Ask almost any academic employee in South Africa what their greatest career fear is and the likely answer will be: “Not getting a journal article published this year.”
This research production, which earns universities money from the education department, is measured primarily by articles in accredited (in the so-called Sapse — South African Post-Secondary Education — list) journals, books, book chapters in scholarly publications and peer-reviewed published conference presentations. The managements of higher education institutions therefore conclude that boosting research “productivity” per employee is a quick way to increase income.
Employees who do not produce any articles get threatened with “performativity” interviews, loss of sabbaticals, no promotion and early retirement. They are regarded as unproductive and “dead wood”.
It is time for some counterarguments against practices that will in the long run harm the growth of knowledge. Let’s list some of the purposes of higher education:
There is a problematic tension between the teaching functions of the first two and the research/scholarship functions of the next two. Faced with huge first-year classes, and many “under-prepared” students, how can the harassed lecturer find time to do the research amid the script-marking and student interaction?
The problem is that the system defines research norms appropriate to the research function of a top university and stretches it across all disciplines and public tertiary institutions. The former technikons always valued a lecturer’s practical career in a technical field. Now these employees do PhDs rather than strengthen links in their field — links that in fact help to anchor their curriculum in the employability of their graduates.
Why can’t the higher education system acknowledge that some prefer a career in teaching rather than in research? By forcing all into a work metric more suitable for research scientists, some higher education institutions are in danger of demotivating their most experienced teachers.
The subsidy criterion of a journal article per employee annually favours natural sciences and engineering (NSE) and forces social sciences and humanities (SSH) into the NSE reward system.
Scientists approach knowledge convergently: they want to discover the truth and share it with their peers internationally so that experiments can be replicated. Scholars in SSH work divergently: they study in particular cultures, regions, languages and literatures and their peers are more specific and local.
This difference is easily measurable in the bibliometric counts that contribute to ranking. Bibliometrics counts the number of times an article is cited by other researchers. The Sapse journal list relies on the Thomson-Reuters Web of Science. The science disciplines predominate: 5 500 science journals, 1 800 in social science and 1 200 in arts.
Molecular biology has the largest number of journals at 261. Citation rates vary between disciplines, from an average rate of 14.6 per article publication in cell biology to 10 for art and architecture. This reinforces the impression that the bibliometrics of publication and citation counting suits the life sciences more than, say, design disciplines.
The Sapse additional list includes some 300 mostly local journals. A 2006 investigation by the Academy of Science of South Africa (Assaf) reported that 77% of Sapse social science articles and 90% of Sapse humanities articles are in the local Sapse journals rather than in international ones.
Assaf calculated that from 1981 to 2004 three disciplines produced more than 1 000 publications each: plant sciences (2 182), animal sciences (2 108) and environmental ecology. Some 16 000 South African researchers produced 7 000 papers — an average of 0.4 per person. These figures suggest that those in the NSE produce more than one paper annually and some in SSH do not see journal article writing as their main intellectual output.
Those alert to cultural hegemony might note that half the world’s research papers, as recorded in Web of Science, are produced by four countries: the United States, the United Kingdom, Germany and France.
One way to achieve more “productivity in SSH would be to improve the quality and quantity of African-produced journals. The bottom-level entry for publishing one’s work is to attend conferences, present a talk to others in the field and then to try to get that written up as a paper, reviewed and published. This process depends on having enough researchers and scholars willing to organise conferences, review papers, edit papers and edit and publish journals — all locally. But the South African research subsidy system rewards the article writers, not the conference organisers, the reviewers or the journal editors. For the next generation of local scholars and researchers, this must change.
But if some SSH academics do not value journals as much as those in NSE, what are they publishing? Economists are involved in “grey” literature — that is, current economic reports, often for local or national governments. Dutch scholars of literature reckon that up to 43% of their writing is to enlighten lay readership.
Outputs are different in different disciplines. A historian might spend 10 years producing one book, whereas a social scientist might be able to recycle survey facts rapidly in several short publications and so play to the reward system. Types of output being disincentivised are textbooks, public reports, fiction, newspaper articles, how-to manuals, and magazines for teachers or other practitioners.
Those reared in NSE career paths do not have any problem with excluding these types of texts from the reward system. To them the dividing line between “academic” and “popular” is clear. But many in the SSH are naturally inclined to work up and down the continuum of texts for different purposes and readership. However, the South African system chooses not to reward those who publish for the public, thereby denying one of the functions of higher education — knowledge dissemination.
Here are two examples of worthwhile publications excluded from the current system’s priorities:
If one considers the possibilities beyond textual output, what is excluded is even worse. What about the three-dimensional output of the design disciplines? What about the curating of exhibitions? And the staging of plays? And the performance of concerts? Some universities have set up a category of “creative” output, but the metrics are hazy.
Finally, isn’t it time that the metrics shifted from textual output to digital output? We are on the cusp of a technological shift, from paper to the web, which is comparable in scale of impact to the setting up of printing presses 500 years ago. Digital technology allows superior visual display and audio output. Fast linking across URLs means knowledge is no longer chained to the sequencing of characters on the printed page.
Digitalisation also means that universities will be better able to collect and preserve knowledge resources for public use, whether text archives such as the Timbuktu manuscripts or the Bleek collection, or pictorial collections of San art, or music collections of early jazz.
Digitalisation that consists of turning hard-copy journals into e-journals is a welcome first step. The bigger advance is in seeing that knowledge production is now more like a wave on the web. Production of websites, either personal ones or departmental ones, might advance and disseminate knowledge better than umpteen journal articles.
Academic purists fear the openness of the web — “too much unverified junk circulating”, “where is the peer review?” But a new type of quality measurement is emerging for digitalised knowledge: webmetrics, which measures the web-links of sites, and Google Scholar. A Spanish team even uses webmetrics to rank universities and has come up with a more diverse list than the “Shanghai” rankings, tied to the Thomson-Reuters bibliometrics.
The competing forces within the economics of knowledge production are, on the one side, those who commercialise knowledge through intellectual property, copyright and subscription barriers. On the other, there are those who want open source, free e-journals, and higher education knowledge dissemination funded as a public good.
The commercialisers include the major international journal publishers, whose subscription rates exclude many African researchers. An alternative model would be to pay the universities from the fiscus to disseminate knowledge through their various websites and reward the academic knowledge producers according to webmetrics.
Meanwhile, can we persuade the education department that its categories for textual output in the higher education funding framework are looking a little old-fashioned? And that they are pernicious in pushing out of higher education some valuable types of knowledge production?
Charlotte Mbali retired from full-time employment in the Centre for Higher Education at the University of KwaZulu-Natal in December 2008. She was national secretary of the tertiary education union Ntesu from 2003 to 2008. This is a shortened version of her forthcoming article in the South African Journal of Higher Education, 2010 (5)