Thursday, December 23, 2010

How to evaluate within the humanities?

The evaluation of projects is a crucial topic today. In fact, (nearly) all paradigms of evaluation have been developed having in view natural sciences. Hence, in order to avoid the unreflected application of this kind of criteria as if they were the only viable options, scholars interested in the humanities should (I believe) try to propose other criteria.
The following ones are the criteria I suggested in order to evaluate the degree of success of a project involving the critical edition, translation and study of the fifth chapter of Jayanta Bhaṭṭa's Nyāyamañjarī:

• Increase of interest among scholars for Indian linguistics and philosophy in general and the NM in particular.

• Publication of the research's results on international and peer-reviewed journals. The journals involved should a wide dissemination of the research's results, to a European, US, Asian and Japanese audience, focusing mainly on South Asia, or on Asia in general, on philosophy and philology. In order to improve the dissemination of the research's results, particular attention should be given to the open access of articles and research reports.

• Dissemination of the research's results through the research's website and partnerships with related websites, such as TEI: Text Encoding Initiative, Perseus, TITUS, GRETIL, SARIT (the present team is already in touch with some of them). As a first presentation of the project, a page should be created also on Academia (

• Dissemination of the research's results and their usage also by scholars external to the present group. In order to evaluate it, I suggest the following criteria: (i) Access to the research's website and its achievement of a leading role among scholars interested in the NM, in (Indian) linguistics, philosophy of language and epistemological investigation on language as instrument of knowledge. (ii) Use of materials elaborated or collated by the present team in further articles, books, research projects by other scholars. (iii) Most importantly, the edition and translation produced should become the standard reference works for scholars working on the NM.

• The research's team becoming the centre of attraction for international research on the NM, (Indian) linguistics, philosophy of language and epistemology, as testified also by the success of the international seminars organised.

• The adequacy of the translation and the quality of the explanatory glosses and commentaries of the NM could be verified also through the degree of usability of the output edition for scholars not provided with a special training in Sanskrit and in Indian philosophy. In order to test it, one should plan to organize some presentations of our results within international conferences devoted to linguistics, philosophy of language and epistemology of language (i.e, on the role of testimony and of language in general as an instrument of knowledge). One should be able to explain the importance of the NM for contemporary disciplines, and to stimulate the interest and the curiosity of scholars working in different fields of humanities.

• Particularly, no Sanskrit philosophical term must remain without a translational equivalent in modern terminology.

• Furthermore, since the final outcome of our project must be an internet site presenting the electronic edition of NM 5, the usual criteria for web sites apply here, that is:
- clearness and completeness in covering the declared scope;
- accuracy of the presentation and authority of the sources;
- possibility to obtain the output edition in other formats than HTML (such as PDF file, CD-ROM, etc);
- feedback with users;
- speed and responsiveness of the user interface;
- searchability of the resource.

• More technically, the outcome web resource must be compliant with the standards involved, such as the XML standard elaborated by the Text Encoding Initiative for cases similar to ours (with special variants intended for the encoding of manuscripts and for morphological tagging of texts), HTML5, Unicode Collation Algorithm, and SQL for database queries. In addition to these, some further formal standards may be taken into consideration in case our project is accepted as a part of a greater project, such as the TITUS electronic database of Indo-European texts (University of Frankfurt, Germany).

Do readers have other suggestions, particularly in different connections?


Dominik Wujastyk said...

It's notable, Elisa, that your list of criteria doesn't include any of the measures that big funding bodies have developed in order to try to measure the importance of research, especially those techniques going under the name "Impact Factors" ( Attempting to objectify measures of importance has led to practices such as counting references in research literature to a particular paper, etc. You may be right to exclude such techniques, but several of the methods you mention would be very hard to assess quantitatively.

elisa freschi said...

Dear Dominik, thanks for this interesting comment and sorry for the long delay in answering.
1. You are right, I am suspicious about calculating the impact factor of studies on Indian philosophy, since the total number of people interested in them is rather low and, hence, highly liable to be influenced by, e.g., the fact that an author belongs to one's own 'school' rather than by her value alone. How do you think this could be avoided, given the small number of Sanskrit scholars? Obviously enough, I am not referring to unquestioned geniuses, who would be cited anyway just because they deserve it.

2. I agree, the methods I mention are hard to be assessed quantitatively. But I think that the specific character of the humanities lies exactly in their ability of taking into account the qualitative aspects of reality, the ones natural sciences are compelled to ignore. Humanities are bound to be 'inexact' and should not (in my humble opinion) try to imitate exact sciences. On the contrary, they should develop their esprit de finesse. As for funding, I do not think that quality in humanistic studies cannot be assessed, although I do think that it has to be assessed qualitatively.

Licenza Creative Commons
Quest' opera è distribuita con licenza Creative Commons Attribuzione - Non commerciale - Non opere derivate 2.5 Italia.