Will this paper ever be cited?

It is an accepted fact of academic life that some papers, in whatever discipline and wherever published, will never be cited. Empirical support for this can be found in studies such as those reported by Glanzel (1992), Rousseau (1994), and Egghe (2000) where the distribution of the time of first citation is investigated. In all cases, it is assumed that never-citation is connected with the perceived obsolescence of any body of papers, where the rate of citation changes over time, eventually declining toward zero (see also Van Raan, 1988). For instance, in his model for the first-citation distribution, Egghe (2000) assumes that there is a fixed but unknown proportion of the collection that will never be cited, and this proportion is therefore built-in as a parameter of the model. In a different approach, to be pursued here, Burrell (2001) proposed a model for the citation process incorporating obsolescence from which he demonstrated that it is inevitable that there will be a proportion that is never cited. Of course, it is not possible to identify beforehand which papers will never be cited. Indeed, no matter how long a particular paper has remained uncited, there is always a chance that it will be cited at some time in the future. On the other hand, it seems intuitively clear that the longer a paper has been uncited, the less likely it is that it will be cited in the future. In this article, we shall determine this probability in a mathematically precise way and confirm the intuitive feeling. We should remark that essentially the same question was investigated, using essentially the same tools, by Burrell (1987) in the context of a policy for relegation of unused library materials. The Model and Its Consequences