A draft report from the recent (January 2008) Global Administrative Law conference in New Delhi is now available on the GAL pages of the IILJ website. The Conference focused on the broad theme of "India, the South and the Shaping of Global Administrative Law"; we hope to begin making the papers presented there available in the near future.
Thursday, February 28, 2008
Wednesday, February 27, 2008
New Global Administrative Law journal symposium published
A number of the papers from the Global Administrative Law Workshop held in Buenos Aires in March 2007, co-sponsored by NYU and the Universidad de San Andrés, have been published in the October -December 2007 issue of Res Publica Argentina, in a mini-symposium on "The Argentine Financial Crisis and Bilateral Investment Treaties".
The papers, written in English or Spanish, include a Spanish translation of the Kingsbury, Krisch and Stewart framing paper, entitled "The Emergence of Global Administrative Law", that launched the project; and various other papers looking at the role of Argentine courts and bilateral investment treaties in the broader context of the financial crisis at the turn of the century.
Full text versions of all of the articles have been made available on the Global Administrative Law pages of the IILJ website.
The papers, written in English or Spanish, include a Spanish translation of the Kingsbury, Krisch and Stewart framing paper, entitled "The Emergence of Global Administrative Law", that launched the project; and various other papers looking at the role of Argentine courts and bilateral investment treaties in the broader context of the financial crisis at the turn of the century.
Full text versions of all of the articles have been made available on the Global Administrative Law pages of the IILJ website.
Monday, February 25, 2008
World Food Programme to Ration Aid?
The BBC is reporting that the UN World Food Programme is considering plans to "ration" its food aid to developing due to a combination of rising prices and inadequate resources. Leaving to one side the odd use of the decidedly emotive idea of "rationing" in this context - one can only assume, given the widespread and enduring persistence of hunger and starvation in the world, the WFP has been "rationing" its aid since its establishment - we are clearly confronted with an issue of interest to any emerging global administrative law: an international administrative agency, clearly exercising a public function, in a manner that will potentially have profound and far reaching effects on some of the world's most poorest and most vulnerable people. The first and most obvious question that arises is: what mechanisms exist to ensure that these people, these interests, are adequately represented in the decision-making process?
This is an area that warrants detailed study, and for my part I can claim no sort of expertise (comments from any readers more informed on this topic would be hugely welcome!). A quick perusal of the Constitution of the WFP (including financial regulations and the Rules of Procedure for the Executive Board), however, suggests that there is really very little in the way of applicable administrative law mechanisms to the decision-making process; and no procedure to contest any decision to restrict or remove aid. Certainly, there are certain "monitoring" provisions, but these pertain mainly to the accountability of the Executive Director to the 36 Member State Executive Board; an annual reporting requirement on the Board to the UN Economic and Social Council and the Council of the FAO; and provision for internal control and external audits in terms of financial matters. The only other administrative law-type provision that I came across was contained in Rule XV of the Rules of Procedure, which deal with the participation of (non-voting) observers in the deliberations of the Executive Board: Member States of the UN/FAO with a "particular interest" in deliberations have the right to participate; NGOs "interested and cooperating with the programme" can also do so, but only on the invitation of Executive Director.
Not, I think, the most robust of administrative law regimes; and this despite the profound human rights implications of the regulatory activity in question, and the high level of institutionalisation of the administrative body carrying it out...
This is an area that warrants detailed study, and for my part I can claim no sort of expertise (comments from any readers more informed on this topic would be hugely welcome!). A quick perusal of the Constitution of the WFP (including financial regulations and the Rules of Procedure for the Executive Board), however, suggests that there is really very little in the way of applicable administrative law mechanisms to the decision-making process; and no procedure to contest any decision to restrict or remove aid. Certainly, there are certain "monitoring" provisions, but these pertain mainly to the accountability of the Executive Director to the 36 Member State Executive Board; an annual reporting requirement on the Board to the UN Economic and Social Council and the Council of the FAO; and provision for internal control and external audits in terms of financial matters. The only other administrative law-type provision that I came across was contained in Rule XV of the Rules of Procedure, which deal with the participation of (non-voting) observers in the deliberations of the Executive Board: Member States of the UN/FAO with a "particular interest" in deliberations have the right to participate; NGOs "interested and cooperating with the programme" can also do so, but only on the invitation of Executive Director.
Not, I think, the most robust of administrative law regimes; and this despite the profound human rights implications of the regulatory activity in question, and the high level of institutionalisation of the administrative body carrying it out...
Labels:
Human Rights,
International Organizations,
UN
Thursday, February 14, 2008
Erasmus Law Review: Call for Papers
The Erasmus Law Review has just issued a call for papers that is of real relevance for anybody seeking to write and publish on global administrative law. It raises a number of important issues, including many as yet left largely unaddressed within GAL literature generally:
Since its rise at the beginning of the 20th century, comparative legal research has gained an influential place in legal research concerning national legal systems. Comparative legal methodology is used to acquire insight into foreign legal systems, to find solutions for problems of a specific legal system, or to promote the unification of law between national legal systems. Its methods consist in a comparison of different legal systems or legal traditions (external comparison) or of fields of law within national legal systems (internal comparison). With the proliferation of regulatory regimes at the international level (e.g. in the context of the United Nations or WTO), comparative lawyers have extended their focus to the field of international law. Consensus, however, has not been reached on the most suitable way of applying comparative law methods to the global context. Can the concepts and methods developed to conduct comparative legal research of national legal systems be transposed to study the international legal system?
As the call for papers goes on to make clear, the journal is in the main interested in those writing from a comparative perspective on this issue. At least two as yet under-studied GAL issues arise clearly: firstly, the differences between different national perspectives on the meaning, scope and content of domestic administrative law, and how these might translate into the global setting (and it is worth bearing in mind here that a criticism often levelled at the GAL project, by European scholars in particular, is that it remains locked within a US "interest representation" model of administrative law); and secondly, the equally vexed question of the complex relationship between the global administrative law project on one hand, and the field of public international law on the other. The development of a body of work addressing either or both of these comparative perspectives could prove to be a very significant advance in terms of the project more generally.
Since its rise at the beginning of the 20th century, comparative legal research has gained an influential place in legal research concerning national legal systems. Comparative legal methodology is used to acquire insight into foreign legal systems, to find solutions for problems of a specific legal system, or to promote the unification of law between national legal systems. Its methods consist in a comparison of different legal systems or legal traditions (external comparison) or of fields of law within national legal systems (internal comparison). With the proliferation of regulatory regimes at the international level (e.g. in the context of the United Nations or WTO), comparative lawyers have extended their focus to the field of international law. Consensus, however, has not been reached on the most suitable way of applying comparative law methods to the global context. Can the concepts and methods developed to conduct comparative legal research of national legal systems be transposed to study the international legal system?
As the call for papers goes on to make clear, the journal is in the main interested in those writing from a comparative perspective on this issue. At least two as yet under-studied GAL issues arise clearly: firstly, the differences between different national perspectives on the meaning, scope and content of domestic administrative law, and how these might translate into the global setting (and it is worth bearing in mind here that a criticism often levelled at the GAL project, by European scholars in particular, is that it remains locked within a US "interest representation" model of administrative law); and secondly, the equally vexed question of the complex relationship between the global administrative law project on one hand, and the field of public international law on the other. The development of a body of work addressing either or both of these comparative perspectives could prove to be a very significant advance in terms of the project more generally.
Friday, February 8, 2008
Global Administrative Law and the ECJ: The Advocate General and Kadi
**UPDATE** The judgment itself has now been handed down. For an account, see here.**
As promised earlier in the week, I wanted to blog in a little more detail from a GAL perspective on the recent (Jan. 16 2008) Opinion of the Advocate General of the European Court of Justice, Poiares Maduro, in the case of Kadi v. Council and Commission. Apologies in advance for the length of this post, but it is a subject of potentially huge importance. The case concerns the Council Regulations passed in order to give effect to the finding of the UN Security Council's 1267 Sanctions Committee, that Mr. Kadi - a resident of Saudi Arabia - is suspected of supporting terrorism, and must therefore have all assets frozen indefinitely.
The appellant has challenged the implementing Regulation on the grounds that his fundamental rights have been violated; of particular interest in terms of GAL is the fact that at no point in the listing procedure did he have the right to be heard (or the evidence against him presented), nor was there any mechanism for reviewing whether the process had indeed violated his fundamental rights. The case is on appeal from the Court of First Instance (CFI), which found that it could only review UN Security Council resolutions for their conformity with jus cogens obligations. Useful background to that judgment, and others of the CFI in related cases, can be found in Chia Lehnhardt's ASIL Insight on the subject.
We thus have, very clearly, an administrative relation established between the UN Security Council and the appellant in this case; and, moreover, one that is recognisably global in nature, given the regulatory body involved. We have what amounts to administrative action by the Security Council (through the 1267 Commtitee), which clearly - and profoundly - affects a set of interests (here, those of a private individual and his right to free enjoyment of his property). This is the basic precondition for anything that can be called global administrative law to exist. The key questions from a legal point of view are thus 1) what are the ex ante rules of administrative law that (at least should) shape, guide and restrict the administrative process in question? and 2) What are the possibilities for ex post review of the process, to ensure that it complied both with the ex ante administrative law requirements, and with other legal obligations more generally?
These latter two GAL questions are, in effect, the central issues at stake in the Advocate General's opinion, and, indeed, in the case more generally. After considering and rejecting the appellant's claim that the Council had no legal basis in Community law for implementing such measures against a private citizen from a third country, the Opinion moves on to consider (in reverse order from the manner in which I presented them above) 1) the possibility for the ECJ to effectively establish itself as an ex post review body over Security Council actions (the question of jurisdiction); and 2) the issue of to which ex ante standards the listing mechanism of the 1267 Committee should be held. I will look briefly at each in turn.
The ex post question
Can the ECJ act in effect to review the legality of measures adopted by the Security Council (even if, technically, this review is only indirect, as it speaks to the legality of the Community implementing legislation in fulfillment of a Security Council-imposed obligation). This was the question that the CFI answered in the positive only insofar as it related to issues of jus cogens. This is clearly correct, as far as it goes: the superiority of UN law over other treaty obligations is itself a treaty obligation (Art. 103 of the UN Charter), and as such cannot prevail over peremptory norms of international law. The case in question, however, deals with human rights obligations that do not have this status, even if they are "general principles of Community [and, indeed, international] law".
The Advocate General's opinion frames the issue in the following - fairly dramatic - manner:
...the present appeal turns fundamentally on the following question: is there any basis in the Treaty for holding that the contested regulation is exempt from the constitutional constraints normally imposed by Community law, since it implements a sanctions regime imposed by Security Council resolutions? Or, to put it differently: does the Community legal order accord supra-constitutional status to measures that are necessary for the implementation of resolutions adopted by the Security Council? (para. 25).
This presents starkly what is at stake - constitutional principles of the European legal order on one hand, and constitutive principles of the international one on the other. The key provision relied upon by the respondents to suggest that UN law should prevail is Art. 307 EC, which states that the treaty shall not affect obligations arising from agreements concluded before 1st January 1958. As Art. 103 of the Charter was concluded before this time, the EC Treaty cannot, it is claimed, override the supremacy of UN law; and thus the ECJ cannot review measures necessary to implement UN obligations for compatibility with Community law (paras. 29-30).
The Advocate General rejects this view, in effect on the basis that the EC Treaty has constituted an autonomous "municipal" legal order that stands in a fundamentally dualistic relation to international law. The reading of Art, 307 EC by the respondents would "break away from the very principles on which the Union is founded, while there is nothing in the Treaty to suggest that [it] has a special status - let alone a special status of that magnitude" (para. 31). Instead, he argues that, as in all dualistic systems,
[w]hile it is true that the restrictions which the general principles of Community law impose on the actions of the institutions may inconvenience the Community and its Member States in their dealings on the international stage, the application of these principles by the Court of Justice is without prejudice to the application of international rules on State responsibility or to the rule enunciated in Article 103 of the UN Charter (para. 39).
There also seems to be a remarkable assertion that Art. 307 flows "in both directions", creating duties on Member States to behave in conformity with the general principles of EC law in all their actions with international organizations, to the extent that
[a]s Members of the United Nations, the Member States, and particularly – in the context of the present case – those belonging to the Security Council, have to act in such a way as to prevent, as far as possible, the adoption of decisions by organs of the United Nations that are liable to enter into conflict with the core principles of the Community legal order. The Member States themselves, therefore, carry a responsibility to minimise the risk of conflicts between the Community legal order and international law (para. 32).
(One can't help but wonder just how far the ECJ would be prepared to push the potentially radical implications of recognising this "obligation"...)
The other important authority cited by the Advocate General is that the ECJ - in a case apparently largely overlooked by the CFI - has already reviewed a measure adopted to implement a Security Council Resolution under Chapter VII. In the Bosphorus case, the Court found found that the interest in preserving peace and security in the former Yugoslavia outweighed the rights of an innocent third party to pursue economic activities using assets leased from a Yugoslav company. It is true that, as the respondents suggested, the Court did not deal directly with the issue of the scope of its own jurisdiction; however, to claim that Bosphorus is therefore no authority in the present case does seem weak. Jurisdictional issues are, in the normal course of things, dealt with first by courts for obvious reasons; any ruling that it lacks jurisdiction in the present case by the ECJ will amount to a finding that it acted ultra vires in the earlier judgment.
The ex ante question
The Opinion then goes on to discuss "the standards of review" - the fundamental rights that the appellant alleges were breached in the administrative relation between himself and the UN Security Council. These were 1) his right to property (the "affected interest" from a GAL perspective); 2) his right to be heard in any administrative proceedings against him (a procedural/participatory GAL requirement); and 3) his right to judicial review of any such administrative proceedings (here, the requirement of an ex post review body functions as an ex ante rule of administrative law).
The respondents suggest that the subject matter of the administrative proceedings - the maintenance of international peace and security - should lead automatically to the application of - if any - only very minimal standards of review (akin to the jus cogens approach adopted by the CFI). The Advocate General does not agree, basing this view not on legal arguments per se, but rather on ideas of the role of law in liberal political theory (fighting, as it were, fire with fire). He agrees that there may be circumstances in which the fight against terrorism justifies limitations on rights, but maintains that individual legal proceedings to establish that such limitations go only as far as is strictly necessary are fundamental to a democratic polity such as the EU.
The Advocate General's view that the EC Regulation implementing the Security Council resolution not only can but also should be set aside is laid out in paragraph 51, worth quoting at some length here for its direct relevance to issues of global administrative law:
...although certain restrictions on that right may be envisaged for public security reasons, in the present case the Community institutions have not afforded any opportunity to the appellant to make known his views on whether the sanctions against him are justified and whether they should be kept in force. The existence of a de-listing procedure at the level of the United Nations offers no consolation in that regard. That procedure allows petitioners to submit a request to the Sanctions Committee or to their government for removal from the list. Yet, the processing of that request is purely a matter of intergovernmental consultation. There is no obligation on the Sanctions Committee actually to take the views of the petitioner into account. Moreover, the de-listing procedure does not provide even minimal access to the information on which the decision was based to include the petitioner in the list. In fact, access to such information is denied regardless of any substantiated claim as to the need to protect its confidentiality. One of the crucial reasons for which the right to be heard must be respected is to enable the parties concerned to defend their rights effectively, particularly in legal proceedings which might be brought after the administrative control procedure has come to a close. In that sense, respect for the right to be heard is directly relevant to ensuring the right to effective judicial review. Procedural safeguards at the administrative level can never remove the need for subsequent judicial review.
Conclusion
There can be little doubt that, should the Court follow the Advocate General on this issue, it would represent a development of real - and possibly snowballing - significance in terms of the emergence and progressive development of global administrative law. The Opinion seems well argued - although some international legal scholars may well bemoan the relatively glib way in which it is prepared to accept stark conflict between the international and European legal orders in terms of human rights protection. See, in this regard, the short but provocative comment by Andreas Paulus to the Opino Juris post that I referred to in my blog on Al-Jedda, in which he argues that preferable to the fragmented "mess" that such a position favours would be a finding that "arguing that this is not only what the ECJ demands, but what the Charter itself requires". We might doubt, however, that the best way to deal with conflicting legal orders is for a key institutional player in one to pretend a capacity to provide authoritative interpretations of the foundational document of the other - particularly where the main institutional player in the latter so clearly disagrees.
While perhaps undesirable in terms of the development of a coherent body of international human rights law and jurisprudence, a stark finding of conflict need not - indeed, should not - be viewed as an undesirable result when viewed from the perspective of global administrative law. As has been frequently noted, such "bottom up" application of "municipal" administrative law rules to the proceedings of global administrative bodies can play a key role in encouraging the latter to "bootstrap" themselves, and develop their own rules on the subject in a "top-down" manner. In this regard, it is worth noting that the Advocate General is, in many respects, calling for the ECJ to position itself in precisely the same relation to the Security Council as the German Constitutional Court did to the Community in the Solange judgment:
Had there been a genuine and effective mechanism of judicial control by an independent tribunal at the level of the United Nations, then this might have released the Community from the obligation to provide for judicial control of implementing measures that apply within the Community legal order. However, no such mechanism currently exists.
Such a course of action would, of course, appear to be in keeping with some observable trends in the field of GAL, such as, for example, the increasing willingness of courts to set aside the immunities of international organizations before domestic courts in employment disputes, where no adequate procedure is provided for by the organisation in question itself - even if it would be a fairly dramatic example of this. Indeed, the final ECJ judgment here will provide us with something of a test case for evaluating some of the key factors in determining the emergence - or otherwise - of global administrative law. It seems reasonable to suggest that, where regulatory objective of a global administrative body is security-related, we would expect strong resistance to the development of GAL in that context. On the other hand, we also expect that, where a powerful municipal court applies its own administrative law to such a body, and strikes down outcomes of procedures that do not boast adequate guarantees in this regard, the body in question will begin to develop its own procedures to avoid this result in the future.
The relative strengths of these different variables will, however, play out at different points in the process: to the extent that the security argument - advanced strongly by the respondents in the Kadi case - is accepted, then the ECJ will shy away from exercising robust review over Chapter VII Security Council actions at all. If it does exercise such review, as the Advocate General has argued it should, then it seems very likely indeed that the Security Council will fairly rapidly come up with an "adequate" procedure of its own. Interesting times...
As promised earlier in the week, I wanted to blog in a little more detail from a GAL perspective on the recent (Jan. 16 2008) Opinion of the Advocate General of the European Court of Justice, Poiares Maduro, in the case of Kadi v. Council and Commission. Apologies in advance for the length of this post, but it is a subject of potentially huge importance. The case concerns the Council Regulations passed in order to give effect to the finding of the UN Security Council's 1267 Sanctions Committee, that Mr. Kadi - a resident of Saudi Arabia - is suspected of supporting terrorism, and must therefore have all assets frozen indefinitely.
The appellant has challenged the implementing Regulation on the grounds that his fundamental rights have been violated; of particular interest in terms of GAL is the fact that at no point in the listing procedure did he have the right to be heard (or the evidence against him presented), nor was there any mechanism for reviewing whether the process had indeed violated his fundamental rights. The case is on appeal from the Court of First Instance (CFI), which found that it could only review UN Security Council resolutions for their conformity with jus cogens obligations. Useful background to that judgment, and others of the CFI in related cases, can be found in Chia Lehnhardt's ASIL Insight on the subject.
We thus have, very clearly, an administrative relation established between the UN Security Council and the appellant in this case; and, moreover, one that is recognisably global in nature, given the regulatory body involved. We have what amounts to administrative action by the Security Council (through the 1267 Commtitee), which clearly - and profoundly - affects a set of interests (here, those of a private individual and his right to free enjoyment of his property). This is the basic precondition for anything that can be called global administrative law to exist. The key questions from a legal point of view are thus 1) what are the ex ante rules of administrative law that (at least should) shape, guide and restrict the administrative process in question? and 2) What are the possibilities for ex post review of the process, to ensure that it complied both with the ex ante administrative law requirements, and with other legal obligations more generally?
These latter two GAL questions are, in effect, the central issues at stake in the Advocate General's opinion, and, indeed, in the case more generally. After considering and rejecting the appellant's claim that the Council had no legal basis in Community law for implementing such measures against a private citizen from a third country, the Opinion moves on to consider (in reverse order from the manner in which I presented them above) 1) the possibility for the ECJ to effectively establish itself as an ex post review body over Security Council actions (the question of jurisdiction); and 2) the issue of to which ex ante standards the listing mechanism of the 1267 Committee should be held. I will look briefly at each in turn.
The ex post question
Can the ECJ act in effect to review the legality of measures adopted by the Security Council (even if, technically, this review is only indirect, as it speaks to the legality of the Community implementing legislation in fulfillment of a Security Council-imposed obligation). This was the question that the CFI answered in the positive only insofar as it related to issues of jus cogens. This is clearly correct, as far as it goes: the superiority of UN law over other treaty obligations is itself a treaty obligation (Art. 103 of the UN Charter), and as such cannot prevail over peremptory norms of international law. The case in question, however, deals with human rights obligations that do not have this status, even if they are "general principles of Community [and, indeed, international] law".
The Advocate General's opinion frames the issue in the following - fairly dramatic - manner:
...the present appeal turns fundamentally on the following question: is there any basis in the Treaty for holding that the contested regulation is exempt from the constitutional constraints normally imposed by Community law, since it implements a sanctions regime imposed by Security Council resolutions? Or, to put it differently: does the Community legal order accord supra-constitutional status to measures that are necessary for the implementation of resolutions adopted by the Security Council? (para. 25).
This presents starkly what is at stake - constitutional principles of the European legal order on one hand, and constitutive principles of the international one on the other. The key provision relied upon by the respondents to suggest that UN law should prevail is Art. 307 EC, which states that the treaty shall not affect obligations arising from agreements concluded before 1st January 1958. As Art. 103 of the Charter was concluded before this time, the EC Treaty cannot, it is claimed, override the supremacy of UN law; and thus the ECJ cannot review measures necessary to implement UN obligations for compatibility with Community law (paras. 29-30).
The Advocate General rejects this view, in effect on the basis that the EC Treaty has constituted an autonomous "municipal" legal order that stands in a fundamentally dualistic relation to international law. The reading of Art, 307 EC by the respondents would "break away from the very principles on which the Union is founded, while there is nothing in the Treaty to suggest that [it] has a special status - let alone a special status of that magnitude" (para. 31). Instead, he argues that, as in all dualistic systems,
[w]hile it is true that the restrictions which the general principles of Community law impose on the actions of the institutions may inconvenience the Community and its Member States in their dealings on the international stage, the application of these principles by the Court of Justice is without prejudice to the application of international rules on State responsibility or to the rule enunciated in Article 103 of the UN Charter (para. 39).
There also seems to be a remarkable assertion that Art. 307 flows "in both directions", creating duties on Member States to behave in conformity with the general principles of EC law in all their actions with international organizations, to the extent that
[a]s Members of the United Nations, the Member States, and particularly – in the context of the present case – those belonging to the Security Council, have to act in such a way as to prevent, as far as possible, the adoption of decisions by organs of the United Nations that are liable to enter into conflict with the core principles of the Community legal order. The Member States themselves, therefore, carry a responsibility to minimise the risk of conflicts between the Community legal order and international law (para. 32).
(One can't help but wonder just how far the ECJ would be prepared to push the potentially radical implications of recognising this "obligation"...)
The other important authority cited by the Advocate General is that the ECJ - in a case apparently largely overlooked by the CFI - has already reviewed a measure adopted to implement a Security Council Resolution under Chapter VII. In the Bosphorus case, the Court found found that the interest in preserving peace and security in the former Yugoslavia outweighed the rights of an innocent third party to pursue economic activities using assets leased from a Yugoslav company. It is true that, as the respondents suggested, the Court did not deal directly with the issue of the scope of its own jurisdiction; however, to claim that Bosphorus is therefore no authority in the present case does seem weak. Jurisdictional issues are, in the normal course of things, dealt with first by courts for obvious reasons; any ruling that it lacks jurisdiction in the present case by the ECJ will amount to a finding that it acted ultra vires in the earlier judgment.
The ex ante question
The Opinion then goes on to discuss "the standards of review" - the fundamental rights that the appellant alleges were breached in the administrative relation between himself and the UN Security Council. These were 1) his right to property (the "affected interest" from a GAL perspective); 2) his right to be heard in any administrative proceedings against him (a procedural/participatory GAL requirement); and 3) his right to judicial review of any such administrative proceedings (here, the requirement of an ex post review body functions as an ex ante rule of administrative law).
The respondents suggest that the subject matter of the administrative proceedings - the maintenance of international peace and security - should lead automatically to the application of - if any - only very minimal standards of review (akin to the jus cogens approach adopted by the CFI). The Advocate General does not agree, basing this view not on legal arguments per se, but rather on ideas of the role of law in liberal political theory (fighting, as it were, fire with fire). He agrees that there may be circumstances in which the fight against terrorism justifies limitations on rights, but maintains that individual legal proceedings to establish that such limitations go only as far as is strictly necessary are fundamental to a democratic polity such as the EU.
The Advocate General's view that the EC Regulation implementing the Security Council resolution not only can but also should be set aside is laid out in paragraph 51, worth quoting at some length here for its direct relevance to issues of global administrative law:
...although certain restrictions on that right may be envisaged for public security reasons, in the present case the Community institutions have not afforded any opportunity to the appellant to make known his views on whether the sanctions against him are justified and whether they should be kept in force. The existence of a de-listing procedure at the level of the United Nations offers no consolation in that regard. That procedure allows petitioners to submit a request to the Sanctions Committee or to their government for removal from the list. Yet, the processing of that request is purely a matter of intergovernmental consultation. There is no obligation on the Sanctions Committee actually to take the views of the petitioner into account. Moreover, the de-listing procedure does not provide even minimal access to the information on which the decision was based to include the petitioner in the list. In fact, access to such information is denied regardless of any substantiated claim as to the need to protect its confidentiality. One of the crucial reasons for which the right to be heard must be respected is to enable the parties concerned to defend their rights effectively, particularly in legal proceedings which might be brought after the administrative control procedure has come to a close. In that sense, respect for the right to be heard is directly relevant to ensuring the right to effective judicial review. Procedural safeguards at the administrative level can never remove the need for subsequent judicial review.
Conclusion
There can be little doubt that, should the Court follow the Advocate General on this issue, it would represent a development of real - and possibly snowballing - significance in terms of the emergence and progressive development of global administrative law. The Opinion seems well argued - although some international legal scholars may well bemoan the relatively glib way in which it is prepared to accept stark conflict between the international and European legal orders in terms of human rights protection. See, in this regard, the short but provocative comment by Andreas Paulus to the Opino Juris post that I referred to in my blog on Al-Jedda, in which he argues that preferable to the fragmented "mess" that such a position favours would be a finding that "arguing that this is not only what the ECJ demands, but what the Charter itself requires". We might doubt, however, that the best way to deal with conflicting legal orders is for a key institutional player in one to pretend a capacity to provide authoritative interpretations of the foundational document of the other - particularly where the main institutional player in the latter so clearly disagrees.
While perhaps undesirable in terms of the development of a coherent body of international human rights law and jurisprudence, a stark finding of conflict need not - indeed, should not - be viewed as an undesirable result when viewed from the perspective of global administrative law. As has been frequently noted, such "bottom up" application of "municipal" administrative law rules to the proceedings of global administrative bodies can play a key role in encouraging the latter to "bootstrap" themselves, and develop their own rules on the subject in a "top-down" manner. In this regard, it is worth noting that the Advocate General is, in many respects, calling for the ECJ to position itself in precisely the same relation to the Security Council as the German Constitutional Court did to the Community in the Solange judgment:
Had there been a genuine and effective mechanism of judicial control by an independent tribunal at the level of the United Nations, then this might have released the Community from the obligation to provide for judicial control of implementing measures that apply within the Community legal order. However, no such mechanism currently exists.
Such a course of action would, of course, appear to be in keeping with some observable trends in the field of GAL, such as, for example, the increasing willingness of courts to set aside the immunities of international organizations before domestic courts in employment disputes, where no adequate procedure is provided for by the organisation in question itself - even if it would be a fairly dramatic example of this. Indeed, the final ECJ judgment here will provide us with something of a test case for evaluating some of the key factors in determining the emergence - or otherwise - of global administrative law. It seems reasonable to suggest that, where regulatory objective of a global administrative body is security-related, we would expect strong resistance to the development of GAL in that context. On the other hand, we also expect that, where a powerful municipal court applies its own administrative law to such a body, and strikes down outcomes of procedures that do not boast adequate guarantees in this regard, the body in question will begin to develop its own procedures to avoid this result in the future.
The relative strengths of these different variables will, however, play out at different points in the process: to the extent that the security argument - advanced strongly by the respondents in the Kadi case - is accepted, then the ECJ will shy away from exercising robust review over Chapter VII Security Council actions at all. If it does exercise such review, as the Advocate General has argued it should, then it seems very likely indeed that the Security Council will fairly rapidly come up with an "adequate" procedure of its own. Interesting times...
Labels:
Europe,
Human Rights,
International Organizations,
Kadi,
UN
Monday, February 4, 2008
The Security Council and Human Rights
I wanted to flag a couple of interesting posts from over at Opinio Juris a short while ago, dealing with the recent (December 07) judgment from the UK House of Lords in R (Al-Jedda) v Secretary of State for Defence. This case, which concerned almost exclusively international law, dealt with the issue of whether the detention of the appellant - a dual British and Iraqi national - by British forces in Iraq, on the basis of "imperative reasons of security" (the Court notes that "[t]he appellant has not been charged with any offence, and no charge or trial is in prospect") constituted a violation of Article 5(1) of the European Convention on Human Rights (the right to liberty and security of person).
This case raises a number of interesting points, many of which are of real relevance to the field of global administrative law. Here, I want to note just two:
1) Firstly - and incredibly - the Court had to consider a claim by the Secretary of State that under international law the detention of the appellant was attributable not to the British Government, but rather to the United Nations. This was based on the claim that, given that the UN Security Council had accorded to the "multinational force" "the authority to take all necessary measures to contribute to the maintenance of security and stability in Iraq" in its Resolution 1546/2004 (amongst others), it thus had assumed "effective control" over the actions of the British and US forces in the country. The legal authority for this claim was drawn entirely from the contentious recent European Court of Human Rights decision in Behrami and Behrami v. France, which decided - on the basis of a disappointingly superficial discussion of the complex international law doctrine on attributability, and a questionable movement from a standard of "effective control" to one of "ultimate control" - that the UN itself was responsible for UNMIK and KFOR activities in Kosovo, not individual states carrying them out, and that the ECHR could not thus be applied.
Thankfully, the House of Lords rejected this argument, distinguishing the facts of the Al-Jedda case from those in Behrami - although, as the interesting discussion following the Opinio Juris posts makes clear, whether there were in fact important and legally relevant differences between the cases is at best an open question. In this regard, that the UK Government was able to claim at all that the UN was responsible for its actions in Iraq seems more a testament to the poorness of the ECtHR's judgment in Behrami than it is an accurate description of the nature of the legal or factual situation in Iraq. If the Office of Legal Affairs at the UN were said to have been angered by the decision in Behrami, one can only imagine their rage at being told they were now also responsible for UK/US actions in Iraq...
2) Neverthless, the House of Lords upheld the decisions of the previous courts, finding that, although the action was attributable to the UK itself under international law, the authorisation to act to preserve security in Iraq contained in several Security Council Resolutions meant that the ECHR was not applicable, as a result of the superiority of UN obligations over all conflicting treaties provided for in Art. 103 of the UN Charter. In the Court's own words, the key questions were whether "the UK became subject to an obligation (within the meaning of article 103) to detain the appellant and, if so, whether and to what extent such obligation displaced or qualified the appellant's rights under article 5(1)".
The Court held that "obligations" under Art. 103 included those generated by Security Council Resolutions, and that, in the present case at least, it also covered "authorizations" to act. In particular, Lord Bingham argued that
... there is a clash between on the one hand a power or duty to detain exercisable on the express authority of the Security Council and, on the other, a fundamental human right which the UK has undertaken to secure to those (like the appellant) within its jurisdiction. How are these to be reconciled? There is in my opinion only one way in which they can be reconciled: by ruling that the UK may lawfully, where it is necessary for imperative reasons of security, exercise the power to detain authorised by UNSCR 1546 and successive resolutions, but must ensure that the detainee's rights under article 5 are not infringed to any greater extent than is inherent in such detention.
(It is perhaps worth noting that this conclusion - the grounds on which the appeal was rejected - cannot but appear, from some perspectives at least, profoundly question-begging. Lord Bingham began his judgment with the observation that "these allegations [of terrorist activity in Iraq] are roundly denied by the appellant, and they have not been tested in any proceedings. Nor is their correctness an issue in these proceedings. The House must therefore resolve the legal issues falling for decision on the assumption that the allegations are true, without forming any judgment whether they are or not". The fact that the Court could not evaluate the substance of the allegations against the appellant meant that it was impossible for it to decide whether or not the detention had in fact been "necessary for imperative reasons of security" - the only condition that would authorise it under UN resolutions and thus evade ECHR applicability. Instead, it was compelled to assume that this was the case, which appears in effect to result in carte blanche for UK forces to detain more or less at will in Iraq - even where such detention is not in itself legal).
Article 103 thus means that obligations on states stemming from Security Council Resolutions are not subject to the provisions of the ECHR or, by extension, any other human rights treaties (excepting, perhaps, those provisions that reflect norms of jus cogens). What is the relevance of this for global administrative law? Simply that, where the Security Council, acting as an administrative body, passes a resolution that places obligations on states to act such actions are essentially unbound by human rights obligations.
Perhaps the clearest - and most currently contentious - example of this type of activity is the UN Security Council's 1267 Committee, responsible for compiling a list of all those suspected of supporting terrorism, against whom the Security Council then obliges states to take a range of measures, including freezing of all assets. If the House of Lords got it right in Al-Jedda, those hoping that measures implementing the Security Council's findings - a result of a process that affords no hearing, fair or otherwise, to those it affects - would be found to violate human rights law and thus invalidated will be disappointed.
All eyes now turn to the European Court of Justice, which is currently considering precisely these issues in the Kadi case, and which will go a very long way to settling the issue once and for all. The opinion of the Advocate General in this case was published in mid-January of this year; I'll post on that in more detail later in the week...
This case raises a number of interesting points, many of which are of real relevance to the field of global administrative law. Here, I want to note just two:
1) Firstly - and incredibly - the Court had to consider a claim by the Secretary of State that under international law the detention of the appellant was attributable not to the British Government, but rather to the United Nations. This was based on the claim that, given that the UN Security Council had accorded to the "multinational force" "the authority to take all necessary measures to contribute to the maintenance of security and stability in Iraq" in its Resolution 1546/2004 (amongst others), it thus had assumed "effective control" over the actions of the British and US forces in the country. The legal authority for this claim was drawn entirely from the contentious recent European Court of Human Rights decision in Behrami and Behrami v. France, which decided - on the basis of a disappointingly superficial discussion of the complex international law doctrine on attributability, and a questionable movement from a standard of "effective control" to one of "ultimate control" - that the UN itself was responsible for UNMIK and KFOR activities in Kosovo, not individual states carrying them out, and that the ECHR could not thus be applied.
Thankfully, the House of Lords rejected this argument, distinguishing the facts of the Al-Jedda case from those in Behrami - although, as the interesting discussion following the Opinio Juris posts makes clear, whether there were in fact important and legally relevant differences between the cases is at best an open question. In this regard, that the UK Government was able to claim at all that the UN was responsible for its actions in Iraq seems more a testament to the poorness of the ECtHR's judgment in Behrami than it is an accurate description of the nature of the legal or factual situation in Iraq. If the Office of Legal Affairs at the UN were said to have been angered by the decision in Behrami, one can only imagine their rage at being told they were now also responsible for UK/US actions in Iraq...
2) Neverthless, the House of Lords upheld the decisions of the previous courts, finding that, although the action was attributable to the UK itself under international law, the authorisation to act to preserve security in Iraq contained in several Security Council Resolutions meant that the ECHR was not applicable, as a result of the superiority of UN obligations over all conflicting treaties provided for in Art. 103 of the UN Charter. In the Court's own words, the key questions were whether "the UK became subject to an obligation (within the meaning of article 103) to detain the appellant and, if so, whether and to what extent such obligation displaced or qualified the appellant's rights under article 5(1)".
The Court held that "obligations" under Art. 103 included those generated by Security Council Resolutions, and that, in the present case at least, it also covered "authorizations" to act. In particular, Lord Bingham argued that
... there is a clash between on the one hand a power or duty to detain exercisable on the express authority of the Security Council and, on the other, a fundamental human right which the UK has undertaken to secure to those (like the appellant) within its jurisdiction. How are these to be reconciled? There is in my opinion only one way in which they can be reconciled: by ruling that the UK may lawfully, where it is necessary for imperative reasons of security, exercise the power to detain authorised by UNSCR 1546 and successive resolutions, but must ensure that the detainee's rights under article 5 are not infringed to any greater extent than is inherent in such detention.
(It is perhaps worth noting that this conclusion - the grounds on which the appeal was rejected - cannot but appear, from some perspectives at least, profoundly question-begging. Lord Bingham began his judgment with the observation that "these allegations [of terrorist activity in Iraq] are roundly denied by the appellant, and they have not been tested in any proceedings. Nor is their correctness an issue in these proceedings. The House must therefore resolve the legal issues falling for decision on the assumption that the allegations are true, without forming any judgment whether they are or not". The fact that the Court could not evaluate the substance of the allegations against the appellant meant that it was impossible for it to decide whether or not the detention had in fact been "necessary for imperative reasons of security" - the only condition that would authorise it under UN resolutions and thus evade ECHR applicability. Instead, it was compelled to assume that this was the case, which appears in effect to result in carte blanche for UK forces to detain more or less at will in Iraq - even where such detention is not in itself legal).
Article 103 thus means that obligations on states stemming from Security Council Resolutions are not subject to the provisions of the ECHR or, by extension, any other human rights treaties (excepting, perhaps, those provisions that reflect norms of jus cogens). What is the relevance of this for global administrative law? Simply that, where the Security Council, acting as an administrative body, passes a resolution that places obligations on states to act such actions are essentially unbound by human rights obligations.
Perhaps the clearest - and most currently contentious - example of this type of activity is the UN Security Council's 1267 Committee, responsible for compiling a list of all those suspected of supporting terrorism, against whom the Security Council then obliges states to take a range of measures, including freezing of all assets. If the House of Lords got it right in Al-Jedda, those hoping that measures implementing the Security Council's findings - a result of a process that affords no hearing, fair or otherwise, to those it affects - would be found to violate human rights law and thus invalidated will be disappointed.
All eyes now turn to the European Court of Justice, which is currently considering precisely these issues in the Kadi case, and which will go a very long way to settling the issue once and for all. The opinion of the Advocate General in this case was published in mid-January of this year; I'll post on that in more detail later in the week...
Labels:
Human Rights,
International Organizations,
Kadi,
UN
Subscribe to:
Posts (Atom)