In Part I of the interview Jürgen Renn, Director at the Max-Planck-Institut für Wissenschaftsgeschichte (MPIWG), spoke about his career and his perception of the relation between science and the history of science. In this sequel he speaks freely about his views on open access. Following the news on the demonstrations of students and professors in Amsterdam and the subsequent resignation of the president of the executive board of the University of Amsterdam, I felt the need to ask Jürgen Renn if he thought that open access can also have potential political consequences. Thereafter, Jürgen Renn explained me his view on the future prospects of the history of science and the possibilities of new digital methods.
Currently, there is much discussion about the open access movement, which is endorsed also at the MPIWG. Some might say that open access means a revolution of the kind of the Gutenberg printing press in 15th century. However, open access could also have some less pretty side-effects. Some expect that open access could give free reign to badly written, hastily produced and unconvincing texts (see, for example earlier blogposts on this subject on Shells & Pebbles: here and here). Some argue that open access demands an easy and fast spread of knowledge. Although this sounds great, the typically slow sciences like history and other humaniora could easily suffer from such a transition. In short, it might stimulate bad scholarship. How would you respond?
First of all, to compare open access with Gutenberg is I think exaggerated. I mean the digital medium, the internet, the web, the possibility of global connectivity and electronic availability, that’s the revolution. Open access is just one form of making use of the potential of this new infrastructure. Open access has many facets. It started with the sciences and the need of scientists to communicate their information as quickly as possible and without many of the obstacles that hamper traditional publication processes. Many journals take a long time before something is published. The high-energy physics community has, for instance,relied for a long time on a preprint server. The actual information flow was dominated much more by the preprint dissemination than by the journals. (The MPIWG itself, has produced hundreds of freely available preprints –PvdH.)
Journals increasingly took on the function of just legitimising authorship and establishing curricula, while the real information flew completely through these preprint services. In other words, these two functions were very much disentangled. In addition, journal prices rose a lot, while people felt that publicly funded science should be publicly available and that people shouldn’t pay twice for the information. Open access was to some extent about the issue of economic viability and justice. Yet, I think that the most important issue of open access is really the connectivity it allows. Artificially imposing open access without boundaries makes it possible to interconnect information available on the web not just by hand but also by automatically supported procedures.
Let me give you an example from the humanities. Let’s say I publish a Chinese text, open access, and I connect it automatically to an openly available dictionary, Chinese-English or Chinese-German. The two documents can be seamlessly connected in one process, where I can click on a word in the text and within seconds look up the dictionary entry. This is only possible when the two documents, the text and the dictionary, are openly available and allow an automatic connection. If you have a barrier—think of having to insert your name or your credit card before you can access it—you cannot establish an automatic link between the text and the dictionary. This is just a very simple example. Semantic searches over great archives—as one might want to do for the history of science—is possible only when you do not have such limits and can cross over immediately from one text to the other. That is, I think, the most important part.
But there is another reason which I think is coming more and more into the picture. If you have a historical interpretation that refers to sources and if the sources are openly available online, you can just click on the footnote and look up the source. This is not just a question of controllability or of reliability, this is also a question of being able to accumulate research in a way that the interpretation becomes an access structure for the primary sources. You can use the interpretation as a door through which you enter the world of the sources. All these novel intellectual achievements cannot be realized without open access.
Now, the economy of open access is a totally different issue. There are costs involved; open access is not free. So the question is: who pays for it? There are various models. A difficult issue indeed, I won’t dispute that. But, one thing remains true. The dissemination of research is part of the mission of research institutions and universities. They have a responsibility to make their knowledge available.
This brings us to the point of quality control. There is simply no reason why open access journals should lower their standards compared to any other journal. One can impose the same peer review mechanisms and open access could even improve upon it. Peer reviewing can be made much more open and much more transparent by involving more people. For example, if I have to judge whether a paper is a new paper, then I will have to check other papers. When all those other papers are freely available, this is a much easier job.
The fickle flow of money increasingly dictates research programs and prospects for researchers and students. Looking at the demonstrations against the executive board of the University of Amsterdam we can conclude that this development can have dramatic results. A management which focuses centrally on academic output and the lowering of production costs may come at the expense of the considerations and interests of the academic community itself. This also involves the quality of education. Do you think that open access intrinsically contributes to a system that is more autonomous, more independent of top-down policy making and more democratic?
It depends. It’s not an automatism, because open access requires money and the question is how that money is being administered and distributed. Open access relies on a completely different cycle. It is no longer necessary, for example, to have somebody who is financially interested like a publisher. Nevertheless, for the production of publications you do need to have some institutional basis. So the question is: how will these institutions be organised and how will they decide on what to publish and what not? So in other words, the economy of an open access based system is not fully defined unless you also define who owns the means of production, to put it in old Marxist terms. Open access allows, of course, for coupling those who own the means of production and the ones who decide what is published. Although such coupling is not a new phenomenon, the system shapes up differently when the universities and institutions do it themselves. There is a degree to which these things are not fully defined yet.
On the other hand, I do believe that the total production costs will in the end be lower and that open access will make it easier to shape the world of publications in a more democratic and more bottom up fashion. It will also be easier to break with some cliché type publications like the classical book that even academic publishers, in particular US academic publishers, hope to sell on a book market. Publishers have their profiles and actually have strong commercial interests. Well, that reflects on the kind of books they accept and promote. When an academic institution doesn’t need to make profit from the selling of books, these constraints fall away. When the mission is wholly to disseminate scientific knowledge of a high quality, you get other regulations on the information flow that may be beneficial.
Could open access by itself empower universities or institutions?
Yes, although you will still need to have a filtering process. Open access publishing must be tuned to the intellectual quality of the output. Thus, it can empower universities, but only if the universities decide among their own members and staff who gets access to publication and who does not. That is a danger as well of course. I would therefore opt for a continuation of the peer reviewing system, but maybe in new forms. It has to be clear that when something is published as a scientific publication that there is some control by the scientific community and I don’t mean necessarily the local scientific community but in some sense the global expertise of the particular field involved. Now, that has to be managed as well. And so I see the necessity for a certain division between on the one hand the technical infrastructure that is necessary for producing open access publications, and on the other hand the decision mechanisms that have to be carried not by local institutions but by a community of experts.
There is a lot that can be improved in the open access world when it comes to peer reviewing. Transparency is the first thing. Instead of closed clubs of people that act basically anonymously you may very well have open peer reviewing with names mentioned and possibilities for authors to respond. You may have different degrees of publication and different stages. I think that these are mechanisms that will ultimately improve the scientific quality of publications, but it has to be set up in the right way.
Costs will in the end be lower, maybe because the technical quality will in some sense also be lower. You won’t find as much technical expertise in a typical university or academic institution as with at least some publishers. Think of how to deal with illustrations, how to deal with copy editing and so on. Yet, I think that an improvement of the peer reviewing process will compensate for these potential deficiencies. If organised in the right way, open access will ultimately enhance the quality of the scientific publications.
Open access facilitates trans-disciplinary research as it increases accessibility and connectivity of results and sources. (See here and here for examples of projects on open access at the MPIWG.) In light of such developments, you encourage to address bigger questions. You have said that this can now be done on an empirical instead of philosophical or essayist basis. Could you explain what this means?
Yes, I think this is very important. By empirical work of historians I mean document-based ways: historical work based on archival material and historical documents, dealing with large amounts of primary data, making databases of persons, artefacts and whatnot. For instance, you can look at the scientific revolution of the early modern period not just from the perspective of an individual scholar like Galileo or Kepler or Thomas Harriot. Premised on the data being basically open, you can now also draw on a very broad array of documents from various countries in various languages and bring them together. I do not mean necessarily to make an all-comprehensive presentation but in order to probe different hypotheses, heuristically; develop ideas, try them out on the material, change your ideas. So the process of interaction between ideas and the available material becomes much more rapid and flexible.
Once you have settled on a particular view or interpretation you want to pursue and substantiate, you can now document that with historical material by establishing direct links between the interpretation and the primary source material. We can now start to do bigger projects. This is what I mean by this broad empirical basis. The interpretation directly becomes a window of access to the sources and through which you can also read those sources. And this also works the other way around. You can start from a source and look at the various interpretations that are available.
What are examples of these bigger questions that were previously answered on a philosophical or essayist basis?
For instance, one question that I believe is really important for the early modern period is: to which extent did scholarly networks play a role and how did these scholarly networks function? Traditionally this can be done by starting from a single correspondence, but this doesn’t give an overall view of the connectivity of the community. Who wrote letters to whom? Who were the hubs in the centre? Where were the information nodes, exchanging information with many people? Which role did publishers play? You can only start to study the role of these exchange structures for major processes taking place in the early modern period on the basis of larger data. (See here, here and here for examples.)
Other examples would be similar network related questions for other periods. For instance the revolution of modern physics, the quantum revolution. Here you have many players, not just one or two or a handful of people, like it was the case with Einstein’s relativity theory. Dozens of people were involved. How did their interactions play out? The relations between specific institutions and specific discoveries is another example that you might think of. One project we are currently pursuing deals with the history of the Max Planck Society. This is a project that deals with fifty years of history and covers over eighty institutes. There are literally kilometres of documents! What would such a history look like if you don’t use digitization over part of these documents and not try to use bibliographic databases or databases with publications or documents that you can search through? You cannot really master them and see how global shifts take place; whether an institution like the Max Planck Society played a leadership role or whether it was lagging behind. You need to play this out on the background of the global context of science.
In order to get these big pictures, one could of course sit down with his present knowledge and start writing, but this simply has no Verbindlichkeit, you know. Such hypotheses would not be grounded in empirical data.
Digital methods help us in establishing such a basis. Yet, there is, of course, always the danger of using this big data digital humanities approach to do encyclopaedic work and that’s very boring.
History without interpretation…
Yes, you can get all this data and kind of categorize and classify it and so on, but the more interesting approach is to try out new hypotheses. What were conditions of creativity in the period under study? How did new fields emerge? What were the main factors? You can do comparative studies on the background of a lot of data that you can only manage if you have it in digital form.
Is this in contrast with the more traditional history of ideas?
Well, think of something that I admire very much and that you are very familiar with like Cassirer’s Erkenntnisproblem (see also: Volume II & III). This is really an attempt to give an overview of a broad intellectual development; a wonderful piece of the history of ideas and the outgrowth of an individual scholarly effort that took many years and a great mind. Now, imagine the same thing, but this time with all the empirical basis, having the primary texts and the secondary literature that Cassirer used on order to write this at your fingertips. You might see that Cassirer was stringing together things that don’t belong together. You can go from Cassirer where he makes an allusion and then directly look at the primary texts, compare them and improve it! Maybe there were unknown authors that he didn’t know of and that you can add to the picture or you can even correct the picture.
We live in a time were unfortunately very few people still try to write such comprehensive overviews and synthetic works as Cassirer has done. Today there is this bifurcation of people who are either very careful and meticulous and base their interpretations on careful studies of primary material and those who write sweeping overviews but with no such grounding. This approach might help to connect them again.
o-o-o
Featured image: Main entrance of the Max-Planck-Institut für Wissenschaftsgeschichte. Picture by Andrew Stuart Reynolds