The one-stop shop for anyone wanting to read completed dissertations is the ProQuest Dissertations and Theses Full Text Database. While working with this resource during this summer, I noticed something strange. Many of the dissertations completed after 2011 were simply unavailable, the access restricted. I knew that completed dissertations are also submitted as hard copies to the University archives, so I went there to read what I was interested in, knowing that everywhere in the world bound dissertations are made available by libraries and archives to whoever wishes to inspect them. But I was wrong. At Harvard, dissertations that authors choose to embargo from being released online are also barred from use in the library in their paper versions. They are effectively made top secret. Why?

Almost three years ago, doctoral students started opposing the automatic release of their dissertations online, due to fears that publishers will not be interested in printing books developed from dissertations anyone can read on the Internet. As I was informed by Professor Robert Darnton, the University Librarian, dissertation embargoes were allowed based on a 1951 decision by the Harvard Corporation, but this was intended to be exceptional—to protect sensitive materials. As a rule, dissertations were made openly available at the Harvard Archives. This changed a few years ago, when the deans, responding to doctoral students’ fears, allowed embargoes broadly. As a result, as a search in ProQuest reveals, an unprecedented number of dissertations (almost one in three) produced at Harvard in 2012 and 2013 are embargoed. These dissertations are now secret and the authors can decide how long to keep them so.

The current regulations declare that dissertations must be made openly available as a proof of achievement because a doctoral candidate “cannot have a degree for making a discovery that is kept secret” and that it is “only in very exceptional cases that access to dissertations is restricted.” But embargoes are still allowed. A dissertation can be embargoed for six months, one or two years, and a longer embargo of unspecified length is also possible. It is, however, unclear what this embargo entails. Does one need to prove that an embargo is necessary? Who has the final say? More importantly, given the number of embargoed dissertations in the last two years, is it completely true that restrictions are allowed only in very exceptional cases?

In July 2013 the American Historical Association made public its position on dissertation embargoes. It supports embargoes to avoid jeopardizing doctoral students’ chances at landing contracts with publishers. But the AHA is also careful to say that dissertations should still be made available in hard copy by the home institutions of doctoral graduates.

Harvard’s Office for Scholarly Communication has criticized AHA’s approach. It advocates not only access to the bound dissertation, but also an overall open-access policy. The OSC argues that concerns about online dissertations rendered unpublishable are unfounded. It quotes a study showing that most publishers would consider such a dissertation. Only four percent would never consider publishing one. The OSC believes this is sufficient to show that a maximalist open-access philosophy is appropriate, with the caveat of “flexibility for students in different circumstances, and the shortest embargoes that will serve student needs.” What those embargoes would entail remains unspecified. In any case, given the high numbers of embargoed dissertations this year, the University doesn’t appear to be doing what the OSC preaches.

As a doctoral student, I obviously wouldn’t want to jeopardize the publication of my dissertation as a book, if I were certain putting it online would have this effect. And there are many students using this argument to advocate for embargoes, although the actual data seems to tell another story. There are nevertheless also voices strongly opposing embargoes. One of my friends wrote the following in response to a Facebook post: “Embargoes really make me sad. Even if limited in time, a total betrayal of the spirit we should enter graduate school with: the fortunate among us get paid in all manner of scholarships, and we produce a contribution to our field, and hopefully humanity.” He would only accept embargoes for protecting still-living human sources and sensitive materials.

Whether embargoes are easily allowed or not makes a world of difference to how students might approach their dissertations. If I knew that my dissertation would be made openly available, I would try my best to make it solid and useful for other researchers. But if I could embargo it easily and make it secret, I would have an incentive to only carefully chisel my title and abstract—what everybody can see. As for the rest, only the three very busy professors on my committee could see it. Which of these two visions comes closest to the ideals of what a dissertation should be?

The University needs to create clear guidelines on embargoes, guidelines which strike a balance between the interests of doctoral students and the interests of research. If Harvard decides that dissertations can be embargoed very easily, this might spread to other schools, creating an unprecedented regime of secrecy. I hope, at the very least, that the issue of whether dissertations should be made available online can be disentangled from the issue of whether dissertations should be made available at all.

Florin-Stefan Morar is a Ph.D. candidate in history of science.