But Arthur Bienenstock, Vice Provost and Dean of Research and Graduate Policy of Stanford University, probably did not expect to provoke controversy with his keynote address on "The Essential Characteristics of Research Universities." That he did surprised at least a few participants at the colloquium.

The eight elements Bienenstock identified as common to high quality research universities and his elaborations on them were neither exceptional nor controversial. The list appeared entirely reasonable and even predictable.

1) High quality faculty committed to research and teaching: Universities should not recruit only their own graduates; this practice unnecessarily narrows the field of eligible candidates and forfeits the chance to bring in new ideas from the outside.

2) High quality graduate students: Necessary because the students must be able to learn not only from the faculty but also from each other.

3) An intellectual climate that encourages scholarship: Ideology or dogma must not inhibit intellectual exploration. Faculty must be free to determine their own research directions and to publish the results of their studies.

4) Facilities to support effective teaching and research: These would include libraries and laboratories appropriate to the teaching and research tasks, and should be available also to junior faculty.

5) Funding for operations and instruction: No major research university covers its operations solely from tuition fees (at least none Bienenstock is aware of).

6) Research funding: Bienenstock favors a process where outside agencies provide funding to individuals or small groups through a competitive, peer review process.

7) Research infrastructure: This would include specialized offices to help the faculty with the legal, administrative and accounting requirements of funding agencies, assist in addressing, environment, safety and ethical concerns, and prepare for possible commercialization of the research results.

8) High quality leadership: Because building a first-rate research university is a major, long-term task, it requires a visionary leader who can manage competing demands of talented people interacting in a complex organization.

These are obviously desirable characteristics. As many presentations at the colloquium also made obvious, they are also beyond the reach of most universities in the developing world. Even venerable European universities feel threatened by the competition for the resources required to maintain first-rate research universities.

Professor Timothy Garton Ash, with appointments at both Oxford and Stanford, noted that only Oxford and Cambridge managed to surface in all global top 10 ranking of universities, dominated by American institutions. Writing in the Guardian on "Research universities’ uncertain future," he wondered "whether Europe will have any worldclass research universities at all in 20 years’ time."

With an endowment fund roughly double that of Oxford, Stanford just launched a drive to raise .3billion by 2011, and has already obtained pledges approaching .2billion. Britain spends only 1.1% of GDP on tertiary education. The US spends 2.6%, 1.2% from public sources and 1.4 from the private sector. Comparisons with developing countries would reveal even greater gaps.

To those for whom the essential requirements for a world-class research university appear unattainable, Bienenstock’s paper was a provocation. It appeared to project the ideal American research university as the norm to which all universities must aspire and the standard upon which to base their value.

Imanol Ordorika, professor at the Universidad Nacional Autonoma de Mexico, warns against getting caught up in the global competition among elite universities. He worries that the qualifications and performance criteria established for the comparative ranking of universities in "league tables" will diminish the credibility of national research universities in the South and distract them from addressing the knowledge needs of their own countries.

Ordorika notes that, even in the United States, few institutions can meet the standards set for the ideal American research university. This is something Bienenstock himself might concede. His paper begins with a cautionary note on the great variety of American higher education institutions. Of over 4000 listed by the Carnegie Foundation in 2000, only 261 came under the category of research universities.

UNESCO deserves credit for focusing attention on the threats to the universities, in North and South, that seek to serve as centers of research and knowledge creation. This is an important issue that must be addressed. But UNESCO also has a larger constituency of educational institutions that aim, not for knowledge creation, but for skills development to meet the human resource requirements of the country.

All countries need both types of learning institutions: Those that will create knowledge and those that will prepare people for the job opportunities that the new knowledge, whether from North or South, will hopefully increase. Unfortunately, both types brand themselves as "universities."

In its terms of reference, the colloquium identifies research as a "key ingredient in the institutional identity of universities." This assumption would exclude from the discussion those representing the great majority of so-called "universities" in the South.

But the colloquium also regards research as "an indispensable prerequisite for a successful program of teaching and public service," which often are the explicit objectives of postsecondary education institutions in developing countries. These institutions could use UNESCO’s help in identifying what kind of research would help promote their teaching and public service mission and how to build sustainable capacity for it.

Better to help these "universities" do a better job with knowledge transmission and skills development than to burden them with the task of knowledge creation for which they know they are not prepared.