RATING PAKISTANI SCIENTISTS

THE NEWS International, 28th October 1999

The Pakistan Council for Science and Technology (PCST) has produced a 133-page document entitled Leading Scientists of Pakistan. The rationale, as propagated in the Foreword, is to “help the administrators of science to evaluate the quality and output of scientists who seek key positions”. Scientists, 681 in total, are rated according to the so-called Impact Factor (IF), which is based on the number of times the journal, which contains their research publications, is cited.

Nowhere in the world of academia are scholars rated as such. It is an established routine that a scholar is judged on the basis of his best work. For instance, Andrew Wiles is ranked high in merit as a mathematician because of his proof of Fermat’s Theorem. But nobody finds it necessary to compare him with Pierre Fermat and judge who is better. The Institute of Scientific Information (ISI) of USA, has produced a long list of about 4,950 scientific periodicals. This SCI Journal Citation Reports lists journals rated on the basis of how many times a journal is cited. This is not meant to be used to rate scientists as such. PCST has misused the SCI Journal Citation Reports.  

Such linear ranking of scientists in the PCST document is rude and objectionable. A number of scientists who are ranked high in the booklet are well known in the scientific community and are respected for their valuable scientific and technological contributions. It does not matter what their IF is. Ranking them linearly would rather be insulting them.

On the other extreme there are funky scientists who may have published a good number of papers but the worth of their work is pathetically low. It does not take much for an intelligent person to judge the calibre of some of these scientists. A scientist who claims himself to be a specialist of a research field A may have been actually publishing papers in journals of research field B. The journals of field B, by and large, may have high IF whereas the journals of field A may, on the other hand, generally have low IF. Obviously he will be ranked high on the basis of IF in field A when he is compared with his co-scientists working in the same field and publishing also in the journals of field A. Whereas he will be ranked low in comparison with the scientists who are publishing their work in the journals of field B.

It is highly likely that two scholars may have separate papers X and Y published in the same journal but the journal is cited ten times because of paper X and is not cited at all because of paper Y. How can one say that the two authors have the same IF? Thus the rating of a journal does not necessarily reflect the scholarship of all those who have published papers in that journal.

The list itself is defective. For instance, there are a number of high standard journals which are not mentioned in the list. The Mathematical Reviews of the American Mathematical Society review papers every month published in some 1,500 mathematical journals whereas the list contains only approximately 163 mathematical journals. Then there are many well-reputed journals which are not included in the list, e.g, the Proceedings of the Cambridge Philosophical Society.

One critical study of the list has revealed that there are 63 journals directly related to chemistry, 5 journals directly related to mathematics, 34 journals directly related to physics, and 430 journals directly related to biology, which have an IF higher than 2. The highest IF of a journal in mathematics is 9.040 and that is of the Proceedings of the National Academy of Sciences, which, publishes papers from all sciences. One notes that in medical sciences, one journal has the IF 38.854. This means that if one publishes one paper in this journal, one gets the IF equal to 38.854 whereas if one of the great mathematicians of the 20th century (for instance Philip Hall) publishes 40 research papers in the best journals of mathematics, he will only get an IF equal to 20.353. 

Across the board comparison of scholars is absurd and unfair. Scholars who do research on topics of local interest, such as in agriculture, will have difficulty in finding an international journal which will be interested in publishing their research outcomes. This does not mean that their work is insignificant but that their work is not of global interest as such. Thus they will never be able to achieve high IF although they might have done research of national interest.

Age is another factor to be considered. A scholar who has just stepped into the domain of research is very unlikely to match a researcher who is senior in age. It will be interesting to compare the cumulative IF of scientists of the same age group. Many scientists will lose their pagris. In a society like Pakistan it has become a paradigm that subordinates, whether they be Research Officers, Lecturers, or M.Phil./Ph.D.Scholars, will inevitably have to co-author their research work with their bosses. Obviously if one has ten such “sub-ordinates” working under his supervision and each worker produces three research papers per year, the boss will certainly have 30 papers to his credit every year and if this “productivity” continues he will have 300 research papers in a decade. Such bosses and subordinates are abundantly available in our scientific culture. Of course there were/are great scholars in the world, e.g. A. Cayley (1821 - 1895), L. Euler (1707 - 1783), J. L. Lagrange (1736 - 1813), A. Cauchy (1789 - 1857), and P. Erdos (1913 - 1996) who have high number of research papers to their credit but they are not great names on the basis of the number of their papers but because of the worth of their work.

In 1983, Simon Donaldson (b. 1957) showed that there is an exotic smooth structure on 4-dimenisonal space, in addition to the standard one. He proved a difficult problem by using ideas born of an interaction between topology, algebra, and mathematical physics. His work is published in the Journal of Differential Geometry, whose IF is only 0.735. The CV of Donaldson lists five publications only. He is a Field Medalist (Noble Prize equivalent in Mathematics) and an FRS and has a number of other international awards but his IF is only 1.763, whereas according to the ranking of PCST, there are many Pakistani mathematicians' whose IF are above Donaldson's.

Another example is that of Philip Hall (1904 - 1982), the main impetus behind the British school of group theory. He was President of the London Mathematical Society and was awarded De Morgan medal, Sylvester Medal, and was FRS. In 1932 he wrote his most famous paper A Contribution to the theory of groups of prime-power order, published in the Proceedings of the London Mathematical Society, whose IF is only 0.735.  Hall has published 40 papers in his career of 47 years. Five of these papers are published in journals which are not listed in the list of SCI Journal Citation Reports. The cumulative IF of Hall is 19.844. How is it that a mathematical giant is ranked so low by PCST standards?

One may not be productive research-wise every year. For instance, Alfred Young, famous for his work on standard tableaux which was published in the 1920s, did not produce intentionally anything for the next twenty years. His cumulative IF of almost twenty years would have been zero.

Pakistan lacks a true scientific culture. It has been successful in achieving nuclear capability but that does not mean it possesses true scientific infrastructure capable of producing high-class fundamental research. There are scientists of international repute who are doing fundamental research but they are few and their research is sporadic. A lot needs to be done and it will take a long time before we can stand as a nation at par with the international community of scientifically developed nations.

PCST’s document has not only projected the pathetic state of science in Pakistan but it could lead to long term damaging repercussions on science in the country. The authors of the PCST’s report have not realized perhaps that the ultimate blame for the poor state of science in Pakistan goes to those who look after education, technology and science. A worth-mentioning example of this so-called IF’s misuse has recently come to the surface. One academic has sued a university for selecting a colleague rather him for a higher position. The plea is that he is a better scientist because his IF is (marginally) higher than the colleague who has been selected.

It would have been useful if PCST had produced a report on the state of science in Pakistan and recommendations for its improvement rather than taking out a booklet in the effort to try to rank scientists linearly. Such a ranking may be of some benefit to those few who are looking for status, recognition, and appreciation but it hardly is going to benefit science in Pakistan. The report produced by PCST criticises certain top-level appointments on the basis that these personalities do not have a high IF. Research is not the only factor for appointments at top levels. There are a number of other job requirements which need to be considered. There is no correlation between scholarship and administrative qualities.