My fundamental concern, however, is that the debate about how best to make sure that the Refugee Convention really dictates practice on the ground may have been effectively ended before it has even begun. There just has not been a fulsome discussion of both the reasons that an enhanced supervisory mechanism is needed to oversee the Refugee Convention, much less a detailed analysis of the lessons that can be learned from experience in supervising international human rights elsewhere, particularly in the UN system. My specific worry is that after having allowed this matter to languish for half a century, there is now a rush to capture at least some kind of a commitment - even if it is only a very minimalist commitment - to overseeing the Refugee Convention. I am concerned that if we allow ourselves to be rushed into embracing a particular model which has no more than very minimalist expectations, because to do so allows us to make at least some progress on this issue, we may find that we have locked ourselves into a model for oversight of the Convention that, in the long run, really is inadequate. While there is of course the possibility that a minimalist project may provide the level of experience and confidence to move in a more ambitious direction in the future, there is also the possibility that states will take the view that after having established a minimalist mechanism they have, so to speak, "dealt with the supervision question." Thus, they might argue that there is no need to revisit the issue, at least not any time soon.

I would hope, then, that we can re-focus our energies for the medium term. We really cannot afford to sell-out the future of refugee protection in a desperate grab to lock-in at least something that looks, more or less, like an oversight mechanism for the Refugee Convention.

So first, let's begin by being clear about exactly what we are discussing. Some commentators seem erroneously to assume that this debate is about how to stay on top of UNHCR as an agency. That is not, or at least should not be, right. UNHCR has a mandate which is, of course, much broader than supervising the Refugee Convention. In recent years, its work as a humanitarian relief agency has, in fact, come to overshadow its core protection functions. Its work on behalf of the internally displaced has in many instances eclipsed its primary duty to protect refugees. It has often taken on roles which put it into the realm of the political, notwithstanding its explicitly non-political mandate. These are all questions of extraordinary complexity and importance. But they are not the same as the question of how best to oversee the Refugee Convention. While there can and should be initiatives more effectively to supervise UNCHR as an agency, these are matters which, to my mind, are logically entrusted to UNHCR's executive committee, or indeed to the ECOSOC itself. We should absolutely not allow the question of how best to oversee the Refugee Convention to be redirected towards difficult but distinct questions of supervising UNHCR' s compliance with its broader statutory mandate, much less of how to monitor the various jobs it has taken on outside of its mandate.

On the other hand, it is equally wrong for UNHCR to attempt artificially to cut off debate on the appropriate range of potential mechanisms to oversee the Refugee Convention by reliance on its institutional authority under Article 35 of the Refugee Convention. As we all know, UNHCR has a special responsibility under the Refugee Convention to "supervise the implementation" of the Refugee Convention. This is a critical role, which should be defended at all costs. Indeed, as the proposal developed by several Geneva-based non-governmental groups suggests, it is important to find a way to reinforce the commitment to protection within UNHCR, and to bring external voices into the process by which its protection agenda is set.

But Article 35 emphatically does not create a monopoly in favor of UNHCR vis a vis oversight of the Refugee Convention. UNHCR's duty is to supervise the implementation of the Refugee Convention, not to oversee the Refugee Convention itself. The Convention, as an international treaty, is the responsibility of the states which signed it. As the mechanisms for enforcement of the Convention itself make clear, it is states which have the fundamental right and duty to ensure that other states actually live up to their obligations under the Refugee Convention. While UNHCR is charged with the day-to-day duty of making sure that interstate commitments result in effective intervention and advocacy on the ground, there is nothing in Article 35 which precludes the states which are both the objects and the trustees of the refugee protection system from deciding to establish an arms-length mechanism to provide general guidance on, and oversight of, the Refugee Convention. Indeed, as I will hope to make clear, a move in this direction is precisely what I believe to be required now.

Thus, it is vital that we not see important and laudable efforts to devise an advisory committee structure to advise the High Commissioner on the exercise of his protection mandate as either inconsistent with, or a substitute for, a genuine program of independent, impartial, transparent, and socially accountable supervision of the Refugee Convention. If we inadvertently send the signal that we see the advisory committee proposal as in any sense an answer to the much more critical question of how best to oversee the Convention itself, we will in my view have lost an extraordinary opportunity to make progress on what should be the real agenda of protection. While the advisory group proposal meets one of Professor's Kälin's four criteria - that is, it is certainly an operationally practicable proposal - it is doubtful that it comes close to any real claim to independence or to objectivity, much less to inclusiveness for all relevant parties.

As we open this discussion of options today, a first question must surely be just why it is that the Refugee Convention, virtually alone among major human rights treaties, still has no free-standing mechanism to promote interstate accountability?

In part, it is a question of history. The Refugee Convention was the second major human rights treaty adopted by the United Nations, having been preceded only by the Genocide Convention. It is noteworthy that the Genocide Convention, like the Refugee Convention is not externally supervised. In part, then, the absence of an external supervisory mechanism for the Refugee Convention is simply a reflection of the historical reality that, in the late 1940' and early 1950's, the entire idea of interstate supervision of human rights was new, potentially threatening, and not truly accepted by states. Yet with the adoption of the human rights covenants and more specialized treaties beginning in the mid 1960's, the establishment of an independent mechanism for interstate oversight of the human rights treaties has become routine. Unless there is some good, principled reason why refugee law should be immune from this general commitment, it is high time to reverse the historical aberration by bringing the commitment to supervision of refugee law into line with the practice in human rights law more generally.

It might be suggested, however, that it was - and is - the existence of a United Nations High Commissioner for Refugees which distinguishes refugee law from every other UN human rights project. Only in refugee law is there an international organization assigned exclusively to oversee implementation of the treaty. At best, other UN human rights treaties can rely on the recently established, generic authority of a (grossly underfunded) UN High Commissioner for Human Rights. Refugee law, however, has its own institutional guardian in the person of the High Commissioner. It might be thought that any additional mechanism for oversight is superfluous.

I believe that this would be a tragic error of judgement. UNHCR performs some absolutely extraordinary and essential function via its supervisory authority codified in Article 35 of the Refugee Convention. In particular, the Department of International Protection has real expertise in assisting governments to draft policy and legislation; in engaging directly and indirectly in defensive case interventions; and of organizing and conducting refugee law outreach and training. DIP's role is complemented by the critical function of UNHCR's Executive Committee, which symbolically reaffirms the commitment of states to refugee law, and provides a form democratic legitimacy to the work of UNHCR. There is therefore no need for a mechanism of international oversight to take on any of these roles.

But there are also some things which are usually understood to be central to a meaningful project of international supervision that UNHCR does less well, and is perhaps not ideally positioned to take on. In practice, neither DIP nor EXCOM has done enough to provide systematic, non-crisis policy guidance on the substance of refugee law, carefully anchored in the real context of protection challenges. There has been a lack of leadership in the design of mechanisms to implement burden and responsibility sharing, so as to enable the imperatives of refugee law duties to be reconciled to the political and social realities of asylum states. There has not really been a genuinely inclusive range of voices, including those of refugees themselves, brought into the policy formation process. And not enough efforts have been made to empower local institutions to make enforcement of rights meaningful in a way that no international institution can ever aspire to do. These are all examples of the sort of work which, in other contexts, is entrusted to an autonomous supervisory body.

Beyond the importance of setting reasonable expectations for the sort of supervisory tasks which UNHCR should itself be expected to take on, there are two more fundamental reasons why vesting UNHCR with sole responsibility to oversee the Refugee Convention is not a credible proposition.

First, particularly during the decade of the 1990', UNHCR has been fundamentally transformed from an agency whose job was, in large measure, to serve as trustee or guardian of refugee rights as implemented by states to an agency which is now primarily focused on direct service delivery. Simply put, UNHCR is no longer at arms-length from the implementation of refugee protection. In most big refugee crises around the world today, UNHCR is - in law or in fact - the means by which refugee protection is delivered on the ground. My point today is not to lament the decision to transform an agency with a once-unique supervisory mandate into yet another humanitarian relief agency. But we cannot deny that UNHCR today is not remotely the same as UNHCR as conceived in 1950. In particular, as the crises in, for example, Rwanda and Kosovo make clear, oversight of respect for the Refugee Convention today in many cases effectively means oversight of some of UNHCR's own activities.

To suggest that we can honor Professor Kälin's insistence on supervision which is genuinely independent and objective by the establishment of a new mechanism, even if quasi-autonomous, within UNHCR itself, is therefore simply disingenuous. UNHCR faces an ethical dilemma, in my view. Either it must return to concentrating on the implementation of its core supervisory responsibilities and leave what has become the majority of its operational mandate to others; or it must concede that it cannot ethically supervise itself, and endorse the establishment of a genuinely arms-length body to oversee the Refugee Convention.

Second, and perhaps even more fundamental, the difficulty with relying solely on UNHCR to oversee the Refugee Convention is that it is simply too easy for states to avoid the meaningful accountability between and among themselves that is at the root of the entire international human rights project. Because states presently take little if any personal responsibility for ensuring that their fellow states live up to international refugee law obligations, the dynamic of persuading, cajoling, and indeed shaming of partner states which has been so critical to the success of the international human rights project in general is absent, for the most part, in refugee law. It is simply too easy to leave the task to UNHCR.

Yet, as we all know, UNHCR is not really positioned to apply meaningful forms of pressure on states. UNHCR is, after all, an entity with a a tiny core budget, and which is effectively dependent on the annual voluntary contributions of a very small number of powerful states, virtually none of which has been predisposed to empower UNHCR to act independently and truly autonomously to advance a strong regime of international refugee protection. Yes, these states have been generous in providing funds for refugee relief and for humanitarian assistance; but all too often they have either avoided or, on occasion, evaded UNHCR's insistence on the importance of protection principles. Recent tragic events off the coast of Australia, and the legally indefensible domestic reaction to the attempt to bring international law to bear on Australia, are more than adequate testimony to this problem.

The effort by some within governments and UNHCR itself to stifle creative thinking about a genuinely responsive and accountable interstate mechanism for oversight of the Refugee Convention is thus deeply regrettable. Indeed, it has been made clear to me by some colleagues within UNHCR that whatever the institutional positions articulated by senior UNHCR officials, these protection officers would personally welcome an external mechanism which could provide the critical support to UNHCR's protection activities so urgently needed to get things right in the field. This is, in my view, the right way to understand the project of overseeing the Refugee Convention. Not as a challenge to UNHCR as an organization, but as a formal recognition of the fact that UNHCR is charged with a multiplicity of tasks, not all of which derive from, or are even consistent with, a fundamental commitment to the fulsome implementation of the Refugee Convention.

Moreover, because UNHCR as an agency is, and will remain, politically and fiscally constrained by design, it cannot reasonably be looked to provide the sort of strong voice in favor of unflinching attention to refugee protection that is now required. Indeed, it has been argued that there may simply be no good reason to compromise UNHCR's on-the-ground efforts to promote implementation of the Refugee Convention - which do frequently require compromise and even expediency in the interest of saving lives - by forcing that same organization to be the source of critique and broad guidance on acceptable international practice under the Refugee Convention. Much less may it be reasonable to expect UNHCR, as an interstate organization, to devise the sorts of political mechanisms - involving international burden and responsibility sharing - which are critical to the continued effectiveness of refugee law in the modern world.

In short, my essential point is that the non-governmental community would be ill-advised to limit its thinking to models which are housed within, or functionally intertwined with, the work of UNHCR as an international organization. And by the same token, UNHCR as an organization would be ill-advised to insist that any mechanism to reinforce oversight of the Refugee Convention be situated within its walls, since to do so may simply constrain its operational effectiveness in protection and other fields, and reinforce the current sense of despair among many UNHCR staff brought on by expectations not matched by either political independence or fiscal autonomy.

It is in this light of these realities that I think the urgent priority now is for all of us to insist on the importance of the draft resolution proposed for adoption at the Ministerial Meeting in which the importance of additional mechanism to promote respect for the Refugee Convention is emphasized. This is a vital commitment to secure on the 50th anniversary of the Refugee Convention, and one which we should without hesitation not only embrace, but actively promote. But let us not then rush from endorsement of that critical commitment to put forward any model which has not been predicated on thorough study of the experience of human rights protection. In particular, the successes and failures of the six major United Nations treaty bodies provide an unbelievable wealth of information, both for and against particular modes of oversight, which we ignore at our peril. Indeed, it would be ironic at a moment when the chairpersons of all of the U.N. human rights treaty bodies have insisted on regular coordination and mutual learning, were those of us in the refugee protection community to rush forward to embrace any model which is not predicated on an intimate knowledge of the range of potential protection options.

It was in the hope of advancing this critical inquiry that the collaboration between the International Council of Voluntary Agencies (ICVA) and the University of Michigan's Program in Refugee and Asylum Law was born. Realizing that non- governmental activists rarely have the opportunity to consider at length the broader lessons of human rights history, we devised a shared project which harnessed the intellect and energy of some of my best students to prepare a series of brief discussion papers that draw together, in at least skeletal form, some of that extremely rich history of oversight of human rights treaties which is, in my view, so important to getting it right in the context of the Refugee Convention.^**

By way of the briefest of overviews, the first four papers in the series consider the usefulness in the context of the Refugee Convention of the most widely used mechanisms of treaty supervision which, as forcefully argued by Harvard's Professor Henry Steiner, are mutually reinforcing and interdependent processes:

• Working Paper No. 1 takes up the question of state reporting requirements, regularly reviewed through a "dialogue of justification" between the supervisory body and states, supported by strong non-governmental input. It emphasizes the value of carefully targeted, thematic reporting, rather than routine, generic reports; and outlines a well-prepared and forward-looking process of review;
• Working Paper No. 2 looks at the possibility of a complaints mechanism under the Refugee Convention, and recommends in favor of a selective, group-based petition system as a means of injecting the voices of refugees directly into the supervisory process;
• Working Paper No. 3 takes up the often-overlooked value of "general comments" issued by other human rights treaty bodies to codify their work on particular legal issues, which have had extraordinary value in influencing the work of domestic courts and tribunals. It recommends a staged process to conceive and review general comments, including an open debate in which NGOs and IGOs would participate actively;
• Working Paper No. 4 proposes that the supervisory body have an auxiliary investigative capacity to supplement its reporting, complaints, and general comments functions. It stresses the importance of direct access to evidence on the ground as critical to the credibility and effectiveness of the supervisory body.

Our essential conclusion is that most discussion to-date has failed to recognize the value of these mechanisms to the design of a system to oversee the Refugee Convention. With creative adaptation, we recommend their use.

The final three papers speak to the process of overseeing the Refugee Convention: