the restructuring of the global order from unipolarity to multipolarity- epitomized by the ascensions of China in 2001 and Russia in 2012 has reconstituted the WTO into a vehicle to facilitate great power competition.[29] This, alongside the rise in populist anti-trade governments in the late 2010s, has made trade conflicts and abuse of the national security exception, more likely than at any point in the past; as the spirit of cooperation that existed in the Cold War is being diminished in favour of nationalism and autarky. The first instance of this was the seemingly arbitrary global steel and aluminium tariffs put in place by the United States under the Trump Administration in March of 2018, and the instigation of its ongoing trade war with China later that year.[30]

Into this geopolitical milieu, the Panel in the 2018 case Russia-Traffic in Transit argued that Article XXI(b) was not totally self-judging based upon the phrasing of the Article XXI(b) and its chapeau which qualifies ‘it considers’ as limited to the scenarios outlined in Article XXI(b)’s subparagraphs.[31] Those being:

b) to prevent any [Member] from taking any action which it considers necessary for the protection of its essential security interests

i) relating to fissionable materials or the materials from which they are derived;

ii) relating to the traffic in arms, ammunition and implements of war and to such traffic in other goods and materials as is carried on directly or indirectly for the purpose of supplying a military establishment;

ii) taken in time of war or other emergency in international relations;

The Panel established an objective test, to be applied on a case-by-case basis, to determine whether an invocation of Article XXI was justified, requiring that the measure in-question relate to one of the circumstances outline in Article XXI(b)’s subparagraphs. Further qualifying Article XXI(b), it stated that although a nation has wide discretion in determining what their ‘essential security interests’ are, this is not unlimited and a subjective test determining whether those ‘essential security interests’ are complaint with Article XXI’s chapeau and had been done in ‘good faith’ was required.[32] It argued that ‘essential security interests’ are narrower than regular security interests and thus are only those relating to ‘quintessential functions of the state, namely, the protection of its territory and its population from external threats, and the maintenance of law and public order internally’.[33] It also highlighted the importance of the word ‘necessary’, stating that there must be a relationship between the measure taken and the ‘essential security interests’ which are threatened.[34] It further created a time bar requiring a measure be implemented during an ongoing ‘war or other emergency in international relations’.[35]

Due to the similarities between the Article XXI(b) and Article XIVbis(b), as much of the language in Article XIVbis was replicated from Article XXI, and the similarities in both Article’s chapeau’s it is likely that the WTO would come to similar findings and institute the same or nearly the same standards for a future case arising from Article XIVbis.[36]

[D] Implementation of Article XIVbis(b)

Having established that the panel is likely to review an invocation of Article XIVbis under a similar standard as that established in Russia- Traffic in Transit for Article XXI of the GATT, it is now possible to analyze the likelihood of the success can occur. Both India and the US banned TikTok from doing business within their respective states, showing a clear violation of the GATS National Treatment obligation has occurred- ergo requiring that an exception be invoked for the measure to stand.[37] This section will analyse Article XIVbis(b), via an examination of whether its subparagraph iii) could relate to TikTok. Employing the tests established in Russia-Traffic in Transit to determine whether these were done within the scope of a ‘war or other emergency in international relations’, relate to an ‘essential security interests’ and were done in good faith.

(1) India- Article XIVbis (b)

(a) Factual Background

Since 5 May 2020, China and India have been engaged in a series of ongoing conflicts along the Sino-Indian Border. Fighting started when the People’s Liberation Army (PLA) troops crossed the LAC into Indian Territory, subsequently occupying forty-to-sixty square kilometres previously held by New Delhi.[38] This was preceded by light skirmishes in the Sikkim region in January 2020 between PLA and Indian National Army (INA) Troops and was followed by a melee skirmish on the 15 June between INA and PLA regulars in the Galwan Valley that led to forty-three casualties, thirty-five dead, and ten INA soldiers -including four officers- being taken into detention by the PLA. Since May, it has been reported that there have been one hundred INA casualties ‘mostly minor injuries’ and fifty PLA casualties. In response to the Chinese incursion, a popular movement against Chinese imports has occurred in India. The government has cancelled a number of contracts with Chinese firms and faces further calls for a nationwide boycott.[39] New Delhi responded by instating a number of restrictions on imports from China, including TVs, tyres, air conditioning, and, at of writing, two hundred plus apps have been banned including all apps owned by Alibaba, Tencent, Baidu, Sina and ByteDance.[40]

(b) War or Other Emergency

Sino-Indian border conflicts are not a new occurrence, the two countries have had skirmishes in 1962, 1967, 1987, and 2017 prior to this. Though, there have not been reported deaths from these since the 1962 Sino-Indian War. In Russia- Traffic in Transit the Panel sought to reflect on what the meanings of ‘war’ and ‘emergency in international relations’ were; broadly defining the later as any “situation of armed conflict, or of latent armed conflict, or of heightened tension or crisis, or of general instability engulfing or surrounding a state” and defining the former as a subcategory of emergency. [41] Though broadly defining war as an “an armed attack or conflict”.[42] The objectivity test states that for a measure to be justified it must have occurred during or in response to one of either. It further indicated that an emergency would ‘give rise to defence or military interests, or maintenance of law and public order interests’.[43]

In Russia-Traffic in Transit, the facts lent themselves to an easy analysis: there was/is an ongoing conflict on the border between the two countries in the Donbass region of Ukraine between the Ukrainian Armed Forces and separatist elements in that region known as the Donbass Peoples Republic; and, arguably, there is an ongoing state of conflict between the two since 2014. When Russian forces took part in the annexation of Crimea and are assisting, with lethal and command support, for the separatists in the Donbass.[44] Further United Nations General Assembly Resolution no. 68/262 and UN General Assembly Resolution No. 71/205 directly reference the ongoing situation between the two nations and denounces Russian aggression into Ukraine, describing it as an armed conflict.[45]  In its report the Panel stated that the ongoing conflict directly ‘affects the security of Russia’s border with Ukraine in various ways’ and that the situation constituted an emergency.[46]

Following this logic, it is likely that the Panel would also find that the ongoing situation between China and India would constitute an emergency in international relations or war. This is because, although there have been talks to return to the status quo ante bellum, no such effort has been successful, skirmishes and tit-for-tat economic retribution are still ongoing.[47] The deaths of uniformed regulars during these skirmishes would constitute an ‘armed attack or conflict’ or ‘the heightening of tension’ between them. Further, the broader conflict over the LAC could constitute a ‘latent armed’ conflict as tensions between the two spikes with some regularity. The continuation of tensions between the two nations indicates that a situation which would ‘give rise to defence or military interests…’ is in existence. Thus, showing that it is likely that the Panel would accept that an emergency in international relations that affects India is ongoing.

(c) Essential Security Interests

The panel decision indicates that ‘essential security interests’ are narrower and should be treated higher than usual security interests; stating that they should be related to the ‘quintessential functions of the states’ mentioned above.[48] However, it acknowledged that the essential security interests of its member states vary and are subjective; therefore, only the member state can determine it.[49] Thus it is likely that the panel would prima facie accept whatever essential security interests India felt were threatened.

(d) Good Faith

The Panel held that invocations of the security exception were required to have occurred in ‘good faith’ and had not arisen from arbitrary, non-security, interests.[50] Member states, therefore, are require to ‘articulate sufficiently’ the veracity of their claims of what an essential interest was and demonstrate, to a ‘minimum requirement of plausibility’ that the emergency in international relations directly relates to the concern for the states essential interest.[51]

In Russia-Traffic in Transit the Panel found that an emergency involving armed conflict on one’s border would constitute a situation in which a state’s essential interests had been challenged.[52] Russia showed that the measures in question[53] directly related to the change-in-government that occurred in 2014 and the subsequent invasion of Crimea by the Russian Federation; as the measure taken specifically related to cargo crossing the Ukrainian-Russian border and had traveling through the location of the emergency- the Donbass and Luhansk oblasts.[54] Showing a ‘minimum requirement of plausibility’ that their was a potential that the cargo could be used in a capacity that could threaten Russia’s ‘essential interests’.[55]

The question then, is did the TikTok ban occur in good faith. In Article XIV of the GATS deference is given for states seeking to ensure data protection and individual privacy. It is likely then, that a Panel would bring this consideration over into questions of a conflict, [56] as the widespread distribution of computers and cell phones has altered traditional battlespaces.[57] Cyberspace is now recognized as the fifth operational dimensional alongside land, sea, air, and space.[58] In wartime, breaches of digital spaces could constitute a ‘potential risk’ to those on the battlefield and within society writ large; as data supremacy can confer a battlefield advantage to a warring army.[59] One such example would be geolocation and geotags, data which TikTok collects. In the Donbass conflict in Eastern Ukraine, cell phone towers and geolocation software have been used by Russian-allied separatists to track the movements of the Ukrainian Army for the purposes of targeting artillery and distributing propaganda.[60] Both of these could be achieved through TikTok. Plausibly its geolocation tracking could be used for targeting purposes, calculating troop movements, and uncovering the location or number of personnel within enemy military bases. Concerns of tracking led the Uttar Pashar and Madhya Prades Police Departments and governments banning Chinese apps prior to national ban.[61] Further, TikTok has been used by the Chinese government for propaganda purposes; reportedly, it has shadow banned Indian accounts discussing the border skirmishes, disrupting internal communication channels within India. A similar, though more effective, tactic was used during the Russo-Georgian War where Russian forces used cyberattacks to bring down fifty-four government communication sites as well as local news in Gori, the provincial capital of the province at the centre of the Russian invasion, to disrupt civilian-governmental communications to obfuscate the ongoing invasion.[62] India therefore could plausibly show that the ‘minimum requirement of plausibility’ necessary for the ‘good faith’ test outline in Russia-Traffic in Transit had been sufficiently met as there are plausible situations in which an essential interest could be threatened. Those likely being the ‘protection of its territory’ or ‘population’; as both are threatened by further Chinese incursions beyond the LAC. Therefore, because the measure taken by India happened during a time of emergency and met the plausibility standard, the Panel is likely to rule that it had acted in good faith.

(e) Conclusion

It is likely that the Panel would grant India an Article XIVbis exception due to the ongoing state of conflict between it and China over the LAC between them, and the measure had been done in ‘good faith’- as it related to the emergency and could plausibly relate to the protection of a security interest.

(2) United States of America- Article XIVbis (b)

(a) Factual Background

During the 2016 campaign, then candidate, Donald Trump proclaimed that the largest threat to the United States was China, declaring that it was a currency manipulator and stating his intent to balance the trade deficit between the United States and it. Starting in 2018, President Trump sought to make good on these declarations and established a number of tariffs between the United States and China- including a 50% ad valorem tariff on washing machines, 30% ad valorem tariff on solar panel parts and a 25% ad valorem tariff on Steel. In response China imposed retaliatory tariffs on the import of American goods including Soybeans. Despite efforts to deescalate the trade war, the average tariff rate for imported goods both ways is ~20% and there are numerous cases at present in the WTO over the disputes.[63] Part of this trade war included an executive order banning TikTok on national security grounds.[64]

These events did not occur in a vacuum as China is considered a rising power and a chief rival to the United States. The 2018 National Security Strategy described China as a ‘revisionist power’ and threat, seeking to reorder global power structures away from the US and towards itself.[65] Leading some to argued that the two are engaged in a Cold War.[66]

(b) War or Other Emergency

Unlike India, the United States is not engaged in any situation which could be construed as a ‘war’ or ‘latent conflict’ between it and China; as there has not been a reported situation of Chinese and American regulars having exchanged fire since the Korean War in 1953.[67] This said, there are geopolitical tensions between the two, though whether that can meet the standard of a ‘heightened tension or crisis’ is uncertain. The two are engaged in an economic or trade war with one another. Evidenced by the aforementioned tariffs, trade restrictions, Chinese quasi-annexation of the disputed nine-dash line territory,[68] forced technology transfers, and de-facto embargo of some Chinese Tech companies including Huawei, ZTE, and TikTok by the United States. Economic warfare is a common strategy used to deny an opposing force or geopolitical rival access to strategic economic material (via blockades, privateering, strategic bombing, the freezing of capital assets, etc.) which could be beneficial to it. [69]  As a strategy, it has been used during warfare and strategic competition throughout history and is a recognized dimension of warfare.[70] Ioachimescu-Voinea argues that because a trade war constitutes a deviation from the status quo it could objectively meet the standard of a situation of ‘heightened tensions’ between the US and China. [71] Though this is unlikely, and a Panel would likely not concur.[72] In Russia-Traffic in Transit the panel stated that:

‘…that political or economic differences between Members are not sufficient, of themselves, to constitute an emergency in international relations for purposes of subparagraph (iii). Indeed, it is normal to expect that Members will, from time to time, encounter political or economic conflicts with other Members or states. While such conflicts could sometimes be considered urgent or serious in a political sense, they will not be “emergencies in international relations”’ [73]

Thus, the Panel would likely not consider that the US-China Trade War constitutes an emergency in international relations, because it is unlikely to be objectively seen as anything but a political or economic conflict.

This said, there is disagreement as to whether Russia-Traffic in Transit requires that the situation of ‘war or other emergency’ needs to be between the two parties of the WTO dispute.[74] Although the emergency of Russia-Traffic in Transit occurred between Russia and Ukraine as well as on the frontier between them, the decision does not attempt to establish any spatial requirements or require that the emergency be between the parties to the dispute as part of the objective test. It defines an ‘emergency in international relations’ as happening either in ‘world’ or ‘global politics’.[75] The Panel only mentions territorial proximity in its discussion of ‘general instability’, as the instability must be either ‘engulfing or surrounding a state’.[76] Similarly the Panel nowhere suggests that that the war or emergency must be between the parties to the dispute, only that an emergency be objectively occurring.[77] In EC, Australia, Canada – Trade restrictions affecting Argentina applied for non-economic reasons, the emergency only existed between the United Kingdom and Argentina; Australia and Canada were not involved with the conflict.[78] In the Cold War era GATT cases arising from the measure taken against Czechoslovakia, there as well, existed no direct situation of conflict between it and the US or Peru. [79] Further illustrating a lack of need for the emergency to be between the parties of the dispute. Therefore, it could be argued that the US is engaged in a ‘war or other emergency’. Since the late 90s the US has been engaged in a number of anti-terror operations across the world, which has colloquially become known as the War on Terror. This includes, but is not limited to, military operations in Libya, Niger, Nigeria, Somalia, Yemen, Iraq, Syria, Afghanistan, Pakistan, and the Sahara. These operations include the regular deployment of uniformed troops to battlespaces where they are engaged in lethal combat with hostile combatants leading to the deaths and detention of American personal. The legal basis of this is the 2001 and 2003 AUMFs which establish that a state of hostilities exists between the United States and: Al-Qaeda, Al-Qaeda’s constituent forces, the Ba’athist Republic of Iraq, and irregulars resisting the occupation of Iraq.[80] Therefore, because the objective test established by the panel requires that ‘the action [only] was “taken in time of”’ emergency to the disputed parties, it is likely that the Panel would establish that objectively the United States is involved in an emergency in international relations.

(c) Essential Security Interests

The analysis for this section would be the same as that had occurred for India, with the Panel prima facie accepting that whatever interests the United States thought was being impacted.

(d) Good Faith

The Good Faith test requires that the state invoking the national security exception demonstrate a clear connection between its measure, the emergency in international relations and that a ‘minimum requirement of plausibility’ exist between the measure and the interest which was threatened. Although the United States claims that the Trade War arose out of national security concerns; it is clear that what motivates it are both political and economic disagreements.[81] The Panel considers these to be outside of the realm of acceptable reasons for an invocation of the national security exception and thus cannot constitute an emergency. If the Trade War did constitute an emergency, it would be more likely the Panel might accept it- but it is still be uncertain at best. This is because it is difficult to articulate a connection between the measure and the national security concerns cited by the US as justification for the Trade War. Those being an industrial gap between the USA and China which could threaten US interests in the event of a war. Further, some speculate that the reason the Trump Administration ordered the banning of TikTok arose from TikTok users having disrupted a rally held during Trump’s run for reelection.[82] The reluctance of the government to follow up on litigation within domestic courts that has arisen from the measure and the pressure it put on ByteDance to sell TikTok of America to a regime friendly firm further illustrates that it may have been acting in pursuit of economic and not security goals. However, the inattentiveness of the government may be due to domestic events- specifically, the transition to the Biden Administration, which may have superseded these proceedings, as the Biden Department of Justice were involved in oral arguments for TikTok v Trump in January. Further, if the emergency were the War on Terror, there exists no foreseeable way that a Panel would accept that a ‘minimum requirement of plausibility’ that links the banning of TikTok to the War on Terror would be met. Therefore, it is unlikely that the United States would pass the subjective test established in Russia-Traffic in Transit as the link between TikTok and the United States ongoing conflicts is tenuous at best.

(e) Conclusion

Thus, it is unlikely that the United States banning of TikTok would be accepted by the Panel and the United States would have been seen to be violating its GATS obligations. This is because it is unlikely that a Panel would accept that the banning of TikTok directly relates to whatever emergency in international relations the United States is currently engaged in; as the War on Terror does not involve China and the Trade War between it and China is unlikely to meet the standard. However, this does not preclude the possibility that during a Panel investigation the USA could present evidence of an ongoing emergency between it and China which is not publicly known, that may meet the standards of article XIVbis(b). However, this may not be accepted as in Russia- Traffic in Transit the Panel adopted Russia’s recommendation that an emergency be ‘publicly known’.[83] Further, it is unlikely that the Panel would find that the United States was acting in good faith, as the Trump Administration seems to have been acting out of narrow economic protectionist interests in the Trade War. Therefore, it is likely that the Panel would find that the United States had improperly invoked Article XIVbis(b) and order than the US reinstate TikTok. However, it is unlikely that the United States would accept this seeing as it is long-held American practice to deny that the WTO has the power to judge the national security exception.[84]

[E] Conclusion

International society is converting from the Post-war order into a new age of multipolarity and global integration not seen since the 19th Century. Partly driven by the decline in armed conflict following 1945 and the establishment of the UN and partly driven by a rising tide of populism and economic protectionism, great power conflicts are bound to arise from trade disputes and an era of trade wars may soon be upon us. The Trump Administration’s aggressive implementation of tariffs, quotas, and the banning of TikTok being the opening salvos. This will challenge the WTO order in a profound and yet unseen manner as it will become a political battlefield in which states will engage in economic competition with one another.

Central to this new era will be the national security exceptions in the TRIPS, GATT, and GATS as they provide member states with a comprehensive tool to circumnavigate their regular obligations vis-à-vis their rivals. However, outside of Article XXI of the GATT, the national security exception is under considered. This article hopes to help develop the literature by demonstrating a possible path for how the decision made in Russia-Traffic in Transit may apply in the GATS context by exploring the legality of the banning of TikTok by the US and India and by exploring how the national’s security exception may apply in situations of data flows. However, it cannot be conclusive in its estimations as situations may differ if taken up by the Dispute Settlement Body.


* 4th Year Law and International Relations (LLB Honours) student at Edinburgh University

[1] B Doyle “TikTok Statistics” Wallarro 1 January 2021, available at

[2] K Poulsen & R McMillen “TikTok Tracked User Data Using Tactic Banned by Google” The Wall Street Journal 11 August 2020, available at

[3] Chinese data collection regulations allow for wide discretion by state security to engage in bulk meta data collection on those within China and those whose data is held by Chinese firms. See S Hennessy & C Miransola “Did China Quietly Authorized Law Enforcement to Access Data Anywhere in the World” Lawfare Blog 27 March 2017, available at As well, there are concerns over censorship of content that is critical of the Chinese Communist Party see A Hern “Revealed: How TikTok Censors Videos that do not please Beijing” Guardian 25 September 2019, available at and F Potkin “ByteDance Censored Anti-China content in Indonesia” Reuters 13 August 2020, available at

[4] N Vigolor “U.S. Military Branches Block Access to Tok Tik App Amid Pentagon Warnings” New York Times 4 January 2020, available at

[5] See Exc. Order. 13942, 85 Fed. Reg. 155 (06 August 2020) and R Chesney “Banning TikTok and WeChat: Another Primer” Lawfare Blog 7 August 2020, available at The Trump Administration concurrently has been seeking to facilitate the purchase of TikTok of America by a consortium of Oracle Services and Walmart and the administration has delayed the implementation of the IEEPA order. However, at time of writing, this merger has stalled over disagreements between the Consortium, ByteDance, the US and China over the terms of the purchase, see J Horwitz & Brenda Goh “Beijing unlikely to approve Oracle, Walmart’s TikTok deal: Global Times” Reuters 22 September 2020,  available at The Committee on Foreign Investment in the United States (hereafter CFIUS) has ordered ByteDance to divest itself from TikTok of America by mid-November or face sanctions, though this has been delayed. Occurring concurrently there is a series of proceedings in the DC, Nineth, and Third Circuits seeking injunctive relief by BtyeDance and TikTok creators arguing that the ban was contrary to the first amendment and was ultra vires see TikTok Inc. v. Trump, Civil Action No. 1:20-cv-02658 (CJN) (D.D.C. 27 September 2020) and Marland, Cosette, Rinab et al v Trump Civil Action No. 2:20-cv-04597 (WB)(E.D.P. 30 November 2020).

[6] Reuters “China says US and Indian bans on TikTok, WeChat broke WTO rules” Reuters 5 October 2020, available at

[7] Panel Report- Russian- Measures Concerning Traffic in Transit, WT/DS512/R and WT/DS512/R/Add.1, circulated to WTO Members 5 April 2019, adopted 26 April 2019. (hereafter Russian-Transit).

[8] N Mishra “The Trade- (Cyber)security Dilemma and its Impact on Global Cybersecurity Governance” Journal of World Trade (Forthcoming), Centre for International Law Working Paper Series 19/05, available at  (hereafter Mishra, “Cybersecurity”).

[9] Though this has not been formalized into one of the treaties as there are ongoing disputes to the IP implications of digital services. See, WTO General Council, Work Programme on Electronic Commerce, S/L/74, Adopted on 27 July 1999, para. 9 and WTO General Council, Work Program on Electronic Commerce, IP/C/18, adopted 30 July 1999.

[10] See, Appellate Body Report- United States-Measures Affecting the Cross-Border supply of Gambling and Betting Services, WT/DS285, circulated 7 April 2005, adopted 20 April 2005 (hereafter US-Gambling), Appellate Body Report- China-Measures Affecting Trading Rights and Distributions Services for Certain Publications and Audiovisual Entertainment Products, WT/DS363, circulated 21 December 2009, adopted 19 January 2010 (hereafter China-Publications and Audiovisual Products) and Panel Report- China – Certain Measures Affecting Electronic Payment Services WT/DS413/10,  circulated 16 July 2012, adopted 31August 2012 (hereafter China-Electronic Payment Services).

[11] US-Gambling para. 6287.

[12] China-Publications and Audiovisual Products para. 71209; China-Electronic Payment Services para. 7180 and 7182.

[13] WTO General Council, Work Programme on Electronic Commerce, S/L/74, Adopted on 27 July 1999, para. 4-6.

[14] Panel Report, US-Section 337 Tariff Act (1989), L/6439 – 36S/345, circulated 19 January 1989, adopted 7 November 1989 para. 5.9.

[15] M Francesca Ferracance “Data Flows and national security: a conceptual framework to assess restrictions on data flows under GATS security exception” (2018) 21(1) Digital Policy, Regulation and Governance at 44-70; M Burri, “The Regulation of Data Flows Through Trade Agreements” (2017) 48 Georgetown Journal of International Law at 407-448 (hereafter Francesca Ferracance “Data Flows and National Security”) ; A Mitchell and J Hepburm, “Don’t Fence Me In: Reforming Trade and Investment Law to Better Facilitate Cross-Border Data Transfer” (2017) 19 Yale Journal of Law and Technology at 182-237.

[16] Ibid.

[17] C. Malmstorm “Trade in a Digital World”, Conference on Digital Trade, European Parliament, 17 November 2016, available at

[18] Francesca Ferracance “Data Flows and National Security”

[19] A Sen 2020 “Chinese app ban: India may invoke GATS security exception at WTO” The Hindu Business Line, 04 July 2020, available at: .

[20] Appellate Body Report, EC-Measures Prohibiting the Importation and Marketing of Seal Products WT/DS400/16/Add.7 WT/DS401/17/Add.7 circulated 22 May 2014, adopted 16 June 2014, para. 5.214; US-Gambling para. 292.

[21] M Nguyen 2018 “Vietnam lawmakers approve cyber law clamping down on tech firms, dissent”, Reuters, 12 June 2018, available at:

[22] R Bhala “National security and International Trade Law: What the GATT says, and what the United States Does” 19(2) University of Pennsylvania Journal of International Law (1998) at 263-317.

[23] B Jordan “The WTO versus the Donald: Why the WTO Must Adopt a Review Standard for Article XXI(b) of the GATT”, 37(1) Wisconsin International Law Journal (2020) at 173-205 (hereafter Jordan “WTO versus the Donald”); M Ioachimescu-Voinea “National Security Exceptions in the WTO- A Carte Blache for Protectionism? Part II- US-Steel and Aluminium Products disputes, Improvements of the Security Test, Conclusion” X(1) Law Review(Romania) (2020) at 3-43 (hereafter Ioachimescu-Voinea, “National Security in the WTO”).

[24] GATT Panel Report, United States – Trade Measures Affecting Nicaragua, L/6053, (1986), unadopted, paras. 5.17 and 5.3.

[25] T Cottier and P Delimatsis, “Article XIVbis: Security Exceptions”, in P Stoll and C Feinäugle (eds), Max Planck Commentaries on World Trade Law, WTO-Trade in Services, vol 6 (2008) 329-348.

[26] J Yeong Yoo and D Ahn “Security Exception in the WTO System: Bridge or Bottle-Neck for Trade and Security” 19(2) Journal of International Economic Law (2016) at 417-443.

[27] Nicaragua and Czechoslovakia turned Communist after their accession to the GATT and Yugoslavia ascended in 1965, See GATT Panel Report, US – Issue of export licenses, CP.3/SR22-II/28 (1949); US – Suspension of Obligations between the US and Czechoslovakia, CP.5/5-II/36 (1951); Peru – Prohibition of Czechoslovakian Imports, L/2844 (1954); US- Imports of Sugar from Nicaragua, BISD/31S/67 (1983); US – Trade measures affecting Nicaragua, L/6053 (1985); EEC – Trade Measures taken by the EC against the Socialist Federal Republic of Yugoslavia, L/6948 (1992).

[28] GATT Panel Report, EC, Australia, Canada – Trade restrictions affecting Argentina applied for non-economic reasons, C/W/402 (1982).

[29] H Zimmermann “Realist Power Europe? The EU in the Negotiations about China’s and Russia’s WTO Accession” 45(4) Journal of Common Market Studies (2007) at 814-832 and A Reich “The Threat of Politization of the World Trade Organization” 11 May 2005.

[30] F Li “Trump’s Steel and Aluminum Tariffs, National Security and WTO Law” 2 China and WTO Review (2018) at 273-300 (hereafter Li “Trump Steel and Aluminum Tariffs”); Jordan “WTO versus the Donald”.

[31] Russia-Traffic in Transit para. 7.72.

[32] Russia-Traffic in Transit para. 7.132.

[33] Ibid. para. 7.130.

[34] Ibid. para. 7.63.

[35] Russia-Traffic in Transit para. 7.77.

[36] Mishra, “Cybersecurity”.

[37] China-Electronic Payment Services para. 7.687.

[38] A Tellis “Hustling in the Himalayas: The Sino-Indian Border Confrontation” Carnegie Endowment for International Peace, June 2020, available at

[39] A Dastidar & R Tiwari “Chinese Firms to lose India business in Railways, Telecom” The Indian Express 18 June 2020, available at and N Gamai “‘Software In A Week, Hardaware In A Year’: Magsaysay Awardee Sonam Wangchuk calls for ‘Boycott Made In China’ The Outlook 30 May 2020, available at

[40] TNN, “After TV sets and tyres, government bans AC imports” The Times of India 16 October 2020, available at

[41]Russia-Traffic in Transit para 7.76 and 7.102.

[42]Russia-Traffic in Transit para 7.76 and 7.102.

[43] Ibid. para 7.75.

[44] Russia-Traffic in Transit para. 7.115 and C Ziegler “A Crisis of Diverging Perspectives: U.S.-Russian Relations and the Security Dilemma” 4(1) Texas National Security Review 2020.

[45] United Nations General Assembly Resolution no. 68/262 (2014), A/RES/68/262 (2014), UN General Assembly Resolution No. 71/205, (2016).

[46] Russia-Traffic in Transit para 7.119, and 7.123.

[47] S Miglani & C Cadell “India says troops had ‘minor faceoff’ with China in Sikkim border area” Reuters 25 January 2021, available at and A Krishnan “China slams India’s ban on 43 more apps” The Hindu 25 November 2020, available at

[48] Russia-Traffic in Transit para 7.130.

[49] Ibid. para 7.131.

[50] Russia-Traffic in Transit para 7.134

[51] Ioachimescu-Voinea, “National Security in the WTO”.

[52] Russia-Traffic in Transit para 7.142-7.146.

[53] A requirement that all Ukrainian cargo traveling to third-party states via Russia needed to be rerouted to Belarus for an interim period and receive special authorization to enter into Russia; and the banning of certain classes of goods from entering Russia, Ibid para. 7.140.

[54] Ibid. para. 7.142-7.146.

[55] Ibid. para. 7.27.

[56] Francesca Ferracance “Data Flows and National Security”.

[57] M Schmitt “Taming the Lawless Void: Tracking the Evolution of International Law Rules for Cyberspace” 3(3) Texas National Security Review (2020) at 33-47.

[58] Summary of the National Defense Strategy of the United States (2018), available at

[59]  Mishra, “Cybersecurity”.

[60] Asymmetric Warfare Group, US Army “Russian New Generation Warfare Handbook” (2016), available at

[61] N Baus “Banning Apps violate WTO rules, will affect employment of Indians: Chinese Embassy” The Print 30 June 2020, available at

[62] D Hollis “Cyberwar Case Study: Georgia 2008” Small Wars Journal (2011), available at

[63] C Bown “US-China Tradewar Tariffs: An Up-to-Date Chart” Peterson Institute For International Economics 14 February 2020, available at US-China Trade War Tariffs: An Up-to-Date Chart | PIIE and C Bown & M Kolb “Trump Trade War Timeline: An Up-to-Date Guide Peterson Institute For International Economics 18 December 2020, available at Trump’s Trade War Timeline: An Up-to-Date Guide | PIIE.

[64] Exc. Order. 13942, 85 Fed. Reg. 155 (06 August 2020).

[65] Summary of the National Defense Strategy of the United States (2018), available at

[66] M Zhao “Is a New Cold War Inevitable? Chinese Perspectives on US-China Strategic Competition” 12(3) The Chinese Journal of International Politics at 371-394.

[67] It could be argued that because an armistice is the cessation of fighting not a treaty returning parties to the status quo antebellum, that a conflict situation may exists between them. However, in common practice an armistice functions as peace treaty. See E Stanley “Ending the Korean War: The role of Domestic Coalition Shifts in Overcoming Obstacles to Peace” 34(1) International Security 2009 at 42-82.

[68] The nine-dash line territory overlaps a vital sea lane with 60% of global trade moving through it a year. China claims the territory however the international community disagrees. See South China Sea Arbitration, Philippines v China, PCA Case No 2013-19, ICGJ 495.

[69] E Sand “Desperate Measures: The Effects of Economic Isolation on Waring Powers” 3(2) Texas National Security Review (2020) at 12-37.

[70] Ibid.

[71] Ioachimescu-Voinea, “National Security in the WTO”.

[72] Something she herself notes, Ibid.

[73] Russia-Traffic in Transit para 7.75.

[74] See Ioachimescu-Voinea, “National Security in the WTO” and Li “Trump Steel and Aluminum Tariffs”.

[75] Russia-Traffic in Transit para 7.73.

[76] Ibid. para 7.76

[77] Ioachimescu-Voinea, “National Security in the WTO”.

[78] Though there was not a conclusive decision in this case. The GATT council decided to shelve the dispute-which became irrelevant after the cessation of the conflict and the status quo was restored, see A Reich “The Threat of Politization of the World Trade Organization” 11 May 2005.

[79] US – Issue of export licenses; US – Suspension of Obligations between the US and Czechoslovakia; and Peru – Prohibition of Czechoslovakian Imports

[80] Authorization for Use of Military Force, Pub. L 107-40, 115 Stat. 224 (2001) and Authorization for Use of Military Force Against Iraq Resolution, Pub. L 107-243, 116 Stat. 1498 (2003).

[81] See Li “Trump Steel and Aluminum Tariffs”, Ioachimescu-Voinea, “National Security in the WTO”, and Jordan “WTO versus the Donald”. All of which concluded that the Trade War was done in bad faith and that the measures done as part of it would be unlawful. Also see US Dept. of Commerce, The Effect of Imports of Steel on the National Security- An Investigation Conducted under Section 232 of the Trade Expansion Act of 1962 (2018) available at; specifically ‘displacement of domestic steel by excessive imports and the consequent adverse impact on the economic welfare of the domestic steel industry….’.

[82] However, the CFIUS investigation started before the Administration writ large took interest. See A Brown “Is this the Real Reason why Trump wants to Ban TikTok” Forbes 1 August 2020, available at

[83] Russia-Traffic in Transit para. 7.119.

[84] Jordan “WTO versus the Donald”, R Lighthizer “Statement by Ambassador Robert E. Lighthizer on Retaliatory Duties” 26 June 2018, available at and Dispute Settlement Body, “US Statements at the Meeting of 23 October 2017” pp. 5 and Russia-Traffic in Transit Annex D-10 “Executive Summary of the Arguments of the United States”.


by Ellie Cumberbatch*



On 3 January 2020, American drones targeted a convoy leaving Baghdad International Airport in Iraq, killing ten individuals including Iranian Major-General Qasem Soleimani.[1] Five days later, Iran responded with missile attacks on the Ayn Al Asad air base in Iraq, resulting in the severe brain damage of over one hundred United States military personnel.[2] These attacks occurred amid growing military tensions between the United States and Iran,[3] and both uses of force have been justified under the umbrella of self-defence. The U.S. government initially justified the strike as a measure to deter imminent attacks on United States embassies and personnel,[4] but later stated in a letter to the Security Council that the United States’s right to self-defence was invoked by Iran’s actions in the months prior to the strike.[5] Iran took a similar approach; in its letter to the Security Council the ambassador to the U.N. noted that Iran’s “armed forces have been the subject of armed attacks by the United States”, resulting in the armed forces taking “measured and proportionate action in self-defence”.[6]

However, whilst both states have claimed that the acts were necessary and proportional acts of self-defence,[7] a variety of legal theories have been advanced to support the claim that neither uses of force were consistent with the principles of jus ad bellum. Broadly speaking, contemporary jus ad bellum prohibits the use of force with the exception of Security Council measures and acts of self-defence.[8] The right to self-defence is not without limitations, and state practice post-9/11 has increasingly pushed the limits of this legal framework in an effort to expand its scope to include provisions such as pre-emptive self-defence and the “unwilling or unable” doctrine.[9] It is with this tension in mind that this essay will provide an assessment as to whether the targeted strike on Soleimani and the subsequent Iranian response can be considered lawful acts of self-defence under international law.

(1) Applicable legal framework

As noted by UN Special Rapporteur Callamard, the killing of Soleimani is “the first known incident in which a state has invoked self-defence as a justification for an attack against a State actor, in the territory of another State”.[10] Given this clear demonstration of the blurring of the lines between war and peace, it has generated a great deal of contention as to the appropriate legal frameworks to regulate such use of armed force. Despite both states denying that they are “at war”[11], there has been discord as to whether the United States and Iran can be considered to be in an international armed conflict. Consequently, a holistic examination of the international legal frameworks which regulate these uses of force is necessary for a complete assessment of their legality.[12] However, given the complexity of this assessment, and that the application of key jus in bello provisions is arguably contingent upon whether the conditions for jus in bello are satisfied[13], this paper will narrow its scope to provide an extensive initial analysis as to whether the uses of force by the United States and Iran can be regarded as lawful under the right to self-defence.

The legal framework of jus ad bello is provided for by article 2(4) of the United Nations Charter, which proclaims that “all members shall refrain from the threat or use of force against the territorial integrity or political independence of any state”.[14] This is recognised as a binding principle of customary international law and as jus cogens.[15] However, the Charter provides for two exceptions to the prohibition of force: force authorised by the Security Council, and the use of force in self-defence under Article 51.[16] Given that drones are not a prohibited weapon under international law and targeted killings may be regarded as legal acts of self-defence in some circumstances,[17] the lawfulness of the strike on Soleimani and the subsequent Iranian response must therefore be determined by the degree of compliance with Article 51 and relevant customary international law. Although the rules pertaining to the right to self-defence remain a widely debated topic and are therefore difficult to apply in practice,[18] two broad requirements for the lawful use of self-defence can be identified: (1) the existence of an ‘armed attack’ against the state;[19] and (2) a necessary and proportionate use of force.[20]

As a means of structuring an analysis of the lawfulness of the use of force by both States under jus ad bello, this essay will first provide an assessment of the extent to which both the United States and Iran can be considered to have suffered an armed attack, and whether this justifies the breach of Iraqi territorial sovereignty. Secondly, the paper will then go on to determine whether the action undertaken in response to these “armed attacks” was necessary and proportionate, thus falling within the required limitations on the use of force in self-defence. Finally, drawing upon these determinations, an analysis of the paper’s determination of the legality of the cases at hand will be provided. Overall, this essay finds that both uses of force fail to comfortably fit within the “rubric” of self-defence given the lack of substantial evidence supporting the existence of an armed attack or the necessity of both responses, particularly with regards to the violation of Iraqi sovereignty.



In assessing the lawfulness of the United States’ strike on Soleimani and the subsequent Iranian response, it is first necessary to evaluate adherence to the key requirement of self-defence as provided for under Article 51 of the Charter: the existence of an armed attack.[21] As outlined in the Armed Activities case, Article 51 “does not allow the use of force by a State to protect perceived security interests beyond these parameters”.[22] However, the flexible language of this provision has generated considerable debate regarding the required thresholds for an act to be considered an “armed attack”.[23] This tension has been further exacerbated by state practice attempting to expand this provision to include attacks which have not yet occured- referred to as preemptive self defence- on the basis that states are unwilling or unable to mitigate imminent security threats.[24] The following section will address these issues, analysing the extent to which the actions of the United States and Iran were justifiable acts of self-defence in response to armed attacks, applying both broad and more restrictive interpretations to the Charter’s requirement that self-defence may only be invoked if an “armed attack” occurs.

(1) Justifications for the use of force

The dilemmas which arise from attempting to determine the criteria for an “armed attack” sit at the heart of assessments of the lawfulness of these uses of force. Iran and the United States justified their measures on two ill-defined grounds: the principle of imminence and the “accumulation of events” doctrine. Whilst several scholars consider the expansion of the term “armed attack” to include these grounds for self-defence to be impermissible,[25] there is a broad acceptance that preemptive acts of self-defence taken to avert imminent attacks is legal under Article 51 of the Charter.[26] With regards to the notion of an “accumulation of events”, there is also general support for the argument that a series of events can constitute an armed attack.[27] However, even if accepting the permissibility of these principles, lack of clarity regarding their criteria complicates assessments of legality. Taking these contentions into account, this section will provide an analysis of the extent to which the strikes on Soleimani and the Ayn Al Asad airbase can be considered to be responses to an “imminent” threat or a “series of armed attacks”. Further analysis will then be undertaken to determine the extent to which these “armed attacks” justified the violation of Iraqi territorial sovereignty.

a) Pre-emptive self-defence: the notion of “imminence”

There is a broad acceptance that the pre-emptive use of force in self-defence must be restricted to ‘“imminent threats”.[28] This principle was referenced by the Trump administration, which stated that the strike on Solemani was aimed at preventing “imminent attacks” on United States diplomats and military personnel.[29] However, the lack of explicit criteria for this provision has generated contention pertaining to the required thresholds for imminence. Firstly, scholars of the liberal school of thought contends that measures of self-defence to prevent an imminent attack are permissible if they meet the standard laid out in the Caroline formulation that a threat must be “instant, overwhelming and leaving no choice of means, no moment of deliberation”.[30] Secondly, expansionist scholars contend that a broader understanding of the right to preemptive self-defence must be utilised to “reflect the capabilities and aims of terrorist groups and rogue states”.[31] Recognising that Iran did not provide the justification of an “imminent threat” for its response[32] and it is not admissible “to ascribe to States legal views which they do not themselves formulate”,[33] this section will draw upon both schools of thought to determine whether the United States strike on Soleimani meets the criteria to invoke the right to preemptive self-defence.

Rooted in the liberal interpretation that states are only allowed to resort to the use of force in pre-emptive self-defence when the threat is real and imminent,[34] several scholars have argued that as the Trump administration failed to provide evidence in support of a tangible threat, the strike on Soleimani was unlawful under Article 51.[35] Drawing upon the ICJ’s judgement in Gabcikovo-Nagymaros that “‘imminent’ is synonymous with ‘immediacy’ or ‘proximity’ and goes far beyond the concept of possibility”,[36] it may be reasoned that “self-defence cannot be exercised merely on the ground of speculations”.[37] On this basis, Milanovic has argued that the lack of facts provided by the Trump administration pertaining to the perceived imminent threat “casts serious doubts on whether the various factual predicates for lawful self-defence could be met”.[38] This argument is supported by Callamard, who, drawing upon the Caroline formulation,[39] notes that the United States failed to provide any evidence of a threat “which left no choice of means, and no moment of deliberation”.[40] Consequently, as the United States’s expanded interpretation of the right to self-defence was utilised to counter an unspecified attack which it has not disclosed tangible evidence of, it may be reasoned that the attack constituted “an unlawful exercise of preventive self-defence”.[41]

However, expansionist interpretations of self-defence recognise that there is an argument that the United States’ justification of the targeted strike as self-defence in response to an “imminent threat” is permissible if Solemani was in fact planning attacks on Americans. As Sofaer has noted, given the difficulty of determining the imminence of a threat from ‘rogue’ states which utilise “weapons that can be easily concealed, delivered covertly, and used without warning”[42]; the traditional narrow interpretation is arguably not applicable to such security threats.[43] This argument appears to be supported by Chatham House, which proposes that an assessment of imminence should include whether the state is in possession of weapons of mass destruction and if the threat is likely to come without warning.[44] Given the widely acknowledged military capability of Iran and that it is considered to be a ‘rogue’ state by the United States[45], it may therefore be conceivable that Soleimani did pose an imminent threat in which “any further delay in countering [would] result in the inability of the defending state to effectively defend against or avert”.[46] However, the lack of evidence provided by the United States in support of this argument renders it difficult to make a conclusive judgement in favour of the lawfulness of the strike on this basis.

b) The accumulation of events doctrine

Whilst the legality of the strike on Soleimani on the basis of an ‘imminent threat’ is disputable, both states justified their uses of force on the basis that they had been “subjected to a series of attacks”.[47] As recognised by the Court in the Armed Activities case, there is considerable support that the ‘accumulation of events’ does affect the possibility of exercising the right to self-defence”,[48] raising the possibility that both states could potentially draw upon this doctrine to claim self-defence.[49] However, its applicability to the United State’s use of force is arguably limited by the lack of evidence that the series of attacks were in fact ongoing. Whilst in such cases it is difficult to determine “whether a series of attacks is continuing or whether it has come to an end with the latest attack”,[50] it may be argued that due to the “series of attacks” being intersected with acts of self-defence by the United States, the drone strike cannot be reasonably claimed as an act of self-defence on a cumulative basis.[51] This perspective is reinforced by Haque, who contends that “if one attack is clearly over, then the legal ‘clock’ resets”.[52] For example, as recognised by Ferro, “the United States already responded militarily on 29 December, thereby exhausting the possibility of again invoking the [accumulation of events] doctrine a few days later”.[53] Similarly, in its letter to the Security  Council Iran recognised that in June 2019, “the armed forces of the Islamic Republic of Iran took measured and proportionate military action in self-defence…when a United States unmanned aircraft system, despite repeated radio warnings, violated Iranian airspace”.[54] Thus, the prior acts of self-defence carried out by both states raises considerable doubts as to the applicability of the “accumulation of events” doctrine to either uses of force.

(2)  Violations of Iraqi sovereignty?

Given that both uses of armed force were carried out within Iraqi territory, it is also necessary to assess the permissibility of the use of force within a third state. With reference to the statement in article 2(4) of the United Nations Charter that “all Members shall refrain in their international relations from the threat or use of force against the territorial integrity or political independence of any state”,[55] it is widely recognised that states cannot take measures within the territory of another state without consent.[56] However, as noted by The International Law Commission, there are some circumstances which negate the illegality of attacks conducted on a foreign territory, including acts of self-defence against a State “responsible for its internationally wrongful acts.”[57] Recognising that consent was not provided for either strikes,[58] this essay will assess the extent to which the actions of both states were consistent with international law on the basis of the armed attacks being attributable to Iraq, or Iraq’s unwillingness to eliminate an imminent threat.

a) State responsibility

It is widely recognised that states may use force against a state which is deemed responsible for the armed attacks of a non-state actor. Whilst there are issues in determining state responsibility,[59] clarity can be ascertained from the United Nations General Assembly resolution on the “Definition of Aggression”, which finds that attacks in which that state had a “substantial involvement” constitute acts of aggression by the supporting state.[60] Further clarity is provided in the Nicaragua case, where the Court determined that the supplying of arms by a state to an armed group that then conducts armed attacks on a neighbour does not meet the threshold of “substantial involvement”, and thus does not invoke the rights of self defence.[61] However, the targets in this case were not non-state actors, but instead state actors and military personnel of Iran and the United States. With this in mind, and given that that both states attributed the “armed attacks” to one another and not to Iraq, it may be reasoned that the United States and Iran did not understand Iraq to be responsible for the attacks, meaning that the justification of self-defence was not available for either state to legally violate Iraq’s territorial sovereignty.

b) Unwilling or unable?

As Iran and the United States did not appear to have the right to violate Iraq’s territorial sovereignty on the basis of state responsibility, it is necessary to assess the extent to which the acts of force were permissible under the “unwilling or unable” doctrine. The United States government is of the view that it can legally use force in states which are “unwilling or unable” to take action to eliminate what it perceives to be imminent threats.[62] Despite the legality of this doctrine being heavily contested and international law providing little guidance to determine when it may be invoked,[63] some scholars have drawn upon the Bethlehem Principles to argue that considering Kata’ib Hezbollah’s close ties to Iraq, no consultation or evidence was required to determine that Iraq was unwilling to address the threat posed to United States personnel.[64] With reference to Principle 12, given that there is a “strong, reasonable, and objective basis” that obtaining consent would increase the risk that Iraq would alert the target and render the use of force ineffective[65], the United States was arguably entitled to operate under the assumption that Iraq would be unwilling to take action or provide consent.[66] Thus, it may be reasoned that the close ties between Iraq and Solemani “fundamentally transformed the legal nature of the situation”,[67] entitling the United States to legally take measures within Iraqi territory without concrete proof of Iraq’s position.

However, given that the Bethlehem Principles do not provide concrete legal guidance as to how determinations of unwillingness are to be made, assumptions of legality on this basis are weak.[68] Several scholars have argued that contrary to the interpretations of Bethlehem, it is necessary to provide substantial evidence to demonstrate a state’s unwillingness. For example, Schmitt asserts that “the victim state must first ask the [host] state to meet its legal duty…only if it is unable to do so.. the victim state may act militarily in self-defense”.[69] This interpretation is mirrored by Callamard, who notes that “given the absence of consultation prior to the drone strike…there was no evidence that Iraq was unwilling or unable to cooperate”.[70] In fact, it may be reasoned that the lack of evidence and consultation resulted in a misjudgement of Iraq’s willingness to mitigate the threat, as demonstrated by reports that in June 2020 Iraqi forces arrested Kata’ib Hezbollah operatives to pre-empt an attack on United States targets in Baghdad.[71] Thus, even if accepting the lawfulness of the doctrine, given the lack of consultation by both states prior to the use of force on Iraqi territory,[72] it may be reasoned that there is a lack of the required evidence for Iraq’s unwillingness to be used as a legal basis for self-defence for either state.



In order to assess the lawfulness of the strike on Soleimani and the subsequent Iranian response, it is also necessary to determine the extent to which the force used was in both cases proportionate and necessary. Although these principles are not mentioned in the Charter, the ICJ has established that it is a principle of customary international law that use of force in self-defence must be “proportional to the armed attack and necessary to respond to it”[73], and thus failure to meet either test would render the use of force unlawful.[74] These requirements are of particular importance to ensuring that the use of force is used for defensive rather than offensive purposes, and thus limited to achieving the objective of ending or repelling illegal armed attacks.[75] Recognising that even if the threats posed were recognised to be armed attacks, the force used in response must still have been necessary and reasonable in the circumstances of the case to comply with international law, this section will assess the extent to which it was permissible to use defensive force in response to the attacks, and whether the extent of the force used was justifiable.

(1) Necessity

To meet the requirement of necessity, states must only use force in self-defence when it is necessary to bring an attack to an end, or to avert an imminent attack.[76] Whilst the ICJ’s evaluation of necessity has been limited, guidance may be found in the Oil Platforms case where the Court directly connected the nature of the target to the necessity of the response.[77] Applying this criteria to the cases at hand, an argument in favour of the necessity of the targeted strike can be drawn from statements by the Trump administration that killing Soleimani was required due to his instrumental role in the planning of a future attack.[78] As noted by Milanovic, “considering KH’s close ties to Iraq, their open pledge of loyalty to Iran and the regular meetings with Soleimani, there may be evidence that Iran, through its Quds forces, planned, ordered and controlled the KH attacks on US forces in Iraq”.[79] Evidence in support of this argument can be found in reports that in the months following the targeted attack, Kata’ib Hezbollah did in fact agree to ceasefire operations against United States interests in Iraq.[80] On the basis of a determination as to whether the target was required to bring an attack to an end, it may be reasoned that in this case, striking the key personnel of what the United States designates as a terrorist group may have in fact been necessary.[81]

However, this argument is weakened by the recognition that whilst it is plausible that killing Soleimani did have a tangible impact on the prevention of future attacks, this is purely conjecture given that the United States has offered no evidence in support of this.[82] Further doubt on the necessity of the strike is generated by reports that United States officials were aware that Iran “had not yet approved any plans by the general for an attack”.[83] Operating under the assumption that these reports are correct, there is little evidence that the strike on Soleimani was necessary to avert an imminent attack. This is further supported by a statement by the Army General Milley that “violent plots by Soleimani might still happen despite his death”[84]. Thus, given the reasonable doubt that the strike on Soleimani was necessary to avert an imminent attack, it may be reasoned that the necessity of the attack by the United States is questionable. A similar argument can be applied to the Iranian missile strike; it is difficult to argue that an attack on an air base used to train Iraqi soldiers would be necessary to bring a “series of attacks” by the United States to an end, and support for this can be found in statements that indicated that the attack was in fact an act of “fierce revenge”.[85] Thus, it may be argued that the principle of necessity was not met by either state given the lack of evidence that there were no alternatives other than the use of force on the specific targets to bring attacks to an end.[86]

Further arguments that the United States’ and Iranian attacks failed to meet this requirement are rooted in the acknowledgement that as they were carried out in Iraqi territory, the necessity of carrying out the attacks in this location must also be demonstrated.[87] As recognised by Milanovic, “the US would need to demonstrate that it had to strike at Soleimani when and where it did… and that it could not wait to strike at Soleimani elsewhere”.[88] Given that the immediacy of the attack is disputable, it is questionable that the United States’ strike had to take place in Iraq during the days that Soleimani was visiting. Furthermore, it was not suggested that Iraq was the location of an imminent attack, and Soleimani was liable to international travel which suggests that he could be targeted anywhere else in the world.[89] On this basis, Callamard notes that there is little evidence to suggest that the condition of necessity was met.[90] With regards to Iran, there is also little basis for the argument that it was necessary to strike the Ain Al Asad air base in Iraq, as attacks would not be prevented by such a measure given that the prior use of force had been carried out with drones which are not constrained to singular locations. Thus, it may be determined that neither military attacks were necessary to either put an end to a series of attacks or prevent an imminent threat.

(2) Proportionality

Whereas necessity determines whether armed force may be used to respond to an actual or imminent armed attack, proportionality requires that the defensive response be no more than necessary to defeat the armed attack and remove the threat of attacks in the future.[91] While proportionality is accepted as a requirement for self-defence under customary international law, no consensus has developed among legal scholars with regards to its exact criteria.[92] Whilst some scholars are of the belief that defensive action must be equivalent with the first use of force,[93] most argue that the test for proportionality comes down to whether the defending state used the amount of force which was proportional to repelling the attack.[94] Accordingly, for the purposes of assessing the proportionality of the cases at hand, a mixed model of proportionality which draws upon both tests will be applied.

With regards to a quantitative assessment of whether the armed attacks and the defence were “commensurate in terms of casualties and damage incurred”,[95] whilst it is difficult to determine what either state perceives to be the “first attack”, it may be reasoned that both uses of force were disproportionate in relation to all prior attacks identified. In its letter to the Security Council, the most substantial attacks identified by the United States were sustained by unmanned aircraft and aerial systems, in addition to the siege upon the United States embassy in Baghdad in which there were no casualties.[96] With the scale of these attacks in mind, the United States drone strike which resulted in the death of ten individuals, including one of the most powerful men in Iran,[97] may be interpreted as a disproportionate response in terms of casualties and damage incurred. In Iran’s letter to the Security Council, two major prior attacks pertaining to violations of Iranian airspace by unmanned aircraft systems were identified.[98] Even if the ‘first attack’ identified were the strike upon Soleimani, the missile attack which resulted in severe brain damage of over one hundred U.S military personnel[99] is clearly unproportionate in terms of scale of damage when compared to the prior attacks incurred.

Under the assumption that the defending state is entitled to engage in the extent of force required to repel an armed attack,[100] it may be argued that the United States’s measures were in fact proportional given that previous lighter approaches had failed to successfully avert future threats. For example, Ferro highlighted that on 29 December 2019 the United States bombed bases of the Kata’ib Hezbollah militia to reduce “KH’s ability to launch additional attacks against United States personnel”, but this was proved to be ineffective given the attacks by Kata’ib Hezbollah on United States personnel in the storming of the embassy in Baghdad a few days later.[101] Thus, given the potential capacity of the future attacks given that the United States recognises Iran as a “rogue” state, it may be reasoned that the extent of the measures taken by the United States can be deemed an appropriate pursuit of the military defeat of the aggressor. However, this argument is once again weakened by the United States failing to disclose knowledge of a threat from Solemani that is tangible, immediate and to the United States in particular; this renders assessments of the proportionality of the strike difficult given the lack of evidence provided with regards to what it was countering.



(1)  Concluding remarks on self-defence

As recognised by Dinstein, under no circumstances can the use of force by both parties to a conflict be lawful under the provisions of jus ad bellum simultaneously.[102] However, following a determination of the compliance of the cases at hand with Article 51 and relevant customary international law, this essay finds that there is a lack of credible evidence that either state meets the criteria for their uses of force to be considered lawful acts of self-defence. Firstly, with regards to both the killing of Soleimani and Iran’s missile strike, there is a strong argument that neither strikes meet the requirements that acts of self-defence must be in response to an armed attack, even under the broadest of interpretations. Secondly, the lack of clarity surrounding the purported threats presents significant difficulties in determining the lawfulness of the strike with regards to the requirements of necessity and proportionality in both cases. On these bases, this essay finds that both the United States strike on Soleimani and the subsequent Iranian missile attack clearly demonstrate a lack of “fit” with the requirements of self-defence under both the Charter and customary law.

Additionally, a clearer argument can be established with regards to the illegality of the United States and Iranian uses of force given that both attacks were carried out in the territory of a third state. It may be reasoned that the lack of reasonable grounds for breaching state’s sovereignty to carry out the attacks on Iraqi territory as opposed to within the territory of the parties to the dispute clearly results in a failure to meet the requirements of self-defence pertaining to the existence of an armed attack or threat attributable to a failing on Iraq’s part, and a necessary use of force. On this basis, both the United States strike on Soleimani and the Iranian missile strike may be considered to be unlawful uses of force under the jus ad bellum framework, constituting an impermissible violation of Iraqi territorial sovereignty under the United Nations Charter.

(2)  A reasonable armed reprisal?

As a result of the failure of the cases at hand to meet the requirements for self-defence, it is necessary to assess whether the use of force by Iran and the United States can be considered to be a “reasonable” armed reprisal.[103] Broadly defined as a breach of international law in response to a violation by another state in order to enforce compliance,[104] whereas self-defence is directed towards establishing future state security, reprisals aim to punish past behaviour.[105] Whilst generally recognised as illegal under both the Charter and customary international law,[106] practice of the Security Council suggests that “while reprisals remain de jure illegal, they have gained de facto acceptance”.[107] Thus, assessments as to whether the cases at hand fall within the “rubric” of self-defence or armed reprisal, whilst not necessarily generating different outcomes in terms of lawfulness, may give rise to differing degrees of acceptability amongst the international community. However, as it is generally accepted that armed reprisals can only be invoked against the actual wrongdoing state and not against third parties,[108] it is evident that, due to the measures being carried out in Iraq, the hostilities carried out by both states fail to “fit” within the scope of both self-defence and a reasonable armed reprisal.



This paper has highlighted the failures of both the United States’ and Iran’s use of armed force to meet the requirements to be considered lawful under the jus ad bellum framework. This is due to a lack of credible evidence that Iraq was either responsible for an imminent attack, or unwilling to address a threat upon either the United States or Iran within its territory, which in turn gave rise to a lack of necessity of the strikes. The findings of this study provide an initial understanding as to why the cases at hand appear not to fall within even the broadest interpretations of the right to self-defence or a “reasonable” armed reprisal, as recognised by Callamard, “even the legality of a strike under Article 51 of the Charter does not preclude its wrongfulness under international humanitarian or human rights law”.[109] Consequently, further research is necessary in order to determine whether (1) the situation amounts to an international armed conflict; (2) the targeted killing fulfilled the requirements of necessity and last resort under international human rights law, and; (3) the targets were considered to be legitimate under jus in bello, and the use of force was governed by the principles of distinction, proportionality and precaution.



*LLM International Law student at the University of Edinburgh

[1] W Martel “United States Drone Strike in Iraq Kills Iranian Military Leader Qasem Soleimani” (2020) 313 AJIL 1

[2] M Nichols “At U.N., United States justifies killing Iranian commander as self-defense” Reuters (London,  9 Jan 2020)

[3] AJIL Contemporary Practice of the United States, “Trump Administration’s Iran Policies Raise Questions About the Executive’s Authority to Use Force Against Iran” (2019) AJIL 845

[4] United States Department of Defense Statement by the Department of Defense (Jan 2 2020)

[5] UNSC Letter dated 8 January 2020 from the United States Ambassador to the UN addressed to the Secretary-General (2020)

[6] Ibi

[7] Ibid

[8] J Moussa “Can jus ad bellum override jus in bello? Reaffirming the separation of the two bodies of law” (2008) 90 IRRC 372 at 965

[9] At M Bothe “Terrorism and the Legality of Pre-emptive force” (2003) 40 EJIL 2 at 227-240

[10] HRC Targeted killings through armed drones and the case of Iranian General Quassem Soleimani (UN Doc. A/HRC/44/38 2020) para. 60

[11] L Loveluck “Trump says Iranian military leader was killed by drone strike ‘to stop a war,’ warns Iran not to retaliate’” The Washington Post (Washington, 4 Jan 2020)

[12] HRC Targeted killings through armed drones and the case of Iranian General Quassem Soleimani (UN Doc. A/HRC/44/38 2020) para. 31

[13] J Moussa “Can jus ad bellum override jus in bello? Reaffirming the separation of the two bodies of law” (2008) 90 IRRC 372 at 963

[14] Charter of the United Nations 1945)1 UNTS XVI article 2(4)

[15] Nicaragua v. United States of America (Merits, Judgement) [1986] ICJ Rep 14 ICGJ 112 para. 69

[16] Charter of the United Nations 1945)1 UNTS XVI article 51

[17] J Paust “Use of Armed Force Against Terrorists in Afghanistan, Iraq and Beyond” (2002) 35 CILJ 3 at 538

[18] D Kretzmer “Inherent Right to Self-Defence and Proportionality in Jus Ad Bellum” (2013) 24 EJIL 1 at 262

[19] Charter of the United Nations 1945)1 UNTS XVI article 51

[20] Request for Advisory Opinion by the General Assembly of the United Nations) [1996] ICJ Rep 226 ICGJ 205, para. 41

[21] Charter of the United Nations 1945) 1 UNTS XVI Article 51

[22] Democratic Republic of the Congo v. Rwanda [2005] ICJ Rep I68 ICGJ 31 para. 148

[23] At Y Dinstein, War, Aggression and Self-Defence (2001) 159-191

[24] L Van den hole “Anticipatory Self Defence Under International Law” (2003) 19 AUIL Rev 1 at 74

[25] At E Erikson Legitimate Use of Military Force Against State-Sponsored International Terrorism (1989) 217

[26] HRC Targeted killings through armed drones and the case of Iranian General Quassem Soleimani (UN Doc. A/HRC/44/38 2020) para. 52

[27] Democratic Republic of the Congo v. Rwanda [2005] ICJ Rep I68 ICGJ 31 para. 132

[28] At L Ferro “Killing Qasem Soleimani: International Lawyers Divided and Conquered” (2021) 53 CWRJ

[29] A Chung “United States ‘self-defense’ argument for killing for killing Soleimani meets skepticism” Reuters (4 January 2020)

[30] The Caroline Case [1840] BFSP 1137 (Sourced from D.J Harris Cases and Materials on International Law 1976)

[31] A Sofaer “On the Necessity of Pre-emption” (2003) 14 EJIL 2 at 221

[32] HRC Targeted killings through armed drones and the case of Iranian General Quassem Soleimani (UN Doc. A/HRC/44/38 2020) para. 61(b)

[33] Nicaragua v. United States of America (Merits, Judgement) [1986] ICJ Rep 14 ICGJ 112 para. 226

[34] Ibid 268

[35] At M Milanovic “The Soleimani Strike and Self-defence against an Imminent Armed Attack” (2020) EJIL 1

[36] Hungary v Slovakia Judgement, Merits [1997] ICJ Rep 7 ICGJ 66 para. 54

[37] Y Dinstein, War, Aggression and Self-Defence (2001) 235

[38] At M Milanovic, “The Soleimani Strike and Self-defence against an Imminent Armed Attack” EJIL (2020) 1

[39] The Caroline Case [1840] BFSP 1137 (Sourced from D.J. Harris Cases and Materials on International Law 1976)

[40] HRC Targeted killings through armed drones and the case of Iranian General Quassem Soleimani (UN Doc. A/HRC/44/38 2020)  para. 52

[41] At L Ferro “Killing Qasem Soleimani: International Lawyers Divided and Conquered” (2021) 53 CWRJ 1

[42] Ibid

[43] A Sofaer “On the Necessity of Pre-emption” (2003) 14 EJIL 2 at 209

[44]  Chatham House “Principles of International Law on the Use of Force by States in Self-Defence” (2005) ILPWP para. 4

[45] J.R. Bolton “Remarks to the Conference of the Institute for Foreign Policy Analysis and the Fletcher School’s International Security Studies Program” (2003)

[46]  E Wilmshurst “Anticipatory self-defence against terrorists” in L Herik (eds), Counter-Terrorism Strategies in a Fragmented International Legal Order (2013) 361

[47] UNSC “Letter dated 29 January 2020 from the Iranian Ambassador to the UN addressed to the Secretary-General” (2020); UNSC “Letter dated 8 January 2020 from the United States Ambassador to the UN addressed to the Secretary-General” (2020)

[48] Democratic Republic of the Congo v. Rwanda [2005] ICJ Rep I68 ICGJ 31 para. 132

[49] F Talmon, “The United States Killing of Iranian General Qasem Soleimani: Of Wrong Trees and Red Herrings, and Why the Killing May be Lawful after All” (2002) Bonn Research Papers on Int’l Law 18 at 7

[50] Ibid

[51] At L Ferro “Killing Qasem Soleimani: International Lawyers Divided and Conquered” (2021) 53 CWRJ at 1

[52] At A Haque “United States Legal Defense of the Soleimani Strike at the United Nations: A Critical Assessment”  (2020) JS 1

[53] L Ferro “Killing Qasem Soleimani: International Lawyers Divided and Conquered” (2021) 53 CWRJ at 1

[54] UNSC “Letter dated 29 January 2020 from the Iranian Ambassador to the UN addressed to the Secretary-General” (2020)

[55] Charter of the United Nations 1945 1 UNTS XVI, article 2(4)

[56] N Duhem The Legitimacy of Anticipatory Self-Defence in Combating Transnational Terrorism” (2020) Universiteit Ghent at para. 56

[57] U.N. International Law Commission “Commentary to Article 21: Responsibility of States for Internationally Wrongful Acts” Supplement No. 10 A/56/10 (2001) para. 8

[58] A Ibrahim “Iraqis denounce Iran missiles as ‘violation of sovereignty’’ Aljazeera (Doha, 8 Jan 2020)

[59] C Martin “Going Medieval: Targeted Killing, Self-Defense and the Jus Ad Bellum Regime” in C Finkelstein Targeted Killings: Law and Morality in an Asymmetrical World (2011)

[60] UN General Assembly resolution Definition of Aggression (UN Doc. A/RES/3313 1974), article 3(g)

[61] Nicaragua v. United States of America (Merits, Judgement) [1986] ICJ Rep 14 ICGJ 112

[62] Stimson ‘Recommendations and Report of the United States Task Force on US Drone Policy’ (2014) 10

[63] Chatham House ‘Principles of International Law on the Use of Force by States in Self-Defence’ (2005) ILP WP for an argument in favour of its necessity post 9/11; At Stimson, ‘Recommendations and Report of the United States Task Force on US Drone Policy’ (2014) for an outline of it opening the door for misuse.

[64] At P Labuda “The Killing of Soleimani, the Use of Force against Iraq and Overlooked Jus Ad Bellum Question”’ EJIL (2020)

[65] D Bethlehem Principles relevant to the scope of a state’s right of self-defense against an imminent or actual armed attack by nonstate actors” (2012) AJIL 106 principle 12

[66] Ibid, principle 11

[67] United States of America v Iran [1981] ICJ Rep 3 ICGJ 124 para. 74

[68] C MartinChallenging and Refining the ‘Unwilling or Unable’ Doctrine” (2019) 52 SSRN 230

[69] M Schmitt Drone Attacks under the Jus ad Bellum and Jus in Bello: Clearing the ‘Fog of Law (2010) 9

[70] HRC Targeted killings through armed drones and the case of Iranian General Quassem Soleimani (UN Doc. A/HRC/44/38 2020) para. 78

[71] Center for International Security and Cooperation ‘Kata’ib Hezbollah’ (2021) FSI Mapping Militants CISAC

[72] A Ibrahim “Iraqis denounce Iran missiles as ‘violation of sovereignty’’ Aljazeera (Doha, 8 Jan 2020)

[73] Request for Advisory Opinion by the General Assembly of the United Nations [1996] ICJ Rep 226 ICGJ 205, para. 41

[74] J Gardam Necessity, Proportionality and the Use of Force by States ( 2004) 16

[75] N Duhem “The Legitimacy of Anticipatory Self-Defence in Combating Transnational Terrorism”, Universiteit Ghent (2012) 7

[76] Chatham House “Principles of International Law on the Use of Force by States in Self-Defence” (2005)  ILP WP 7

[77] Islamic Republic of Iran v. United States of America [2003] ICJ Rep 161 ICGJ 74 para. 51

[78] United States Department of Defense “Statement by the Department of Defense” (Jan 2 2020)

[79] M MilanovicThe Soleimani Strike and Self-defence against an Imminent Armed Attack”(2020) EJIL 1

[80] J Davison “Iraqi militias say they have halted anti-United States attacks” Reuters (London, 11 October 2020)

[81] United States Department of State “Terrorist Designations and State Sponsors of Terrorism: Foreign Terrorist Organisations” (Feb 7 2009)

[82] HRC Targeted killings through armed drones and the case of Iranian General Quassem Soleimani (UN Doc. A/HRC/44/38 2020) para. 61(b)

[83] H Cooper “As Tensions with Iran Escalated, Trump Opted for the Most Extreme Measure” New York Times (4 Jan 2020)

[84] M Hosenball “Trump says Soleimani plotted ‘imminent’ attacks, but critics question just how soon” Reuters (London, 3 Jan 2020)

[85] A Rubin “Iran Fires on United States Forces at 2 Bases in Iraq, Calling it ‘Fierce Revenge” New York Times (London, 7 Jan 2020)

[86] O Schachter “Self-Defense and the Rule of Law(1989) 83 AJIL 267

[87] HRC Targeted killings through armed drones and the case of Iranian General Quassem Soleimani (UN Doc. A/HRC/44/38 2020) para. 68

[88] M MilanovicThe Soleimani Strike and Self-defence against an Imminent Armed Attack (2020) EJIL 1

[89] HRC Targeted killings through armed drones and the case of Iranian General Quassem Soleimani (UN Doc. A/HRC/44/38 2020) para. 73

[90] Ibid para. 61(a)

[91] A Deeks “Unwilling or Unable: Toward a Normative Framework for Self-Defense” (2020) 52 Va. J. Int’l L 494

[92] S Etezazian “The nature of the self-defence proportionality requirement” (2016) 3 Journal on the Use of Force and Int’l Law 261

[93] J Gardam Necessity, Proportionality and the Use of Force by States (2004) 17

[94] Y Dinstein War, Aggression and Self-Defence (2001) 743

[95] L Van den Herik “Counter-Terrorism Strategies in a Fragmented International Legal Order(2013) 389

[96] UNSC, “Letter dated 8 January 2020 from the U.S Ambassador to the UN addressed to the Secretary-General” (2020)

[97] W Martel “United States Drone Strike in Iraq Kills Iranian Military Leader Qasem Soleimani” (2020) 114 AJIL 313

[98] UNSC “Letter dated 29 January 2020 from the Iranian Ambassador to the UN addressed to the Secretary-General” (2020)

[99] W Martel “United States Drone Strike in Iraq Kills Iranian Military Leader Qasem Soleimani” (2020) 114 AJIL 313

[100] Y Dinstein War, Aggression and Self-Defence (2001) 743

[101] L Ferro “Killing Qasem Soleimani: International Lawyers Divided and Conquered(2021) 53 CWRJ 1

[102] Y Dinstein War Aggression and Self-Defence (2001) 743

[103] A Haque “United States Legal Defense of the Soleimani Strike at the United Nations: A Critical Assessment”(2020) JS 1

[104] M Weller The Oxford Handbook of the Use of Force in International Law (2017) 881

[105] B Levenfeld “Israel’s Counter-Fedayeen Tactics in Lebanon: Self-defense and Reprisal Under Modern International Law” (1982) 21 Colum. J. Transnat’l L 37

[106] N Rontzitti “The expanding law of self-defence” (2006) 11 JSCL 354

[107] D Bowett “Reprisals Involving Recourse to Armed Force” (2017) 66 AJIL at 1-36

[108] U.N. International Law Commission “Commentary to Article 21: Responsibility of States for Internationally Wrongful Acts” Supplement No. 10 A/56/10 2001 article 14

[109] HRC Targeted killings through armed drones and the case of Iranian General Quassem Soleimani (UN Doc. A/HRC/44/38 2020) para. 31


by Abdullah Murad*




(1) Context


An unusual rule in Scots property law ensures that the right of ownership in all abandoned property falls to the Crown, the result is that all persons can abandon property except the Crown (barring an exception in the Companies Act 2006).[1] The principle is known as Quod nullius est fit domini regis.[2] G.J. Bell states that this “public expediency” is to reduce fraud, litigation, and adds to public revenue.[3] It is arguable that the rule has reduced fraud, as criminals will not be able to claim property they have stolen to be abandoned. The property is the Crown’s if it was abandoned and the owner’s if it was lost. Most of the items discarded are of no value. However, the government is interested in treasure trove, which includes items of historical and cultural significance. These are of great financial and symbolic interest to the government; it is easy to understand why the rule has not been altered for centuries.

There have been recommendations by the Scottish Law Commission to modify this principle. The recommendation would make most abandoned property ownerless yet allow the Crown to keep possession of future treasure troves.[4] These recommendations have not been adopted by the Scottish parliament at the date that this paper is written (January 2021).

It would follow that this unusual rule may have applied to heritable property and moveable property alike. This notion was rejected by the Court of Session in Scottish Coal.[5] The decision had the effect of confirming that one was disallowed to abandon corporeal heritable property in which they held a registered real right. It is surprising that such an important clarification to the law was made in 2013, 396 years after the establishment of the Register of Sasines.[6]


(2) Aim of the Paper


This paper will aim to answer two questions.


The first question is: to what extent can property rights be abandoned in Scotland?  This paper will seek to describe the manner and extent to which property rights can be abandoned. The subject of abandonment in this paper will only concern the casting away of ownership or subordinate rights. The rights that can be cast away will be referred to as property rights, or real rights in property. The right of ownership is difficult to describe. However, Erskine’s definition is suitable for this project, he says “the right of using and disposing of a subject as our own, except in so far as we are restrained by law or paction”.[7] The doctrinal inconsistencies within the law will be examined as well.


The second question is: how satisfactory is the law here? To answer this, the adequacy of the law will be analysed and compared to the positions abroad. An attempt will be made to recommend new or modified laws, which would fit best in Scotland. It is the opinion of the author that the law regarding the abandonment of property in Scotland is inadequate for a jurisdiction with such a rich legal history. The current legal policy does not favour the Crown or the people, and it is unnecessary for the Quod nullius rule to remain in effect (barring protection of treasure trove for the Crown). The final section will explore the importance of the modifications to the law.




At the heart of the law governing abandonment of Corporeal moveables is the difficult distinction between lost property and abandoned property. If it were possible to differentiate all lost and abandoned property by labelling them with stickers, then drafting law in this area would be simple. The majority of abandoned property could simply become owned by occupatio, while treasure trove will be kept for the Crown; and all lost property will be handed into the authorities. As this is not the case, the law has to balance the interests of the ‘original’ owners, the finders, and the Crown. The law wants to provide owners who have lost their property to be able to recover it, but it has to balance that with the interests of the finders of abandoned property who should not be made to wait decades to acquire legal ownership. This can only be done through a balancing act, the tools of which are positive prescription, negative prescription, and occupatio.

In relation to the abandonment of property rights in heritable property, it would seem that the concern lies in the fact that unrestricted abandonment would allow owners to escape duties and obligations to others. However, as far as Scottish law is concerned the Scottish Coal case demonstrated that the Court of Session gave multiple reasons to restrict abandonment. [8]  In any reformation of the law, the most important consideration should be balancing the interests of creditors, owners, and the Crown.




(1) The Extent to Which Property Rights can be Abandoned


At the most basic level, an ownership right in a moveable may be abandoned by a coupling of a physical and mental act. The physical act to abandon, and an intention to abandon.[9] It follows that if one is not in possession of a moveable item, then only a mental act is needed to renounce the ownership right. For instance, if x wishes to abandon a guitar he had left at the train station, he would only need to mentally confirm its abandonment. As mentioned earlier, any ownerless property comes to be owned by the Crown. However, some ownerless property does not belong to the Crown (for example wild animals). It is possible to abandon ownership of property and have it become ownerless again, this is only possible in the case of wild animals that have been acquired through occupatio.[10]

It is to be noted that negative prescription may be able to extinguish ownership rights in moveables. This is not confirmed verbatim by the legislation, although there is academic consensus.[11] This area of the law is quite understudied, which is understandable as rarely any valuable items are abandoned.[12]

The abandonment of subordinate rights in moveable property may be done expressly or through implication.[13] In general, a physical act is not required as the owner of the property will usually have possession. Security rights in valuable moveables such as jewellery are often expressly discharged due to their importance. Once subordinate rights are abandoned, they cease to exist, therefore they cannot be passed on to the Crown through the Quod nullius rule.

Negative prescription will extinguish subordinate rights in moveable property. If the right is unenforced or unexercised for twenty years it will be extinguished due to the Prescription and Limitation (Scotland) Act 1973 s.8. However, renunciation of a subordinate real right in moveables is instantaneous and one need not rely on negative prescription. These rights are able to be abandoned without any limitations.


(2) The Satisfactoriness of The Law


The law is in an unsatisfactory state for the abandonment of ownership rights of corporeal moveables. There are doctrinal and practical inconsistencies that can be remedied through reform. First, it is peculiar that the principle of occupatio exists in Scots law but can only be used to occupy property that has never been owned (exception for wild animals). [14] Property that has never been owned cannot have been said to be abandoned in the first place, resulting in the fact that occupatio is impossible in the case of abandoned property. It would be expected that a legal system with a weak concept of occupatio would have robust positive prescription to cure this inconsistency. Scotland is not granted that benefit, as it is unlikely that any statutory scheme would apply to gaining ownership in moveables through positive prescription.[15] There may be a common law rule that grants positive prescription of the ownership of moveables after forty years.[16] This rule is uncertain and too long to be of much use in granting finders the ownership to abandoned property. The Scottish Law Commission recommend that “corporeal moveable property should cease to be subject to negative prescription.”[17]

Furthermore, the law should ensure social reality matches legal reality. The Crown is uninterested in most of the abandoned property and makes no effort to possess their moveable goods. In contrast, abandoned property is possessed and used by finders. This could be seen as a criminal act, however in the case of Kane v Friel prosecutors did not rely on the Quod Nullius rule to prove mens rea in the accused.[18] As long as the Crown’s interest in treasure trove is protected, it is not in the interest of any party that the Quod Nullius rule should remain for the majority of abandoned property. A seemingly minor, however a significant inconsistency is created by the negative prescription of ownership rights in moveables, and the Quod Nullius rule. As mentioned above, the ownership rights in moveables are negatively prescribed in twenty years. As a result, the Crown loses and regains their ownership rights instantaneously. This does not create practical difficulties, but theoretical consistency can aid in understanding and studying the law.


(3) The Reform to be Implemented


To recommend what reform should be implemented, the policy objectives should be stated. In a non-exhaustive list, the primary policy objectives in the drafting of this law should include: procedures within the law to ensure that owners of lost property are able to recover, finders of abandoned property are able to become owners, the Crown to retain their interests in treasure trove, and deterrence of thieves to seize lost property (through rewards for finders, and sanctions for criminals).

Numerous countries have well-developed systems of law to govern the abandonment of corporeal moveables. As is observed in the Scottish Law Commission’s report, countries have implemented laws that regard abandoned property as ownerless, rather than belonging to the Crown.[19] These legal systems include Austria, Spain, the Canadian Province of Quebec, and Germany.[20] Legislation implementing this change would have no negative effects in theoretical or practical terms. The Crown would only have minimal losses (as their rights to treasure trove would be kept intact), while the finders of abandoned property would be able to retain what they have found.

If this change is implemented and abandoned property is available for occupatio, it is possible that thieves could appropriate lost property and present the law as a defense.[21] The Civic Government (Scotland) Act 1982 provides that a finder of lost or abandoned property must report it to the police in a timely fashion.[22]  When reform is eventually put in place, there should be a requirement to claim ownership of any abandoned property, it must first be handed in to the proper authorities. Other systems of law also impose a requirement to return property to authorities if found. For instance, the Italian Civil Code requires a finder to return property to its rightful owner, and if unable, to return the property to the mayor.[23] This would not allow thieves an excuse for their actions and ensure the possibility of owners recovering their property.

Once property has been deemed to be ownerless after abandonment, the next step will be to determine in which circumstances the finder can become an owner. In the cases of treasure trove, finders should not become owners. The Crown is in a much better position to take care of and use property of historical and cultural significance. An elegant solution can be found in the laws of the former state of Czechoslovakia (law cited by Carbonnier).[24] The law seeks to categorize abandoned property in between their monetary value; items of low monetary value are returned to the finder, however items of higher value are taken by the state and sold with a reward being paid to the finder. This law keeps intact the treasure trove exception but is able to add to the Crown’s claim of valuable property after the Quod Nullius rule would be abandoned. Further inspiration can be taken from the German Civil code, within this system the finder of the moveable will be able to take ownership rights six months after reporting the find to the competent authorities.[25] The six month period is ample time for the owner to realize their mistake and claim their moveable from the authorities, the rightful owner waiting longer than this period is in essence abandoning their property. In contrast, the French system allows the rightful owner to claim the property within three years after it has been lost.[26] In cases of theft this time period seems short, in cases of lost property it seems lengthy. It seems bizarre to retract possession of a moveable from someone who has possessed it legally for three years. Scotland should adopt the period of time in which the Germans allow the finder to gain ownership rights.

The final piece of reform concerns prescription of moveables. It is not clear if ownership rights can be abandoned through negative prescription due to the lack of case law, although there is an academic consensus. This is not satisfactory; the law should be as concrete as possible. An amendment to the Prescription and Limitation (Scotland) Act 1973, or new legislation would be necessary to remedy this issue (Prescription and Limitation (Scotland) Act 2018 does amend the older act, but not for this purpose).




(1) The Extent to Which Property Rights can be Abandoned

Ownership rights in heritable property are usually valuable and difficult to attain. Many people will spend their entire lives making payments for a single piece of land. In fact, one piece of land in Scotland is being marketed for £8,000,000[27], and the average sale price of a piece of property in Scotland is £157,056.[28] When one wishes to depart from the ownership rights of one’s land, the usual method is transfer through sale. However, there are instances in which sale would not be viable. Take for instance, ownership rights of a piece of land which needs extreme repairs to be of any use. These rights will be of low value, or indeed of no value at all. Several situations of this nature exist and make abandonment of the property an attractive route for the owners.

It was previously unclear whether heritable property could be abandoned.[29] This was clarified in the Scottish Coal case.[30] The court decided that there is no legal process by which a person can abandon their ownership rights in heritable property (without transferring to another or ceasing to exist). A person is allowed to physically ignore property and leave it to become derelict but cannot abandon their ownership rights. The dereliction of the property had significant consequences for the parties involved in the Scottish Coal case, as the ones attempting to abandon land were liquidators seeking to salvage assets of an insolvent firm. Due to environmental regulations, the land would have cost £500,000 per month to keep from being a safety hazard. In response to this the liquidators attempted to abandon the land.[31] The court said there were several methods in which ownership rights may be terminated. For instance, the land can be destroyed through coastal erosion; or the owner may cease to exist or be dissolute (if the owner has left no heirs or creditors to take the property, then it would fall to the Crown under the Quod Nullius rule). There were only two further methods in which ownership rights can be abandoned. In the first scenario, the law divests the owner through formal legal processes, this could include compulsory purchases or land attachment. However, this would transfer the property to another person, not leave it truly ownerless. The final method of transferring is through simple voluntary transfer to another person. In effect, if one wishes to truly abandon heritable property (in the sense of leaving it ownerless) the only method in which to do so is to not leave any heirs and cease to exist. Unsurprisingly, this method is not attractive for owners. As the court put it, “Counsel were not able to find any authority which supported the idea that an owner could abandon land in Scotland”.[32]

In order to cast away ownership rights of heritable property to another, writing must be used. The Requirements of Writing (Scotland) Act 1995 s. 1(2)(b) outlines that a contract or unilateral obligation for the creation, transfer, variation or extinction of a real right in land must be in a written document. There are additional requirements that need to be fulfilled in regard to the execution of the document, they lie outwith the scope of this project.

Subordinate real rights are able to be abandoned without theoretical limitation. One is able to renounce their subordinate rights in heritable property and they will cease to exist. Scots law recognizes a limited amount of real rights. Of those only four are strictly subordinate real rights in heritable property. They consist of: rights in security over heritable property (‘standard securities’), proper liferent, servitudes, and leases.[33] It is possible to abandon these rights through negative prescription.[34] This method of abandonment is most relevant to the right of servitude.

In relation to servitudes, it is possible to abandon them through acquiescence as well. If a benefited proprietor does not interject and lets the burdened proprietor carry out work which would not allow the exercise of the servitude, then the right to oppose the work may be lost.[35] According to the courts, this outcome will be more likely if the works required considerable expenditure.[36]

It is to note, that any abandonment of these rights will require to be in writing (other than negative prescription).[37] This is probably due to the fact that these rights are valuable and written documents can highlight their importance. These rights are able to be abandoned and are only limited by the writing required.


(2) The Satisfactoriness of the Law


The law regarding the ownership rights of heritable property is in an unsatisfactory state in Scotland. It is telling that only in 2013 the law regarding abandonment of ownership rights was clarified. The ambiguity surrounding this distinct, yet inevitable scenario could be due to the recently abolished feudal system, and the rarity of negative value land ownership rights.[38] There is doctrinal inconsistency in the fact that a civilian system of law that follows Roman principles does not allow an owner to abandon ownership rights. Lord Hodge found that the lack of precedent in Scotland allowed an analogy to Roman Dutch law to act as a strong argument for the decision in allowing abandonment of property.[39] Due to the strong similarities in two legal systems, this is a convincing argument. In fact, Professors Reid and Gretton noted that private law rights may be abandoned, and if the ownership of land cannot be abandoned, the system is ‘asymmetrical’.[40] As mentioned above, the Inner House argued that there was no authority which supported the idea that ownership rights in land could be abandoned. This is a strong argument, but it does create a theoretical inconsistency with the general system of property law in Scotland. There are policy considerations to take into account if a legal system was to allow the abandonment of ownership rights. The Scottish coal case demonstrates that if allowed without restriction, then owners could escape payments. This issue is not difficult to solve. Courts could regulate the abandonment of property, and only allow it in cases where there was no abuse. 

In addition, those within private practice have observed several problems which arise after the final decision in the case.[41] It is clear that reform could significantly improve the position on a doctrinal and a policy level.


(3) The Reform to be Implemented


The law regarding the abandonment of ownership rights in heritable property was obscure until the Scottish Coal case. The law is not well developed and the majority of it is encapsulated within one case. There is a need for further development.

The current law heavily restricts abandonment, and several US common law jurisdictions restrict abandonment like Scotland (albeit with different underlying principles).[42] The Pocono Springs case is factually similar to the Scottish Coal case mentioned above. In the Pocono Springs case, a married couple attempts to abandon land in order to avoid paying fees associated with ownership.[43] The Superior Court of Pennsylvania found that the property could not be abandoned as the couple held a perfect title, and title could only be relinquished if one ceased to exist or transferred it to another. The result here creates the same effect as the law in Scotland. Furthermore, the court in Cristofani concluded that the abandonment of ownership was not possible and there was no authority that would support a contrary conclusion.[44]  This was also noted by the Inner House in the Scottish Coal case. It is safe to say that the law’s effects are quite similar. As noted by one academic, it seems obvious that the American cases should be overruled.[45] The complete ban on abandonment of property serves no policy purposes. There is a fear of owners abandoning negative value property. This can easily be remedied through the courts regulating abandonment, and only allowing abandonment when the property is of positive value. It is clear that in some circumstances one should be able to abandon ownership rights in property.

The civil code in Switzerland allows one to abandon ownership in property.[46] The reform in Scotland should take inspiration from the procedure in Switzerland. The model is straightforward and uncomplicated. The owner must deliver a written document confirming their abandonment to the proper authority, after which their entry in the register will be deleted. This renders the property ownerless. Theoretically the first individual to appropriate the land would become owner. However, there is a caveat in the code which allows for the ‘Cantons’ (subdivisions of a country for political purposes) to claim ownership before private individuals. This system would be appropriate to fit in Scots law, barring the ability for private individuals to be able to claim the land.

Once a system in which land is able to be abandoned is created, the assumption would be that it becomes ownerless. However, for policy purposes the Crown should be granted ownership (similar to the Quod Nullius rule). Jurisdictions such as Argentina, Chile, Quebec, and Germany grant the state ownership of the abandoned property.[47] Scots law should follow a similar path, because if the state is granted ownership this would avoid possible conflict that may ensue to claim the abandoned property. Moreover, the abandoned property will increase the funds of the Crown. The government in Scotland already operates a division to claim ownerless property (Queen’s and Lord Treasurer’s Remembrancer). It is not hard to imagine that this organization could easily adapt to adopt heritable properties.

The proposed reforms should be clear, concise, and should balance the interests of the parties involved. An owner of heritable property that is of positive value should be able to unilaterally abandon their land, this is in their own interests and that of the Crown’s. In contrast, the owners of negative value land should not be able to abandon their land in order to escape liabilities, this would be patently unfair to their creditors. Policy considerations should account as higher priority than theoretical and doctoral considerations. 




The law of abandonment may not be as significant as Land Registration or Rights in Security; however, it does have an important role to play in the framework of society.


(1) For Moveable Property


As mentioned earlier, academics have noted that the abandonment of moveables is not as well analysed as other areas of property law (probably because abandoned moveables are usually of low value). Despite the low average price of abandoned items, the people who appropriate abandoned property are in fact taking property that is the Crown’s. The case of Kane v Friel demonstrated that prosecutors would not rely on this technical doctrine, but the existence of the Quod nullius rule is unnecessary and possibly harmful.  


(2) For Heritable Property


Much like the law for moveable property, the law of abandonment for heritable property is given insufficient attention.[48] This is probably because the instances of individuals attempting to abandon ownership rights in heritable property are few and far between. However, the importance of the law is quite clear. Due to the lack of clarity within the law, the taxpayers of Scotland could have paid £73,000,000 of the costs incurred by the Scottish Coal Company. [49] Laws that govern the abandonment of heritable property must be concrete and clear.




(1) For Moveable Property Rights


In answering the first question regarding the extent of abandonment of property rights, moveable property rights can be abandoned without any significant restrictions. Depending on the circumstances, a physical and mental act are needed, or only a mental act is required. Technically speaking, once ownership in moveables is abandoned, one is unable to gain it back because the moveable now belongs to the Crown. Negative prescription in the Prescription and Limitation (Scotland) 1973 act may provide an avenue in which ownership in moveable property can be abandoned.

In answering the second question regarding the satisfactoriness of the law on abandonment, the law is quite unsatisfactory. Indiscriminately granting the Crown all abandoned property through the Quod nullius rule should be abandoned. If treasure trove rights are protected, then it is in public interest to allow occupatio of abandoned moveables. To prevent thieves from using the defense of occupatio for what they believed to be abandoned goods, the duty to contact the proper authorities to gain rightful ownership should be present.


(2) For Heritable Property Rights


In answering the first question, the abandonment of ownership in heritable property is heavily restricted by Scottish law. The only manner in which one is able to abandon property is when they cease to exist (either through dissolution or death). However, subordinate real rights in heritable property are easily abandoned and without limit (including through negative prescription). It is to be noted, that to abandon subordinate real rights, this must be done in writing as set out in Requirements of Writing (Scotland) Act 1995 s. 1(b).

In answering the second question, the law is unsatisfactory in this regard as well. At the fundamental level, there seems to be no argument against allowing property owners to abandon their land if it does not have adverse effect on others. A system in which abandoned land is vested within the Crown is beneficial to all parties involved.


(3) The Future


The law of Scotland is constantly changing and improving. There have been significant developments within property law in recent years (Abolition of Feudal Tenure etc. (Scotland) Act 2000, Title Conditions (Scotland) Act 2003). The Scottish Law Commission has a report that recommends reform within the abandonment of Corporeal Moveables.[50] The SLC recommendations on abandonment of moveable property should be adopted completely, with the addition of the German law which prescribes that a finder of property become owner in six months. Furthermore, Scotland should adopt a system similar to Switzerland in the law of abandonment of heritable property. As part of the reform, ownership rights in land which are of positive value should be able to be unilaterally abandoned. There is hope for reform in the law of abandonment within Scotland.  


*Third year law (LL.B Hons) student at The University of Edinburgh.

[1] Companies Act 2006 s. 1013.

[2] Erskine Inst II, 1, 12; Bankton Inst I, 3, 16.

[3] Bell, Prin §1291.

[4] Scottish Law Commission, Report on Prescription and Title to Moveable Property (Scot Law Com No 228, 2012) para 5.6.

[5] Scottish Environmental Protection Agency v Joint Liquidators of the Scottish Coal Company Ltd 2013 CSIH 108 [103].

[6] Registration Act 1617.

[7] Erskine Inst II, 1, 1.

[8] Scottish Coal CSIH.

[9] Kenneth Reid, The Law of Property in Scotland (Law Soc. of Scotland 1996) para 4; Lord Advocate v University of Aberdeen and Budge 1963 SC 533; Mackenzie v MacLean 1981 SLT (Sh Ct) 40.

[10] Erskine Inst II, 1, 10.

[11] Reid, Property para 675; David Johnston, Prescription and Limitation (1st edn, W. Green, 1999) paras 7.08, 7.14; David C Miller, David Irvine, Corporeal Moveables in Scots Law (2nd edn, W. Green 2005) para 7.05.

[12] Scottish Law Commission, Discussion Paper on Prescription and Title to Moveable Property (Discussion Paper No 144) para 9.1.

[13] Reid, Property para 9.

[14] Bell Prin §1287.

[15] Reid, Property para 565.

[16] Aberscherder Parishioners v Parish of Gemrie 1633 Mor 10972; Ramsay v Wilson 1666 Mor 9113. 

[17] Scottish Law Commission, Report on Prescription and Title to Moveable Property para 3.41.

[18] 1997 JC 69.

[19] Scottish Law Commission, Report on Prescription and Title to Moveable Property para 5.5.

[20] Austrian Civil Code article 349; Spanish Civil Code article 610; Quebec Civil Code article 935; German Civil Code §959.

[21] Scottish Law Commission, Report on Prescription and Title to Moveable Property para 5.6.

[22] Civic Government (Scotland) Act 1982 s. 67(1).

[23] Italian Civil Code article 927.

[24] Jean Carbonnier, Droit Civil, vol 3 Code (1950) article 119 285ff.

[25] German Civil Code §973.

[26] French Civil Code article 2279.

[27] ‘Seton Castle’ <> accessed 5th January 2021.

[28] ‘UK House Price Index Scotland: June 2020’ <>.

[29] Reid, Property para 9.

[30] Scottish Coal CSIH para 103.

[31] Ibid para 1.

[32] Scottish Environment Protection Agency v Joint Liquidators of the Scottish Coal Company Ltd 2013 CSOH 124, para 22.

[33] Reid, Property para 4.

[34] The Prescription and Limitation (Scotland) Act 1973 s. 8.

[35] Reid Elspeth, Blackie John, Personal Bar (W. Green 2013).

[36] Davidson v Thomson (1880) 17 R 287, 290.

[37] The Requirements of Writing (Scotland) Act 1995 s. 1(2)(b).

[38] Malcom Combe, ‘Abandonment of Land and the Scottish Coal Case: was it Unprecedented?’ (2018) Edinburgh Law Review 301.

[39] Scottish Coal CSOH para 22. 

[40]Kenneth Reid, George Gretton, Conveyancing (4th edn, W. Green, 2013).

[41] Scottish Coal decision: Inner House finds that Scottish liquidators do not have a power to abandon onerous property’ <> accessed 5th January 2021.

[42] For Maryland: Cristofani v. Bd. Of Educ. 632 A.2d 447, 450 (Md. Ct. Spec. App. 1993); For Pennsylvania: Pocono Springs Civic Ass’n v. MacKenzie 667 A.2d 233 (Pa. Super. Ct. 1995); For Mississippi: Waldrop v. Whittington 57 So.2d 298, 300 (Miss. 1952).

[43] 667 A.2d 233 (Pa. Super. Ct. 1995).

[44] 632 A.2d 447, 450 (Md. Ct. Spec. App. 1993) para 103.

[45] Lior Strahilevitz, ‘The right to Abandon’ (2010) 158 U. Pa. L. Rev. 355, 418.

[46] Swiss Civil Code article 666, 964.

[47] Argentinean Civil Code article 2376; Chilean Civil Code article 590; German Civil Code §928; Quebec Civil Code article 936.

[48] Eduardo Peñalver, ‘The Illusory Right to Abandon’ (2010) Cornell Law Faculty Publications 209 192.

[49] ‘Liquidators escape £73m clean-up bill for Scottish Coal’ (BBC News, 17 July 2013) <> accessed 8th January 2021 ;Scottish Environment Protection Agency v Joint Liquidators of the Scottish Coal Company Ltd 2013 CSOH 124 para 22.

[50] Scottish Law Commission, Report on Prescription and Title to Moveable Property.




By Frauke Tepe*




It has long been recognized that words can be as hurtful as physical harm, which is why Hate Speech has been subject to discussion for a long time. In recent times, more communication has shifted to online formats than ever before. This new environment brings light to new challenges and gives a new platform to Hate Speech in a dimension that has never been seen before. Speech is available worldwide, across jurisdictions and on platforms that are in the force of private companies that have their registered offices in different countries than their users. Online communication does not only take place between friends and acquaintances, but also between complete strangers. While some of those communicants act under their full name, others choose to stay anonymous. The anonymity of the internet provides a shield for the speaker to spread hate. While Hate Speech has always been a sensitive topic and subject to many judicial rulings in different jurisdictions, the 21st century has brought to light new challenges on how to address Hate Speech.

Especially with respect to the rise of new media, the calls for banning Hate Speech are increasing in volume. Intermediaries are being asked to immediately delete Hate Speech: e.g. the Netzwerkdurchsetzungsgesetz (Network Enforcement Act) in Germany makes intermediaries liable if illegal content is not deleted within a certain period of time.[1]

Such efforts are to be positively recognized. However, one question is yet to be solved, namely what falls under the notion of Hate Speech exactly. There is no uniform definition to be found in international treaties or similar agreements. One of the consequences of this fact is that private individuals have the power to de facto decide which speech is prohibited – namely social networks such as Facebook and Co. The line between Hate Speech and criticism is often very thin and it can be difficult to distinguish between them. For example, Hate Speech can fuel racism and be discriminating, but publishing a scientific article on the correlation of race and crime can do the same.[2] Sellars wonders, “[w]ithout a clear definition, how will scholars, analysts, and regulators know what speech should be targeted?”[3]

This essay will draft a definition of Hate Speech that takes into account the realities of the 21st century. First, the necessity for a definition will be discussed, in order to build a foundation. When defining the notion of Hate Speech, different aspects must be considered such as the forms of expression, the harm, and the targeted group. One aspect to be considered is the different understanding of limited speech around the world. Different challenges with regards to freedom of expression will be considered leading to a final definition. In addition to examining definitions in international agreements and academic literature, but also closely consider definitions of social networks – those platforms that accommodate Hate Speech the most.




So far, there is no uniform definition of Hate Speech in any legally binding treaties. The term is not used in international human rights law. It is, for example, not enshrined in the European Convention on Human Rights (ECHR); furthermore, the European Court of Human Rights (ECtHR) has avoided limiting itself to a definition.[4] However, some legal texts address the issue of Hate Speech while not calling it by that name. The International Covenant on Civil and Political Rights (ICCPR) requires parties to prohibit “[a]ny advocacy of national, racial or religious hatred that constitutes incitement to discrimination, hostility or violence” by law.[5] In similar fashion, Article 4(a) of the International Convention on the Elimination of All Forms of Racial Discrimination (ICERD) requires the parties to “declare an offence punishable by law all dissemination of ideas based on racial superiority or hatred, incitement to racial discrimination, as well as all acts of violence or incitement to such acts against any race or group of persons of another colour or ethnic origin, and also the provision of any assistance to racist activities, including the financing thereof”. The Rabat Plan of Action has tried to bring further clarity to the formulation of Article 20(2) ICCPR.[6] Despite these efforts, the term Hate Speech is still widely used as a “generic term” for threats to individuals and groups, but also for “cases in which people may be simply venting their anger against authority.”[7] This underlines the need for a uniform definition. The scope of the obligations under those provisions varies, so that no uniform framework can, as yet, be achieved.[8] Additionally, a common definition would also lower the risk of manipulation by governments, which is a common fear, especially in times of elections.[9]

Naturally, the question arises as to how much benefit an international agreement would bring, especially in light of the fact that different cultures/nations differ in their understanding of freedom of expression. For the desired effectiveness of the definition, the political will to implement it is required. All parties must eventually agree with the definition. This poses a challenge, as the interests of protection sometimes diverge widely. An international agreement has already been sought by the EU and is also advocated in literature, but the USA in particular has opposed such regulation.[10] Protected by the First Amendment of the U.S. Constitution, freedom of expression is “the most cherished American constitutional right”.[11] Its main objective is to protect the citizen from interference from the state.[12] Whereas the U.S. understanding of freedom of speech is shaped by individualism and libertarianism, in Western countries, on the other hand, honour and dignity build the foundation for freedom of speech.[13]

There have been several cases illustrating the high standing of freedom of expression in the U.S.  One of such cases is R.A.V. v City of St Paul in which a cross was burnt in the yard of a black family.[14] The U.S. Supreme Court held that an ordinance which prohibited racially motivated “fighting words” was unconstitutional. This was limited by Virginia v Black pursuant to which the criminalization of cross burning is constitutional if such is made with “intent to intimidate”. The cross burning itself cannot, however, be taken as prima facie evidence for such intent.[15] Thus, the U.S. system prohibits content-based restrictions on freedom of speech, whereas for example, the German Grundgesetz allows such restrictions.[16]

The difficulty of combining these two approaches is even greater in the online environment, as several jurisdictions may be affected by one hateful post: e.g. a user posts a hateful comment on a social media platform in one country that has its headquarters in a different country; additionally, the content is then available worldwide. In the Yahoo! case, the problem of differing understandings of protection in Europe versus the USA becomes apparent.[17] A French Court held that Yahoo! was liable for the sale of Nazi merchandise which was considered Hate Speech in France. Yahoo! argued that, since the content was uploaded in the U.S., the French jurisdiction wasn’t applicable; the French court dismissed that claim. Yahoo! then went on to a court in the U.S. and successfully sought a judicial ruling that the enforcement of the French decision would violate Yahoo!’s First Amendment protection.[18]

The case highlights the tension involved in regulating online speech extraterritorially and the powerlessness of states when it comes to cross-jurisdictional enforcement of court rulings.[19] An internationally valid definition of Hate Speech could overcome these differences. Cultural particularities would still have to be taken into account, but if, for example, social networks worldwide took a uniform definition as a basis, enforcement difficulties would no longer arise.  

In general, the transnational reach of private companies like Facebook could be an effective way to combat the problem of Hate Speech.[20] Especially in light of laws such as the German Netzwerkdurchsetzungsgesetz that imposes the obligation to delete Hate Speech, among other things, on private platforms, a uniform definition is appropriate. Otherwise, the risk arises that allowed speech is also banned since the networks want to avoid sanctions and thus take precautions by deleting every critical posting, leading to private censorship. By means of a definition, principles are established at international level, e.g. within the EU, by which Hate Speech can be determined.

This can even have a positive effect on freedom of speech. One example for that is the EU Code of Conduct on Countering Illegal Hate Speech Online.[21] In 2016 Facebook removed the content in 28.3% of received notifications, whereas in 2018 the number rose to 82.4%.[22] The numbers prove that, to a certain degree, a non-regulatory approach can be helpful.[23] They also suggest that a common understanding of Hate Speech laid down in the EU Code of Conduct is beneficial,[24] even if the individual definitions of the platforms differ. This does not only apply to intermediaries, but also traditional courts. A uniform definition would require courts to be more rigorous in their analysis, which could then lead to greater protection of freedom of expression.[25]

Having established the necessity of a definition that provides guidance for states, but also intermediaries, the wording of such a definition will be discussed in the following.




In drafting a definition of Hate Speech, it is important to balance the fact that the freedom of expression is a fundamental right and of importance for every individual’s personality and democracy as a whole.[26] It should be noted that freedom of expression “is applicable not only to ‘information’ or ‘ideas’ that are favourably received or regarded as inoffensive or as a matter of indifference, but also to those that offend, shock or disturb the State or any sector of the population.”[27] Thus, one wants to be careful to not define the notion of Hate Speech too broadly, so that protected speech is not included.


(1) Forms of speech


There have always been different ways of expressing oneself. This does not necessarily require oral or written speech; certain actions may also express resentment and hatred. This includes, for example, cross burning[28] and Nazis marching through areas with a predominantly Jewish population (Skokie[29]). Symbols such as swastikas and the robes of the KKK convey powerful messages without saying something out loud.[30] Using the term Hate Speech creates the erroneous impression that only spoken words should be regulated.[31] It should be clear from the definition that this is not the case.

With new possibilities that arise online, even more forms are being added. On some platforms, communication occurs only through videos or photos. So-called memes and gifs are very popular at present and communication only through emojis is also possible. There are endless possibilities to express oneself and to communicate hatred. This should be taken into account in a definition, so that all forms of hateful expression can be considered Hate Speech.


(2) Addressee


Traditionally, the characteristics of race, ethnicity, religion and nationality are taken as distinctive characteristics for Hate Speech. This is, for example, the case in Article 20(2) ICCPR. Article 6 of the Audiovisual Media Services Directive (AVMSD) at least prohibits “any incitement to hatred based on race, sex, religion, or nationality.” However, especially in today’s times, a focus on this alone is not enough. Since the AVMSD took effect in 2010, the understanding of discrimination has developed due to contemporary social justice movements and a broader field of addressees is necessary. Sex, gender, sexual orientation and disabilities are other characteristics that serve to exclude people. These have yet to be included in international standards.[32]

When choosing the addressee of Hate Speech, a distinction is made between a group-based approach and equal protection clauses.[33] The group-based approach is, however, criticized for being ‘under-inclusive’.[34] Google+ uses a mixture of both forms. It first lists different characteristics and then uses the open clause of “any other characteristic that is associated with systemic discrimination or marginalization”.[35] The wording still limits the group of possible addressees, but highlights the difference to cyberbullying, which can turn against anyone and differs in consequences.[36] Such a formulation is appropriate in order to be as inclusive as possible, while still taking into account the particularities of various cultures. For this reason, this wording is adopted for the proposed definition.

The speech must always be directed at an individual or an individualizable group.[37] Taking the claim “soldiers are murderers” as an example: the claim is not directed at a particular soldier, but at soldiers working for the government. Rather than construing this expression literally, this claim constitutes an opinion and is thus constitutionally protected. The German Constitutional Court, who had to decide this case, came to the conclusion that an interpretation of the slogan results in the understanding that it casts soldiers as much as victims as it does as killers. [38] If the statement was referred to an individual or an individualizable group, it would be more likely to be understood as Hate Speech.


(3) Harm


In order to justify limitations to the freedom of expression, the harm that individuals suffer from Hate Speech have to be considered.[39] This forms the most important part of the definition. Through careful wording, a line can be drawn between mere offensive/insulting statements and Hate Speech.

Especially problematic with Hate Speech online is that the connection between hateful expressions online and actual harm in the real world is difficult to trace.[40] This is true for any media, but the peculiarities of the internet, such as the anonymity of users and the speed with which messages are distributed to large audiences, make this particularly complex.[41]

Again, the difference between the U.S. and European approaches constitutes another challenge  in this regard. The U.S. Supreme Court has clarified in its past rulings that restrictions on the freedom of speech are only possible in the case of “incitement to violence”,[42] whereas the ECtHR considers incitement to hatred also sufficient.[43] However, when defining the scope of harm, it should be remembered that physical harm and psychological harm can have the same effect on individuals.[44] Manifestations of hatred can constitute “a degradation of human dignity” without causing violence.[45] Here it should be noted that even small words can be given meaning in a definition, so that “or” should be used instead of “and”. Violence and hatred are two possible harms that are suffered, but the exclusion of minorities can have consequences as serious as violence and hatred, which is why intolerance and discrimination are also included in the definition.

The Recommendation R(97)20 of the Council of Europe introduces a broad definition understanding Hate Speech as “all forms of expression which spread, incite, promote or justify racial hatred, xenophobia, anti-Semitism or other forms of hatred based on intolerance, (…).”[46] The definition does not limit itself to incitement, but also considers spreading, promoting and justifying racial hatred as sufficient. Other definitions by academics use the phrase “promote hatred”, which, however, carries a less imminent standard than incitement.[47] Although such formulations are generally to be advocated due to the severity that hateful statements may carry, this is hardly compatible with the U.S. understanding. For a successful enforcement of the definition, the standard should thus be kept stricter than the one of Recommendation R(97)20 and other definitions.

Moreover, the likelihood of harm needs to be discussed. Incitement is by definition an inchoate crime.[48] Specifically, the “action advocated through incitement speech does not have to be committed for said speech to amount to  a crime.”[49] This requires some degree of risk of harm to be identified. The Rabat Plan of Action thus suggests requiring a “reasonable probability that the speech would succeed in inciting actual against the target group”.[50]  In the U.S. suppressing Hate Speech because it has a ‘bad tendency’ to cause harmful effects is prohibited, whereas in most other countries the connection between speech and harm can be rather loose.[51] Brandenburg v Ohio requires the speech to be “likely to incite or produce such action”[52] in order to be prohibited, which is understood to only apply when incitement is imminent, nearly inevitable.[53] The linkage between cause/speech and harm should not be interpreted too loosely; the causation should rather be direct.[54]


(4) Intent/purpose


In addition to the aforementioned aspects, the intention of the speaker is important. Article 20 ICCPR anticipates intent,[55] while, pursuant to Article 4 ICERD, the mere “dissemination of ideas based on racial superiority or hatred” is punishable. Here too, a uniform line must be followed.

In the case of Jersild v Denmark, the ECtHR decided in favour of the journalist that conducted and edited an interview with a radical xenophobic group in Denmark, during which derogatory statements were made.[56] By fining Jersild for publishing racist statements, the Danish authorities violated Article 10 ECHR. In order to avoid further cases such as this, some sort of requirement of intent should be included. This would be in line with the U.S. approach to protecting the freedom of expression as it limits the possibilities for speech to be considered Hate Speech.

Ward defines Hate Speech as “any form of expression through which speakers primarily intend to (…) incite hatred against their targets.”[57] He argues that, in this way, it is taken into consideration “whether the speakers’ desire to communicate ideas” other than hate outweighs “the desire to injure their victims.”[58] Marwick/Miller require the “speaker’s message to only promote hatred (…) without communicating any legitimate message.”[59] They do not require direct intent, but question the purpose of the speech. Here, again, the special problem arises that intent online is difficult to identify.[60] Instead of using a strict intent requirement, one could require that the speech has “no redeeming purpose”.[61] Thus the use of Hate Speech for other purposes such as journalism, education or similar is still allowed.[62] Proving such use is easier, so that the evaluation of Hate Speech does not fail because of intent.


(5) In public


In order to be understood as Hate Speech, the speech should be by definition made in public. A statement in a private space, e.g. at home, should continue to be protected as long as it is not immediately directed against a person with protected characteristics. Otherwise this would go too far and would not be compatible with U.S. ideologies. It rather leads to a regulation of the freedom of thought.[63] Here, the special nature of the internet should be considered. Direct messages may superficially be considered private. However, exchanges in forums to which one can gain access by signing up and publication of private conversations cannot be protected under the guise of privacy; they need to be seen as public speech. Some private conversations even take place on Twitter for everybody to publicly follow. This cannot be considered “private” in this context as each user has the chance to read and further disseminate the messages.[64] A publicity requirement such as this is already laid down in individual legal texts, e.g. the Canadian Criminal Code.[65]


(6) Context/medium


Hate Speech can be expressed in many different ways. Regardless of the format, Hate Speech can be expressed in “offensive, angry, abusive and insulting language”, but at the same time also “subtle, moderate, nonemotive, even bland; its message conveyed through ambiguous jokes, innuendoes, and images.”[66] This is, however, not the only clue into the nature of Hate Speech. As stressed before, there are different approaches to protecting freedom of expression. In general, there are cultural differences which must be able to be taken into account despite a uniform definition.

Facebook discusses the importance of context in determining Hate Speech and explains that using the term ‘fag’ can be offensive, while in Britain it is a common designation for cigarette.[67] On a more striking level, in Germany the downplaying of acts committed under the rule of National Socialism are especially fined.[68] Every culture has their own social and historical understandings which must be reflected in the definition. The European Court of Justice considered the “everyday language of the terms (…), while also taking into account the context in which they occur (…)” when interpreting Article 22a AVMSD.[69] The Rabat Plan of Action, as a multi-stakeholder process, recognises the great importance of context. It states that “[a]nalysis of the context should place the speech act within the social and political context prevalent at the time the speech was made and disseminated”.[70] It also highlights that context can serve as indicator for a person’s intent.[71]

Furthermore, the medium chosen also deserves recognition when assessing Hate Speech. The ECtHR stressed this fact, inter alia, in the decision of Gündüz v Turkey where the statements were made during a public live broadcast.[72] By spreading hate messages through media, they can have an even more serious impact on individuals due to their larger outreach.[73] If a hate message is posted online, it is accessible worldwide by millions of people. It might be deleted on one website, but could have been shared dozen times or might even be available on a new website by then. Until the message is deleted from there, a new process must be undergone, which takes time and allows further dissemination.

These requirements should not be included in the main body of the definition, but should nevertheless be part of an international treaty in order to provide guidelines for decision-makers such as courts or intermediaries that are obliged to delete illegal content under the law or pursuant to their own Codes of Conduct when assessing the proportionality.


(7) Definition


Taking into account the above discussed, the following definition can be proposed:

Hate Speech are all forms of expression that are made without redeeming purpose and are likely to incite to violence, hatred, intolerance, or discrimination, directed at a certain or identifiable group or member thereof distinguished by protected characteristics. The speech must be made either publicly or immediately directed at a member of a protected group.

Protected characteristics are race, religion, ethnicity, national origin, sex, gender, gender identity, sexual orientation, and disability, or any other trait that can be associated with systemic discrimination or marginalization.

In determining the proportionality, decision-makers shall consider, inter alia, the context and medium of the speech.

To keep the definition understandable, it was divided into individual sentences and the protected characteristics were listed separately. It captures the points that have been discussed over the course of the essay and can thus be used as international benchmark.

Nevertheless, the definition is not unassailable. The foregoing analysis highlights the challenges to drafting a definition for an international treaty with participating parties that have differing understandings of the freedom of expression. For example, the definition remains open to interpretation due to its open-ended wording. A uniform application of the definition in different legal circles can thus not be granted. U.S. American courts might still enforce the definition less strictly than, for example, European courts. However, such risk of slight deviations is to be accepted in order to find a definition that is internationally recognised. In any case, the definition ensures a common ground for addressing Hate Speech. Moreover, it gives legal certainty to social networks operating internationally. In this way, they can enforce a uniform notion of Hate Speech in their networks world-wide and do not have to fear prosecution by authorities for not complying with their legal duties.




Because of the different understandings of freedom of expression and thus also of the notion ‘Hate Speech’, a definition is favoured at international level. In particular, due to the new challenges that the online world presents to us as the “new frontier”[74] for the dissemination of Hate Speech messages, the definition should be applied cross-jurisdictionally. An internationally binding definition of Hate Speech will not only help courts in deciding on the legitimacy of speech; it will also act as a step before intermediaries such as social networks that are held responsible for the spread of Hate Speech on their platforms. Hate Speech is by now seldomly available in only one jurisdiction and, as proven by the Yahoo! case, this can lead to difficulties in enforcement. A uniform definition will facilitate the unified fight to combat Hate Speech.

In its current wording, the here-specified definition sets out clear requirements, but at the same time is open to contextual dynamics. This is particularly important in order to appreciate the specificities of each culture. As stressed by McGonagle, “[t]he approach must combine sensitivity and strategy in order to conduct a fine balancing act between freedom of expression and minority rights.”[75] The different levels of harm that can be caused by Hate Speech are not thereby overlooked, while freedom of expression continues to retain its full strength. States and private players should jointly ensure effective implementation of the definition in order to combat Hate Speech and address the challenges that come with the online environment.


*LL.M. in Innovation, Technology and the Law at the University of Edinburgh Class of 2020; trainee lawyer at the Regional Court of Cologne, Germany.

[1] Netzwerkdurchsetzungsgesetz 1 September 2017, BGBl I 2017 S3352.

[2] R Post, “Hate Speech”, in I Hare and J Weinstein (eds), Extreme speech and democracy (2009) 123 at 134 et seq.

[3] A F Sellars, “Defining Hate Speech” The Berkman Klein Center for Internet & Society Research Publication Series. Research Publication No. 2016-20 (2016) 4, available at

[4] T McGonagle, “Minorities and online ‘Hate Speech’: a parsing of selected complexities” (2010) 9(1) European Yearbook of Minority Issues Online 419; M Oetheimer, “Protecting freedom of expression: the challenge of Hate Speech in the European Court of Human Rights” (2009) 17(3) Cardozo Journal of international and Comparative Law 427 at 428 et seq.

[5] International Covenant on Civil and Political Rights art 20(2).

[6] Report of the United Nations High Commissioner for Human Rights on the expert workshops on the prohibition of incitement to national, racial or religious hatred, “Rabat Plan of Action on the prohibition of advocacy of national, racial or religious hatred that constitutes incitement to discrimination, hostility or violence”, 5 October 2012.

[7] I Gagliardone, D Gal, T Alves, G Martinez, “Countering online Hate Speech” (2015) UNESCO Series on Internet Freedom 7.

[8] T Mendel, “Does international law provide for consistent rules on Hate Speech?”, in P Molnar and M Herz (eds.), The content and context of Hate Speech: rethinking regulation and responses (2012) 417 at 418.

[9] I Gagliardone et al (n7) at 10.

[10] J Banks, “Regulation Hate Speech online” (2010) 24(3) International Review of Law, Computers & Technology 233 at 236; for example the EU made huge efforts to include Hate Speech in the Council of Europe’s Convention on Cybercrime, which was then amended because of refusal by the U.S.

[11] M Rosenfeld, “Hate Speech in constitutional jurisprudence: a comparative analysis”, in P Molnar and M Herz (eds.), The content and context of Hate Speech: rethinking regulation and responses (2012) 242 at 247.

[12] Ibid.

[13] Ibid at 259.

[14] R.A.V. v. City of St. Paul [1991] 505 U.S. 377.

[15] Virginia v Black [2003] 538 U.S. 343.

[16] M Rosenfeld (n11) at 267.

[17] La Ligue Contre La Racisme et L’Antisemitisme (LICRA) and Union Des Etudiants Juifs De France (UEJF) v. Yahoo! Inc. and Yahoo! France [2000] Tribunal de Grande Instance de Paris, 22 May 2000 and 22 November 2000; J Oster, “European and international media law” (2017) ch. 3 at 98.

[18] Yahoo! Inc. v. La Ligue Contre Le Racisme et l’antisemitisme (LICRA) [2006] 433 F.3d 1199.

[19] J Banks (n10) at 233.

[20] I Gagliardone et al (n7) at 15.

[21] The EU Code of Conduct on countering illegal Hate Speech online (2016), available at

[22] European Commission, “Factsheet – 4th monitoring round of the Code of Conduct” (2019), available at

[23] Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions, “Tacking Illegal Content Online – Towards an enhanced responsibility of online platforms” COM(2017) 555 final (28 September 2017) 4.

[24] The Code of Conduct refers to the definition developed in the framework decision 2008/913/JHA of 28 November 2008 on combating certain forms and expressions of racism and xenophobia by means of criminal law and national laws transposing it: “all conduct publicly inciting to violence or hatred directed against a group of persons or a member of such a group defined by reference to race, colour, religion, descent or national or ethnic origin.”

[25] T Mendel (n8) at 428.

[26] See for example: Handyside v UK [1976] App No 5493/72, Council of Europe: European Court of Human Rights.

[27] Handyside v UK [1976] App No 5493/72, Council of Europe: European Court of Human Rights at para. 49.

[28] As in R.A.V. v City of St. Paul and Virginia v Black.

[29] National Socialist Party of America v. Village of Skokie [1977] 432 U.S. 43.

[30] M J Matsuda, “Public response to racist speech: considering the victim’s story” 87(8) Michigan Law Review (1989) 2230 at 2365 et seq.

[31] J Waldron, “Why call Hate Speech group libel?”, in J Waldron, The harm in Hate Speech (2012) 34 at 37.

[32] T McGonagle (n4) at 423.

[33] Example for an equal protection clause: Universal Declaration of Human Rights art 7; the ICCPR instead follows a group-based approach; see: P Cavaliere, “Digital platforms and the rise of global regulation of Hate Speech” 8(2) Cambridge International Law Journal (2019) at 282, available at SSRN:

[34] E Heinze, “Viewpoint absolutism and Hate Speech” 69(4) Modern Law Review, (2006) 543 at 565; P Cavaliere (n33).

[35] Terms and Policies for Google+, available at

[36] A F Sellars (n3) at 25.

[37] B Parekh, “Is there a case for banning Hate Speech?” in P Molnar and M Herz (eds.), The content and context of Hate Speech: rethinking regulation and responses (2012) 37.

[38] Tucholsky II [1995] case no. 1 BvR 1476/91, Bundesverfassungsgericht (German Constitutional Court); M Rosenfeld (n11) at 270 et seq.

[39] P Cavaliere (n33).

[40] I Gagliardone et al (n7) at 54; see also: A F Sellars (n3) at 28.

[41] I Gagliardone et al (n7) at 54.

[42] Brandenburg v Ohio [1969] 395 US 444.

[43] P Cavaliere (n33); see for example Directive 2010/13/EU OJ  L 1995, 15.4.2010 (Audiovisual Media Services Directie – AVMSD) art 6.

[44] Ibid.

[45] As argued by the Yugoslavian representative in U.N. General Assembly, Third Committee, U.N. Doc. A/C.3/SR.1079, 20 October 1961, para. 9; S Farrior, “Molding the matrix: the historical and theoretical foundations of international law concerning Hate Speech” 14(1) Berkeley Journal of International Law (1996) 1 at 26.

[46] Recommendation No. R. (97) 20 of the Council of Europe Committee of ministers to member states on “Hate Speech”, 30 October 1997.

[47] A F Sellars (n3) at 17.

[48] Report of the United Nations High Commissioner for Human Rights on the expert workshops on the prohibition of incitement to national, racial or religious hatred (n6) at para. 29(f).

[49] Ibid.

[50] Ibid.

[51] R Post (n2) at 134 et seq.

[52] Brandenburg v Ohio [1969] 395 US 444 at 447.

[53] A F Sellars (n3) at 28.

[54] Report of the United Nations High Commissioner for Human Rights on the expert workshops on the prohibition of incitement to national, racial or religious hatred (n6) at para. 29(f).

[55] Ibid at para. 29(c).

[56] Jersild v Denmark [1994] European Court of Human Rights, Grand Chamber, App. No. 15890/89.

[57] K D Ward, “Free speech and the development of liberal virtues: an examination of the controversies involving flag-burning and Hate Speech” 52(3) University of Miami Law Review (1998) 733 at 765.

[58] Ibid.

[59] A Marwick and R Miller, “Online harassment, defamation, and hateful speech: a primer of the legal landscape” CLIP Report (2014) 17.

[60] A F Sellars (n3) at 28.

[61] Wording used in A F Sellars (n3) at 30 et seq.

[62] See also YouTube’s Hate Speech policy, available at:

[63] A F Sellars (n3) at 29; see for example Stanley v Georgia [1969] 394 U.S. 557.

[64] A F Sellars (n3) at 29.

[65] s 319(3); also the Australian Racial Discrimination Act 1975 s18C(2).

[66] B Parekh (n37) at 41 et seq.

[67] R Allan, “Hard questions: who should decide what is Hate Speech in an online global community?” (2017) Facebook Newsroom, available at:

[68] Strafgesetzbuch (German Civil Code) s 130(3).

[69] Case C-244/10 Mesopotamia Broadcast A/S METV and Case C-245/10 Roj TV A/S v Bundesrepublik Deutschland [2011] ECR 2011 I-08777 at para. 40; R Craufurd-Smith et al., Media law text, cases, and materials (2014) ch. 5.9 Content requirements I: The protection of minors and Hate Speech at 221.

[70] Report of the United Nations High Commissioner for Human Rights on the expert workshops on the prohibition of incitement to national, racial or religious hatred (n6) at para. 29(a).

[71] Ibid; see also: R Allan, (n67).

[72] Gündüz v Turkey [2003] ECtHR App. no. 35071/97 at para. 49.

[73] Recommendation No. R. (97) 20 of the Council of Europe Committee of Ministers to Member States on “Hate Speech”, 30 October 1997; similar: M Rosenfeld (n11) at 281.

[74] J Banks (n10) at 234.

[75] T McGonagle (n4) at 440.