[This is the text of a talk I gave last week, as “Speculative Collections and the Emancipatory Library,” to close a symposium honoring Dan Hazen, about the future of academic library collecting. See also #HazenatHarvard tweets as assembled by Merrilee Proffitt, a video of the presentation from Harvard Library, and an excerpt from a prior talk (“Alternate Futures/Usable Pasts“), which introduces the concept and offers some paths in.]
[Edited to add: and please read the wonderful “Liberatory Archives,” by Jarrett M. Drake, which takes up many of the same themes as the talk below, and was apparently delivered on the same day! Maybe it’s time? Hurry up please it’s time.]
[2019 update: an edited version of this talk was published in The Routledge International Handbook of New Digital Practices in Galleries, Libraries, Archives, Museums and Heritage Sites, with a green OA copy of the author-accepted manuscript immediately available as a PDF file, here. Other OA chapters are available as well.]
Reproducibility. Openness. Transparency. Rationality. Interoperability, and an orientation toward interdisciplinary problem-solving. Mine is a non-exclusive list, to be sure, but you might recognize these as values driving data management in the sciences and social sciences, and underlying the creation of collections, interfaces, and infrastructure in what we call “data-driven” fields. They have their problems of positivism, these values—and it has become the necessary project of many thinkers in the library and information science community to demonstrate how underlying assumptions of neutrality and universality in them—and therefore in our practices of selection and description, our design of search mechanisms, and even in many libraries’ public service policies and stances around them—are in fact decidedly non-neutral expressions of dominant, sometimes oppressive ideologies.
But I’ll risk the ire of friends to say that—taken together—the value-sets of open science represent a quality I find disappointingly, maybe even irresponsibly absent from digital library interface design and collection-building. They represent a forward-looking temporal orientation. And I think we feel the absence of that orientation, particularly, now that we are so decidedly past the era of collecting “on spec”—past, that is, being able to hold an image of libraries un-stuck in time, libraries on the long tail, libraries with a far, far future reach—where we invest in and gather materials that may have no immediate use-value.
While administrative imagination slowly catches up to the logic of the network—and while we work to realize “collective collections” that might mitigate this problem—local pressures move inexorably in, and train our attention on contemporary, not future needs: on meeting needs (as we say) “just in time.” Please don’t misunderstand. I do not propose that we adopt the values of open science wholesale (it will be a cold day in Hell when “reproducibility” takes hold in English departments—and “openness” itself has different valences and dangers across communities and fields). Instead, I suggest that we consider the cumulative effect of underlying value sets like these in terms of their temporal orientation—the degree of forward-lookingness and open-endedness inherent in the concepts we hold dear—and what that means for the systems we build. Continue reading “speculative collections”