The article on citation counting ("A measure of humanities?", 5 November) mentions two reasons why this method is ill-suited to assessing the quality of work produced in the arts and humanities: disregard for non-English publications, and the preference for books over journals. But it did not mention a more fundamental reason why this method simply doesn't work as a measure of worth.
Scholars in the arts and humanities cite work as much to expose failings as they do to commend excellence (indeed, perhaps rather more). A reference to a piece of research may therefore indicate that it is hasty, ignorant or foolish as much as that it is thorough, scholarly and intelligent. It seems slightly surprising that this situation doesn't also obtain, to some degree, in the sciences; but be this as it may, only someone who does not really read much research in the arts and humanities would suppose that bibliometrics reflects brilliance.
Richard Serjeantson, Trinity College, Cambridge.