"All academics have drafts of things that one day may get published. Then the lorry hits you, and that's that."
This may be a worst-case scenario, but it serves to underline the importance of sorting and archiving the data scholars painstakingly collect over the course of their careers. Yet for those working in politically sensitive fields, archiving can prove to be an ethical minefield, discouraging some from approaching it at all.
Wendy James, professor of social anthropology at the University of Oxford, is investigating how her own material can be presented in a way that satisfies the competing requirements of accessibility and confidentiality, should the lorry intervene.
"My plan is to leave it in an accessible and user-friendly form, so that people can press a button to bring up everything to do with the lives of children, for example," she said. "Leaving it in that form is actually much more of a challenge than just arranging one's written materials. But it would be a terrible shame if it got left in a box because it was too difficult to sort out."
Since some of her information was collected in Sudanese conflict zones, it is not possible for all of it to go online lest it endanger any of the people involved.
Instead, Professor James has in mind a two-level system, where non-sensitive footage of singing and dancing, for example, can go on the web, while other material is archived in a protected section and made available only with permission.
A similarly tiered system of archiving data for reuse is already being used elsewhere, for example in the Timescapes project at the University of Leeds, which is exploring how personal relationships and identities unfold over people's lifetimes.
Another issue in the sharing of data is interpretation. Pat Caplan, emeritus professor of anthropology at Goldsmiths, University of London, explores the difficulties in the journal Anthropology Today.
"Using someone else's material when they're not around is quite difficult," she told Times Higher Education. "It is not the same as having done the interviews oneself, actually being there, smelling it, seeing it, listening to it and so making sense of it afterwards. It doesn't quite have the same immediacy."
It is not only anthropologists who have to be sensitive to such challenges. British philosopher and archaeologist R.G. Collingwood, for example, did not want his data to be used for anything he had not prepared himself. But after his death in 1943, his daughter Teresa Smith and other scholars collated, edited and published some of his work.
However, social science approaches make archiving a thorny issue. "Anthropologists get talking to people in open-ended ways, or make friends with them," Professor James explained. "Sociologists or economists don't often do this. Their data are 'data' and can be archived without any qualms."
Such quantitative data can be codified and compiled into publicly available corpora - also a valuable resource in linguistic research.
Whatever the discipline, possessiveness is a trait found among many researchers when it comes to sharing data - at least until it has been published. And when the data are ready to be archived, resources are required to get the job done.
From the research councils' point of view, archiving adds value to the research since it allows others to benefit from existing material.
According to Professor Caplan, scholars applying for funding should put in for additional money to ensure that archiving time is factored in. "It's extremely labour-intensive," she said.
Another challenge comes with the increasing level of ethical regulations on consent imposed by the research councils. Some warn that blanket requirements are not compatible with all disciplines.
"These ethics structures are made on the medical model, such as the agreement that you mustn't experiment on a human being without their consent," said Professor James.
"This becomes extremely restrictive if you apply it to all social sciences where you are just photographing or talking to people."
As research becomes increasingly interdisciplinary and demand for public accessibility grows, the debate on archiving is becoming pertinent to researchers everywhere.
"It's very hard to lay down absolute rules," said Professor Caplan. "You can only lay down guidelines, and then people have to use common sense as much as anything."