Legal justifications for gender parity on the bench of the International Court of Justice: An argument for evolutive interpretation of Article 9 of the ICJ Statute

Abstract

The UN General Assembly and the UN Security Council should amend their rules of procedure to create gender parity on the bench of the International Court of Justice. Only 3.7 per cent of all judges on the ICJ have been women. The UN Charter, ICJ Statute, and long-standing practice of the Court underscore the importance of representation, but the focus has been on geographical representation. Using the law of international organizations, combined with the law of treaty interpretation and international human rights law, this article argues that Article 9 of the ICJ Statute should be interpreted to include a requirement of gender parity. Established practice, subsequent practice, and the UN’s multi-decade gender parity in staffing policy establish an evolutive interpretation of what is required to fulfil equality at the UN and the ICJ. The nomination and election procedures for ICJ judges are sufficiently flexible to facilitate this interpretation.

1. Introduction: A gender-unequal bench at the ICJ

‘Each time that I gaze out at the delegations representing parties … I am struck that their composition bears too much resemblance to the groups of persons who gathered in 1945 to draft the Charter of the United Nations and the Statute of the Court. Very few of the counsel are from developing countries and almost all, regardless of nationality, are men. This is an unsatisfactory situation.’

President of the International Court of Justice, Joan E. Donoghue, 16 April 2021¹

As the UN’s principal judicial organ and a key mechanism for peaceful settlement of inter-state disputes, the International Court of Justice (ICJ) remains a uniquely important player in public international law (PIL). What it is not, however, is a beacon for gender equality.² Decades after women have entered legal studies and the legal profession in equal or greater numbers than men, employment in legal posts continues to skew male.³ While other international and supranational courts, such as the International Criminal Court (ICC) and the African Court of Human and People’s Rights (ACtHPR), have made strides towards or have achieved parity on the bench, the ICJ lags behind. Since its establishment almost 80 years ago, only four of the 108 permanent judges (3.7 per cent) have been women.⁴ Amongst the ad hoc judges, the numbers are worse, with only four women out of 117 judges (3.4 per cent).⁵

While newer supranational courts and tribunals have had a chance to address this issue during their creation, ICJ judicial elections for one-third of the bench occur every three years,⁶ creating regular opportunities for significant change. The ICJ bench is 20 per cent women (presently three ICJ members are women),⁷ and would need at least four additional women to achieve parity. With the announcement of eight candidates for five positions on 29 June 2020,⁸ it became clear that parity could not be achieved in the most recent election cycle, as only three candidates were women, and two of those were already on the bench.⁹ On 12 November 2020, the United Nations Security Council (UNSC) and United Nations General Assembly (UNGA) re-elected the four incumbent judges and Mr. Georg Nolte, maintaining the court’s gender ratio of three women and 12 men.¹⁰ Notwithstanding calls from civil society to consider gender in the 2020 ICJ election,¹¹ the UNSC and UNGA chose not to elect all three women candidates. Similarly, the 2017 and 2018 elections were missed opportunities, where men filled all five vacant spots, despite three of the nine nominees being women.¹² The UN Charter (Charter),¹³ the ICJ Statute (Statute),¹⁴ and the long-standing practice of the Court underscore the importance of representation,¹⁵ but the focus has been on geographical representation.¹⁶

This article assesses the possibility of expanding the well-established convention of equitable geographic distribution to include gender parity. It contends that the laws of established practice and subsequent practice have coalesced to require an updated interpretation of the Statute of the ICJ and the UN Charter, and that this evolutive interpretation mandates gender parity on the ICJ bench. Just as subsequent practice can modify treaty agreements for states, established practice of an international organization (IO) – and in this case the gender parity practice of the UN – can modify the meaning of the terms of the IO’s constituent instruments (Article 9 of the Statute and Article 8 of the Charter). Arguing that the obligation born of this evolutive interpretation is both an obligation of conduct and result¹⁷ and falls on UN organs and likely UN member states, the article suggests amending the UNSC and UNGA rules of procedure to achieve this outcome.

Parity on judicial benches matters. Women’s participation in the judiciary is key in part because ‘[t]he judiciary influences society at all levels’.¹⁸ The ICJ impacts everything from environmental law, to human rights, to the rule of law, and remains uniquely important regarding its impact on general PIL as a field.¹⁹ Scholars have published excellent work regarding gender equality at supranational courts.²⁰ Little has been written regarding gender parity on the ICJ bench,²¹ and no parity arguments have been made relying on the combination of international human rights law (IHRL), the law of international organizations (IO law), and treaty interpretation. 2020 marked the UN’s seventy-fifth anniversary, prompting reflection regarding whether the UN has fulfilled its promises.²² The UN Secretary General (UNSG) Antonio Guterres utilized this anniversary to re-emphasize the UN’s gender parity policy.²³ This article posits that via a combination of IO law and treaty interpretation, there are legal justifications for extending the UN’s gender parity policy to the ICJ bench.

To build this argument, Section 2 of the article defines gender parity in the UN and ICJ context, explains the benefits of parity on judicial benches, and suggests that the best way to achieve parity is with a numerical, consequence backed rule. Section 3 analyses how the long-standing requirement of gender equality has evolved into gender parity in the UN context, and unpacks how the UN’s decades-old gender parity policy in staffing can be understood as an evolutive interpretation of gender equality requirements in the Charter. On this basis, Section 4 builds a case that the previously examined UN parity practice amounts to binding IO law obligating UN organs and member states. Further, Section 4 links established practice to subsequent practice and examines the implications this has for interpretation of the ICJ Statute, leading to Section 5, which argues that gender parity can be read into Article 9 of the Statute via an evolutive interpretation. After these legal arguments have been made, Section 6 puts forward a proposal for how gender parity on the ICJ bench can be achieved in accordance with current ICJ judicial election procedures (Section 6.1), suggesting that the internal rules of both the UNSC and UNGA permit these organs to require gender parity in candidate lists and voting (Section 6.2). Section 6.3 considers how such changes to the rules of procedure would impact member states, and how they might be brought on board. Finally, Section 7 concludes that the UNSG, UNSC, UNGA, and member states all likely have obligations both of conduct and result to achieve gender parity on the ICJ bench.²⁴ The article ends with a call to translate its legal arguments into political persuasion.

2. The benefits of a rule mandating numerical gender parity

Gender parity means ‘an equal number of women and men’.²⁵ The UN Commission on the Status of Women interpreted it to mean no less than 50 percent women.²⁶ As the ICJ has 15 sitting judges, remedying its long history of inequality might begin with an eight women/seven men distribution, to be changed to an eight men/seven women distribution and then back again at a future time. Consideration should also be given to the representation of non-binary genders/gender-diverse persons and a protocol drawn up to address this.

Judges significantly impact international law’s development and content.²⁷ Projects such as ‘Feminist Judgments in International Law’ imagine how law would differ if the people writing the judgments changed.²⁸ Judicial identity and diversity influence trial outcomes,²⁹ and the different contexts, perspectives, and approaches judges bring to the bench³⁰ impact the legal content of their decisions.³¹ While a gender-balanced judiciary is not a panacea for all problems,³² the benefits of parity on benches are numerous. Positives include increased legitimacy, representation, public confidence, access to justice, fairness, good governance, equality, rule of law, sustainable development, and much more.³³ It is no wonder, then, that the call for parity on international benches is decades old.³⁴

This article assesses the best options for achieving gender parity on the ICJ bench. After reviewing practice at other supranational courts, it concludes that a consequence backed rule to achieve numerical parity is the most certain path. The ACtHPR did not achieve parity until it implemented a penalty for failing to do so.³⁵ The Protocol to the African Charter called for ‘adequate gender representation in [the] nomination process’³⁶ and that ‘in the election of the judges, the Assembly shall ensure there is adequate gender representation’.³⁷ Yet, law alone did not result in this outcome, and neither did repeated requests from the court’s Office of Legal Counsel asking that states nominate women judicial candidates.³⁸ The Office had to introduce a punitive measure disqualifying states that did not submit at least one woman candidate for parity to be achieved.³⁹

Requiring precise numerical parity and backing failure to achieve this with a penalty provides a guarantee that ‘softer’ rules do not. Grossman noted that ‘[f]or [international] courts where states were required by statute to take sex into account when nominating or voting for judges, a higher percentage of women sat on the bench’ as compared to courts that did not have such a rule.⁴⁰ These results, while improved, were far from parity, totalling 32 per cent women judges at courts with ‘gender balance’ rules compared to 15 per cent women judges for tribunals without such rules.⁴¹ Notably, courts such as the ICC and the European Court of Human Rights (ECtHR), which have rules promoting women’s empanelment that stop short of parity, have not achieved it, or if they have, have not always maintained it. Article 36(8)(1)(iii) of the Rome Statute of the ICC requires that states parties select ‘[a] fair representation of female and male judges’.⁴² The absence of a precise numerical requirement⁴³ in this ‘fair representation’ rule has meant that at times that the percentage of women has been as high as 60 per cent and once there was even an all-women bench.⁴⁴ However, consecutively between 2015 and 2018, women’s representation dropped to almost a third of the overall number of judges,⁴⁵ reminding one that without firm floors, ‘[t]here is no “acquis”, i.e. the advancement of women may at any time fall behind existing achievements’.⁴⁶ For the ECtHR, a 2004 resolution stating that the Parliamentary Assembly that elects the judges would no longer ‘consider lists of candidates where … the list does not include at least one candidate of each sex’⁴⁷ has resulted in at most 40 per cent empanelment of women.⁴⁸ The resolution was weakened via an ECtHR advisory opinion,⁴⁹ and presently women’s representation is 34 per cent.⁵⁰ These examples suggest that a 50/50 penalty-backed parity rule would be beneficial for achieving and maintaining numerical gender parity on the ICJ bench. This article argues that achieving precise numerical parity is necessary to fulfil the right to gender equality. To support this claim, the next section builds a case that, in the UN context, gender equality has evolved to require gender parity.

3. How gender equality has evolved to require gender parity at the UN

3.1 Gender parity as an apex interpretation of gender equality

Gender equality is a long-established legal right,⁵¹ and gender parity is a more recent interpretation of this right.⁵² Gender equality and the related prohibition on gender-based discrimination are represented in every major human rights treaty⁵³ and are international rules with customary status.⁵⁴ Some treaties address gender equality as a non-derogable right⁵⁵ that states have a ‘mandatory and immediate obligation’ to fulfil.⁵⁶ Such obligations would apply equally to the UN and its organs⁵⁷ and prohibit member states and the UN from discrimination based on sex.⁵⁸

Gender parity arises from gender equality in part because over time the prohibition on sex discrimination under international law, a negative obligation, has evolved to require substantive equality under international law,⁵⁹ a positive obligation. In other words, an obligation of conduct (not to discriminate) has evolved into an obligation of result (ensure gender parity). Human rights, including those in the Charter, were designed to evolve.⁶⁰

There are several legal interpretations of equality: formal, substantive, and transformative. ‘Equality before the law’ has historically meant formal equality, but IHRL has long emphasized the need to achieve substantive equality.⁶¹ The Committee on Economic, Social and Cultural Rights stated that securing equality entails achieving equality in fact and eliminating indirect discrimination in law, policies, and/or practices.⁶² Duty bearers, whether states or IOs, are obligated to positively fulfil these rights.⁶³

This interpretation of equality is bolstered by transformative equality, ‘which sees full and genuine equality as likely to be achieved only when the social structures of hierarchy and dominance based on sex and gender are transformed’.⁶⁴ Transformative equality is an appropriate doctrine through which to approach parity on the bench because it focuses on including historically excluded groups and requires structural change to achieve this inclusion.⁶⁵ Gender parity on the ICJ bench corresponds to transformative equality’s aims, such as overcoming cycles of disadvantage and promoting social and political inclusion and participation.⁶⁶ Because transformative equality represents cutting edge IHRL, some might counter that it is lex ferenda. However, the latest legal developments in human rights, such as the Convention on the Rights of Persons with Disabilities, which promotes transformative equality in almost every article, communicates that what was once a novel understanding of equality is better understood as the standard interpretation.⁶⁷ State support for this treaty – there are currently 182 states parties – underscores that transformative equality is now accepted as integral to IHRL.⁶⁸

That equality can result from parity requirements is supported by what IHRL calls ‘special’ or ‘specific measures,’ and domestic law calls quotas. Because formal equality means ‘the state should not give preference to any one group and that people should be treated exclusively on their individual merits and regardless of group membership’,⁶⁹ several treaties create an exceptional basis for such preferences in order to achieve substantive equality.⁷⁰ The CEDAW Committee’s General Recommendation No. 25 notes that special measures should ‘accelerate the equal participation of women in the political, economic, social, cultural, civil or any other field’.⁷¹ For decades, the UNSG has used special measures in furthering its gender parity staffing policy.⁷² The UN’s internal justice system has confirmed through multiple decisions that special measures creating gender parity in UN staffing are valid.⁷³

Gender equality rights extend to gender on the bench.⁷⁴ International law enshrines women’s rights to participate equally in public life including public service.⁷⁵ Member states and the UN must prevent both direct and indirect discrimination as a way to ensure greater gender diversity on the bench,⁷⁶ and ‘take all appropriate measures’ to ensure gender-equal representation at the international level,⁷⁷ including at courts.⁷⁸ Both the Committee on the Elimination of Discrimination against Women and the Working Group on The Issue of Discrimination against Women in Law and in Practice have emphasized that gender equality includes equal access to and participation in the international judiciary.⁷⁹ This is because substantive equality requires both representation and participation,⁸⁰ in part because courts are both political and representative.⁸¹

In addition to aspiring to impartiality, rule of law, and judicial independence, courts are also political agencies where judges might act as political agents.⁸² This is at least partly true at the ICJ, where voting patterns can map to a judge’s national origin and/or favour the state that nominated the judge,⁸³ and where judicial elections can be a political process,⁸⁴ albeit one moderated by institutions and procedures intended to depoliticize.⁸⁵ Structural bias in international law and at the ICJ is well documented.⁸⁶ The underrepresentation of women in the international judiciary contributes to this and ‘legitimate[s] the unequal position of women around the world rather than challeng[ing] it’.⁸⁷ Thus, gender parity on judicial benches is both a fulfilment of representation and participation in and of itself, and a means for securing more of it based on the decision-making patterns of women judges.⁸⁸

3.2 The UN’s interpretation of gender parity as fulfilling gender equality in the Charter

The UN’s commitment to gender parity in staffing is decades old, with early indications dating to 1970,⁸⁹ and the initial target year for achieving gender parity in UN staffing set at 2000.⁹⁰ The UNSG called parity ‘fundamentally a right’⁹¹ and stated that ‘[g]ender parity is … a crucial first step to orienting the system more strongly to deliver on gender equality’.⁹² The UN Working Group has described parity ‘as the ultimate measure of equality’.⁹³ In 2020 the UNSG reemphasized the urgent need to achieve gender parity in UN staffing.⁹⁴ The additional examples presented infra in this section suggest that, in the UN context, gender equality cannot be fulfilled without gender parity.

The UN’s gender parity commitment is articulated in numerous ways, including the Nairobi Forward-looking Strategies of 1985,⁹⁵ the Beijing Platform,⁹⁶ various UNSC Resolutions calling for women’s equal participation in decision-making,⁹⁷ and UNGA Resolutions such as 33/143 calling upon member states to ‘assist the United Nations’ in increasing the proportion of women employees ‘by nominating more women candidates’ to positions.⁹⁸ Parity practice is clear in the Human Rights Council’s Consultative Group’s guidelines on gender parity in international appointments,⁹⁹ and the creation of the Working Group on the issue of Discrimination against Women in Law and Practice.¹⁰⁰ The UN’s parity strategy document states that parity is ‘an imperative requested by Member States’,¹⁰¹ reflecting both state practice in favour of gender parity and the links between customary international law and the customary law of IOs.¹⁰²

Further, there is evidence that parity should apply to UN courts and tribunals and is compatible with geographical distribution. Ebrahim-Carstens notes that the UN’s internal justice system – the United Nations Dispute Tribunal (UNDT) and the United Nations Appeals Tribunal (UNAT) – has always had a significant women’s presence on the bench, and even at times a majority of women.¹⁰³ Both Tribunals’ Statutes state that the appointment of judges shall take both geographical representation and gender balance into account.¹⁰⁴ Since 2015 both the UNSG and the President of the UNGA instruct member states to consider ‘geographical distribution and gender balance’ when appointing UNDT/UNAT judges.¹⁰⁵

The UN’s internal administrative tribunals are significantly distinct from the ICJ, including that the administrative judges are appointed by the UNGA and ICJ judges are elected. The ICJ is of much greater political significance to states than the UNDT/UNAT. Further, gender ‘balance’ is not the same as parity and often results in less than 50 per cent empanelment of women. Despite this, there are several takeaways from the UNDT/UNAT example relevant to parity at the ICJ. The administrative tribunals demonstrate that there is precedent for both the UNSG and UNGA, key players in the ICJ judicial nomination and election process, to call for a gender balance – and to interpret this as requiring numerical parity – in judicial appointments within the UN system. Similarly, it would not be unusual for the UNSG and the UNGA to instruct member states to take gender balance into account when nominating judges, and for member states to comply. And, the UNDT/UNAT example demonstrates that geographical distribution can coexist with gender balance. In a nutshell, internal UN practice supports not only gender parity in staffing but gender parity in judicial appointments.

The UN’s System-Wide Strategy on Gender Parity does not expressly mention ICJ judges. This is important, because a staffing policy is distinct from the political elections that empanel ICJ judges. How an internal administrative policy could create binding rules that apply to elections is the subject of Sections 5–7. For the purposes of this section, it is interesting to consider that, based solely on a fair reading of the plain text of the staffing policy, the position of ICJ Judges might be classed as a senior level of leadership within the UN. This interpretation can be made on the basis that ‘the Strategy is intended to have system-wide application’.¹⁰⁶ The strategy stresses the need to look at the so-called ‘top’ positions because ‘there is an inverse relationship across the [UN] system between seniority and women’s representation—the higher the grade, the larger the gap in gender parity’.¹⁰⁷ It states that ICJ staffing as a whole has achieved more than 50 per cent employment of women,¹⁰⁸ and yet its bench lags behind.

The UN’s System-Wide Strategy on Gender Parity also sheds light on the relationship between equitable geographic representation and gender parity. The UN’s staffing policy holds geographic diversity and gender equality as ‘parallel goals’ that ‘should be mutually reinforcing’ and enacted ‘simultaneously’.¹⁰⁹ Further, ‘geographic representation cannot be used as an excuse not to achieve gender parity’.¹¹⁰ Section 7 examines the ICJ’s long-standing practice of equitable geographical distribution. The UN’s parity policy’s coupling of these targets indicates their equal importance. It also implies the ICJ’s geographical representation achievements are not enough to fulfil equality under the Charter. A UNSG temporary special measure from 1996 allowing exceptions to geographical parity requirements in favour of gender parity in hiring demonstrates that, where there is tension between the two objectives, at least in some cases the UN’s parity policy prioritizes increasing women’s employment.¹¹¹

Viewed holistically, the UN’s gender parity policy demonstrates its intention to evolve the meaning of gender equality under the Charter.¹¹² While IOs are not parties to their constituting documents, ‘it is generally accepted that constituent instruments may be a source of international legal obligations for organisations.’¹¹³ If the Charter is interpreted pursuant to the familiar tenets of treaty interpretation codified in Article 31 of the Vienna Convention on the Law of Treaties (VCLT), there is a case that the Charter contains obligations to uphold gender equality and therefore parity.¹¹⁴ The strictest reading of the Charter regarding parity would result in obligations for the UNSG, and consequently all organs.¹¹⁵ This would affect the way the UNGA/UNSC handle ICJ nominations and elections. These impacts are the focus of Section 8.

An evolutive interpretation of Article 8 of the Charter would entail that ‘all United Nations organs should be obliged to respect Art. 8’.¹¹⁶ The Charter has rarely been amended, but this has not stopped significant modifications,¹¹⁷ aided by the Charter’s ‘open texture’.¹¹⁸ Article 8 implicates staffing at the ICJ in particular, as equal participation should occur ‘in principal and subsidiary organs’.¹¹⁹ Building on the fact that decades of UN practice demonstrate the existence of a belief in gender parity in staffing as a legal requirement, the next section analyses how the law of established practice transforms these repeated actions into binding rules of both conduct and result for both UN organs and UN member states.

4. Established practice, subsequent practice, and gender parity requirements

4.1 So-called established practice

This section examines the so-called established practice, or institutional practice,¹²⁰ of IOs in light of gender parity on the ICJ bench. Established practice should be distinguished from subsequent practice, with the former pertaining to the organization and the latter to states parties and/or member states.¹²¹

While established practice is less frequently discussed than subsequent practice,¹²² it is well-settled that established practice falls within the rules of an IO. In Personal Work, the Permanent Court of International Justice recognized that reference to international practice can help elucidate rules, characterizing resort to practice for ascertaining rules as ‘not unusual’.¹²³ This early reliance on IO practice for rule ascertainment has been expanded and formalized. Multiple treaties reference IO’s internal rules,¹²⁴ indicating that established practice of the organization is envisaged as falling within the rules of the organization.¹²⁵ The 2011 Draft Articles on the Responsibility of International Organizations¹²⁶ reflect an understanding of IO established practice as a source of binding rules for the IO and potentially its member states.

Established practice is a binding, quasi-customary law of an IO, based largely on the organization’s secondary law.¹²⁷ It can be understood as a third source of IO law, potentially modifying the IO’s constituent instruments.¹²⁸ The relevance of established practice as a source of binding law is particularly important for the argument infra regarding an evolutive interpretation of Article 9 of the Statute.

Established practice likely comprises ‘the combined effect of numerous consistent acts’.¹²⁹ In The Wall Advisory Opinion, the ICJ considered a wide range of organ practice to find established practice.¹³⁰ An IO’s established practice must not be ‘uncertain or disputed’ if it is considered binding.¹³¹ At the same time, established practice should be viewed as evolving rather than static.¹³²

Section 3.2 of this article outlines the UN’s numerous, consistent parity acts and a wide range of organ practice. These acts have continuously increased over time. As noted previously, the UN describes its gender parity policy as fulfilling the Charter’s purpose. This is important for understanding established practice regarding parity as a binding IO rule: if the UN asserts its action is taken to fulfil its Charter, this is presumed to be correct,¹³³ particularly when the UN is acting to fulfil the competence of its staff.¹³⁴ On this basis, it is likely that, via the extensive pattern of practice coupled with prolonged acquiescence from member states, the parity in staffing policy has coalesced into ‘Charter law’.¹³⁵ The duration of the parity policy aids this conclusion: the original target for gender parity staffing was 2000, member states supported this, and the policy has expanded in the last 20 years. Enduring state acceptance of IO practice can create binding obligations for states arising from this practice.¹³⁶

Thus, there is a case to be made that the UN gender parity policy is a customary rule for the organization based on established practice. This means the UN parity policy, even if not originally intended to do so, could bind member states and UN organs. While the Charter envisaged gender equality and did not discuss parity, the UN’s established practice demonstrates that, in the context of the UN, gender equality now requires gender parity. This could impact the ICJ bench in part because established practice’s scope is broad: ‘it has the potential to at least add substantive rules to the law of the organization, which have not been included into the constituent instruments.’¹³⁷ Established practice is lex specialis, making internal IO rules preeminent when interpreting the IO’s constituent instruments.¹³⁸ This is important for those who may wish to rebut such a rule, as it will create an additional challenge. Quayle notes that the ICJ has imbued established practice with strength and staying power, requiring new institutional practice to displace an existing practice-based rule.¹³⁹ IO established practice may generate CIL,¹⁴⁰ and may also modify how subsequent practice is understood.¹⁴¹ VCLT Article 5 does not contain or define the words ‘established practice’ nor explicitly state that such practice could displace subsequent practice or modify a treaty’s terms that pre-date the practice. However, The Wall suggests such an application.¹⁴² The next section examines how established practice and subsequent practice interact.

4.2 Subsequent practice through the lens of established practice

Interpreting the Charter requires considering member states’ subsequent practice and evidence of UN established practice. Considering such evidence means that how Charter provisions are interpreted may evolve over time. Subsequent practice is a time-honored tool for treaty modification.¹⁴³ It is vital in the context of the Charter, which is unable to change dynamically owing to its elaborate internal amendment requirements.¹⁴⁴ Such practice can inform the Charter’s application by circumventing international politics and power struggles between member states at the voting stages in the UNGA and UNSC.¹⁴⁵

In the Wall Opinion, the ICJ confirmed that the interpretation of a treaty could evolve over time by the practice of the IO formed under the treaty’s provisions. In this case, the ICJ noted that Article 12 of the Charter was initially interpreted as precluding the UNGA from considering matters while the UNSC was still dealing with them.¹⁴⁶ However, Article 12’s meaning evolved based on the subsequent practice of the UNGA, which repeatedly adopted recommendations for matters while these appeared on the Council’s agenda.¹⁴⁷ The ICJ considered that the ‘accepted practice’ of the UNGA had ‘evolved’ and was consistent with Article 12(1),¹⁴⁸ allowing the UNGA to refer the Palestinian matter to the Court for its advisory opinion.¹⁴⁹ Organ practice stood in for the practice of states parties regarding evolutive interpretation.

A significant proportion of states parties must agree to the change for subsequent practice to modify the meaning of a treaty provision,¹⁵⁰ creating a higher threshold of state practice than is needed to find established practice.¹⁵¹ It would be challenging for this article to quantify all UN member state positions on the UN’s gender parity staffing policy, gender parity on judicial benches, or gender parity on the bench of the ICJ. Less challenging is noting that member states have endorsed or at least allowed the UN parity policy over several decades. That the UN’s gender parity policy originated in the 1980s, and that the original target was 2000, are relevant in assessing the strength of the rules of the organization and member state reactions. Absence of member state opposition to ‘consistent institutional practice’ implies acquiescence to it and can substitute the need for agreement of all parties under Article 31 VCLT.¹⁵² VCLT Article 5 makes clear that an IO’s rules can modify and precede general rules, including Article 31(3)(b).¹⁵³ This means that established practice could change the way one weighs subsequent practice when considering the interpretation of an IO’s constituent instruments.¹⁵⁴

Academics have debated whether subsequent practice can modify a constituent treaty rather than simply provide evidence that is relevant to interpreting or applying its provisions.¹⁵⁵ ICJ jurisprudence implies that organ practice can have law-creating effect even without a high threshold of member state support/with strong member state opposition.¹⁵⁶ Additionally, when interpreting subsequent practice vis-à-vis established practice, the balance changes: ‘the established practice of the UN could have created a rule of the organization with the content that its subsequent practice does not strictly require the agreement of all the Member States’.¹⁵⁷

Some argue that subsequent practice is unsuitable for changing the composition of UN organs or the Charter’s fundamental underpinnings.¹⁵⁸ It would be difficult to claim that gender parity changes the Charter’s nature when the policy is ‘reflective of core values that are as old as the Organization itself’.¹⁵⁹ This is the case even if one asserts that the parity policy is limited to the UNSG and thus only one UN organ, because the ICJ has confirmed that one organ’s practice can establish agreement of parties.¹⁶⁰ Beyond the UNSG, the jurisprudence of the UNDT/UNAT links the parity policy to Article 8 in the Charter.¹⁶¹ Ultimately, the combined established and subsequent practice impacts how gender parity can be read into the ICJ Statute. The next section applies the UN’s established practice on gender parity to an evolutive interpretation of Article 9 of the ICJ Statute.

5. An evolutive interpretation of Article 9 of the ICJ Statute

If a term is deemed generic in nature and particularly if operation of the treaty in question is not time-bounded, the parties may be presumed to have intended the meaning of the relevant term to evolve over time.¹⁶² The text of Article 9 is broad, flexible, and suggestive of the possibility for evolutive interpretation, as does its drafting history.¹⁶³ Article 9 reads:

At every election, the electors shall bear in mind not only that the persons to be elected should individually possess the qualifications required, but also that in the body as a whole the representation of the main forms of civilization and of the principal legal systems of the world should be assured.

Reference to the Charter is appropriate to understand Article 9’s meaning,¹⁶⁴ as ‘the International Court of Justice [is] established by the Charter of the United Nations as the principal judicial organ of the United Nations’.¹⁶⁵ The distinction between the Statute and the Charter is ‘technical’ and not an indication of legal separation between the two.¹⁶⁶ The emphasis on gender equality in the Charter is significant, expressed in both the instrument’s preamble and substantive terms. These clauses indicate that the UN as a whole and the institutions that comprise it ought not only to espouse certain values but also to embody them. This is reinforced by Article 8 of the Charter: ‘The United Nations shall place no restrictions on the eligibility of men and women to participate in any capacity and under conditions of equality in its principal and subsidiary organs.’¹⁶⁷ As Section 4 articulates, established practice of the UN indicates that gender parity is necessary to fulfil Article 8 of the Charter and that this is an obligation of both conduct and result. Thus, Article 9 is subject to the ‘Charter law’ developed by UN gender parity policies and must be interpreted in accordance with it.

Article 9 of the Statute must be read ‘in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose’.¹⁶⁸ The key phrase is ‘the representation of the main forms of civilization … of the world’. Regarding gender parity, one should question what was intended by the words ‘representation’ and ‘main forms of civilisation’, and – more importantly – their meaning today. The Oxford English Dictionary states that representation refers to ‘[t]he fact or process of standing for, or in the place of, a person, group, institution, etc., esp. with the right or authority to speak or act on behalf of these’. This is the same meaning often given in political science and political sociology.¹⁶⁹ The question is, who may legitimately provide such representation? Arguably, this is someone who should be capable of ‘standing in the place of’ the people comprised from the ‘main forms of civilisation’.

The term ‘civilization’ is fraught, as is its colonial history in international law.¹⁷⁰ Tzouvala rightly observes that ‘civilization’ is not ‘a unitary legal concept lending itself to conclusive definition but … a mode of international legal argumentation’.¹⁷¹ Yet, while many things are open to debate, this demographic fact is not. Rather, given women’s historical exclusion from international law, focusing on women’s global prevalence as a basis for representation on the ICJ bench is aligned with transformative equality’s call to reconfigure institutions in order to re-centre the marginalized. A fair reading is that Article 9 cannot be fulfilled without gender parity, because there cannot be a representation of the main forms of civilization if 50 per cent of the members of each civilization are absent from the bench. Thus, the existence of a substantial gender imbalance on the ICJ’s bench suggests that the pure requirement of representation is not being met.¹⁷²

This article proposes expanding the well-established convention of geographical distribution used to ensure diverse representation on the ICJ bench to include gender parity.¹⁷³ The Charter’s text is suggestive that both gender equality and geographical representation should be considered together,¹⁷⁴ and the UN’s parity policy promotes this interpretation. Originally, the reference to the ‘main forms of civilization’ was a means of allaying Great Power fears of their possible future exclusion from the bench.¹⁷⁵ The wording was left sufficiently open to permit smaller states to envisage their equal involvement, while the Great Powers saw Article 9 as an exception to the principle of state equality.

Today, equitable geographical representation is tied to the UN’s system of regional groups.¹⁷⁶ The regional groups are another example of an extra-textual system that has been grafted onto vague Charter provisions. The history of regional groups, and how they evolved from an informal system to a formalized, powerful organizing principle that affects the composition of all of the most important UN offices, is relevant to gender parity on the ICJ bench. It evidences institutional flexibility at the UN, informed at least in part by legal norms, and in part by balance of power considerations. Article 9’s wording is similarly vague, contrasting with the Rome Statute’s Article 36(8)(a)(ii), which expressly requires ‘equitable geographical representation’ for ICC judges.¹⁷⁷ The convention of equitable geographical representation at the ICJ and the UN demonstrates established practice of reading in extra-textual representation requirements, providing a legal precedent for adding gender parity to the notion of representation in ICJ judicial nominations and elections.

Analysing the practice of equitable geographical distribution within the European Court of Justice (ECJ), Kenney examined how geographical parity in practice underscores why states historically have insisted on such representation. She asks,

If one need not “stand for” in order to “act for” in Pitkin’s definition—that is, if men can represent women and Italians can decide cases for Germans—why are Member States so determined to have their own representatives on the ECJ?¹⁷⁸

In Pitkin’s framework,¹⁷⁹ in representation through gender parity women judges will be ‘standing for’ women rather than ‘acting for’ women, the latter referring to how elected legislators might ‘act for’ their constituents by directly representing their interest. Kenney concludes that states feel that equitable geographical representation is necessary for protecting their interests, and that this representation ensures compliance with the court’s decisions by imbuing them with legitimacy.¹⁸⁰ This logic can be extended to gender parity on the bench. Even setting aside the substantive decision-making of women versus men judges discussed in Sections 2–3, women have no less a claim than member states of various IOs to equal representation on the basis of protecting their interests and ensuring the legitimacy of judicial decisions. Given this, the next section outlines a proposal for how to secure gender parity on the ICJ bench.

6. A proposal for achieving gender parity on the ICJ Bench

There are multiple ways that gender parity on the ICJ bench could be achieved. This article proposes practical pathways to secure this outcome. The subsequent sections consider how existing nomination and election procedures at the ICJ could incorporate gender parity; suggests changing the UNGA and UNSC rules of procedure; and addresses how member states might be brought on board with the proposal.

6.1 Incorporating gender parity into the ICJ’s existing nomination and election procedures

The nomination and election procedures for ICJ judges are general and flexible. As with geographical parity, the Statute’s broad stipulations allow for adopted conventions. This is a good thing for gender parity: the process for modifying the Statute is arduous, requiring an amendment to be approved by a two-thirds majority of the UNGA and ratified by two-thirds of the members of the UN, including all five permanent members of the UNSC.¹⁸¹ To date, the ICJ Statute has never been amended.¹⁸² Therefore, an alternative model might be sought.

National groups at the Permanent Court of Arbitration (PCA) are key players in the nomination process.¹⁸³ Articles 4 and 5 of the ICJ Statute state that the UNSG initiates nominations by requesting national groups to nominate between two and four candidates to the ICJ bench. At this stage, the UNSG could encourage gender-equal submissions in the same language as is done for the internal UN justice system. The Statute provides that in preparing their shortlists a state should ‘consult its highest court of justice, its legal faculties and schools of law, and its national academies and national sections of international academies devoted to the study of law’.¹⁸⁴ The Statute offers no guidelines or best practices and is silent on procedure at the municipal level, although academic writing sheds light into some national group consultations.¹⁸⁵ This absence of a formally codified procedure may facilitate new rules enshrining gender parity. It is worth noting that the research of Mackenzie et al.¹⁸⁶ and in person conversations between national group members and this article’s author suggest that in some cases, candidates are picked by governments.¹⁸⁷ Still, national groups remain an important focal point for securing gender parity on the ICJ bench.

After national groups nominate candidates, the UNGA and the UNSC vote separately on the candidates.¹⁸⁸ Candidates receiving an absolute majority of votes from both organs are elected.¹⁸⁹ Under Article 9, electors should keep representation in mind as they vote. Article 9 thus presents an opening for creating a gender-parity rule that could guide and constrain UNGA and UNSC voting. Such a rule might be achieved through a UNGA resolution, similar to the ECtHR rule established by the Parliamentary Assembly. Civil society organizations have lobbied for this option.¹⁹⁰ However, UNGA resolutions are non-binding on states, and UNSC resolutions are only binding on states via Chapter VII, which is not relevant to gender parity on the ICJ bench. An alternative is that each organ could modify its internal rules to require both gender parity in candidate lists and gender parity in voting. The next subsection discusses why the UNSG/UNGA might be obligated to do so and how this can be done.

6.2 Changing UNGA and UNSC rules of procedure to create gender parity on the ICJ bench

Because the UNGA and UNSC cast the final votes for ICJ judges, modifying the voting obligations of the UNGA and UNSC can create judicial gender parity. Amending the UNGA and UNSC rules of procedure is one avenue through which this could be achieved. Both the UNGA and UNSC have the power to make such changes, and there is precedent for them to do so.

Article 21 of the Charter gives the UNGA authority to adopt its own rules of procedure and the UNGA has amended these numerous times throughout the years.¹⁹¹ Article 30 of the Charter empowers the UNSC to adopt its own rules of procedure, which it has done. The UNSC may have even more flexibility regarding amending its rules of procedure given that it has chosen to keep these rules ‘provisional’.¹⁹² Unlike the UNGA’s procedural rule 163, there is no written rule regarding how to amend the UNSC rules of procedure, although in practice adoption of such rules are made via affirmative vote of at least nine UNSC members, per Article 27 of the Charter.¹⁹³ The text of Article 30 across multiple languages affords significant flexibility to the UNSC in how it decides its own rules of procedure,¹⁹⁴ and it may do so on an ad hoc basis; ‘in the case of formally established rules of procedure it would require the formal adoption of amendments to the existing wording’.¹⁹⁵ Thus, UNSC amendments regarding gender parity on the ICJ bench could be submitted and voted on per the normal rules. Such changes to the UNGA/UNSC rules of procedure would be legally binding upon each organ,¹⁹⁶ offering a strong foundation for both securing and maintaining parity.

Additional aspects of the Charter are suggestive that such amendments regarding ICJ elections would be both permissible and required. Article 103 of the Charter provides that the Charter shall take precedent over any internal procedural regulations, giving it primacy over both the UNGA and the UNSC’s rules of procedures. Given that the UNSG has interpreted the Charter to require gender parity in all UN staffing, and that both organs have endorsed this view, modification of their rules of procedure to achieve gender parity on the ICJ bench is indeed appropriate.

Thus, both the UNGA and UNSC could modify their internal rules of procedure regarding the election of ICJ judges and create a candidacy system similar to that of the ICC’s in which candidate lists that fail to contain sufficient numbers of women will lead to a request for a new list and another round of voting,¹⁹⁷ and with disqualification penalties for non-conforming lists like at the ACtHPR. The creation of the regional groups via a series of UNGA resolutions between 1957 and 1978 demonstrates the power of the UNGA to modify voting procedures at the UN for the purposes of equitable representation.¹⁹⁸ The Statute is flexible enough to allow the UNGA and UNSC to create rules that both reject candidate lists that do not meet gender parity thresholds and require gender parity thresholds to be met when casting election votes. For example, if seats remain unfilled after the first meeting held for election, the Statute contemplates that second and third meetings may take place, and if necessary, the UNGA and UNSC may form a joint conference for choosing names for remaining vacant seats.¹⁹⁹ This inherent flexibility renders it more likely that a gender parity rule would be compatible with the Statute.

6.3 Gaining member state support for gender parity on the ICJ bench

It is possible that, as was initially the case at the ECtHR,²⁰⁰ member states might oppose gender parity requirements and the idea of updating the UNGA/UNSC rules of procedure. Amending the rules of procedure would require member state support, as the UNGA/UNSC are comprised of member state representatives. This is compounded by the fact that UN member states are inclined to work via consensus in both the UNGA and UNSC,²⁰¹ creating a higher bar to achieve parity. However, arguably there are legal grounds for requiring member states to support gender parity on the ICJ bench. Additionally, trends towards gender parity rules domestically and supranationally indicate growing state practice in favour of parity on judicial benches. Further, lobbying at the municipal, regional, and international levels can create momentum that fosters member state support for gender parity on the ICJ bench.

Member states are bound by the fundamental human right of gender equality and are specifically instructed to create gender-equal access to the international plane.²⁰² Numerous states have employed these notions of substantive gender equality and gender parity to their domestic governments²⁰³ and via their roles in drafting gender balance rules at supranational courts, indicating their acceptance of this formulation of gender equality. While some states initially rejected these obligations on the international plane,²⁰⁴ in recent years more have championed gender parity in international representation.²⁰⁵ Since states must nominate individuals to the selection process for the ICJ, logically, obligations stipulated in human rights treaties to ensure equal representation apply to states in this context. States’ human rights duties necessitate ending discrimination through initiating and implementing affirmative measures for equal representation in international courts.²⁰⁶ There is evidence that IHRL applies to aspects of the appointment process that are under the exclusive jurisdiction of the state.²⁰⁷ When member states vote for appointments to international bodies, they cannot vote in a manner that contradicts or violates their international treaty obligations.²⁰⁸

Additionally, IO law such as the UN’s parity staffing policy may transfer UN obligations to member states. Organ practice can both create rules that bind member states and be indicative of state practice in favour of such rules.²⁰⁹ Under the theory of transfer of powers, member states have imbued the UN with the power to bind them.²¹⁰ Member states remain partially or fully responsible for an IO’s breach of international law and therefore obligated to remedy this breach.²¹¹ Consequently, if the UNGA and UNSC, given the roles they play in electing ICJ judges, were in breach of their gender equality obligations, states would be responsible for this and therefore required to remedy it. Further, if a state supports the organization’s breach through its actions – for example by failing to nominate women or enshrine the obligation to do so in policies or rules – this support independently gives rise to state responsibility and therefore the need for remedy.²¹² There is also the argument that states must prevent such breaches by IOs in order to uphold the integrity of international law as a system.²¹³

There are several reasons why the UNGA and UNSC may be obligated to ensure gender parity on the ICJ bench and that therefore member states would need to act to ensure this. ‘IO organs are themselves subject to international human rights law’,²¹⁴ and thus the obligation falls on the UNGA and UNSC. Moreover, the UNGA has explicitly confirmed its commitment to the UNSG’s gender parity policy on numerous occasions.²¹⁵ One could argue that the UNSC has shown its commitment to gender representation in decision making through nine different resolutions on women, peace, and security.²¹⁶ A key point is that each organ’s established practice indicates its understanding of the Charter,²¹⁷ and that UNSC/UNGA practice supports gender parity as a Charter requirement. As explained in Section 5, these policies have coalesced into established practice and therefore binding rules for the organization as a whole.

As these organs have not yet taken action to create gender parity on the ICJ bench, if one accepts their obligation to do so, they are in breach and must supply a remedy.²¹⁸ Responsibility for a breach presupposes such things as legal personality, that those deemed responsible had some say in the matter, and that they can meaningfully regulate their behaviour.²¹⁹ The UNGA and UNSC have a say in the matter and can regulate their behaviour because they cast the final votes for ICJ judges and can amend their internal rules that govern this voting; they have a direct impact on the ICJ’s bench and modifying their voting obligations can create gender parity on it. Reparation for Injuries has made clear that IOs are required to provide remedies when in breach of an obligation.²²⁰ While not expressly listed amongst possible remedies,²²¹ a direct method for rectifying the breach could be achieved via modifications by the UNGA and UNSC to their internal procedural rules. New rules requiring that candidate lists present gender parity, and that voting must take gender parity into account, can rectify the imbalance on the ICJ bench.

Even setting aside the notion that member states have gender parity obligations regarding nominations to the ICJ, the UNGA and UNSC have the power to create a new requirement to this effect. Member states might protest that such a change in the nomination and voting procedures creates new international law by transferring UN parity requirements to states, but organ power to this effect is well accepted.²²² Certain Expenses dictates that ‘each organ must, in the first place at least determine its own jurisdiction’.²²³ If the UNGA/UNSC were to modify their internal rules to create gender parity on the ICJ bench, they would be assumed in the first instance to have the competency to do this. State endorsement of the UNGA’s role in the UNDT/UNAT judicial elections per Section 3 shows state practice that could favour this conclusion.

States might object that this interferes with their domestic affairs. However, while states have since the advent of the UNGA’s first session asserted that Article 2(7) of the Charter prevents the UNGA from intervening in the domestic jurisdiction of a state including regarding human rights issues, the UNGA has repeatedly rejected this argument and declined to address so-called domestic matters on this basis except extremely exceptionally.²²⁴ Rather, expansion of UNGA powers is common and settled law and the UNGA and UNSC have confirmed their key role in protecting and fulfilling human rights.²²⁵ It is not unusual for the UNGA’s rules of procedure to impact member states beyond the UNGA itself, for example regarding accreditation, admissions of new members, and elections.²²⁶ In general, IOs and their organs have ‘considerable autonomy’ regarding their internal rules,²²⁷ and this is so even when changes influence member states and even when they object.²²⁸ When it comes to interpreting the constituent documents, and in this case, interpreting if the Charter requires gender parity on the ICJ bench, ‘the balance of power shifts away from Member states to organs’.²²⁹ This further empowers them to modify their internal rules even if member states were opposed.

Member states might argue that it would be impossible for them to comply with gender parity candidate requirements because they were incapable of locating and nominating sufficient numbers of qualified women. States could not be held responsible for an unworkable task.²³⁰ However, this so-called ‘pipeline’ or ‘pool’ argument has been robustly debunked.²³¹ Further, state practice belies such reasoning: if member states can repeatedly make judicial nominations for the ICC, the ACtHPR, and UNAT/UNDT that satisfy gender parity, correspondingly they can for the ICJ. Against the specter that there are simply not enough women candidates for the job, it is important to bear in mind that currently only four additional qualified women need appointing to create gender parity.

Furthermore, while not always accessible for every region, data on women in the legal profession suggests that women comprise a sizeable proportion of high-level legal posts from which to draw ICJ judicial candidates. Women in OECD countries constitute more than 54 per cent of the judiciary,²³² and more than 30 per cent of all Supreme Court judges in Latin America are women.²³³ Hennette Vauchez’s analysis of the CVs of women candidates put forth by states for appointment to the ECtHR reveals both qualified women and the possibility that states overlook additional capable women candidates in their selection process.²³⁴ Beyond numbers, Grossman has pointed out that the data on the ‘pool’ does not tell the whole story.²³⁵ The political and opaque nature of ICJ judicial nominations, which often do not involve a candidate list or pool, but rather one well-connected individual,²³⁶ is much more at fault. Thus, claims that the organs cannot affect the candidate list because it impinges on member state rights or creates an impossible task would likely fail on multiple grounds, paving the way for procedural change if the UNSC/UNGA – and the member states that comprise them – chose this route.

Despite these legal arguments, member states or other actors at the UN might oppose the suggested changes; identifying legal obligations does not guarantee compliance with them. It will be important to couple the doctrinal reasons articulated in this article with persuasion.²³⁷ How much and what must be done to persuade states is an open question, as is where in the ‘life-cycle’ of a norm – norm emergence, norm cascade, or norm internalization²³⁸ – gender parity on the ICJ bench sits. To secure parity on the ICJ bench, a critical mass of supportive states is needed.²³⁹ Lobbying efforts, such as those conducted by GQUAL, the campaign for gender parity in international representation,²⁴⁰ might play a part in encouraging key national governments to support gender parity on the ICJ. Research suggests that the support of less than one third of states is needed to create such change; the support of ‘critical’ states and the ‘socialization’ of other states can turn a minority position into consensus.²⁴¹ Gender parity in international appointments remains salient and urgent. It is highly unlikely that the call for gender parity on international benches will disappear, increasing the possibility that member states will support parity on the ICJ bench.

7. Conclusion: Compelling legal reasons to create gender parity on the ICJ bench

There are compelling reasons to consider gender parity on the ICJ bench and multiple legal, not just policy or moral, arguments in favour of parity. This article has posited that, over time, the customary law of the UN has evolved to support gender parity on the ICJ bench. This in turn affects the interpretation of Article 9 of the Statute regarding judicial selection, so that UN organs such as the UNGA and the UNSC are either legally permitted or legally obligated to ensure gender parity on the ICJ bench. Not only are there grounds for the UNGA and the UNSG to take parity measures ‘establishing real accountability, backed by consequences’²⁴² such as those seen at the ACtHPR, the UNGA and UNSC can achieve this by modifying their internal rules of procedure. There is ample precedent for them to mandate such obligations via their procedural rules, even bearing in mind that this would impact member states. Such an outcome is supported by conventions on equitable geographical distribution. Gender should be given equal weight as geography: there is evidence that the two goals can be achieved symbiotically, and that it is at times appropriate to favour gender parity over geographical distribution. Importantly, only through clear and enforceable measures has parity thus far been achieved and protected. For these reasons, the UNGA and UNSC should act now to modify their rules and create a gender equal ICJ bench.