Current publishing landscape

Commercial publishers have been gradually increasing their market share in scholarly publishing since the end of the Second World War (Crow 2006; Edgar and Willinsky 2010; Hawkins and Battin 1998; Tenopir and King 1997). The increased power of the major publishers has meant less leverage for negotiation and subsequent price increases for libraries, which resulted in the “serials crisis” (McGuigan 2004; Schmidle and Via 2004; Thomes and Clay 1998). In recent years, libraries have taken a greater lead in the publishing world, establishing presses and publishing programs. Part of this has been in response to the growing oligopoly of major publishers, offering academic journals an alternative to existing commercial publishers and as a means to promote OA publishing (Bonn and Furlough 2015; Harboe-Ree 2007).

In Canada, support for non-commercial scholarly journals extends beyond libraries. For example, of note is the Canadian government’s grant program for scholarly journals in the humanities and sciences (see the SSHRC Aid to Scholarly Journals program) as well as the non-profit hosting platform Érudit and Coalition Publica. The SSHRC Scholarly Journals program provides financial support of up to $30,000 a year for eligible journals to assist in production, online dissemination, and the application of new technologies. As of 2018, the program supports over one hundred journals whose funding is subject to renewal and reapplication every three years. Coalition Publica is a partnership between the non-profit hosting platform Érudit and the Public Knowledge Project (PKP) (producers of the open source publishing software Open Journal Systems) to support scholarly publishing in Canada. Under this partnership, eligible Canadian journals may receive financial support as well as professional publishing services. In order to receive funding, both SSHRC and Coalition Publica require journals to make their content OA within at least twelve months of publication.

Despite efforts to grow alternative publishing models, the major commercial publishers continue to be an attractive option to journals. The journal Cultural Anthropology, a major publication of the Society for Cultural Anthropology (SCA), notes the varying services commercial publishers can offer. Among them are managing production processes, handling file management systems, preparing and editing metadata, supporting and maintaining the online publishing platform, and general administration (Elfenbein 2014). Although the SCA made the decision to leave the commercial publisher Wiley-Blackwell to pursue OA for their publication, former managing editor Tim Elfenbein admits:

This is one of the hardest parts about setting off on your own. When a large commercial press such as Wiley adds another journal to its collection, it will in all likelihood already have a well-tested platform to sit on. The cost of building that platform and the rest of its publishing systems is spread across a huge number of journals, which means that the company can afford dedicated technology and infrastructure services. . . . For the SCA, everything is being built anew, and our systems are breaking all the time. It could take us years and a whole lot more investment to fix all the bugs in our systems, and that signals a level of vulnerability we did not have when we were publishing with Wiley. (294)

The vulnerability of journals which are not attached to commercial publishers was also expressed in a 2017 report by a working group of the Canadian Association of Research Libraries (CARL): “Canadian scholarly societies are sometimes drawn to major for-profit publishers, attracted by their well-developed infrastructure, extensive marketing reach, and guaranteed revenue, among other things” (4). Furthermore, the shift to OA has been particularly challenging for many of these journals. Heather Morrison (2016) interviewed fifteen participants from small, scholar-led journals to understand the effects of open access on their publications. Morrison found that many of them were uncertain and “concerned about survival . . . in an open access future” (83). In such an environment, the services and reach offered by the major commercial publishers will continue to be tempting as the funding of non-commercial journals remains precarious and patchwork.

Not surprisingly, the myriad of threats for these journals has meant that many independent journals cease publication. Bo-Christer Björk, Cenyu Shen, and Mikael Laakso (2016) discovered that of 250 open access “indie” journals, only half were still in publication twelve years later.

Journal surveys

Given the previous literature, scholars and other stakeholders have been aware of the vulnerability of independent Canadian journals for over a decade. This has led to several studies which reviewed the Canadian publishing landscape and socioeconomic condition of these journals.

In 2015, the non-profit hosting platform Érudit undertook a survey of Canadian journals in order to assess their socioeconomic situations (Paquin 2017).¹ The focus of the survey and interviews was to better understand the economic structure of these journals as well as their income and expenditures. In a report of the survey’s results, Émilie Paquin (2017) does not state how they assembled the list of journals, but the researchers sent 337 journals a survey (69 completed) and conducted 36 interviews. The journals included open access and subscription journals as well as commercial and non-profit journals. The report noted a great heterogeneity amongst the journals in terms of size of editorial structures, business models, and revenues. Paquin (2017) noted the influence of associations and universities as journal publishers (6) as well as a significant category that fell into “Other.” The report concluded with various recommendations to stabilize revenues for these journals, including gathering more detailed information about their financials and the development of a partnership model between Érudit and the CRKN (Paquin 2017).²

An earlier study that reviewed the financial situation and composition of Canadian journals was undertaken by Rowland Lorimer and Adrienne Lindsay (2004). The focus of this study was to understand the readiness and ability of humanities and social sciences journals to adapt to an online environment. In addition to collecting demographic information about the discipline and the affiliation with other organizations, Lorimer and Lindsay retrieved information about the total number of subscribers and subscription fees.

The most recent analysis of Canadian scholarly journals was undertaken in 2020. In an unpublished report, Marc Couture (2020) compiled and reviewed a list of open access journals in Canada. In particular, he was interested in comparing the demographics of journals which were indexed in the DOAJ versus those that were not. The author collected similar demographic information about the journals as did Lorimer and Lindsay (2004) and Paquin (2017) (e.g., discipline, affiliation, number of articles published, etc.).

Although not Canadian specific, Brian D. Edgar and John Willinsky (2010) surveyed 998 journals publishing on the Open Journal System (OJS) platform. They noted that over half were published or sponsored by an academic department, whereas non-profit publishers and research units made up 16% and 10%, respectively. Independent groups were represented in 10% of the responses, whereas scholarly societies published 32% of the journals in their sample.

Comparison between these sources is difficult as each has different inclusion criteria: OA (Couture 2020), humanities and social sciences (Lorimer and Lindsay 2004), or hosted on OJS (Edgar and Willinsky 2010), whereas Érudit’s survey was broader in focus (OA and subscription journals from all disciplines). They also used differing methodologies, such as surveys, interviews, and website analysis, leading to differences in the size of their sample sets (from 69 to 998). With this in mind, the researchers hope to fill a gap in reviewing the websites of all independent scholarly journals in Canada. This would allow the researchers to be broader in their coverage (e.g., not just those using the OA model or supported by OJS) while focusing on an understudied but important subset of Canadian journals: those operating outside of the main commercial publishers.

This analysis will also permit the researchers to account for the recent changes in the Canadian publishing landscape, such as the launch of the Tri-Agency Open Access Policy on Publications in 2015 and the changes to the eligibility criteria of the SSHRC Aid to Scholarly Journals. By including non-open access journals in the analysis, the researchers hope to obtain a broader picture of the Canadian, non-commercial landscape that builds on the work of Couture (2020).

Gérard Boismenu and Guylaine Beaudry (2004) noted that non-profit publishers have a significant role to play in the Canadian publishing landscape and that they are “instruments of recognition, legitimation, acknowledgement, dissemination, and enhancement of scientific knowledge” (346). Boismenu and Beaudry (2004)saw these journals as important counterpoints to the increasingly commercial operations of scholarly publishing. As such, the researchers hope to add to the body of knowledge about these important instruments of scholarly dissemination in Canada.

Data coding

Journals were assigned one of four subject areas. For ease of analysis, journals were tagged with only one subject code.

• Health and life sciences (HLS)

• Humanities and social sciences (HSS)

• Interdisciplinary (INTER)

• Science, technology, engineering, and mathematics (STEM)

In cases in which journals were assigned a subject heading in Ulrich’s Periodicals Directory, this was mapped to one of the four subjects above. In the absence of a subject classification in Ulrich’s, a determination was made by the researchers, consulting subject headings for the journal in sources such as OCLC’s catalogue WorldCat. When more than one subject was dominant in the journal’s description, it was coded as interdisciplinary.

Journals were also assigned one of three classifications for access type:

• Open access (OA)

• Delayed open access (DELAY)

• Not open access (CLOSED)

Delayed open access was applied to any journal whose archives were open after a certain period (e.g., twelve months, twenty-four months, etc.).

The researchers also collected data about the size of the editorial teams, breaking them down into two divisions: named and unnamed. Editorial positions were counted as named if the person held a title (e.g., editor in chief, copyeditor, webmaster, etc.). Positions were counted as unnamed if a person was given no title but was listed as part of an advisory board, scientific committee, and so forth.

Journals were additionally tagged according to five possible affiliations. Journals could have more than one affiliation.

• Non-profit publisher/host (e.g., Érudit, the University of Toronto Press, etc.)

• Library (the journal was hosted by a university library)

• University (the journal was affiliated with a department/centre/unit at a university, not including the library)

• Association (the journal was affiliated with a society, association, etc.)

• Other

Affiliations were determined by reviewing the journal’s website(s). An affiliation was counted if an entity was named on the website. Common sections this information was located under was “About” and “Journal Sponsorship.”

In cases in which a journal had more than one website, the site which had been more recently updated was used as the primary source of information.

Access type

With the growth of open access, examining how many journals support an OA model is increasingly relevant. In the dataset, over half were fully open access (n = 292), while the remaining were evenly split between closed journals and journals with a delay (see Figure 1). Of the journals which had a delay, the dominant delay period was twelve months (n = 86). A minority had a delay longer than twelve months (n=7), while only four had a delay of six months.

Although APCs are common in OA publishing, only 10 of the 485 journals had these fees.

Subject area

Journals in the humanities and social sciences (HSS) represented the greatest number in the dataset (n = 367; see Figure 2). Given the dominance of HSS journals, it is not surprising that 22% of all journals in the dataset currently receive funding from SSHRC’s Aid to Scholarly Journal program (2018) (n = 108). STEM and HSL journals were in the minority, accounting for 12.6% (n = 61) and 8.7% (n = 42), respectively. Only fifteen journals were coded as interdisciplinary, and these typically fell in the intersection of health and social sciences (e.g., Canadian Journal of Aging, Drogues, santé et société).

Affiliation

As noted in the data coding section, affiliation is defined loosely, requiring only that the journal identify an organization on its website. This could take the form of language on the journal’s website that they were “supported by” or “affiliated with” or “hosted by” a certain organization. While we are unable to determine the extent and nature of these affiliations, it is clear that outside organizations do have a role to play in Canadian independent publishing. For example, out of 485 journals, only 8 had no listed affiliation. The majority (n = 287) had two affiliations, and 19 had three affiliations.

Overall, the spread of affiliations was equally divided between associations, libraries, non-profit publishers, and universities (see Figure 4). “Other” typically included institutes which were not part of a university but occasionally represented other organizations which could not be classified under the other headings.

The most common pairings were:

• Universities with non-profit publishers/hosts (n = 74)

• Associations with non-profit publishers/hosts (n = 76)

• Libraries with universities (n = 60)

• Associations with libraries (n = 47)

Of the affiliations, there were several which had a large number of journals associated with them. This included the non-profit hosting platform Érudit and Canadian Science Publishing (n = 127 and 25, respectively).

Editorial size and composition

Finally, the researchers wanted to examine the characteristics of the journal’s editorial team in terms of size and composition. As labour is typically the primary input for a journal’s activities, understanding the makeup of its editorial team is critical to understanding the size and scope of the work invested in a journal.

Editorial positions were counted as named if the person held a title (e.g., editor in chief, copyeditor, webmaster, etc.). Positions were counted as unnamed if a person was given no title but was listed as part of an advisory board, scientific committee, and so forth.

Editorial size and open access

Although preliminary, the finding that OA journals have smaller editorial team sizes is worth further study. Several scholars have noted the absence of data on labour in OA journals, while some credited labour as the primary glue holding these journals together. For example, Björk, Shen, and Laakso (2016) noted in their case study on the independent, scholar-led OA journal Electronic Journal of Information Technology in Construction that “[w]hat probably has saved the journal (which has no income but is fully based on volunteer work and the access to a free server) is the tiered managerial structure of up to ten co- and junior editors, which helps in spreading the workload of overseeing the review of submitted manuscripts”(para. 33). Models or additional case studies investigating the labour structure of OA journals may illuminate further how these journals operate, particularly in contrast to commercial or subscription-based, closed journals.

Multiple affiliations

Although none of these journals had a commercial publisher, calling them “independent” may also be misleading as the majority had at least one affiliation, and more than half had two. This indicated that the support and organizational structures of these journals are complex and not single tiered. Although the current study is unable to demonstrate the exact nature of these relationships, this is also an area of future investigation.

The multi-tiered affiliation structure of the McGill Journal of Education may give insights into what these relationships look like. In an article on the history of the journal, the authors note that “[w]e are currently supported directly by a SSHRC Aid to Scholarly Journals award and indirectly by our partners: Open Journals Systems (OJS), the McGill Library, and McGill’s Faculty of Education. . . . We belong to Érudit (a consortium of journals in Québec that distributes our content worldwide) as well as associations such as the Canadian Association for Learned Journals. We are also supported through our editorial alliance with French universities in Québec, which allows us to enjoy a wider bilingual presence provincially, nationally, and internationally” (Strong-Wilson et al. 2019, 8).

A similar string of affiliations was noted in the 2008 article about the Canadian Journal of Sociology, which mentioned support from the University of Alberta Department of Sociology and the SSHRC Aid to Scholarly Journals program, as well as publishing and subscription management by the University of Toronto Press and hosting services by the University of Alberta Libraries (Haggerty 2008).

These layers of support and affiliation combine to provide a strong, foundational base for the McGill Journal of Education and the Canadian Journal of Sociology and underline how interlocking ties can form a strong knot. It would be interesting to see if multiple affiliations correspond to journal success and longevity and, in the absence of multiple affiliations, case studies on how these journals continue to publish and exist in the long term.

Libraries as publishers

Given the growth of libraries as publishers, it is perhaps not surprising that libraries are affiliated with as many journals as associations, university departments, and scholarly societies. While earlier studies on the Canadian scholarly landscape focused on the role of associations, universities, and presses in publishing, teasing out separately the role of libraries may be a path forward for partnerships and sustainability. This study affirms the earlier findings by CARL that “library-based publishing services can be a key component in a more comprehensive national sustainability strategy that provides cost-effective publishing support for Canadian journals while also advancing a more viable open access publishing model” (Canadian Association of Research Libraries 2017, 14).

Open access

The dominance of open access in the dataset is reflective of changing trends in scholarly publishing. In 2017, only 25%, or 8, of the journals in Paquin (2017) were open. This could be due to a difference in sample inclusion (e.g., Paquin included commercial journals). It could also be due to the change in the SSHRC Aid to Scholarly Journals program which requires journals must now be open within twelve months. The change in SSHRC’s funding eligibility would have greatly influenced the HSS journals in the sample.

Even though the majority of the journals operated in an OA model, very few (n = 10) charged APCs. This is perhaps not surprising as APCs are less common in HSS and are often higher, and more common, in commercial publishers (Willinsky 2017). However, the absence of APCs for the OA journals should lead to further inquiry into how these journals operate and manage any expenses or costs. For example, do they run entirely on volunteer labour and in-kind contributions? As more OA advocates argue for the “diamond open access” model of publishing (i.e., no charges to authors or readers), understanding how existing journals operate in this model can provide important lessons for other journals pursuing a no-fees, OA publication.

Dominance of humanities and social science journals

Humanities and social science (HSS) journals represented the majority of journals in the dataset by a wide margin (60%). It is unclear why HSS journals represent a significant portion of Canadian, independent, scholarly journals. A quick search of peer-reviewed journals by discipline in Ulrichsweb suggests HSS journals are more common than health or STEM journals (which represent 23% and 32% of Ulrich’s indexed journals, respectively, compared to 44% for HSS journals). This difference may be due in part to the SSHRC Aid to Scholarly Journals program. Another possibility is that HSS journals produce fewer issues and receive fewer submissions, thus they do not require the large, dedicated resources of scale offered by commercial publishers. There is some evidence that researchers in the humanities and social sciences produce fewer papers than researchers in engineering, natural sciences, and health and medical sciences (Shin and Cummings 2010). It would have been interesting if the researchers had collected data on the journals’ article output over a specific time frame to assess if there was any correlation to discipline. Regretfully, this data was not collected due to time constraints. Future studies may look at the publishing patterns of independent journals based on discipline and determine the feasibility of independent publishing for journals with large outputs.

Additional observations

Although anecdotal, the researchers made several other observations which may lead to areas of further inquiry. For example, the researchers noted that francophone journals, particularly those in Québec, were more likely to be affiliated with a university department and more likely to include a set of unnamed positions in the form of a scientific committee. In general, francophone periodicals were strongly represented in the sample, suggesting an area for further inquiry into the specific publishing landscape in Québec as compared to the rest of Canada. Further investigation could be undertaken to examine the publishing culture in Québec and francophone Canada and whether their models for affiliation and support differ.