The provisional measures order recently published by the International Court of Justice (ICJ) in the ongoing dispute between South Africa and Israel has widely been characterized as a warning to States that they risk violating the order and eventually being held complicit in genocide for continuing to provide security assistance to Israel.
A legal advisor for the Middle East and North Africa program of the International Commission of Jurists, for example, reportedly told Al Jazeera that the ICJ “considers there to be a serious risk of genocide in Gaza.” This seemingly triggers the duty of all States that have ratified the Genocide Convention “to take concrete steps to prevent genocide, including by ceasing arms sales and exports and other assistance that could facilitate genocidal acts.”
In a recent blog post here on EJIL: Talk!, Yussef Al Tamimi likewise concludes that States providing security assistance to Israel “will now have to be particularly cautious of” possibly violating certain obligations reflected in the Genocide Convention such as failure to prevent genocide or even potential complicity in committing genocide themselves. This recent analysis begins by presenting commentary from a number of sources reaching similar conclusions.
Given the ongoing debate regarding the provision of security assistance to Israel, what effect will the ICJ provisional measures order have domestically regarding compliance with international law obligations? Do States providing security assistance to Israel genuinely risk complicity in genocide if they continue providing security assistance to Israel?
In short, no—at least, not based on the ICJ provisional measures order alone. That is, the order does not present an elevated risk that States will be complicit in genocide or will violate the obligation to prevent genocide by supplying weapons and other security assistance to Israel as a matter of international law. This is the case primarily for two separate reasons.
First, the “plausibility” determination at the provisional measures stage is situated so low on the standard of proof spectrum that the order is not alone a sufficient basis to find a serious risk exists genocide is being committed and that States must take action to prevent it. Second, concluding it is plausible that some rights claimed by the applicant must be protected does not involve the same legal methodology utilized to determine whether an offender is committing genocide.
Another factor to consider regarding the practical effect of the ICJ provisional measures order is that the Court merely—and seemingly rather deliberately—reiterates and reinforces Israel’s existing international law obligations. If calling for Israel to implement the provisional measures order means precisely the same as saying Israel must comply with obligations that existed before the order was published, then the ICJ provisional measures order has no actual effect in practice.
Each of these considerations is addressed in greater detail below, starting with clarifying the standard of proof required to support a determination that a State party to the Genocide Convention has failed to comply with the obligation to prevent genocide.
Comparing Standard of Proof for Provisional Measures and Merits at ICJ
Judges at the ICJ have gradually developed different standards of proof for various legal requirements. Because the standards are derived from the bench, they can be adjusted when necessary. However, significant methods of assessment or standards of proof generally remain static unless a compelling reason for adjustment is identified.
In response to a request from a party to issue provisional measures, the Court has determined that it may do so “only if it is satisfied that the rights asserted by the party requesting such measures are at least plausible” (para. 35, emphasis added). This standard was applied in the recent ICJ provisional measures order in the ongoing South Africa v. Israel case (para. 54).
Assuming the dispute eventually reaches the merits stage and the Court renders a decision regarding South Africa’s claim that Israel is committing genocide, the requirement for inferring specific intent will be inordinately higher. As Justice Georg Nolte of Germany noted in his individual declaration, the intent to destroy a group in whole or in part a protected group “can only be inferred from ‘a pattern of conduct’ if this is the ‘only reasonable inference that can be drawn’ therefrom” (para. 8, emphasis added).
The chasm between “at least plausible” at the provisional measures stage and “only reasonable inference” is vast. To put the gap into perspective, it is useful to visualize conceptually how the two relate to some common standards of proof utilized in the domestic context.
Situating Relevant Requirements on a Typical Standard of Proof Spectrum
Although different jurisdictions utilize slightly different terminology or standards for various legal applications, generally the spectrum can include standards such as (from least demanding to most): some credible evidence, reasonable suspicion, probable cause, preponderance of the evidence, clear and convincing evidence, and beyond reasonable doubt.
If the two standards of proof developed in ICJ jurisprudence that are relevant to the present inquiry were grafted onto this spectrum, “at least plausible” would be included slightly before “some credible evidence,” while “only reasonable inference” would appear no lower than “beyond reasonable doubt.”
The “only reasonable inference” and the “beyond reasonable doubt” standards must be compared by analogy since the former has been developed for the particular purpose of determining, at the merits stage, whether the specific intent exists that is required for a determination that genocide is being committed, while the latter is a prescribed standard of proof that is required for a finding of guilt in a criminal case (see, for example, Rule 87(A) of the ICTY RPE and Article 66(3) of the Rome Statute).
Before comparing the “only reasonable inference” and “beyond reasonable doubt” requirements, it should be noted that some controversy remains regarding the precise meaning of “beyond reasonable doubt” as a standard of proof in International Criminal Court jurisprudence. This controversy largely reflects divergent approaches in national jurisdictions from a comparative perspective.
Comparing “Beyond Reasonable Doubt” and “Only Reasonable Inference” Standards
In ICTY jurisprudence, the meaning of “beyond reasonable doubt” requires a finding “that there is no reasonable explanation of the evidence other than the guilt of the accused” (para. 220, emphasis added). This formulation is similar to the “only reasonable inference” standard developed by the ICJ in the non-criminal context of an affirmative finding for the specific intent to commit genocide.
Whatever the precise meaning of “beyond reasonable doubt” in the international criminal law context, then, the “only reasonable inference” method of determining intent can be conceptualized by analogy as no less demanding. Even if the standards are considered to be conceptually equivalent, they are both situated on the opposite end as “some credible evidence” on the standard of proof spectrum.
Though it is almost certain that a final decision will not be rendered by the ICJ until long after the current stage of the conflict in Gaza is over, a finding on the merits that the requisite intent is the “only reasonable inference” would unquestionably put States on notice of a potential breach of the obligation to prevent genocide.
Currently, the ICJ has determined that “at least some of the rights” for which South Africa is seeking protection in relation to the population of Palestine “are plausible” (para. 54). All States are on notice of this determination reflected in the ICJ provisional measures order, but this is about as far away on the standard of proof spectrum as conceptually possible from a finding that Israel is in fact committing genocide in Gaza.
Standards of Fault for Complicity or Failure to Prevent Genocide
The ICJ has developed two separate standards of fault that could potentially apply to States providing security assistance to Israel—one for being complicit in genocide and another for failing to prevent genocide. As the Court observed in the Bosnia v. Serbia judgment when distinguishing between the two, “complicity results from commission” of genocide, while “violation of the obligation to prevent results from omission” (para. 432).
Of the two, complicity is more difficult to establish since it requires the accomplice to be “aware of the specific intent (dolus specialis) of the principal perpetrator” to commit genocide. Regarding the prevention of genocide, the Court has determined this obligation is activated “at the instant that the State learns of, or should normally have learned of, the existence of a serious risk that genocide will be committed.”
A provisional measures order from the Court is not alone a sufficient legal basis upon which to assert there is a “serious risk” that Israel is committing genocide in Gaza, thus potentially activating the obligation for other States to prevent genocide, for at least two primary reasons.
First, the finding at the provisional measures stage is on the protection of rights rather than the commission of genocide. Analyzing these standards requires divergent methodology. The process for determining whether rights must be protected is centered on the affected population, while evaluating whether genocide has been committed focuses on the (specific) intent of the alleged perpetrator. Therefore, finding that “at least some of the rights” South Africa claims must be protected on behalf of Palestinians in Gaza “are plausible” at the provisional measures stage does not in itself constitute a “serious risk” that persons taking action on behalf of Israel demonstrate the specific intent required for an affirmative finding that genocide is being committed.
The conceptual distinction between the plausibility that some rights must be protected at the provisional measures stage and a determination that genocide has in fact been committed at the merits stage is emphasized in the provisional measures order and in several separate individual opinions. On this account, the explanation Judge Nolte provides for his vote in favor of all provisional measures is particularly concise.
As Judge Nolte describes, he is “not persuaded that South Africa has plausibly shown that the military operation undertaken by Israel, as such, is being pursued with genocidal intent” (para. 13, emphasis added). Nonetheless, in light of select statements of Israeli government officials and the conditions on the ground in Gaza, Judge Nolte expresses the view that South Africa has “shown that some, but not all, of the rights which it has alleged are plausible at the present preliminary stage of the proceedings” (para. 17, emphasis added) (referencing para. 54 of the Order).
The provisional measures indicated by the Court, as Judge Nolte emphasizes, respond “to certain plausible risks for the rights of Palestinians in the Gaza Strip deriving from the Genocide Convention” (para. 17, emphasis added). This conceptual switch, from the intent to commit genocide to the protection of rights, occurs only at the provisional measures stage.
On the merits, the specific intent to commit genocide must be found to be the only reasonable inference that can be drawn from the factual record presented to the Court. That inquiry will not involve a conceptual switch from the intent of Israel to the rights of Palestinians, which means indicating provisional measures does not alone put States providing security assistance on notice of a “serious risk that genocide will be committed” by Israel in Gaza.
The second reason the ICJ provisional measures order does not alone present a “serious risk” that States providing security assistance to Israel may now be violating the obligation to prevent genocide as a matter of international law is the disparity between standards of proof utilized at the provisional measures and merits stages. As analyzed above, because the determination that it is plausible that at least some rights must be protected is situated so far down on the standard of proof spectrum, asserting this qualifies as a “serious risk” that Israel is committing genocide and States now have an obligation to intervene is unsustainable.
Therefore, the assertion that States risk a future determination that they are violating the obligation to prevent genocide for continuing to provide security assistance to Israel because the ICJ has now issued a provisional measures order is not supportable as a matter of international law.
Comparing Provisional Measures Directed by ICJ with Those Requested by South Africa
The provisional measures order published by the ICJ in the ongoing dispute between South Africa and Israel has been widely hailed as a landmark decision that puts all States “on notice” of the prospect of being complicit in genocide for providing security assistance to Israel. However, this assertion does not withstand critical scrutiny.
Comparing the provisional measures actually directed by the Court (para. 86) with the measures requested in the initial application (para. 144) suggests that this order from the ICJ should be genuinely regarded as a diplomatic setback for South Africa. Instead of the measures requested by South Africa, the Court merely reinforced the status quo that existed before the provisional measures order was published.
Provisional measures 1, 2, and 3 reaffirm obligations that already exist pursuant to the Genocide Convention. The ICJ has previously determined the content of provisional measure 5 (regarding preservation of evidence) to be by its “very nature…aimed at preserving rights” that are already reflected in the Genocide Convention.
Measure 4 reiterates obligations on all parties established by UN Security Council Resolution 2720 (2023), and this Resolution is cited as a reference in the introduction to the ICJ provisional measures order. Aside from provisional measure 6, requiring a report within 30 days of publication (that was reportedly recently submitted by Israel), all obligations directed by the Court existed prior to publication of the order.
Calls for Israel to comply with the ICJ provisional measures order—and for other States to pressure Israel to do so—have been a fixture of public discourse since the order was published (for example, here, here, and here). If the order merely reinforces obligations that existed before it was published, demands for Israel to “comply” with the provisional measures order are precisely equivalent to calling on Israel to fulfil obligations that existed before the order was published.
Although an official for the governing African National Congress of South Africa has hailed the decision as “the first step to end the violence… against the innocent Palestinian people,” in reality the ICJ order has had no practical effect on the ongoing hostilities since the provisional measures directed impose no new obligations—aside from filing the 30-day report.
In light of the provisional measures requested by South Africa but not adopted by the Court, and considering that the published measures simply reinforce the status quo, the most favorable light for South Africa that can be realistically cast on the order is that the ICJ elected not to grant Israel’s request to dismiss the application altogether. Advancing a claim across the initial threshold and into court is something, but it is nowhere near a final determination on the merits that Israel is in fact committing genocide in Gaza.
Conclusion
Notwithstanding the high profile status of the International Court of Justice and the gravity of the accusations made by South Africa against Israel, the recent provisional measures order published by the Court does not alone increase the risk that a State may be deemed complicit in genocide for providing security assistance to Israel.
Separating policy preferences from legal requirements, and fostering an accurate understanding of the doctrinal application of actual legal obligations, is vitally important for encouraging a productive and balanced debate regarding the advisability of continuing to provide security assistance to Israel. This is especially true in light of the undeniable humanitarian catastrophe that persists as a result of the conflict in Gaza.
There are many important considerations to weigh as the debate continues regarding the advisability and permissibility of continuing to supply weapons to Israel as the conflict in Gaza rages on. Because of the extraordinarily low standard of proof utilized and the actual substance of the provisional measures order recently published by the ICJ in the ongoing dispute between South Africa and Israel, this ICJ order should be regarded among the least of these concerns.
Evaluating Security Assistance to Israel Following ICJ Provisional Measures Order
Written by Brian L. CoxThe provisional measures order recently published by the International Court of Justice (ICJ) in the ongoing dispute between South Africa and Israel has widely been characterized as a warning to States that they risk violating the order and eventually being held complicit in genocide for continuing to provide security assistance to Israel.
A legal advisor for the Middle East and North Africa program of the International Commission of Jurists, for example, reportedly told Al Jazeera that the ICJ “considers there to be a serious risk of genocide in Gaza.” This seemingly triggers the duty of all States that have ratified the Genocide Convention “to take concrete steps to prevent genocide, including by ceasing arms sales and exports and other assistance that could facilitate genocidal acts.”
In a recent blog post here on EJIL: Talk!, Yussef Al Tamimi likewise concludes that States providing security assistance to Israel “will now have to be particularly cautious of” possibly violating certain obligations reflected in the Genocide Convention such as failure to prevent genocide or even potential complicity in committing genocide themselves. This recent analysis begins by presenting commentary from a number of sources reaching similar conclusions.
Given the ongoing debate regarding the provision of security assistance to Israel, what effect will the ICJ provisional measures order have domestically regarding compliance with international law obligations? Do States providing security assistance to Israel genuinely risk complicity in genocide if they continue providing security assistance to Israel?
In short, no—at least, not based on the ICJ provisional measures order alone. That is, the order does not present an elevated risk that States will be complicit in genocide or will violate the obligation to prevent genocide by supplying weapons and other security assistance to Israel as a matter of international law. This is the case primarily for two separate reasons.
First, the “plausibility” determination at the provisional measures stage is situated so low on the standard of proof spectrum that the order is not alone a sufficient basis to find a serious risk exists genocide is being committed and that States must take action to prevent it. Second, concluding it is plausible that some rights claimed by the applicant must be protected does not involve the same legal methodology utilized to determine whether an offender is committing genocide.
Another factor to consider regarding the practical effect of the ICJ provisional measures order is that the Court merely—and seemingly rather deliberately—reiterates and reinforces Israel’s existing international law obligations. If calling for Israel to implement the provisional measures order means precisely the same as saying Israel must comply with obligations that existed before the order was published, then the ICJ provisional measures order has no actual effect in practice.
Each of these considerations is addressed in greater detail below, starting with clarifying the standard of proof required to support a determination that a State party to the Genocide Convention has failed to comply with the obligation to prevent genocide.
Comparing Standard of Proof for Provisional Measures and Merits at ICJ
Judges at the ICJ have gradually developed different standards of proof for various legal requirements. Because the standards are derived from the bench, they can be adjusted when necessary. However, significant methods of assessment or standards of proof generally remain static unless a compelling reason for adjustment is identified.
In response to a request from a party to issue provisional measures, the Court has determined that it may do so “only if it is satisfied that the rights asserted by the party requesting such measures are at least plausible” (para. 35, emphasis added). This standard was applied in the recent ICJ provisional measures order in the ongoing South Africa v. Israel case (para. 54).
Assuming the dispute eventually reaches the merits stage and the Court renders a decision regarding South Africa’s claim that Israel is committing genocide, the requirement for inferring specific intent will be inordinately higher. As Justice Georg Nolte of Germany noted in his individual declaration, the intent to destroy a group in whole or in part a protected group “can only be inferred from ‘a pattern of conduct’ if this is the ‘only reasonable inference that can be drawn’ therefrom” (para. 8, emphasis added).
The chasm between “at least plausible” at the provisional measures stage and “only reasonable inference” is vast. To put the gap into perspective, it is useful to visualize conceptually how the two relate to some common standards of proof utilized in the domestic context.
Situating Relevant Requirements on a Typical Standard of Proof Spectrum
Although different jurisdictions utilize slightly different terminology or standards for various legal applications, generally the spectrum can include standards such as (from least demanding to most): some credible evidence, reasonable suspicion, probable cause, preponderance of the evidence, clear and convincing evidence, and beyond reasonable doubt.
If the two standards of proof developed in ICJ jurisprudence that are relevant to the present inquiry were grafted onto this spectrum, “at least plausible” would be included slightly before “some credible evidence,” while “only reasonable inference” would appear no lower than “beyond reasonable doubt.”
The “only reasonable inference” and the “beyond reasonable doubt” standards must be compared by analogy since the former has been developed for the particular purpose of determining, at the merits stage, whether the specific intent exists that is required for a determination that genocide is being committed, while the latter is a prescribed standard of proof that is required for a finding of guilt in a criminal case (see, for example, Rule 87(A) of the ICTY RPE and Article 66(3) of the Rome Statute).
Before comparing the “only reasonable inference” and “beyond reasonable doubt” requirements, it should be noted that some controversy remains regarding the precise meaning of “beyond reasonable doubt” as a standard of proof in International Criminal Court jurisprudence. This controversy largely reflects divergent approaches in national jurisdictions from a comparative perspective.
Comparing “Beyond Reasonable Doubt” and “Only Reasonable Inference” Standards
In ICTY jurisprudence, the meaning of “beyond reasonable doubt” requires a finding “that there is no reasonable explanation of the evidence other than the guilt of the accused” (para. 220, emphasis added). This formulation is similar to the “only reasonable inference” standard developed by the ICJ in the non-criminal context of an affirmative finding for the specific intent to commit genocide.
Whatever the precise meaning of “beyond reasonable doubt” in the international criminal law context, then, the “only reasonable inference” method of determining intent can be conceptualized by analogy as no less demanding. Even if the standards are considered to be conceptually equivalent, they are both situated on the opposite end as “some credible evidence” on the standard of proof spectrum.
Though it is almost certain that a final decision will not be rendered by the ICJ until long after the current stage of the conflict in Gaza is over, a finding on the merits that the requisite intent is the “only reasonable inference” would unquestionably put States on notice of a potential breach of the obligation to prevent genocide.
Currently, the ICJ has determined that “at least some of the rights” for which South Africa is seeking protection in relation to the population of Palestine “are plausible” (para. 54). All States are on notice of this determination reflected in the ICJ provisional measures order, but this is about as far away on the standard of proof spectrum as conceptually possible from a finding that Israel is in fact committing genocide in Gaza.
Standards of Fault for Complicity or Failure to Prevent Genocide
The ICJ has developed two separate standards of fault that could potentially apply to States providing security assistance to Israel—one for being complicit in genocide and another for failing to prevent genocide. As the Court observed in the Bosnia v. Serbia judgment when distinguishing between the two, “complicity results from commission” of genocide, while “violation of the obligation to prevent results from omission” (para. 432).
Of the two, complicity is more difficult to establish since it requires the accomplice to be “aware of the specific intent (dolus specialis) of the principal perpetrator” to commit genocide. Regarding the prevention of genocide, the Court has determined this obligation is activated “at the instant that the State learns of, or should normally have learned of, the existence of a serious risk that genocide will be committed.”
A provisional measures order from the Court is not alone a sufficient legal basis upon which to assert there is a “serious risk” that Israel is committing genocide in Gaza, thus potentially activating the obligation for other States to prevent genocide, for at least two primary reasons.
First, the finding at the provisional measures stage is on the protection of rights rather than the commission of genocide. Analyzing these standards requires divergent methodology. The process for determining whether rights must be protected is centered on the affected population, while evaluating whether genocide has been committed focuses on the (specific) intent of the alleged perpetrator. Therefore, finding that “at least some of the rights” South Africa claims must be protected on behalf of Palestinians in Gaza “are plausible” at the provisional measures stage does not in itself constitute a “serious risk” that persons taking action on behalf of Israel demonstrate the specific intent required for an affirmative finding that genocide is being committed.
The conceptual distinction between the plausibility that some rights must be protected at the provisional measures stage and a determination that genocide has in fact been committed at the merits stage is emphasized in the provisional measures order and in several separate individual opinions. On this account, the explanation Judge Nolte provides for his vote in favor of all provisional measures is particularly concise.
As Judge Nolte describes, he is “not persuaded that South Africa has plausibly shown that the military operation undertaken by Israel, as such, is being pursued with genocidal intent” (para. 13, emphasis added). Nonetheless, in light of select statements of Israeli government officials and the conditions on the ground in Gaza, Judge Nolte expresses the view that South Africa has “shown that some, but not all, of the rights which it has alleged are plausible at the present preliminary stage of the proceedings” (para. 17, emphasis added) (referencing para. 54 of the Order).
The provisional measures indicated by the Court, as Judge Nolte emphasizes, respond “to certain plausible risks for the rights of Palestinians in the Gaza Strip deriving from the Genocide Convention” (para. 17, emphasis added). This conceptual switch, from the intent to commit genocide to the protection of rights, occurs only at the provisional measures stage.
On the merits, the specific intent to commit genocide must be found to be the only reasonable inference that can be drawn from the factual record presented to the Court. That inquiry will not involve a conceptual switch from the intent of Israel to the rights of Palestinians, which means indicating provisional measures does not alone put States providing security assistance on notice of a “serious risk that genocide will be committed” by Israel in Gaza.
The second reason the ICJ provisional measures order does not alone present a “serious risk” that States providing security assistance to Israel may now be violating the obligation to prevent genocide as a matter of international law is the disparity between standards of proof utilized at the provisional measures and merits stages. As analyzed above, because the determination that it is plausible that at least some rights must be protected is situated so far down on the standard of proof spectrum, asserting this qualifies as a “serious risk” that Israel is committing genocide and States now have an obligation to intervene is unsustainable.
Therefore, the assertion that States risk a future determination that they are violating the obligation to prevent genocide for continuing to provide security assistance to Israel because the ICJ has now issued a provisional measures order is not supportable as a matter of international law.
Comparing Provisional Measures Directed by ICJ with Those Requested by South Africa
The provisional measures order published by the ICJ in the ongoing dispute between South Africa and Israel has been widely hailed as a landmark decision that puts all States “on notice” of the prospect of being complicit in genocide for providing security assistance to Israel. However, this assertion does not withstand critical scrutiny.
Comparing the provisional measures actually directed by the Court (para. 86) with the measures requested in the initial application (para. 144) suggests that this order from the ICJ should be genuinely regarded as a diplomatic setback for South Africa. Instead of the measures requested by South Africa, the Court merely reinforced the status quo that existed before the provisional measures order was published.
Provisional measures 1, 2, and 3 reaffirm obligations that already exist pursuant to the Genocide Convention. The ICJ has previously determined the content of provisional measure 5 (regarding preservation of evidence) to be by its “very nature…aimed at preserving rights” that are already reflected in the Genocide Convention.
Measure 4 reiterates obligations on all parties established by UN Security Council Resolution 2720 (2023), and this Resolution is cited as a reference in the introduction to the ICJ provisional measures order. Aside from provisional measure 6, requiring a report within 30 days of publication (that was reportedly recently submitted by Israel), all obligations directed by the Court existed prior to publication of the order.
Calls for Israel to comply with the ICJ provisional measures order—and for other States to pressure Israel to do so—have been a fixture of public discourse since the order was published (for example, here, here, and here). If the order merely reinforces obligations that existed before it was published, demands for Israel to “comply” with the provisional measures order are precisely equivalent to calling on Israel to fulfil obligations that existed before the order was published.
Although an official for the governing African National Congress of South Africa has hailed the decision as “the first step to end the violence… against the innocent Palestinian people,” in reality the ICJ order has had no practical effect on the ongoing hostilities since the provisional measures directed impose no new obligations—aside from filing the 30-day report.
In light of the provisional measures requested by South Africa but not adopted by the Court, and considering that the published measures simply reinforce the status quo, the most favorable light for South Africa that can be realistically cast on the order is that the ICJ elected not to grant Israel’s request to dismiss the application altogether. Advancing a claim across the initial threshold and into court is something, but it is nowhere near a final determination on the merits that Israel is in fact committing genocide in Gaza.
Conclusion
Notwithstanding the high profile status of the International Court of Justice and the gravity of the accusations made by South Africa against Israel, the recent provisional measures order published by the Court does not alone increase the risk that a State may be deemed complicit in genocide for providing security assistance to Israel.
Separating policy preferences from legal requirements, and fostering an accurate understanding of the doctrinal application of actual legal obligations, is vitally important for encouraging a productive and balanced debate regarding the advisability of continuing to provide security assistance to Israel. This is especially true in light of the undeniable humanitarian catastrophe that persists as a result of the conflict in Gaza.
There are many important considerations to weigh as the debate continues regarding the advisability and permissibility of continuing to supply weapons to Israel as the conflict in Gaza rages on. Because of the extraordinarily low standard of proof utilized and the actual substance of the provisional measures order recently published by the ICJ in the ongoing dispute between South Africa and Israel, this ICJ order should be regarded among the least of these concerns.
Share this:
Related
Categories
Tags
Leave a Comment
Comments for this post are closed
Comments
Nicolas Boeglin says
March 7, 2024
Dear Professor Cox
Many thanks for this post.
Considering that South Africa submitted its first application to the ICJ on 29 December after the adoption of UNSC Resolution 2720 (a text which was extremely weak due to the threat of a US veto), and considering that yesterday, 6 March 2024, South Africa submitted a third urgent request on provisional measures to the ICJ, after the third US veto observed last 20 February, I would like to know, if in your view, part of South Africa's actions can be explained by the inability observed at UNSC to stop Israel's destructive impetus in Gaza, in open violation of the most basic rules of international humanitarian law.
In the request of South Africa registered yesterday March 6, we read that Bosnia vs Serbia case you referred is also used by South Africa:
"33. South Africa fears that this Application may be the last opportunity that this Court shall have to save the Palestinian people in Gaza already dying of starvation, and now “one step” from famine.
34. In the Bosnian Genocide case, the Court declined to order the additional provisional measures requested on 27 July 1993. Within two years, approximately 7,336 Bosnians in the so-called ‘safe area’ of Srebrenica had been slaughtered, in what this Court retrospectively determined to have been a genocide.
Here, South Africa respectfully calls on this Court to act again now — before it is too late — to do what is within its power to save Palestinians in Gaza from genocidal starvation".
Source:
https://www.icj-cij.org/sites/default/files/case-related/192/192-20240306-wri-01-00-en.pdf
Yours sincerely
Nicolas Boeglin
Brian L. Cox says
March 8, 2024
Nicolas,
My view of the subsequent requests submitted by South Africa is they indicate this application is not actually about preventing an alleged genocide. The Genocide Convention is a convenient hook because of the compulsory jurisdiction clause, and the ICJ itself opened the door to these questionable applications by accepting erga omnes jurisdiction in Gambia v. Myanmar. Based on the content of the initial request for provisional measures and the subsequent correspondence from SA, my assessment is that the case is actually an attempt to generate global political pressure on Israel - and the only way to get across the threshold and into court is to characterize Israel's conduct in Gaza as a genocide.
That assessment of the genuineness of the application is supported by the nature of the rhetoric presented in the passages you quote from their most recent request. This excerpt in particular stands out to me on that point: indicating additional measures may be the "last opportunity that this Court shall have to save the Palestinian people in Gaza." The objective of saving the Palestinian people in Gaza is, of course, a laudable goal - but it is centered entirely on a preferred outcome. The undeniably horrendous conditions experienced by the Palestinian population in Gaza can exist independently from the question of the presence or absence of the specific intent by Israel to commit genocide. Stopping the conflict (by specifically requesting the ICJ to indicate a cease fire in the PM application) for the benefit of the Palestinian population has, I think, been a subsidiary goal all along for SA (along with the primary goal of generating global political pressure), but the only feasible way to get the Court to take up the application is to characterize Israel's conduct as a genocide.
Regarding the projected outcome of the request for additional measures, I think it's incredibly unlikely that the Court will supplement what has already been ordered. A consistent theme across the individual separate opinions was an emphasis on distinguishing the standard at the PM stage from that at the merits stage. In the main analysis above, I draw only on J. Nolte's opinion for that point simply because his reasoning was particularly concise and because drawing the same reasoning from the rest of the individual opinions would be redundant. The message the individual opinions are collectively trying to convey, along with the main order that merely reemphasizes the existing status quo, is that the judges are not convinced that SA will be able to present sufficient evidence to support a finding that Israel is in fact committing genocide. This is a message that many people prefer not to interpret or accept for partisan or ideological reasons, but my assessment is that the ICJ indicated PMs only because of the intense reputational pressure on the Court to do so. If you're a judge (or a panel of judges, collectively) who feels the need to say something but you don't think you should actually say anything, the best way to accomplish that objective is to indicate measures that are already binding. If that assessment is accurate, I would not expect any additional measures to be indicated.
One last point to make just briefly given the length of this comment already. The question you pose seems premised on the assertion that the conduct of hostilities by Israel to date constitutes an open violation of the most basic rules of international humanitarian law. There is a broader discussion to be had regarding the extent to which the most basic rules of LOAC have been widely distorted to support partisan narratives regarding the conduct of hostilities thus far. Although characterizing actions in warfare as unlawful based widely on outcomes rather than on process, even in the absence of a doctrinal assessment of those actions, is expedient for the effort to delegitimize conduct and generate political pressure on non-belligerent States to in turn pressure Israel to end hostilities in Gaza, I assess that this strategy will be incredibly damaging to the legitimacy of international law in the long term. This is a deleterious outcome I would hope partisans who insist on engaging in this strategy take more seriously - but I have encountered little evidence in that regard to bring cause for optimism. That is a topic for a much longer discussion in some future date, but for now I contest the characterization that sufficient evidence exists to support the conclusion that the conduct of hostilities by Israel constitutes an open violation of the most basic rules of international humanitarian law.
Nicolas Boeglin says
March 8, 2024
Dear Professor Cox
Many thanks for your answer to my question.
We will see in a few days if, as you indicate "If that assessment is accurate, I would not expect any additional measures to be indicated" or if ... ICJ judges have another view. I personnally wouldn´t be as sure as you seem to be.
The discussion on LOAC is not really an open one, and I' m pretty sure you are also aware of a narrative from Israel trying to convince all of us that its military operations in Gaza are made in absolute conformity with applicable LOAC. We heard this very same official narrative from Israel in 2009 and 2014, but ... UN Commissions of Inquiry and Investigation created after 2009 operation in Gaza and 2014 summer operation in Gaza found opposite conclusions.
If as you write "contest the characterization that sufficient evidence exists to support the conclusion that the conduct of hostilities by Israel constitutes an open violation of the most basic rules of international humanitarian law", may I refer you to the conclusions of the two commission of inquiry and investigations mentionned before.
The mere fact that Israel has intentionally and deliberately eliminated 95 journalists in Gaza since Octobre 7th is to me a sign that the evidence that the press can gather about the way it conducts its military operations is considered as a threat by Israel military authorities.
(see last CPJ with the number of 95 journalists killed available at: https://cpj.org/2024/03/news-outlets-express-solidarity-with-journalists-in-gaza/ )
Yours sincerely
Nicolas Boeglin
Note: may I refer you to a note (In Spanish, sorry) on the third request of South Africa before ICJ, available at:
https://derechointernacionalcr.blogspot.com/2024/03/gaza-israel-proposito-de-la-tercera.html
Adil Haque says
March 8, 2024
Hi Brian,
I'm a bit puzzled by your post. The ICJ found "a real and imminent risk that irreparable prejudice will be caused to the rights found by the Court to be plausible, before it gives its final decision" including "the right of the Palestinians in Gaza to be protected from acts of genocide and related prohibited acts."
So the question would seem to be whether a "real and imminent risk ..." justifying provisional measures is equivalent to a "serious risk" activating the obligation to prevent genocide. You don't really engage with that question.
Your post seems focussed on the plausibility threshold, but that seems like an analytical mistake. Plausibility is formally about the rights claimed, and informally about the violations alleged. In contrast, risk is about what may happen next. The Court found that genocide may happen next.
On a separate note, I'm a bit puzzled by your claim that the Court simply reaffirmed Israel's pre-existing obligations. As you note, the duty to prevent genocide is activated “at the instant that the State learns of, or should normally have learned of, the existence of a serious risk that genocide will be committed.” If the first two provisional measures reflect Israel's obligation to prevent genocide, then they necessarily reflect a finding of a serious risk of genocide. Such a serious risk would activate Israel's obligation to prevent genocide, and the same serious risk would activate the same obligation of every other State. But that seems not to be your point.
Adil
Brian L. Cox says
March 9, 2024
Adil,
Excellent points – thank you for checking in to seek clarification, and as always it is wonderful to hear from you. The question you pose at the beginning of your comment is actually what prompted my study of the practical effect of the PMs. I started with the question: Is a “real and imminent risk that irreparable prejudice will be caused to the rights found by the Court to be plausible” the conceptual or functional equivalent of a “serious risk” that genocide is or may be committed, such that the obligation to prevent genocide is activated by States providing security assistance to Israel?
The answer I derived is, “no”, for two primary reasons. The first is the conceptual divergence between “at least plausible” and “only reasonable inference.” The second is the functional difference in methodology at the PM stage (prejudice to *rights* of Palestinians) and the merits stage (specific *intent* of Israel).
So, I do agree with you that plausibility is formally about the rights claimed. And risk is undoubtedly about what happens next. But the point at which your view and mine diverge, I think, is regarding what the Court has found so far at the PM stage. Your take is that the Court found genocide may happen next. My analysis above of the PM order concludes differently.
If we were to connect the “serious risk” standard, which is the standard that applies to the obligation to prevent genocide, to what the Court has found to date, this is what we would find: there is a “serious risk” that “irreparable prejudice will be caused to the *rights* found by the Court to be plausible.”
The obligation to prevent genocide is activated “at the instant that the State learns of, or should normally have learned of, the existence of a serious risk that genocide *will be committed*.” Assessing whether genocide is or might be committed involves the specific intent of the prospective perpetrator. This is fundamentally different than assessing whether it is plausible that rights should be protected.
So, all States are now on notice that it is plausible that the rights of Palestinians should be protected, but this is not the same as determining a serious risk exists that Israel is, or will, commit genocide.
Your assessment, if I understand correctly, is that the Court has now found that genocide may happen next. My analysis concludes differently. Instead, my assessment is that the Court has found *only* that it is plausible that rights should be protected – not that genocide is or will occur.
In the post, I cited J. Nolte’s separate opinion to support that understanding. In relevant part, he says, he is “not persuaded that South Africa has plausibly shown that the military operation undertaken by Israel, as such, is being pursued with genocidal intent” (para. 13). Nonetheless, in light of select statements of Israeli government officials and the conditions on the ground in Gaza, Judge Nolte expresses the view that South Africa has “shown that some, but not all, of the rights which it has alleged are plausible at the present preliminary stage of the proceedings” (para. 17, referring to para. 54 of the Order).
I felt it would have been redundant to cite similar expressions from individual opinions other than J. Nolte’s – but since my analysis of the divergence between rights (at the PM stage) and intent (at the merits stage) still seems to be unclear, here are some other expressions of the same (text between asterisks indicates my emphasis):
___________________________
- J. Bhandari (declaration).
9. “Again, the Court is not at this point deciding whether, in fact, such [genocidal] intent existed or exists. *All it is deciding* is whether *rights* under the Genocide Convention are *plausible*. Here, the widespread nature of the military campaign in Gaza, as well as the loss of life, injury, destruction and humanitarian needs following from it — much of which is a matter of public record and has been ongoing since October 2023 — are by themselves capable of supporting a *plausibility* finding with respect to *rights* under Article II.
10. Taken together and, bearing in mind the *lower standards* that apply in respect of *provisional measures* as opposed to the *merits*, the evidence on the record at this stage in the proceedings is such that, in the circumstances of this case, the Court was justified in granting provisional measures in the terms it did.”
- J. Barak (separate opinion).
43. “Regarding the conditions for the Court to indicate provisional measures, for the reasons stated above, I am not persuaded by South Africa’s arguments on the plausibility of rights, since there is no indication of an *intent* to commit genocide.”
- J. Sebutinde (dissenting opinion).
17. “I am not convinced that all the above criteria for the indication of provisional measures have been met in the present case. In particular, South Africa has not demonstrated, even on a prima facie basis, that the acts allegedly committed by Israel, and of which the Applicant complains, were committed with the necessary genocidal intent and that, as a result, they are capable of falling within the scope of the Genocide Convention. Similarly, when it comes to the *rights* that the Applicant asserts and for which South Africa seeks protection through the indication of provisional measures, there is no indication that the acts allegedly committed by Israel were accompanied by a *genocidal intent* and that, as a result, the rights asserted by the Applicant are plausible under the Genocide Convention.”
(J. Xue’s declaration not included here since she does not address the distinction between rights and intent and, instead, focuses on the justification for recognizing erga omnes partes in this case.)
_________________________
Taken together, we have two individual judges (J. Nolte & J. Bhandari) who emphasize the difference between rights and intent when explaining why they voted in favor of PMs, and two judges (J. Barak & Sebutinde) who do not support a positive finding regarding plausibility of rights *because* the applicant, in their view, has failed to prove genocidal intent.
These individual opinions seem to support my conclusion, from analyzing the PM order, that a positive finding regarding plausibility of *rights* is not the same as a serious risk that genocide is or will be committed since the latter query is centered on the intent of Israel rather than the rights of Palestinians. This, I believe, is where our views on the matter diverge.
This divergence, incidentally, also informs my response to your second point. Your understanding is that the first two provisional measures necessarily reflect a finding of a serious risk of genocide. I think that conclusion is informed by the previous point regarding serious risk, though I reach a different conclusion regarding serious risk – as the analysis above in this comment suggests.
PMs 1, 2, and 3 are drawn directly from the Convention, and my assessment is that these, therefore, merely reinstate and reinforce obligations that already exist in the Convention – but do no more than that. Then, as I suggest in the post, PM 5 is based on ICJ jurisprudence interpreting the Convention, while PM 4 is drawn from UNSCR 2320. Aside from PM 6 (30 day report), then, the PM order creates no *new* obligations. If that is accurate, saying “comply with the PM order” is precisely equivalent to saying “comply with obligations that already existed before the order was published.”
Apologies for the lengthy explanation – but I hope this clarifies any uncertainty regarding the analysis and conclusions from the post. Thanks again for engaging, and very best wishes to you, Adil!
Brian
Adil Haque says
March 11, 2024
Hi Brian,
Thanks for your patient (and lengthy) reply. Here is how I see it. The right to be protected from acts of genocide is a right to be protected from acts committed with genocidal intent. Irreparable prejudice to that right occurs when acts are committed with genocidal intent. A real and imminent risk of irreparable prejudice to that right is a real and imminent risk that acts will be committed with genocidal intent. The right is defined in terms of intent, so the two cannot be separated.
The separate and dissenting opinions differ on the plausibility of South Africa's allegations that Israel has already committed acts with genocidal intent. Only Nolte addresses future risk, finding that statements by Israeli officials "may contribute to a 'serious risk' that acts of genocide other than direct and public incitement may be committed, giving rise to Israel’s obligation to prevent genocide." Nolte makes some analytical mistakes along the way, which I will pass over. The point for now is that Nolte is concerned with the risk that acts will be committed with genocidal intent.
Just food for thought. No need to keep this exchange going. We all have other things to do.
Take care,
Adil