1. Over the last 12 years or so, our understanding of the international legal obligations arising from the International Covenant on Economic, Social and Cultural Rights (ICESCR) has deepened. Several factors have contributed to this progress, not least the work of the United Nations Committee on Economic, Social and Cultural Rights and the increasing attention devoted to economic, social and cultural rights by non-governmental organizations (NGOs) and the academic literature. / Among the especially useful NGO/academic initiatives have been the Limburg Principles on the Implementation of the International Covenant on Economic, Social and Cultural Rights (1986), the Bangalore (ICJ) Declaration and Plan of Action (1995), and the Maastricht Guidelines on Violations of Economic, Social and Cultural Rights (1997)./

2. Despite these advances, however, there remains considerable uncertainty about the nature and extent of the legal obligations arising from ICESCR. These doubts persist for a number of reasons, one being the wording of article 2 (1) which sets out the general legal obligation of States parties under the Covenant:

"Each State Party to the present Covenant undertakes to take steps, individually and through international assistance and cooperation, especially economic and technical, to the maximum of its available resources, with a view to achieving progressively the full realization of the rights recognized in the present Covenant by all appropriate means, including particularly the adoption of legislative measures."

3. This statement of obligation includes some notoriously elusive phrases and concepts. Here I draw attention to two which have particular relevance to indicators and benchmarks: States parties undertake to progressively realize the enumerated rights to the maximum of their available resources. Scott Leckie warns against the progressive realization provision serving as "an escape hatch (for) recalcitrant states". / S. Leckie, "Another Step Towards Indivisibility: Identifying the Key Features of Violations of Economic, Social and Cultural Rights", Human Rights Quarterly, vol. 20, No. 1, February 1998, pp. 81, 94./ After describing the maximum available resources phrase as "difficult", Robert Robertson remarks: "'Maximum' is the sword of human rights rhetoric; 'available' is the wiggle room for the state" / R. Robertson, "Measuring State Compliance with the Obligation to Devote the 'Maximum Available Resources' to Realizing Economic, Social and Cultural Rights", Human Rights Quarterly, 1994, pp. 693, 694./ - while Audrey Chapman rightly observes that evaluating progressive realization within the context of resource availability "considerably complicates the methodological requirements". / A. Chapman, "A 'Violations Approach' for Monitoring the International Covenant on Economic, Social and Cultural Rights", Human Rights Quarterly, vol. 18, No. 1, February 1996, pp. 23, 31./

4. Both phrases - progressive realization and resource availability - have two crucial implications. Firstly, they imply that some (but not necessarily all) States parties' obligations under the Covenant may vary from one State to another. Second, they imply that, in relation to the same State party, some (but not necessarily all) obligations under the Covenant may vary over time. These variable elements of States parties' obligations under ICESCR contribute to the sense of uncertainty which remain a feature of international economic, social and cultural rights.

5. Critically, human rights indicators and benchmarks can help all interested parties identify and monitor these variable or shifting State obligations. Before sketching a process by which indicators and benchmarks might be used in the context of the right to education, I want to make some additional general observations about the nature and extent of States parties' legal obligations under the Covenant.

6. Consideration of the Covenant and the jurisprudence of the Committee, discloses three interrelated and overlapping dimensions to States parties' legal obligations. Here, I deal only very briefly with the first two. My focus is the third dimension which, as we will see, has particular relevance to indicators and benchmarks.

(a) Obligations applying uniformly to all States parties. It must be emphasized that some obligations under ICESCR are not subject to notions of progressive realization and resource availability. These obligations apply uniformly around the world to all States parties whatever their stage of economic development. To take just one example: the principle of non-discrimination. According to article 2 (2) of the Covenant, States parties must not discriminate on the basis of sex, race and other prohibited grounds. Thus, if a State party excludes girls from any State school, it breaches the Covenant. The Covenant and Committee are clear: in this context, notions of progressive realization and resource availability are irrelevant. / See General Comment 3./

(b) A minimum core content for each right. According to the Committee, it is incumbent upon every State party "to ensure the satisfaction of, at the very least, minimum essential levels" of each right set out in the Covenant. / Ibid., para. 10./ Without minimum core obligations, the Covenant "is largely deprived of its raison d'être. / Ibid./ Much work still has to be done to define the minimum core content of each right. The Committee, for example, has yet to identify the minimum core content of the right to education. / On the core content of the right to education, see F. Coomans, "Clarifying the Core Elements of the Right to Education" in F. Coomans and F. van Hoof (eds.), The Right to Complain about Economic, Social and Cultural Rights, 1995, p. 11./ Once defined, however, it will apply to all States parties whatever their stage of economic development. In other words, the minimum core content will not be subject to the notions of progressive realization and resource availability.

(c) The variable dimension. Given article 2 (1), it is difficult to resist the conclusion that States parties' obligations under the Covenant have a variable or shifting dimension. Because of the progressive realization and resource availability phrases, the precise content of at least some State obligations is likely to vary from one State to another - and over time in relation to the same State.

7. In these circumstances, it is important to establish effective techniques and processes which are able to identify and monitor this variable dimension of State obligations. Without such arrangements, article 2 (1) can provide "recalcitrant States" with the "escape hatch" which Leckie warns us about. / Leckie, supra note 2./ In my view, indicators and benchmarks have a role to play in both identifying and monitoring the variable component of States parties' obligations under the Covenant. / The usefulness of indicators and benchmarks is not necessarily confined to this third, variable dimension of States parties' obligations. They might also have a role, for example, in the definition of minimum core content./

8. This also seems to be the position of CESCR which has remarked:

"it may be useful for States to identify specific benchmarks or goals against which their performance in a given area can be assessed ... global benchmarks are of limited use, whereas national or other more specific benchmarks can provide an extremely valuable indication of progress." / General Comment 1, para. 6./

However, if "global benchmarks are of limited use", how can appropriate national benchmarks, consistent with States' obligations under the Covenant, be identified and utilized? The answer to this question is provided by a three-step process. In the following narrative, I sketch this process with reference to the right to education. / For present purposes, I confine my remarks to the right to education provisions of ICESCR (arts. 13 and 14). However, my comments bear upon other right to education provisions, such as those under the Convention on the Rights of the Child, the Convention on the Elimination of All Forms of Discrimination against Women, and the International Convention on the Elimination of All Forms of Racial Discrimination. Moreover, the three-step process might be applied beyond the right to education to other economic, social and cultural rights./

Step one: selecting key indicators

9. The first step is for the international community to agree upon a few - 10 or so - key right to education indicators. For example, they may include:

Expenditure for educational institutions as a percentage of GDP;

Annual expenditure per student at primary, secondary and tertiary levels;

Literacy rates.

Wherever appropriate, each indicator must be disaggregated.

10. The existing literature provides a wealth of education indicators. Consider, for example, UNESCO's World Education Report, UNDP's Human Development Report, the World's Women 1995: Trends and Statistics published by the United Nations, and Social Watch published by a consortium of NGOs. The recently published Education at a Glance: OECD Indicators uses no less than 41 education-related indicators. Education indicators, however, might not be the same as right to education indicators. On this point, guidance can be obtained from the global human rights conferences of recent years, including the World Conference on Education for All which took place in Jomtien, Thailand, in 1990. / Conference participants adopted the World Declaration on Education for All and Framework for Action to Meet Basic Learning Needs. In relation to indicators see, for example, the section "Goals and Targets" of the Framework for Action (paras. 4-9)./ Further, the reporting guidelines of treaty bodies also consider right to education indicators and benchmarks. / To take one example in the Committee's guideline on article 13 (1) States parties are asked: "What time-related goals and benchmarks has your Government set in this respect?" (E/C.12/1991/1, p. 18)./ Moreover, with particular reference to the Convention on the Rights of the Child, UNICEF and others have begun valuable work on, specifically, right to education indicators. / For example, the UNICEF meeting to discuss indicators for global monitoring of child rights held in Geneva, 9-12 February 1998./ I am not suggesting the task will be easy but, drawing upon this rich experience, it should be possible to formulate a few key right to education indicators. / There is a lot of literature on indicators and benchmarks in the human rights context. For a brief introduction, see P. Hunt, Reclaiming Social Rights: International and Comparative Perspectives, 1996, pp. 123-125. Also see Chapman, supra note 4./

11. Significantly, the exercise will demand collaboration between different treaty bodies and different specialized agencies. / In 1993, Thomas Hammarberg, in a paper submitted to a United Nations seminar on appropriate indicators to measure the realization of economic, social and cultural rights, wrote: "There is a need for close cooperation between the various treaty bodies on this [indicators] project." (see HR/GENEVA/1993/SEM/BP.6)/ For example, to the extent that the provisions of ICESCR and the Convention on the Rights of the Child overlap, the Committee on Economic, Social and Cultural Rights should employ the same key right to education indicators. Of course, textual differences between human rights treaties might also generate some indicators which do not apply to all instruments. As for specialized agencies, as the preceding paragraph suggests, the right to education bears upon the mandate of more than one agency. Thus, when trying to identify education indicators, collaboration between the relevant agencies is highly desirable.

12. In a sense, attempting to identify key right to education indicators is like trying to agree upon a new elementary vocabulary for speaking about human rights. If each treaty body formulates different education indicators, we will have missed an important opportunity to develop a new common language which would have been especially useful in the context of economic, social and cultural rights. The rudimentary vocabulary of indicators must be both consistent and simple, otherwise it will not win widespread currency.

Step two: setting national benchmarks

13. Having agreed upon a few key right to education indicators, step two sets the appropriate levels - or benchmarks - for each indicator in relation to each State party, taking into account progressive realization and resource availability, as required by article 2 (1) of the Covenant. Setting these national benchmarks might be done in the following way.

14. As part of the periodic reporting process, the Committee could invite each State party to address each of the key right to education indicators. Let us say that one of the key indicators is disaggregated literacy rates. The Committee could ask the State party to report on its current literacy rates. Secondly, it could invite the State party to set an appropriate projected literacy rate - or benchmark - for five years' time. The State party would undertake this exercise in relation to each of the key right to education indicators.

15. Given its responsibility for monitoring the implementation of ICESCR, the Committee would comment on the appropriateness of the State party's projected benchmarks. Thus, if a State party's projected benchmark appears to be overly modest, the Committee could ask for an explanation and invite the State party to reconsider. In an extreme case, the Committee might not only record the State party's excessively modest projected benchmark, but it might also set a higher benchmark which it considers is more consistent with the proper interpretation of the State's obligations under the Covenant.

16. An NGO might take the view that a State's projected benchmark is too unambitious. For example, if a key education indicator is disaggregated literacy rates, a State might peg its boys' literacy rate benchmark at (say) 80 per cent and its girls' literacy rate benchmark at (say) 70 per cent. Let us assume that the Committee accepts these benchmarks are within the State's "margin of appreciation". An NGO, however, might think they are much too low and declare its own benchmarks of (say) 85 per cent for boys and girls. The NGO could use these alternative benchmarks in the manner indicated in the next session.

17. This highlights a point made earlier. The agreed key indicators can become a new vocabulary for speaking about economic, social and cultural rights. Naturally, in the context of a particular State, there will be differences of opinion about the most appropriate national benchmark in relation to a specific key indicator. The Committee, a State party and NGOs might all have different preferred national benchmarks. But at least they will be using a common language - the rudimentary vocabulary of key indicators - to discuss the issues and monitor the Covenant's provisions. In the long term, the development of this new basic vocabulary of human rights might prove to be a significant contribution to the better promotion and protection of economic, social and cultural rights.

18. Finally, it would be in a State party's interests to set realistic but challenging national benchmarks. If a projected benchmark is too modest, this could be interpreted as governmental complacency. On the other hand, if a projected benchmark is over-ambitious, the State will be exposed in five years' time when it next reports to the Committee. In appropriate cases, a realistic but challenging benchmark is more likely to attract the technical support of specialized agencies and the financial assistance of donors.

Step three: monitoring the national benchmarks

19. Once generated within the Committee's periodic reporting process, the 10 right to education benchmarks could then be used, by all interested parties, to monitor that particular State's progress over the next five years. At the end of five years, when it next reports to the Committee, the State would report whether or not it has reached its 10 right to education benchmarks, and comment on problems - such as the introduction of a structural adjustment programme - it might have encountered.

20. As for the Committee, consistent with its monitoring role, it would comment on the State's progress. It may make recommendations to the State -for example, it may recommend that the State seeks technical assistance from relevant specialized agencies. The Committee would also ask the State to identify new, and (probably) higher, benchmarks for the next five-year period. And then the process repeats itself as part of the Committee's ongoing periodic reporting procedure.

21. While the Committee will usually monitor the benchmarks only every five years, other interested parties could monitor progress on a more frequent basis. NGOs, for example, could monitor progress every one or two years and publish their findings. An NGO might monitor the benchmarks set by the State during the Committee's reporting process and, in addition, the NGO could monitor any alternative more challenging benchmarks it has identified. Further, depending upon its mandate, a national human rights institution might wish to undertake a similar benchmark-monitoring exercise. The benchmarks could be integrated into national human rights - or social justice - action plans which some States might wish to formulate. Again, they could also be utilized by United Nations specialized agencies and other international bodies. In other words, the national benchmarks set in the course of the Committee's periodic reporting process are not reserved for the Committee's exclusive use once every five years.

22. Nonetheless, it should be acknowledged that non-State actors, struggling with very limited resources, may not be able to monitor effectively all benchmarks on an annual or biennial basis. For example, an NGO might not have access to annual nationwide data on disaggregated literacy rates. On the other hand, some of the benchmark data will often be found, on an annual basis, in governmental publications - such as the expenditure for educational institutions as a percentage of GDP. So the non-State monitoring of national benchmarks between a State's five-yearly reports to the Committee might not extend to all of the benchmarks. / When engaging in step one: selecting key indicators, one important factor to be taken into account is the availability of data needed to monitor those indicators and their accompanying national benchmarks./ Even this, however, could be a useful annual or biennial monitoring exercise.

23. I have already suggested that the key indicators should become a new, common, rudimentary vocabulary for talking about economic, social and cultural rights. Another way of viewing the key indicators and national benchmarks is to see them simply as a tool which anyone interested in economic, social and cultural rights - human rights treaty bodies, States, NGOs, national human rights institutions, specialized agencies and so on - can pick up and use if and when they so wish.

Specialized agencies

24. Three years after ICESCR entered into force, Philip Alston wrote:

"the decisive element in determining the success or failure of the Covenant will be the extent to which the UN and the specialized agencies are able to cooperate effectively in its implementation". / P. Alston, "The United Nations Specialized Agencies and Implementation of the International Covenant on Economic, Social and Cultural Rights", Columbia Journal of Transnational Law, vol. 18, 1979, p. 79./

Certainly, ICESCR bestows a crucial role upon the United Nations specialized agencies and, for the most part, / Among the specialized agencies, the ILO has the most creditable record of engagement with the Committee./ they have not responded positively to the challenge - at least, not yet. However, there are now some encouraging signs that some of the United Nations specialized agencies and programmes are beginning to take the Covenant more seriously. / By way of example, see UNDP's recent policy document Integrating Human Rights with Sustainable Human Development (1998).

25. This is not an appropriate place to examine these issues in detail. But one point deserves emphasis here. The three-step process outlined above -selecting key human rights indicators, setting national benchmarks, and monitoring these benchmarks - would benefit enormously from the effective participation of United Nations specialized agencies and programmes for they have an abundance of pertinent data and expertise. For example, United Nations specialized agencies and programmes might:

(a) Help identify the 10 (or so) key right to education indicators;

(b) Help a State identify achievable benchmarks for the next five years;

(c) Most crucially, help a State reach those benchmarks;

(d) Help a State monitor the entire process and report to the Committee;

(e) Following the Committee's consideration of the State's periodic report, the agencies might help a State implement the Committee's recommendations.

26. The adoption of a system of key human rights indicators and national benchmarks provides excellent opportunities for fruitful collaboration between specialized agencies, programmes and the Committee, consistent with the letter and spirit of the Covenant.

Conclusion

27. In my view, human rights benchmarks have a significant role to play in the promotion and protection of economic, social and cultural rights. Nonetheless, it is important to keep their role in perspective. Even if we were to perfect the use of human rights benchmarks, they would only provide us with part of the story. They might be able to tell us what is happening, but they cannot usually tell us why it is happening. If a State's human rights record is a feature film, benchmarks are an inadequate snapshot of human rights at a specific moment. Successive frames, frozen at regular intervals, can tell us something about the film, but not the whole story. So it is with human rights benchmarks: they can tell us part of the story but not all of it.

28. Thus, while they may enhance our understanding, benchmarks are no substitute for other ways of monitoring a State's human rights record. What are these other ways? At the international level, they include the treaty-based periodic reporting process, the preparation of "shadow" reports by NGOs, the elucidation of "minimum core content", appropriate complaints procedures. Charter-based "special procedures" such as the decisions by the Commission on Human Rights in 1998 to establish an independent expert on the right to development and a special rapporteur on primary education, and so on. The combination of all these devices - in conjunction with benchmarks - can provide us an increasingly complete account of a State's human rights policy and practice.

29. In summary, we should not neglect existing devices which address both the individual and structural dimensions of economic, social and cultural rights. Further, we should remain open to the possibility of new approaches to these issues. Equally, however, we need to use human rights indicators and benchmarks - to develop this new human rights vocabulary. Indicators and benchmarks have an especially important role in relation to the shifting State obligations implicit in the concepts of "progressive realization" and "maximum available resources". Indeed, without human rights indicators and benchmarks, it is difficult to see how these elusive concepts can be effectively monitored.