Within the current debate on the evaluation of research in Social Sciences and Humanities (SSH) a recent discussion has been opened on the need for using multiple languages.
The argument is that SSH use different communication channels than STEM disciplines. While researchers in STEM almost exclusively use international, peer-reviewed scientific journals in English language, scholars in SSH publish books in national languages. In addition, scholars in SSH need to interact more closely with local audiences that do not necessarily understand English language.
Following this argument, the criteria for research assessment should not penalize publications in national languages, but adopt a broad and comprehensive attitude.
I believe this argument is well-grounded and should be recommended as an international guideline for research assessment agencies and national ministries.
At the same time, I feel there is a need to establish better foundations for the principle of multilingualism. The main arguments for defending multilingualism are developed with a focus on science communication: the need for multiple languages is defended from the side of the listeners, that is, the readers of books, or users of knowledge from SSH.
I will restrict my analysis to Humanities, while I have discussed at lenght epistemic issues in Social Sciences in a recent contribution.
I believe we can make a better work in developing a more comprehensive framework, which focuses more explicitly the epistemic issue- or the conditions of scientific validity of the knowledge produced by scholars.
As a starting point, I argue that we need a deeper work on the epistemology of Humanities. Two observations here. On the one hand, in the XX century the philosophical analysis of scientific knowledge has almost esclusively focused on hard or natural sciences, primarily on physics (from the Wiener Circle to Popper, down to Lakatos, Kuhn and Feyerabend), only recently on biology (Mayr, Dupré), neurosciences or computer science (Clark, Fodor and others). Contributions in philosophy of history and of historiography, or philosophy of art, or philosophy of literature have not taken central place. Just give a look at the Stanford Encyclopedia of Philosophy, or recent surveys on the topics pursued by professional philosophers, and you will find confirmation to this point. A large share of scholarly research, that is, the research that takes place in Humanities, is left outside the mainstream epistemological reflection of the XX century. On the other hand, Humanities do have their own epistemology, but these epistemologies are regional and idiosincratic: we see a debate between philologists and anti-philologists, or between analytic and continental philosophers but there is no effort to build up a comparative epistemology of Humanities. My point is that we definitely need a comparative epistemology, upon which to establish some general principles of validity of knowledge produced by research in Humanities.
If we were to attempt such an enterprise, we would discover some of the following points.
First, the object of Humanities is intrinsically historical, that is, rooted in time. Given that time, at the human scale, is irreversible, the objects of Humanities cannot be reproduced experimentally. Scientific validity cannot be established on the basis of criteria that are logically impossible to achieve. Notions of causality in Humanities must be different than those we can demand to laboratory sciences (although even in some of these fields, in some cases, history matters in complicated ways).
Second, Humanities work on natural languages, because these are the ways if expression of humans in historical time. Natural languages are intrinsically, as Paul Ricoeur has shown, multiple meaning systems. In all known languages words are polysemic. Interpretation is the effort to infer a univocal message coming from the speaker from the intrinsic polisemy of words she is using. This interpretation takes necessarily place in a context, at the minimum between the speaker and the listener. The pluralism of interpretations, therefore, is not the result of a relativistic or nihilistic attitude (in the sense of “anything goes”) but a necessary consequence of the object of study. This does not mean that knowledge in Humanities cannot meet the standards of scientific rigour, as the neopositivism has argued (see for example Hempel).
What is the method for establishing validity of knowledge in Humanities? It is the endless quest for comparing any single object, a unique object, to the whole of existing relevant knowledge, both synchronically and diachronically. Good researchers in Humanities undertake an endless search for historically rooted documents to illuminate the meaning of the object they study (artwork, image, text). According to a famous statement of Erwin Panofsky, scholars in Humanities derive the meaning of their objects from “the largest possible number of other cultural documents”.
How is this “largest possible number” determined? Here we cannot rely on any closure assumption, as we can do in hard sciences. No statistical reasoning can be applied. But scholars in Humanities are all engaged into this endless inquiry. They form an agreement on whether the collection of “other cultural documents” used by an author is adequate or not to interpret the object of study. Disagreement is not an indicator of the lack of scientific method. Quite the contrary is true: disagreement is a sign of healthy scientific work. Disagreement among scholars in Humanities is driven by the need to enlarge or reinterpret the collection of “other cultural documents”. Scholars agree asymptotically and provisionally during the endless search. Their agreement we call historical truth. Its scientific validity is ensured by the same kind of inter-subjective agreement which is found in hard sciences.
Third, the objects of Humanities have invariably a materiality which establishes a deep relation with the natural world and its laws. Texts are written in manuscripts and drawings are designed on tables, while kings and paysans use material resources for their life. Materiality can be studied with hard sciences, because its is subject to laws of invariance which are not found at the time scale of humans. Humanities make extensive use of that kind of scientific knowledge (the hard science) for which the historical dimension is negligible, or less relevant, because the phenomena follow a longer time scale.
From this argument I suggest that we should not drastically separate Humanities from Natural sciences, as the idealistic tradition of XIX century has done, from Dilthey to Windelband, down, to a certain extent, to the radicalization of the distinction between explanation and understanding in Max Weber. Humanities must read and make intense use of hard sciences as part of their explanatory strategy. I consider the scientific exchange between hard sciences and Humanities as one of the most promising direction (think for example to the use of chemistry or materials science in archeology).
From these general remarks, that we should certainly develop further, we derive the following implications about the issue of multilingualism.
1. For scholars in Humanities the need to speak and understand several languages is intrinsic in their scientific methodology. It is not a problem of communication, it is an epistemic necessity. Object of Humanities are embedded into languages. Their translation into other languages, or into English as “lingua franca” is usually possible but requires itself a scientific work.
2. The use of national languages is fully justified when the “cultural documents” needed for research can be interpreted better, or even exclusively, in a specific language. The relevance of contributions in national languages for the largest scientific communities in Humanities should be evaluated in terms of relevance, generality, and rigour, which are not necessarily reduced by the use of a language of narrower circulation.
3. Scientific journals should accept manuscripts in several languages. Ex ante peer review should become the standard for all scientific journals. Scholars in Humanities should publish, whenever possible, in several languages.
4. Given the relevance of multilingualism and the intensive use of books, peer review is the single most important research evaluation methodology. Peer review should be continuously improved by means of exchange of practices, training, intellectual debate. Referees should be deeply embedded into the language in which contributions are written.
5. Humanities are under-represented in the policy debate and in the allocation of research funding. To argue in favor of Humanities we should not imitate trivial arguments about the purported social or economic impact of research. Rather we should engage our minds into the construction of a comparative epistemology of Humanities, as a first step towards a full scale epistemological framework. We no longer need arguments about the instrumental value of Humanities, for example for cultural heritage. We need arguments about why knowledge produced by Humanities is scientifically valid. And necessary.
6. As a necessary condition for fully representing research in Humanities it is mandatory to continue the effort to integrate national datasets and bibliographic archives, in order to achieve a Europe-level, accessible, fully digital and interoperable repository of metadata including publications in all national languages. In this direction the pioneering work of ENRESSH should be recognized and supported.