Ewekniebeoordeling vertoon fundamentele gebreke, soos wisselende oordele en 'n onvermoë om werklik nuwe kennis te waardeer. Pogings om ewekniebeoordeling te verbeter skep net verdere probleme. As oorsaak word kruipende subjektiwiteit aangevoer en dit dui op onaanvaarbare aannames oor werklikheid, mens, kennis en wetenskap. Hierdie artikel is 'n poging om 'n nuwe benadering voor te stel. Kollektiewe intelligensie is die vertrekpunt. Nie alleen is 'n groep slimmer as 'n individu nie, maar so kom verskillende benaderings, kriteria, prioriteite en so meer in spel en onder die vergrootglas. Terselfdertyd plaas dit wetenskaplike navorsing weer midde-in die wetenskapsgesprek en word navorsing werklik kennisvermeerdering. Beoordeling vereis veel meer as blote vakkennis. Navorsing het as doel om deur kennisvermeerdering 'n bydrae tot die wetenskap te lewer. Die beoordeling van navorsing vra dan ook kennis en vaardighede op hierdie twee terreine, dit wil sê kennis van kennis en van wetenskap. Omdat beoordeling nie 'n aktiewe deel van vakopleiding uitmaak nie, behoort beoordelaars opgelei te word om te kan beoordeel. Bekendheid met die aard en vereistes van beoordeling word bloot veronderstel, met die gevolg dat allerlei onhoudbare idees hieromtrent taakvervulling bemoeilik. Konseptuele reoriëntasie oor "beoordeling" is nodig en insig in belangrike aspekte daarvan, soos die aard van kriteria, die onderskeid tussen feite en waardes, en die struktuur van waarde-argumente, is nodig en vereis opleiding. Ons bespreek die belangrikheid van gesprek as 'n deel van beoordeling, en van beoordeling as 'n bron en vertrekpunt van gesprek, in aansluiting by die logiese band tussen wetenskapsbeoefening en gesprekvoering. Ons beskou wetenskapsbeoefening as deelname aan 'n deurlopende en nimmereindigende gesprek. Beoordeling behels verder byna altyd die lees van 'n geskrewe navorsingsverslag. Dit beteken die leeshandeling beklee 'n sentrale plek in beoordeling. Die hervorming van beoordeling is dus afhanklik van die verbetering van leeshoudings en -style. Lineêre lees moet vervang word deur inventiewe lees, sodat beoordelaars 'n deel van die navorsingsproses word en 'n positiewe en konstruktiewe bydrae tot die navorsingsprojek kan lewer.
This is the fourth article in a series of four under the general theme, "Towards quality science or towards the best possible science" in the same journal, LitNet Akademies (for the previous articles see 7(2):129-64; 7(3):323-81; 8(3):242-78]. The content of this article is the culmination of the deliberations and reflections on different aspects of this general theme, which include: the thorough studying of the literature on, and the questioning of, the current processes of peer evaluation; the serious critical consideration of the place and role, or lack of role, and even the severe incapacity of, scientific management to promote what is claimed to be promoted, namely scientific excellence; and a strong emphasis on the conditions to be complied with if we are serious about our commitment to the best possible science.
Peer evaluation suffers from systemic deficiencies such as a lack of consistency in evaluations, an inability to appreciate new knowledge when it is presented for evaluation, and even more seriously the inability to attend to and deal with paradigmatic deviations from the stereotyped approaches. The influence of subjective factors is blamed for such lapses. The claim that mere sound management of science holds the key to science's salvation is considered to be unfounded, counterproductive and even a bit far-fetched. Therefore suggestions along these lines to improve the system by simply improving evaluation procedures and managerial strategies prove to be unsuccessful. This will remain the case unless the current, but unacceptable and inadequate, assumptions about reality, human nature, knowledge and science are incorporated in a drastically revised form in order to work towards the radical and comprehensive revision of peer reviewing. This remains, for us, the non-negotiable condition for any kind of progress towards quality science.
This article is an attempt, against the background of the previous three articles, to present a new approach to evaluating research. Collective intelligence provides the starting point. Similar to a series of linked networks, a group is cleverer than an individual, and also incorporates different approaches, different criteria, different priorities, a variety of equally important but different insights, and the opportunity to take a critical look at these insights. The initiative should be to encourage and inspire sharing and connectivity that will eventually lead to the constitution of intelligent communities. Such an approach also puts scientific research squarely in the discursive domain and ensures its acceptance as an advancement of knowledge. In fact, if this aspect of science is ignored, as is currently the case to a great extent, nothing is left but techniques and skills. At the same time, however, this will drastically reduce the ability of science to participate effectively in the desperate efforts worldwide to solve the immense problems facing different societies, but also mankind in general.
