Multinational Corporate Accountability: The Supreme Court of Canada Leaves the Door (And a Few Questions) Open in Nevsun Resources Ltd. v Araya

Kees Westland, 3L, Volume 78 Senior Editor

Last month, the Supreme Court of Canada rendered judgment in Nevsun Resources Ltd. v Araya, 2020 SCC 5, an appeal about whether claims brought on behalf of a class of Eritrean workers against a Canadian mining company should proceed to trial. Seven justices agreed that at least some claims could go forward. A five-judge majority also held that pleadings relating to customary international law (“CIL”) norms disclosed a reasonable claim and should not be struck. 

Nevsun has generated an unusual amount of interest both in Canada and abroad. Days after it was rendered, the case had already made its way onto American syllabi. One American scholar tweeted that Nevsun read “like a SciFi rendition of a parallel legal universe, in which our doctrine went another way.” If so, the world-building has been left for future courts. The Supreme Court issued three separate sets of reasons in a sharply divided judgment. Important questions remain open. Ultimately, trial judges will have the unenviable task of sorting through these reasons and adjudicating the merits of these and similar claims. 

Background

The respondents in Nevsun were refugees and Eritrean nationals who claimed they were conscripted into forced labour for an indefinite period at the Bisha mine in Eritrea. They alleged that the government of Eritrea committed human rights abuses contrary to international law – including forced labour, slavery, cruel, inhuman and degrading treatment, and crimes against humanity. The Bisha mine was owned by the appellant, Nevsun, a Canadian company. The respondent workers requested damages from Nevsun based on a number of domestic torts as well as damages based on novel claims for breaches of CIL “as incorporated into the law of Canada.” [1]

Nevsun moved to strike the pleadings as being “scandalous, frivolous or vexatious.” [2] The British Columbia Supreme Court dismissed its motion, and the Court of Appeal dismissed the appeal. The corporation sought leave to appeal to the Supreme Court of Canada on two main issues. First, it submitted that the “act of state” doctrine precluded consideration of the alleged human rights abuses. Second, it submitted that the claims based on customary international law had no reasonable chance of success. Other arguments that the corporation advanced at the courts below, including that BC was not the appropriate forum for the dispute, were not argued before the Supreme Court. 

Nevsun also pleaded that the case was governed by Eritrean law, under which the conduct alleged was lawful. In any case, Nevsun claimed, it was protected by a corporate veil with respect to the tort claims. The Supreme Court did not decide the choice of law issue, nor did it pronounce on whether liability could pierce the corporate veil. These issues will be decided when the proceedings continue.

Furthermore, because the Supreme Court heard the case on a motion to strike, the allegations of human rights abuses were not contested. Instead, the question was whether, assuming that the facts were true, the workers’ claims had a chance of success. However, a Report by a UN Commission of Inquiry has found that the government of Eritrea frequently uses conscripts as forced labourers [3] and that the use of torture is widespread. [4]

Decision

The majority dismissed the appeal in reasons by Abella J. It held that the claims could go forward in BC.

Act of State

“Act of state” is an English common law doctrine, which is applied in other common law jurisdictions but has never been directly applied in Canada. The doctrine limits subject matter jurisdiction in cases where the court would be required to “adjudicate upon the lawfulness or validity of sovereign acts of foreign states,” as Lord Neuberger put it in the leading English case of Belhaj and another v Straw and others, [2017] UKSC 3. It is based on the principle that the domestic courts of one country should not usurp the executive’s authority to conduct foreign relations by “sitting in judgment” over the conduct of a foreign sovereign in its own territory. Unlike sovereign immunity, which protects a state from direct suit in the courts of another state, act of state extends to civil proceedings against parties other than that state.  

Nevsun submitted that the doctrine applied in this case because a Canadian court would have to decide, as a condition precedent to adjudicating the claims against Nevsun, whether Eritrea had committed breaches of CIL norms. As the trial judge held, these pleadings constituted a “novel defence,” because the act of state doctrine has never been applied in Canada. However, Nevsun submitted that the doctrine had been received into BC law in 1858 by the Law and Equity Act and should apply to Araya’s claims. 

After surveying the history of the act of state doctrine and the leading cases in England and Australia, Abella J held that principles underlying it were completely addressed within Canadian conflict of laws and “judicial restraint” jurisprudence. [5] While act of state was part of English common law, there was no continuing need for it in Canada. 

Brown and Rowe JJ concurred in this portion of the majority judgment. [6] Côté J, writing for herself and Moldaver J, dissented – meaning that the court split 7-2 on this issue. The dissenters would have allowed the appeal in its entirety and dismissed the respondents’ claims. Côté J disaggregated the act of state doctrine into two distinct branches: conflict of laws and justiciability. [7]

 Under the “justiciability” branch, the question was whether a court has the institutional capacity to adjudicate a matter and whether it is “legitimate” for it to do so. [8] In the judgment of Côté J, “[a] court has the institutional capacity to consider international law questions, and its doing so is legitimate, if they also implicate questions with respect to constitutional rights, the legality of an administrative decision or the interface between international law and Canadian public institutions.” [9] On the other hand, Canadian courts would not be competent to adjudicate private matters “where the outcome depends on a finding that a foreign state violated international law.” [10] Consequently, Cote J would have held that none of the respondents’ claims were justiciable, since adjudicating them would interfere with the executive prerogative power over foreign affairs. [11]

Customary International Law

The CIL pleadings focused on the possibility that a court might recognise new torts to reflect the gravity of the alleged wrongdoing. The respondent workers submitted that existing torts did not go far enough: torture, they submitted, is something more than battery; slavery is something more than a hybrid of unlawful confinement, unjust enrichment, and battery. [12] The workers submitted that a judge might incrementally develop the common law by recognising new nominate torts based on international law norms. By contrast, Nevsun submitted that the CIL claims were bound to fail and cautioned that recognising new nominate torts would have “complex and far-reaching effects” [13] that were inconsistent with an incremental approach.  

The majority held that the appellants had not established that the respondents’ CIL claims had no reasonable chance of success. Abella J emphasised that the motion to strike must be used with care. [14] In Abella J’s judgment, it was not for the Supreme Court to determine on a motion to strike the novel question of how the CIL claims should proceed. Recognition of new nominate torts was only one possible route; [15] a good argument was also available that a violation of CIL could lead directly to a common law remedy. The human-centric turn in public international law meant that CIL norms had evolved to include personal rights and duties, which might have been adopted into Canadian law. Notably, this theory appears to be at odds with the submissions in the Respondents’ Factum, which takes pains to point out that “to be clear, the respondents do not contend that the adoption of jus cogens norms into Canadian law leads automatically to a civil remedy for the violation of those norms.” [16] Rather, the workers submitted that their claims must satisfy the test for recognition of new common law torts. [17]

Brown and Rowe JJ dissented on this issue. They would have held that the claims based on customary international law could not succeed. They parted ways with the majority in two significant respects. First, they would have held that customary international law did not have “horizontal effect” in common law — in other words, it did not apply as between private parties. [18] Accordingly, the workers could not obtain civil damages directly for breaches of customary international law norms. Second, after a useful review of the test for recognising new nominate torts, the dissenting justices would have held that at least some of the customary international law claims could not qualify. While slavery and forced labour could conceivably meet the test for identifying new common law torts, it would be inappropriate for the courts to recognise these new torts for the first time in litigation relating to conduct that took place in a foreign country. 

Côté and Moldaver JJ concurred in the reasons of Brown and Rowe JJ on this issue. The Court therefore split 5-4 on whether these pleadings disclosed a reasonable cause of action. On the other hand, the dissenting justices had no difficulty accepting that courts may look to customary international law as they interpret and incrementally develop the common law. This confirms that the holding in R v Hape, 2007 SCC 26 extends to private law. [19] One way or another, customary international law is “incorporated” into the common law. To the extent that the LeBel J’s comments in Hape may have been in obiter, [20] Nevsun is also helpful in reasserting them.

Unanswered Questions

Justiciability/Judicial Restraint

The distance between the judgments of Abella J and Côté J should not be overstated. While Côté J would have confirmed the existence of the act of state doctrine, whether the dissent is truly committed to a self-standing act of state doctrine is open to doubt. [21] Rather, like the majority, the minority casts “act of state” as composed of two distinct branches, each of which corresponds to a pre-existing jurisprudence. In particular, Côté J agreed with her colleague that conflict of laws jurisprudence overlaps with one branch of the doctrine. [22] However, whereas the majority locates the other part of the doctrine in “judicial restraint,” the dissent emphasises “justiciability.” Just how much these categories differ remains to be seen. 

At this stage, it is unknowable what the majority’s idea of “judicial restraint” includes. We do know that judicial restraint means that courts will “refrain from making findings which purport to be legally binding on foreign states.” [23] One way to interpret this is that the question of justiciability is simply folded into judicial restraint: claims may be non-justiciable if adjudicating them would negate a foreign state’s claims to sovereignty.

For example, in the English leading case of Blad v Bamfield, [24] which the Nevsun minority considered under the non-justiciability branch, the Lord Chancellor declined to hear proceedings against Blad for seizing Bamfield’s property on the high seas. The seizure was carried out under the authority of letters patent, issued by the King of Denmark, which gave Blad a monopoly to trade in Iceland . In staying the action, Nottingham LC observed that “to send it to a trial at law, where either the court must pretend to judge of the validity of the King’s letters patent in Denmark or of the exposition and meaning of the articles of peace; or that a common jury should try whether the English have a right to trade in Iceland, is monstrous and absurd.” [25] By hearing the case, the High Court of Chancery would have been asserting a power to authorise English traders to trade in Iceland despite Denmark’s claims to sovereignty. 

Similarly, in Buttes Gas and Oil Co v Hammer (No 3), [26] a more recent leading act of state case, the House of Lords was asked to consider conflicting grants of offshore oil concessions in the Arabian Gulf. The case involved the competing territorial claims of three states. Adjudicating the private law claims in Buttes Gas would necessarily have involved negating the claim of a sovereign state by deciding that one or another corporation had a right to explore and exploit natural resources, notwithstanding that state’s claims to certain territorial waters. In this sense, the House would have “purported to bind” the competing sovereigns.

The majority seems to hold that decisions that purport to bind other states are not within the institutional competence of Canadian courts to make. Accordingly, matters like those that arose in Blad and Buttes Gas would presumably be non-justiciable in Canada. However, the precise limits of “judicial restraint” will have to be mapped out in future cases.

Customary International Law

The most dramatic outcome of Nevsun was that the pleadings based on CIL were not struck. For now, the existence of new torts based on CIL remains an open question, as does whether such torts might assist in allowing courts to disregard corporate personality. [27]

However, trial judges in future cases will have to contend with the forceful dissenting reasons of Brown and Rowe JJ on this point. Indeed, it is open to question whether Canadian courts should be in the business of developing common law rules about human rights abuses, such as slavery, which afflict least-developed countries but are relatively unknown to Canada. While private international law can fill a “governance gap” [28] by holding powerful transnational corporations accountable for human rights abuses abroad, the courts must nonetheless work within the decolonising trend in private international law established during the Morguard revolution in the nineties. As the dissenting justices observe, Canadian courts “have no legitimacy to write laws to govern matters in Eritrea, or to govern people in Eritrea.” [29]

Applying choice of law principles to Nevsun, Eritrean law, the lex loci delicti, would ordinarily govern. However, there may be reasons of domestic public policy to apply Canadian law. Either way, a trial court should be cautious: the fact finding necessary to make out the elements of a domestic cause of action (even a novel tort) against Nevsun must not amount to an indirect indictment of Eritrea’s sovereign conduct. This sort of adjudication would seem to go beyond the institutional competence of our domestic courts; it would also run counter to the principle of international comity and potentially intrude into the executive power over foreign affairs.

In theory, the road to recognising claims for civil damages for breaches of customary international law norms also remains open. However, at least two significant hurdles stand in the way of making these sorts of claims. First, customary international law historically binds states, not private individuals, and the majority’s view that CIL has evolved to be “human-centric” in the field of international human rights law appears to be contentious. Second, even if international human rights law does bind a corporate actor, there is no clear path from a violation of customary international law to a civil remedy of damages. 

This part of Nevsun is perhaps best understood within its procedural context. At the end of the day, the Supreme Court’s conclusion in this part of the judgment is only that the litigation may proceed to be heard on its merits in BC. The Court has not authoritatively pronounced on the quality of the customary international law claims, other than to say it was not plain and obvious that they should fail. The majority’s musings on the state of public international law explain why the pleadings met that low bar and survived a motion to strike. They will do less to assist a trial judge who must assess the CIL claims on their merits.

Constitutional Law and Private Right

A final theoretical issue merits discussion. In their partially dissenting opinion, Brown and Rowe JJ address at some length the civil law doctrine of “horizontal effect.” This doctrine attempts to account for the effect of fundamental rights in disputes between private parties (the label of “horizontal” is applied in contradistinction to the “vertical” effect of fundamental rights between individuals and the state). 

The perfunctory treatment of this issue in the judgment may have the unfortunate effect of discouraging Canadian courts and scholars from referring to a sophisticated body of foreign jurisprudence. Brown and Rowe JJ (dissenting on this point) write that “the majority’s approach amounts to saying that the doctrine of adoption has what jurists in Europe call “horizontal effect.” [30]  This would be surprising, because (according to the dissent) the Court rejected the idea that the Charter applies between parties to a private dispute in RWDSU v Dolphin Delivery Ltd, [1986] SCR 573. It would be astonishing if CIL applied directly when the Charter does not.

Even accepting that this is true, the holding in Dolphin Delivery that the Charter does not apply to a court order resolving a private dispute between private parties does not necessarily mean that the Charter does not have a “horizontal effect.” Some of the most important contributors to the seminal literature of the horizontal effect in Germany reject the proposal that constitutional rights give rise directly to private obligations. [31] However, the leading authorities propose that the constitution may still have “horizontal effect,” in that it provides a mandatory source of norms for interpreting and applying private law. [32] This position would not be controversial in Canada, and it appears to be consistent with the partially dissenting reasons in Nevsun.

As Brown and Rowe JJ recognize, “conceptually, these are deep waters.” [33] The discussion of the horizontal effect in Nevsun is unlikely to be helpful in navigating them. By contrast, the minority’s endorsement [34] of the reasons in Jones v Tsige, 2012 ONCA 32 may help push Canadian law forward at the interface of private law and constitutional rights.  

Notes

[1] Nevsun at para 60.

[2] Supreme Court Civil Rules, BC Reg 168/2009, Rule 9-5.

[3] Report of the Commission of Inquiry on Human Rights in Eritrea, at 1518-1519

[4] Ibid. at 1515

[5] Nevsun at para 44. 

[6] Ibid. at para 135.

[7] Ibid. at para 275.

[8] As the Court most recently recognized in Highwood Congregation of Jehovah’s Witnesses (Judicial Committee) v. Wall, 2018 SCC 26, [2018] 1 SCR 750, at para 32.

[9] Nevsun at para 296.

[10] Ibid. at para 273.

[11] Ibid. at para 313.

[12] Nevsun Resources Ltd. v Araya, “Respondent’s Factum” (30 November 2018) at para 98. 

[13] Nevsun Resources Ltd. v Araya, “Factum of the Appellant Nevsun Resources Ltd.” (4 October 2018) at para 79.

[14] Nevsun at para 66, citing R. v. Imperial Tobacco Canada Ltd., 2011 SCC 42, at para 21.

[15] While the judgment of Abella J did not explicitly endorse claims in tort, her reasons at para 131 should be read as not excluding them. 

[16] Respondents’ Factum at para 117.

[17] The broader theory of the case in the judgment of Abella J corresponds to the Factum and oral submissions of the Joint Interveners, Amnesty International Canada and the International Commission of Jurists.

[18] Nevsun at para 269, per Côté J, concurring with the judgment of Brown and Rowe JJ on this issue.

[19] Nevsun at para 176.

[20] This was suggested in Nevsun Resources Ltd. v Araya, “Factum of the Intervener, Mining Association of Canada” (30 November 2018) at para 12.

[21] See Nevsun at para 276.

[22] Ibid. at para 285.

[23] Ibid. at para 47. This language was likely influenced the submissions of the Intervener, International Human Rights Program, University of Toronto Faculty of Law.

[24] (1674) 3 Swans 604, 36 ER 992.

[25] Emphasis added.

[26] [1980] 3 All ER 475, [1980] 3 WLR 668 (“Buttes Gas”).

[27] This possibility is discussed in the Respondent’s Factum at para 106. 

[28] Penelope Simons and Audrey Macklin, The Governance Gap: Extractive Industries, Human Rights and the Home State Advantage (Abingdon: Routledge 2014) at 246-247.

[28] Penelope Simons and Audrey Macklin, The Governance Gap: Extractive Industries, Human Rights and the Home State Advantage (Abingdon: Routledge 2014) at 246-247.

[29] Nevsun at para 259.

[30] Ibid. at para 210.

[31] See, for example, G Dürig ‘Grundrechte und Zivilrechtsprechung’ in T Maunz (ed.), Von Bonner Grundgesetz zur gesamtdeutschen Verfassung (Munich, 1956) 117; C-W Canaris, ‘Grundrechte und Privatrecht’, Archiv für die civilistische Praxis (1994) 201; U Diederichsen, ‘Das Bundesverfassungsgericht als oberstes Zivilgericht — Ein Lehrstück der juristischen Methodenlehre,’ 198 Archiv für die civilistische Praxis 171 (1998); R Alexy, A Theory of Constitutional Rights (J Rivers trans, OUP 2002). 

[32] The locus classicus for this doctrine, which is called the indirect horizontal effect, is the Federal Constitutional Court’s decision in the Lüth case: 7 BVerfGE 198 B, II, 1.

[33] Nevsun at para 210.

[34] Ibid. at para 243.