This essay is a response to a study by Gareth J Johnson, entitled ‘Cultural, ideological and practical barriers to open access adoption within the UK Academy: an ethnographically framed examination’, previously published in Insights.1

Except where otherwise indicated, all quotations are taken from the article in question.

What was the goal of this study?

Johnson’s article seeks to offer some answers to a long-standing question: why are scholarly and scientific authors not adopting open access (OA) publication practices more willingly and in greater numbers? Or, as the author puts it, how can it be that ‘despite over a decade of concerted effort by librarians, repository managers, scholars and other actors’, we still find that ‘local academic communities [show] limited or reluctant engagement with the various emerging open dissemination forms’ – this despite the widely held view among OA advocates that ‘OA represents an ideological “self-evident” good’?

Clearly, a number of answers to this question are possible, as are some combinations of them. However, I believe each of the possible answers can be characterized under one of three attitudinal categories, each of which maps to one of the factors (‘cultural, ideological and practical’) listed in the title of the paper under consideration.

First, authors who experience a ‘practical’ barrier to OA adoption might be characterized as willing but unable. These authors would be happy to make their work available on an OA basis, but are prevented from doing so by one or more external factors: promotion and tenure requirements that pressure them into publishing in toll-access venues that may or may not permit self-archiving; lack of funds to cover article processing charges (APCs); anti-OA bias on the part of colleagues who have control over promotion and tenure opportunities, etc. (Clearly, in this category there may be a direct connection between the ‘cultural’ and the ‘practical’, to the degree that it is academic culture that erects practical barriers between the author and OA adoption.)

Second, authors who experience a ‘cultural’ barrier to OA adoption might be characterized as unwilling due to misunderstanding. These authors are not anxious to make their work OA, but might be more willing if they did not harbor misunderstandings of what OA is, or subscribe to irrational and unfounded biases arising from their academic culture. For example, they may have been given the false impression that OA journals are, by definition, non-selective, or that the only way to make their work OA is to pay APCs.

Third, authors whose resistance to OA arises from an ‘ideological’ barrier might be characterized as unwilling due to disagreement. It is possible for authors to oppose OA in principle – or, perhaps more likely, simply to be unconvinced that making their work OA is always and necessarily the only correct course. For example, such authors might want to retain more of their exclusive rights than OA regimes allow; or they may be content to communicate their findings to a limited audience, at least in the short term; or it might simply be that they do not believe there is anything fundamentally wrong with publishers charging for access to content. Any of these positions would undermine an author’s desire to make his work OA.

While other specific examples and hypothetical scenarios of resistance could be spun out, these three categories of unwillingness seem to exhaust the logical possibilities to explain an author’s failure to publish on an OA basis. Either the author wants to do so and is somehow being prevented, or the author chooses not to do so (for reasons that may or may not be based on an accurate understanding of OA).

Given that most scholarly and scientific authors continue to publish in toll-access venues, and that convincing them to publish OA is necessarily an important part of the OA community’s various strategic agendas, figuring out what is stopping authors from doing so is clearly an important project – one with which Johnson’s study is designed to assist.

How was it conducted?

In order to gain insight into the reasons why more authors are not embracing OA, Johnson undertook investigative fieldwork intended to ‘provide a baseline of current cultural publishing practices and norms’ by means of ‘an ethnographically framed critical exploration of the institutional OA practitioner community’s perceptions of scholarly engagement, resistance and comprehension of OA praxis’.

The author’s fieldwork consisted of interviews with ‘OA practitioners’ – that is to say, people ‘tasked with promoting and enabling open access via policy, advocacy and practical work’, most of them in university libraries.2 Informants from 81 universities were interviewed using a ‘qualitative semi-structured cultural interviewing method’, a ‘conversational, non-confrontational and naturalistic method which can yield considerable in-depth insights alongside contextual information’. The interviews were recorded and transcribed, following which the author subjected the transcriptions to ‘close reading then qualitative content analysis (QCA), before analysis through ideological critique’.

Johnson acknowledges that the ethnographic approach (which ‘incorporates a variety of participatory and observational methods, and focuses on uncovering overlooked, mundane and everyday knowledge and behaviours which expose a community’s inner workings’) is controversial as a method of empirical study, and is seen by some as ‘valueless, suggesting results demonstrate researcher bias’. For the sake of argument, however, we will accept that this approach has clear merit in the present context – while at the same time acknowledging legitimate concerns regarding any method that calls on the researcher to identify and explore ‘implicit nuances and latent meanings’ in his or her informants’ responses. Meanings that are latent are, by definition, not directly expressed by the informant, and nuance (let alone ‘implicit nuance’) is highly dependent on inference by the researcher. One need not be in thrall to ‘reductionist, quantitative approaches’ to see the obvious risk here: all of us, to varying degrees, see and hear what we expect to see and hear when others respond to our questions, and this is no less true of ethnographic researchers than it is of anyone else. A research method that requires the researcher not only to perceive his informants’ unexpressed meanings, but also to detect nuances that are themselves implicit in the things his informants are not saying, will run a higher risk than others of leading to confirmation bias and other errors of inference.

What did the author conclude?

Johnson’s interviews with his various informants led him to conclude that there are real systemic and structural barriers impeding author uptake of OA, including ‘an array of potential mechanistic, policy or legal blocks’ and distorted incentives arising from ‘higher education’s ongoing neoliberalization’. However, it is ‘the academic community’s knowledge of and attitudes towards OA that were shown to present the greatest obstacles’. In other words, his findings suggest that authors are not mainly avoiding OA because they are being externally prevented from embracing it (willing but unable), but rather because either their knowledge is lacking (unwilling due to misunderstanding) or because they simply do not wish to engage (unwilling due to disagreement).

And here is where we encounter a fundamental problem exposed by the findings of this paper, one that is reflected as a larger problem within the OA community as a whole. The problem is illustrated by one particular passage in Johnson’s report. ‘Despite the endeavours of OA practitioners who were devoted to advocacy’, he writes, ‘the majority of scholars’ understanding or embrace of openness within research dissemination practice was found to be “patchy”, “ill-informed”, or “confused”.’ [Italics mine.] Now consider the next sentence: ‘Consequently, a picture of the UK academy was presented wherein the reluctance of academics to engage with OA was predicated on an underlying lack of sound information about it.’

This explicit linkage of the concepts of ‘understanding’ and ‘embrace’ is both significant and instructive. Underpinning this linkage is the assumption that to understand OA is to embrace it – or, in other words, that there are in reality only two reasons that an author might not embrace OA: either he understands it and is willing but unable to embrace OA praxis, or he fails to understand it and is therefore unwilling due misunderstanding. Although the language is not always as clear as it might be, Johnson’s report seems to indicate that this assumption substantially informs his interviewees’ understanding of authors’ orientation towards OA.

But this assumption is clearly problematic. On what basis can we dismiss out of hand the possibility that an author might fully understand OA, and be in a position to embrace it, and yet still freely choose not to do so? At no point do any of Johnson’s informants seem to have considered this possibility – although since the reader has no access to the interview texts, or even to any direct quotes from the informants, it is not possible to say for certain whether this is the case, and some informants did reportedly express concern about their own ability to ‘perceive publication issues from the academic community’s perspective’. Approaching his interview data through the prism of Marxist/Foucaultian analysis, Johnson himself suggests that academics’ unwillingness to engage with OA may be a manifestation of ‘false consciousness’ – in Marxist theory, the term for a sort of socio-economic Stockholm Syndrome, whereby the oppressed worker is lulled into a false sense of identification with the system that oppresses him.3 Rejecting OA because of false consciousness would, obviously, be a variant on unwillingness due to misunderstanding (in this case, the author’s failure to recognize his own oppression at the hands of the dominant system, or, in Johnson’s words, ‘a blinding of the self to economic and social rationality whilst adhering to a more “accepted” norm’). At the same time, Johnson understands his informants to be suggesting that authors’ ‘scholarly publishing practices [are] increasingly shaped through time pressure and performance measure, rather than ideological desires’ (willing but unable).

And here we see the second fundamental problem, one which is inherent to research of the kind being reported here: Johnson has, first of all, relied on third parties to tell him what authors are thinking; and, second, chosen a lens of interpretation that filters out any possibility of informed, principled disagreement with the OA practitioners’ agenda on the part of those authors. Furthermore, the parties on whom he has relied for information about authors’ attitudes are people whose job it is to advocate for OA and encourage authors to adopt it. Since these practitioners are in the business of promoting OA, they are arguably more likely than other third-party observers to see reluctant authors as unreasonable or uninformed in their resistance to the program.

But here it is also important to point out that authors’ failure to engage with OA does not necessarily arise from any objection to OA in principle. If an author is unwilling due to disagreement to engage actively in OA praxis, it does not automatically follow that the author disagrees with OA practitioners’ belief that ‘OA represents an ideological “self-evident” good’. The author may agree that OA is a good thing, and merely disagree that OA is the only acceptable approach. Authors exhibit this attitude, for example, when they freely choose to publish selectively, rather than exclusively, in OA journals. Of course, if one dismisses such a belief as necessarily arising from false consciousness, and therefore frames it as fundamentally a matter of misunderstanding or self-deception, then one can conveniently avoid dealing with the annoying possibility of genuine and principled disagreement as to the unique and universal desirability of OA. (And of course if one rejects the idea of free choice, then this whole discussion becomes moot; in that case there is no real point in trying to understand authors’ perspectives, since, lacking meaningful agency, their views and desires are irrelevant.)

None of this is to suggest that Johnson’s research is not useful and instructive; on the contrary, he makes a number of interesting and compelling points as a result of this study. However, these are mainly useful for what they tell us about the population he interviewed – their concerns about the effectiveness of OA advocacy, their beliefs as to what influences drive authors’ decision-making, their perceptions of how the structure of the Academy affects publishing behaviors, etc. They tell us little about the people whose attitudes are under consideration – and whose attitudes will shape the future of scholarly publishing, to the degree that they remain free to make choices about how and where to publish.

What would be tremendously interesting would be a similar article based on actual conversations with the authors themselves – and one that makes not only the interpretation of those conversations but also the raw (and anonymized) text of them openly available for examination. Hopefully such a study will be forthcoming.