African university presses and the institutional logic of the knowledge commons

This article investigates the current status and challenges faced by university presses in Africa, looking particularly at the institutional perspective. Four case studies from Ethiopia, Kenya, and South Africa show how different presses adapt their practices and adopt new technologies. Interpreted through an institutional logics perspective, the status of the university presses is described according to established editorial and market logics, to which a third, hypothetical logic of the knowledge commons is added. The logic of the knowledge commons accounts for changes advanced by the digitization of content, peer‐to‐peer networks as the basis for production, the rise of open access, and an emerging social capitalism. In two cases, we find university presses constrained by traditional editorial logics, while a third one exhibits a hybrid editorial–market model with the purposive adoption of new technologies. Only the fourth, recently established press has embraced the new logic of the knowledge commons wholeheartedly. Thus, if there is a second transition of the academic publishing industry underway, it is in its early stages, partial, and limited in the African context. We thus show that the logic of the knowledge commons provides a useful theoretical lens for studying the far‐reaching and rapid ongoing changes in international academic publishing in Africa and further afield.


INTRODUCTION
In 2016, Zell (2016) pointed out the paucity of empirical evidence on the state of academic publishing in Africa at a time when a better understanding of the African landscape of university presses was particularly important. On the one hand, higher education is currently expanding faster in Africa than it has in any other world region at any time in history (Luescher, 2017;UNESCO, 2015). This rapid expansion is accompanied by an increasing differentiation among universities (Blom, Lan, & Adil, 2016), which is also signalled in the recent establishment of the African Research University Alliance (McGregor, 2015). On the other hand, globally significant technological innovations of the last decade provide new tools for university presses to propel themselves into international markets, thus making the most of current market conditions. According to Dougherty (2012), these technological innovations include supranational online book retailers such as Amazon; print-on-demand; online publicity exploiting Web 2.0 platforms and services; and library aggregation services, including the increase in bulk subscriptions to digital content.
In a quest to better understand the landscape and state of university presses in Africa, in 2016, we initiated a research project asking a number of interrelated questions: What is the composition of the university press landscape in Africa? How are university presses faring under the current market conditions?
What do university presses make of the challenges and opportunities presented in the academic publishing realm within their respective contexts? Are they deploying the technological changes in production, distribution, and marketing made possible by digitization and the network effects of the internet? (Van Schalkwyk & Luescher, 2017).
In the process of our studies, we produced a database of the landscape of university presses in Africa, in which we mapped and profiled 52 university presses across the continent (Van Schalkwyk, Rasoanampoizina, & Warren, 2016). From this database, we identified a group of 15 active university presses, from which 4 were selected for in-depth investigation. These case studies are meant to provide insights into the diversity of ways that very different university presses in Africa deal with the changes and challenges encountered within their respective local contexts and the new developments in academic publishing that can be observed internationally and that offer new potentials for university presses in Africa to contribute to the global knowledge project.
Institutional logics and social capitalism: Towards a logic of the knowledge commons?
One way of conceptualizing the impact of historical technological changes on the academic publishing industry is the institutional logics perspective (Thornton, Ocasio, & Lounsbury, 2012). Thornton and Ocasio (1999), in their sector-specific study of the key characteristics of the academic publishing industry in the USA, provide evidence of a shift from a professional or editorial logic to a market logic in academic publishing between the late 1950s and the early 1990s. Empirically, this shift involved changes in key practices implicating the typical characteristics of academic publishing, such as the organizational identity of the publisher, its internal organization, governance, mission, strategy, and so forth (Thornton & Ocasio, 1999, p. 809; see also Friedland, 2012). At the end of the transition from the editorial logic of operation to the market logic, publishing had become a different world: the shift inserted academic publishing into the system of market capitalism, turned it from a profession into a business where market position and share value became the dominant sources of legitimacy, and its collective mission became one of building a competitive market position for the firm by accumulating surplus in a context of resource competition. The CEO as head of the corporate hierarchy had replaced the founder-editor and his or, more rarely, her personal networks as the source of authority. Ownership by shareholders replaced private and often family-based ownership. In the end, academic publishing had transformed into another industry driven by the logic of the capitalist economy.
Looking at more recent developments in academic publishing, it is evident that new technological changes have brought to the fore new opportunities and challenges (Taubert & Weingart, 2017). Along with recent technological innovations in Information and Communication Technologies and those specific to the publishing industry, such as print-on-demand, is the growing knowledge commons and associated shift towards open access publishing. Furthermore, networks in which everyone is connected to everyone are replacing hierarchical arrangements as the structural basis for social organization (Castells, 2009;Stalder, 2006). Considering the economics of open-source software development, crowdsourcing, and so forth, Shaughnessy (2012) suggests that we are seeing the emergence of a new kind of economics, which he calls 'social capitalism': This social capitalism is driven by collaboration and sharing, by the replacement of command and control by self-organising ecosystems of independent producers of software, apps and even components.
The new economics involves inter alia a shift in the market from 'what to consume to what to make' (Shaughnessy, 2012). There is already evidence in the ICT sector in Africa of the emergence of social enterprisesbusiness-like ventures motivated more by social good than by profit (Mejabi & Walker, 2016). Globally, amidst the challenges, new university presses are being established, often departing from traditional publishing models and pushing the boundaries of open access publishing, so much so that some of these university presses have been labelled 'radical' (Adema & Moore, 2018).
It therefore seems reasonable to ask whether the industry is in the midst of a second transition from a market capitalist institutional logic to one that accounts for the changes precipitated by the digitization of content, the ubiquity of the internet, networks as the basis for production and social relationships, and the rise of economies of sharing and social capitalism?
In keeping with these conceptions, we look at the academic publishing industry in Africa through the lens of institutional logics and add to Thornton and Ocasio's two historical institutional logics a third, hypothetical one. This 'logic of the

Key points
• African university presses are constrained by their institutional support and outlook.
• Younger, emerging, African university presses are more able to adopt the logic of the knowledge commons rather than presses that follow the older editorial or market models.
• African university presses are well aware of opportunities afforded by new technologies but are not making full use of these opportunities.
• Technological opportunities are mainly understood in terms of creating marketing and distribution channels complementary to the existing print-based model focused on local markets. knowledge commons' expands our view to be able to consider whether the university presses in Africa are affected by an emerging social capitalism and are likely to become part of the sharing economy: a technologically enabled synthesis of capitalist and socialist economics, which expresses itself in the development of knowledge commons; one that views publishing as a collective social innovation enterprise rather than a business; one that organizes in self-organizing peer-to-peer networks, who invest their cultural capital in the global open access knowledge commons with the aim of collectively producing shared knowledge and value, thereby 'socializing' problems and solutions.
In our view, open access is one of the indicators of the new sharing economy and the emerging shift to a new institutional logic of academic publishing; we therefore propose that the transition to the logic of the knowledge commons is driven specifically by key developments in distribution. Our characterization assumes that factors that contributed to the erstwhile decline of an editorial logic and rise of market logic during the earlier transition (Thornton & Ocasio, 1999, p. 810) are once more facing contradictions that are 'resolved' by means of new practices and norms that indicate a knowledge commons-oriented institutional logic. In the broader perspective, this reflects an industry-specific expression of an emerging social capitalism. Table 1 characterizes the three logics, including the original institutional logics elaborated by Thornton and Ocasio (1999) and our characterization of an emerging logic of the knowledge commons.

Design and methods
To determine the ways in which African university presses are positioning themselves in relation to various local and technological changes and opportunities using the institutional logics framework, empirical research was needed. The diversity of contexts and conditions across the continent, along with the sparsity of existing data, invites a comparative case study design as a useful approach to investigate the institutional logics. Furthermore, to maximize the opportunity to learn (Stake, 1995(Stake, , 2000, we applied a most dissimilar case design. Thus, from the database of Each case includes a basic profile that draws principally on three sources: the landscape database developed for the purpose of the wider project; interviews with the heads of the university presses and, in some cases, other staff; and primary documents from the university press, particularly their websites, annual reports, budgets, and catalogues. While the database includes basic information about each press, the interviews provide indepth, context-specific knowledge, as well as an opportunity to verify what was captured in the database. We specifically focused on four empirical dimensions of practice and innovation: marketing and promotion, production, distribution, and revenue models.
We conducted all interviews with due regard to the ethics of social research. All participants provided fully informed voluntary consent for participation, waived all offers for anonymity and confidentiality, and had an opportunity to comment on the draft case study report (see Van Schalkwyk & Luescher, 2017 Most important in the appointment of head of the university press is someone who can help the press continue to make contact with publishers outside, authors outside and in the university, and someone who is interested to help the university press to go forward.
[…] I don't have special editorial, managerial experience and this was also the case with previous appointments. It is taken to the senior faculty to do the job. (Y. Arsano, AAUP, interview, August

5, 2016)
Thus, the key criterion in appointing the head of the press is academic seniority, academic management experience, and access to networks to attract authors and to establish links with other publishers. The part-time Director has a modest staff, consisting of three full-time editorial staff as well as administrative staff who ensure the day-to-day operation of the press. The press is fully funded by AAU and operates with a budget provided by the university (which, in turn, is financed primarily by the Ethiopian fiscus). AAUP is not expected to make a profit; indeed, it is not even expected to cover its costs and has only limited control over costs and income. Rather, the mandate or 'deal' is a different one as the Director explicates: The price of our books is much lower than other publishers in town. The reason is, it's not for making profit. It's possible to make money but that money goes into the university income system. It circulates. We are not required to sell books to the extent of covering our costs. That is not the deal. The deal is we publish as many quality books as possible [...] and make it possible for the public to read the books, and if there is a deficit of costs and budget then that is where the university financial system comes in. (Y. Arsano, AAUP, interview, August 5, 2016) For AAUP, the biggest budget item is printing, which is outsourced. The AAU printing services (which fall under the university's general management) as well as printing companies from outside the university submit tenders for the printing work.
All payments are authorized by the AAUP Director but paid through the university finance system.
In its 50 years of existence, the AAUP has accumulated an impressive catalogue and backlist comprising about 100 titles. According to the AAUP Director, the plan is to accelerate the production of new publications 'at a fast, commercial pace' (Y. Arsano, AAUP, interview, August 5, 2016). However, there are a number of challenges in this regard, mostly organizational shortcomings and a high turnover of qualified staff who can secure better salaries in the private sector than as civil servants on the government salary system (which covers non-academic staff at AAU). To address the problem of staff turnover, Arsano plans for editorial staff to be appointed as academic staff (and thus employed on the academic staff track within faculties), thereby allowing the editorial staff of the press to benefit from the relatively higher remuneration, social benefits, and other privileges associated with an academic post.
The AAUP follows a very traditional marketing and distribution model. To market its books, the press participates in book fairs in Addis Ababa (there were four in 2016), and it holds its own book fair on the university's main campus. In addition, AAUP Source: Columns 1-3: Thornton and Ocasio (1999, p. 809) with additions and adaptations by authors; Column 4: developed by the authors. ICTs, Information and Communication Technologies.
prints copies of new book covers for distribution as promotional material across the city. Digitally, the presence of AAUP is limited. Its website is easy to find as a subsection of the university website and includes listings of published and forthcoming books.
But this is the only web presence of the AAUP; it has neither Facebook nor Twitter accounts. It is also not possible to buy books from the AAUP website. While there have been some efforts by the press to include digital formats of its publication and to leverage digital marketing and distribution, none of these efforts have come to fruition.
Overall, the press then has a limited, mostly Addis Ababa-based, distribution network for its books. There are 12 booksellers in Addis, who sell their books on consignment. There are also buyers who come directly to the AAUP offices on the AAU campus to buy books. Apart from book retailers, book fairs, and in-house sales, the AAUP does not have any national, Africa-continental, or international distribution channels in place. The international distribution agreement with the African Books Collective is inactive. The copublishing arrangements that AAUP has entered into for selected titles are for the sole purpose of making those titles affordable in the Ethiopian market. Thus, distribution beyond the local market is not currently high on the agenda of the press.
There are also no agreements with library aggregation services. The focus of the press is entirely on print publishing, and when a title is out of print and demand is evident, a reprint may be put out to tender. There is also no open access policy in place even though AAU has been a signatory on the Berlin Declara- The expectation is that, once capacity problems at the editorial staff level have been resolved, the press will be in a better position to concentrate on other matters, including digitization and e-publishing. The need for a full-time Director of the press was expresseda more operationally involved manager to ensure that what is articulated in senate legislation and in the strategic plans of the press are implemented and that the entire press runs more efficiently and with its own income and budget allocation. There are several criteria by which Wits UP measures its success: sales numbers (of which the most successful tend to be the less scholarly books), citations (Wits UP is the first university press in Africa to be indexed in the Web of Science Book Index), and reputation. Klipp sees the press as being in a very strong position currently, both financially and in terms of its reputation.
Wits UP has a digitally driven marketing strategy. It markets its publications by sending information about new titles by e-mail; the press has its own website, which is search-optimized for Goo- In this regard, author-based funding is seen by the press as a viable approach, but it is concerned that some publishers are charging high page fees.
While the press does not have to make a profit, it is meant to cover its own costs, including salaries. As a cost centre of the university, it presents its budget to the university; in some years, it manages without a subsidy, but in other years, it receives a subsidy. The press's income is mainly from sales; some titles are also supported by author-based funding. For prescribed school books, Wits UP has a licensing agreement in place with two educational publishers, which is an important source of revenue. The press has no grant income, although it did seek donor funding for its digitization strategy. Case 3: University of Nairobi Press, Nairobi, Kenya

Ms
The UONP was registered as a private company wholly owned by the University of Nairobi (UON) in 1984, but it only became fully operational in 1990. Its mandate is to serve as a publishing house for UON research outputs, that is, scholarly books of all disciplines as well as textbooks at secondary and tertiary levels and some biographies. It does not publish any journals. Currently, about 80% of its authors are scholars associated with the UON.
The press's multidisciplinary catalogue (which is available from its website) is somewhat skewed towards the humanities and social sciences. To date, UONP has over 50 scholarly titles in its backlist. The press negotiates its publishing programme on an annual basis with its host institution, typically to publish about six titles per annum. It has, at times, exceeded that number of publications.
UONP is described by its host university as being 'semiautonomous' (University of Nairobi, 2016). In reality, however, the UONP as a company is non-existent, and the press is fully integrated into the university's governance, management, planning, funding, and HR system. It is de facto treated as an administrative unit of the university, and UONP staff are staff members seconded to it by the UON and occasionally deployed to do work elsewhere within the university. The UONP company itself has no staff of its own, and its Board of Directors has been defunct for over a decade. This has created some insecurity in terms of governance and the strategic direction of the UONP.
As part of the UON's drive in the late 1990s/early 2000s to ensure the university's financial sustainability, the future of the university press was uncertain. The press did not fit the idea of an 'income-generating unit'. It was therefore a boon for the press when a commitment was made to subsidize the press based on the argument that the UONP is crucial to the realization of the academic core function of the university to disseminate and preserve knowledge and to promote the global reputation of the university. While Oxford University Press is mentioned as an 'aspirational peer', this is not seen as realistic by the UONP staff.
Indeed, the UONP Head sees it as an achievement that the press has survived to date when there is no university press in Kenya that is able to operate as it does, and other major university presses in the region, such as the presses of Makerere and Dar es Salaam universities, have ceased operations.
At the time of research, the Editorial Board was not meeting, and its function was executed by its chair, that is the university's DVC: Research, Production and Extension. Notwithstanding, all scholarly publications are peer reviewed. In addition, manuscripts are entered through plagiarism detection software. Editors (staff ) can also recommend the rejection of a manuscript. While a number of manuscripts are rejected every year, the press cannot publish all manuscripts that are accepted due to resource constraints.
The UONP has nine staff members seconded to it by the university. They comprise two editors as well as the managing editor/head of the press, one designer, a marketing manager, as well as an accountant, a secretary, a driver, and a messenger. Especially at the level of design and editing, the Head of the Press, Mr Josphat Kirimania, feels that the press is strained for resources and would benefit from another four to six staff members. In his view, the expansion of the press is circumscribed mainly by funding and human resources.
As for its finances, the press is, in principle, expected to cover its own costs, and there are pressures for it to commercialize, often with reference to international presses like Oxford University Press or Cambridge University Press. Currently, the UON pays for the staff, offices, and facilities of the press, as well as providing a small operational budget. The press generates revenue from mainly three sources: sales of books, author contributions (which are not compulsory), and partnerships. The latter are agreements with organizational partners for which the UONP produces publications on request. UONP currently receives no external/donor grants.
The marketing and distribution network of the UNOP has, at its core, sales by campus-based bookshops and national distributors in Kenya. UNOP also handles direct enquiries and orders itself or through the Nairobi campus bookshop. UNOP markets its books through its catalogue, the website, by means of exhibitions, and by In terms of digitally driven publicity, the UNOP has a working website, and digitally driven marketing has been piloted in the past with a blog and a Twitter account, but both were not sus- The dilemma for the head of the press is that, on the one hand, the UON library would like the press to go fully open access, but on the other hand, authors say that 'they need something out of it, otherwise I am not giving you my manuscript' (J. Kirimania, UONP, interview, May 17, 2016). Thus, while in general terms, the UNOP sees the opportunities of technological innovations in a positive light and has certainly been able to experiment and adapt to some changes, it appears constrained by key beliefs and existing practices to make effective use of them.
UONP measures its success in reputational terms linked to the quality and relevance of its publications. While in terms of sales, the textbooks and manuals that UONP publishes are the most successful, as a university press, it discounts sales in favour of the scholarly 'seriousness' of a publication. Visibility and the extent to which a book is referred in public discourse, citation counts, and 'hits' on the websites, enquiries, etc. are all mentioned, but there is currently no firm measure in place to account for success in these quantitative terms. The university management, in turn, has a different perspective on what success is: 'The first thing that they look at is not how many books have they done; they look at how much you have generated, and then we start a big debate and a big fight there' (J. Kirimania, UONP, interview, May 17, 2016). The head of the press, therefore, feels that he constantly has to defend the performance of the press. and it established a press to support its scholarly functions in this ambitious goal. WUP was the initiative of the first president of WU, and its establishment and first years of operation have been coordinated by the current Editor-in-Chief. WUP aims to publish monographs, academic journals, textbooks, and reference books across all disciplines and for all levels of education, for pupils, students, teachers, and academics, for policymakers as well as general readers. It seeks to reach an Ethiopian and global audience with its publications.

Case 4: Wollega University Press, Nekemte, Ethiopia
Given its recent establishment, by mid-2016, WUP had only published the conference proceedings of the university's annual national research symposium as well as five volumes of its flagship Science, Technology and Arts Journal (STAR Journal). The STAR Journal has been able to attract authors from across Africa (25% papers from WU academics; 75% from outside WU and abroad) and established itself as a widely indexed, multidisciplinary, quality, open access journal. Three more journals are being planning, respectively specializing in agriculture and natural resources; medicine and health sciences; and Afaan-Oromo (i.e. the local language). In addition, WUP monographs (one in economics and one in cultural studies) and a textbook in physiology have been submitted to the press, all of which were in the editorial process at the time this research was being conducted. All publications are double-blind peer reviewed, and there is an emphasis on ensuring high-quality publications. Furthermore, there is a plan to buy software to perform systematic plagiarism checks on manuscripts.
The press has no employees. Dr H. L. Raghavendra, an expatriate Associate Professor of Biochemistry, was tasked by the first president of WU to establish the press and to act as its Editor-in-Chief. He is supported by the WU top management and their staff. The Editorial Board of the press, which is de facto also the editorial board of the STAR Journal, works closely with Raghavendra. Effectively, the success of the press is a team effort by all top manager-academics of WU. Thus, WUP is able to run a 'tight ship' due to the high-level commitment from the top of university management, who are all relatively young, research-involved manager-academics with a strong commitment to building the university.
At the time of research, the Editor-in-Chief was being reassigned to a management position as 'Director: Publication and Dissemination' in acknowledgement of his role at the university; yet, he asserted that he would nonetheless continue his research and teaching. In addition, two coordinators for publication and dissemination, respectively, would eventually constitute the formal structure to be put in place in support of the press. Given its structure to date, the press has not needed a budget of its own. WUP's direct expenses are minimal, and part of the research budget is negotiated annually with government and under the control of the university Vice-President. Payments are approved as they arise. WUP has not explored any external funding nor has it generated any income of its own. While there is no expectation from the side of the university to generate revenue, it is clear that the press will need to move towards financial sustainability within the next 10 years. Thus, a business model will need to be developed, which may include subscriptions, but it is unlikely to include an author-pay system as this is seen as prohibitive to Ethiopian authors. The current approach is that WUP plays an important role in popularizing WU research; that it is part of establishing the university's brand identity; and that the quality and relevance of its publications will attract funding from regional and federal governments in the future.
WUP's few publications are all available open access from its website, and the journal in particular is co-hosted by the Directory of Open Access Journals and African Journals Online. Print copies are available for purchase directly from the press. Currently, WUP prints 200 copies of all its publications. Given the lack of high-quality printers in Nekemte (which is some 330 km from Addis Ababa), along with the administrative and time burden that is involved in obtaining quotations from printers in Addis and supervising the printing process, WU has purchased digital printing equipment to print and bind books on campus. WUP distributes three free copies of its publications to all Ethiopian university libraries, research centres, agricultural colleges, and government ministries. There are no agreements in place with national or international distributors. The Editor-in-Chief is, however, aware that this is the way to go, especially once WUP starts to publish books.
WUP licenses the content of the STAR Journal under a Creative Commons Attribution license. The licensing and compensation model for authors of books is under discussion. Given that the WU library, which has several branches across the various WU campuses, is digitizing its holdings to ensure better accessibility of knowledge to students and staff, and has set up a digital research repository, there is a commitment by WU to ensure that WUP publications remain accessible.
The STAR Journal has been indexed widely to increase its international visibility (including DOAJ, WorldCat, ProQuest, OAJI, SHERPA-ROMEO, and the like). It will also apply for inclusion in indices such as SCOPUS and the Web of Science. The press also approached JSTOR, but it had not yet received a satisfactory response at the time of research. All STAR Journal articles have DOIs to facilitate online indexing and discovery. Furthermore, the press has a well-functioning website from which PDF versions of its publications can be downloaded and print publications can be ordered. The press does not have Facebook, Twitter, or other social media accounts. The overall ICT infrastructure available in Nekemte was described as being inadequate for WUP's needs; the lack of social media presence is also due to the repeated social media blackouts in Ethiopia.
In the next 5 years, WUP is expected to grow organically to a stable of five to six quality journals; publish relevant books, including textbooks, authored by academic staff at the university; have in place standardized procedures to improve the overall workflow; enhance the WU brand; and become known as a centre of excellence in academic publishing and knowledge dissemination on the continent. Correspondingly, WUP considers the success of its publications in two terms. The success of the STAR Journal is measured in terms of citations as indicators of academic quality, accessibility, and relevance. Other measures of success, particularly linked to the notion of serving the community with relevant knowledge and technology transfer, include who publishes in the journal, on what topics, and how this is immediately relevant to the local community.

DISCUSSION
The in-depth case studies presented above illustrate the diversity of ways in which local publishing practices at four African university presses have been shaped in response to the local and global environments in which they operate. The first case of AAUP shows a press that has put any digitization efforts or efforts at globalizing its publishing on the backburner. It is a press that operates along traditional, editorial lines and retains a focus and primary orientation on the printed book as its core object of attention and key source of legitimacy. As the case study illustrates, this is not entirely out of choice, and there are plans for the future development of the press that are not articulated in explicit market-or commons-oriented terms, even if the notions of 'productivity' and 'efficiency' articulated by the AAUP's director may, in time, create pressure for a more market-oriented way of operating.
When prompted, the opportunities presented by technological innovations in the publishing industry are mainly understood in terms of creating marketing and distribution channels complementary to the existing print-based model focused on the local market. The defining characteristics of our knowledge commons logicsuch as open access distribution, creative commons licensing, focus on use value, and so onare not part of the thinking of the AAUP leadership at present, even though the commitment to access (currently implemented only in terms of affordability) and the AAUP revenue model (which does not require the press to cover its own costs) would allow it to seriously consider such a logic in a way that can balance the interests of authors and reach a much wider audience.
In comparison, the three other presses exhibit more of the characteristics of the market or knowledge commons institutional logics. Wits UP's institutional logic can be interpreted as a hybrid as it straddles both the editorial and market logics. Wits UP has a strong niche position in the local publishing market in terms of sales revenue and reputation. It has been able to keep up with key technological developments in the publishing industry, follow global trends, and expand its market reach through international distribution agreements. Yet, at the same time, it remains a traditional university press in terms of its core identity and mission, its ownership and governance model, its non-profit revenue model, and its ability to grow organically in a niche market with limited resources. Thus far, Wits UP seems to be able to make the best of the technological innovations in the publishing industry: it has a digitally driven publicity strategy that leverages social media and complements traditional forms of publicity such as book fairs and book launches. A fully open access publishing strategy would require a radical change from its currently successful business model, and its experimentation in this area has been cautious. There is, however, an acute awareness that others are experimenting with or are making a wholesale shift to open access. This does not, however, suggest a departure from its hybrid editorialmarket logic; it is strengthening the hybrid logic to make the press sustainable in the face of changing industry and market trends.
Conversely, the institutional logic of the UNOP is more firmly rooted in the editorial logic, even if this is contrary to long- teristics of the knowledge commons logic). Both of these rationalesreputation and relevanceare sufficient for WU to commit to its press. It governs and finances the press as an integral part of the university's research function, and the university's top management form, in many respects, a 'peer cooperative' in the way they are integrally involved in the operations of the WUP and committed to its development. Although our research project did not specifically solicit evidence from university leaders, the interviews we conducted with staff of the four university presses suggest that university managements hold their own beliefs about the function of a university press. These beliefs may either align with those of the press (as shown especially in the case of Wollega and also at Wits) or they conflict as in the Nairobi case. Institutional pluralism may be the default setting in which an organization operates within multiple institutional spheres (Frølich et al., 2013). Organizations are confronted with diverging or multiple institutional logics embedded in different regulatory regimes, normative orders, and cultural logics. Yet, notwithstanding the acuteness of divergent logics in the case of Nairobi University Press, it appears that, regardless of whether the logic of the press is in alignment with that of university management or not, the dominant logic is the one manifesting in the practices of the UP, and it is here where resistance to change and innovation is evident.
This argument can also be made with respect to WU, which as a new university is clearly not immune to pressures to conform to existing taken-for-granted norms and values as indicated by its 'reputation project'. Nonetheless, the press was afforded the opportunity to develop its own practices without the undue influence of experiences shaped in the earlier phase of the history of the university press. It can be expected that these practices may foster over time to firmly root a logic of the knowledge commons, as it has in the cases of AAUP and UNOP, where editorial logics are enabled and constrained by the established administrative apparatus, practices, and logics of their host universities. Finally, there is some hybridization at Wits UP, with evidence of marketoriented practices co-existing with that of an editorial logic. This could be attributed to the fact that Wits UP has access to a larger market than the other presses by virtue of being located in South Africa where there is a larger book-buying public; there are more and financially stronger university, public, and corporate libraries; and more general and academic bookshops.
Finally, although it presents the most classic case of a university press steeped in an editorial logic, the case of AAUP seems particularly relevant when considering the prospects of a transition from one logic to another: AAUP is well positioned to explore open access as a viable publishing model because this aligns with expectations of its host, which has signed the Berlin Declaration, and its current 'deal', whereby the press is not expected to be profitable but rather to make the university's knowledge products accessible and affordable. Holding back a transition to open access is the expectation of authors to generate income from royalty payments that are dependent on book sales, an expectation that is characteristic of the editorial logic.
But if there is no expectation for the press to be profitable, the press could explore alternative methods of remunerating authors, such as once-off, upfront payments to name but one example (Van Schalkwyk, 1998). This would resolve the tension between sales-dependent royalty payments and open access. For now, the press seems unable to devise solutions that are outside of the taken-for-granted editorial logic, but it is clear that moving towards a logic of the knowledge commons is within the realm of the possible.

CONCLUSION
We set out to investigate the state of university presses in Africa and the way they deal with the changes and challenges encountered within their local contexts and the opportunities presented by technological innovations in academic publishing internationally. Using the established theoretical perspective of institutional logics, amended by our conception of a logic of the knowledge commons, we analysed key practices and the characteristics of four university presses in Africa.
We found that university presses in Africa are well aware of new technologies and the opportunities that they afford the press. Some are using technology to improve their visibility and the reach of their publications. But as the baseline study showed and our case studies confirmed, university presses in Africa are not yet making full use of these technologies, particularly when it comes to shifting to new publishing and distribution models.
Established practices and related institutional logics appear constraining. Thus, if there is indeed a second transition of the academic publishing industry underway towards a new logic of the knowledge commons, it is in its early stages, partial, and limited in the African context. The case studies show that the institutional logics approach provides a useful theoretical framework and set of concepts for doing empirical research aimed at furthering our understanding of the academic publishing landscape in Africa and further afield.
There is ample opportunity for further research across multiple contexts to evaluate whether the logic of knowledge commons is a viable ideal type within the institutional logics perspective of institutions: the establishment of the African Research Universities Alliance and the proliferation of institutional repositories and of several research centres of excellence are notable within the African context, as well as the continued survival and development and established and new university presses. Further afield, surprisingly, many new university presses have emerged in the UK despite austere conditions (Adema & Stone, 2016;Lockett & Speicher, 2016), while across the Atlantic, university presses in the USA are being forced to close as they are not permitted to or are unable to adapt (Sherman, 2014;Straumsheim, 2017). It is clear there is no such thing as a typical university press, and presses are experimenting with and modifying emerging models such as open access in a context in which funders of research are increasingly mandating open access publishing for grantees (Adema & Moore, 2018;Cond & Rayner, 2016;Schonfeld, 2016). Evidence of the emerging logic of the knowledge commons may well be found in these contexts and warrants further research.