ACO*HUM report 1998 part 2.3

Financial agreement number: 26030-CP-2-97-1-NO-ETN

2.3 Evaluation of the work undertaken

Difficulties

The most important cost-effectiveness problem was that the large effort needed to write the final report, in particular the declaration of expenses (part 4), was beyond proportion to the grant. Since the amount of the grant is only a small fraction of the total project cost, it is perceived as unreasonable that a specific and costly accounting of the total project expenditure is required. Due to the large number of participants in the project activities, the effort needed to write up the declaration of expenses was complicated and consumed a large portion of the project efforts, thereby significantly reducing the project's cost-effectiveness.

The main organizational difficulty was the large effort necessary to obtain additional funding for the conference. Due to the rather limited proportion of funding obtained from the EC and the coordinating institution (University of Bergen), additional funding was sought at an early stage. A sufficient level of financial support was secured only in the spring of 1998, which meant that the conference had to be postponed and the available time for definitive planning was extremely short.

Achievements, elements of good practice, side-effects, discoveries

Many elements of good practice were discovered and identified for presentation at the conference. We refer to the conference abstracts (included) for an overview.

The most surprising achievement was the success of the conference in terms of depth and breadth of the themes discussed. There does not seem to be any other forum for discussing the important and wide-ranging effects of the digitization of the humanities. A Europe-wide discussion on humanities in the digital age has been provoked and will certainly manifest itself further in future cooperation programmes.

The activities of the current project increasingly reveal the importance of reflexion on the nature of humanities and its place in the educational system. Looking at the high student numbers, one has the impression that the humanities are in good shape. On the other hand, the humanities are also in a crisis. The advent of the digital age necessitates new alliances between disciplines (see below) and thorough overhauls of traditional disciplines. Based on experiences in this project, it seems that the humanities are perhaps the area where the limitations of the traditional university system are manifesting themselves most urgently.

An important discovery in the working group on Computational linguistics and language engineering is that its cooperation with another TNP, that on Speech Communication Sciences, is continually getting closer, somewhat at the expense of cooperation within the other working groups in the present TNP. This seems to indicate the need to reflect on TNP structuring. On the one hand, linguistics is often situated in Humanities faculties (Arts faculties), at least in Europe. On the other hand, computational linguistics is situated in different faculties in different countries, sometimes in Humanities faculties, sometimes in Computer Science, sometimes in Psychology. With speech communication sciences, computational linguistics is typically an important interdisciplinary field in need of an educational embedding which does not limit it to one faculty. Together, the present TNP and that on Speech Communication Sciences have organized a workshop on September 28, 1998, on Perspectives for international cooperation in language and speech.

Divergence from the original conception of the project

Aims and objectives. There were no divergences with respect to the adjusted aims and objectives as stated in the previous report (1996-1997). The intended extensions, consisting of adding working groups on Formal methods in the humanities and Computing for non-European languages, were carried through. However, the time scale and resources of the project were too limited to open up the project at this time to further areas within the humanities, including languages, archaeology, philosophy and logic. Some of these areas were addressed at the conference, but no working groups were established for them.
Workplan and calendar. There were no divergences with respect to the list of planned activities in section B.1 of the renewal application for 1997-1998. There were, however, divergences with respect to the workplan in section C.2 of the renewal application for 1997-1998. The Policy Symposium was held in November 1997 instead of October. The Conference was postponed from May to September 1998, but falls within the planned activities of the academic year covered by this report (1997-1998). Due to this change, the planned assessment study was also postponed, until after the conference, and is not completed at the time of submitting this final report. The planned drafting of the tentative common ground document was not effectuated because it was judged premature by the Steering Committee.
Partnership. New partners have joined the network for the new working group on Computing for non-European languages.
External environment. No divergence.

Additional feedback

It is felt that this TNP is acting like a catalyst in the field, such that our attention becomes focused on important issues for humanities in the next century. However, the TNP typically points to more problems than it can bring solutions. If the recommendations produced by the TNP are not taken up by further efforts within cooperation programmes, the work will have been for nothing. We expect therefore that a more direct cooperation with other actions could prove fruitful.

Finally, we point out the need for future structures to give natural language and speech sciences their own forum for inernational educational cooperation. A joint venture of speech communication sciences with computational linguistics would be a formidable starting point for future work in Natural language and speech sciences, where both written and spoken language communication are studied. The need for cooperation in this field is witnessed by the importance of the area in research and education worldwide. Human-computer communication is becoming increasingly sophisticated in the extent to which machines can handle written and spoken language. Europe is currently a leader in the field of natural language and speech processing. However, a recent conversation with Jo Lernout (Lernout & Hauspie) revealed that the scarcity of trained language engineers is a threat to long-range sustained development.

In order to face challenges like this and to stimulate new alliances, it is important to think in a creative way and not let future cooperative structures and measures be grounded in traditional domains or according to faculty divisions. This is perceived to be of strategic importance for future thematic network projects.