Citation indexes, you ask? The theory is that the best research gets referred to most frequently in subsequent research. By totting up the number of references to Professor Fixit’s seminal paper on cold nuclear fusion, you can rank the significance of his work. There are indeed online league tables that will tell you the names of the world’s most cited scientists in each discipline. These are no doubt very useful to superprofs in their salary negotiations, but even among hard scientists there are considerable reservations about the value of such indexes. For one thing, Professor Fixit’s paper might have a thousand citations as an example of failed research. And for another, if you give more significance to citation, then everybody will start mentioning their mates’ work in their footnotes so as to bump up the count. You cite my paper and I’ll cite yours. One imagines the establishment of clandestine internet “citation rings”.

According to the “ISI Web of Knowledge” the database that will be used for the new Research Excellence Framework (“REF”), my top citation hit is a little book that I dashed off in a few weeks as a polemical squib — it far outscores the considered monographs over which I laboured for years in the great research lib­raries of the world. In the humanities, there are few established hierarchies of journals, and much of the most significant work is done in books, scholarly editions and other forms that elude the citation databases. Gordon Brown has not backed down on his metric basket case, but he has conceded that it will not work for the humanities — not because of reservations about the principle, but because the metrics aren’t sophisticated enough. The new mechanism will go ahead, but there will be some kind of fudge for the arts and humanities disciplines, a yet-to-be-determined “light touch peer review process informed by quantitative indicators”.

What the reform of the RAE conspicuously fails to address is the way that the whole research assessment process has deformed the delicate ecology of higher education in the humanities. There certainly was deadwood that needed pruning in the 1980s. But in the humanities, productivity is not synonymous with publication of one article a year in the Journal of English and Germanic Philology. There is an elephant in the room of the debate about how to fund university research, namely the separation of research from teaching.