Authorship and e‐Science: Balancing Epistemological Trust and Skepticism in the Digital Environment

Abstract

In this article I consider the role of authorship in balancing epistemological trust and skepticism in e‐science. Drawing on studies of the diagnostic practices of doctors in British breast care units and the gate‐keeping practices of a Californian publisher of (professional and amateur) horticultural works, I suggest that conventions of authorial designation have an important role to play in nurturing the skepticism essential for scientific rigor within the framework of epistemological trust that pragmatism and morality require. In so doing I question the assumption of contemporary scholars that scientific works are determinate in fact, while nonetheless supporting the idea of semi‐indeterminate authorship as a goal. I then consider the theoretical and practical consequences of that view with an analysis of the Anglo‐Australian legal constraints on attributions of authorship, and the complex of other relevant (authorial and public) interests and rights that may require different models of attribution—oriented not around authors themselves, but rather their employer, their manager or supervisor, or some (other) pseudonym deemed to be an appropriate organizing mechanism for the works in question.

Keywords:

• Authorship
• e‐Science
• Epistemological Trust

Acknowledgements

This article is part of the Oxford e‐Social Science Project supported by a grant from the Economic and Social Research Council (RES‐149‐25‐1022). I am grateful to Maria Jirotka and her eDiaMoND coauthors, and to Karen Yeung, for their comments on earlier drafts.

Notes

[1] By “law” in this context I mean the entire system of the State’s imposed and enforced rules of conduct. Note the distinction here, drawn explicitly below and central to my analysis as a whole, between “trusted” and “trustworthy”; my suggestion being not that law makes people trustworthy, but rather significantly less likely— and capable of being trusted not—to act in certain ways.

[2] For further information regarding CalFlora and eDiaMoND see http://www.calflora.org/ and http://www.ediamond.ox.ac.uk/ respectively. The eDiaMoND study has been chosen because of its importance generally as “a flagship U.K. e‐science project”, and specifically to the Oxford e‐Social Science Project for which this article was written. (See Ralph Schroder and Jenny Fry’s “Social Science Approaches to e‐Science: Framing an Agenda” at http://jcmc.indiana.edu/vol12/issue2/schroeder.html (accessed 31 August 2008)). The CalFlora study has been chosen because of its use in the eDiaMoND study, and my suggestion that this use overlooks important differences between the eDiaMoND and CalFlora environments from which the insights in this paper are derived.

[3] On the use and value of intellectual property law as a source of default rules, policy and public interest safeguards in the regulation of science, see Dreyfuss (2002, esp. 1216–1227).

[4] The relevant legislation is the Copyright, Designs and Patents Act 1988 (CDPA) in the UK, and the Copyright Act 1976 (USC Title 17) in the USA (Copyright Act).

[5] Claims in this sense are akin to ideas, which are not works in which copyright subsists. Whether the copyright subsisting in works protects the ideas which those works contain is a separate question. In the USA it does not by virtue of US Code Title 17 §102(b), which states that “[i]n no case does copyright protection for an original work of authorship extend to any idea, procedure, process, system, method of operation, concept, principle, or discovery, regardless of the form in which it is described, explained, illustrated, or embodied in such work”. In the UK, as stated by Lord Hailsham in LB (Plastics) Ltd v Swish Products Ltd [1979] RPC 551 (HL) 629, it “all depends on what you mean by ideas” (quoted by Lord Hoffman in Designer’s Guild v Russell Williams Textiles Ltd [2000] 1 WLR 2416 (HL) 2422).

[6] On the distinction between acts and statements of authorship and its theoretical basis see generally Heymann (2005).

[7] This aspect of the CalFlora study is not considered here for reasons of space.

[8] Jirotka et al. also consider issues relating to the works’ protection in the digital environment, including specifically the concerns of healthcare workers to restrict access to medical information in order to comply with their legal and ethical obligations to protect patient privacy and confidentiality. The specific means they employ to that end include the use of tracking sheets to record the various stages in the information’s production, rules restricting use and disclosure, and open‐plan office design to make unauthorized or secret access to the information difficult. According to Jirotka et al., healthcare workers have confidence in their system because they believe it to be ethical. Faced with the eDiaMoND proposal, their confidence is threatened by a new concern with the future of their system. It is interesting to consider the reasons for that concern. Intuitively, information seems harder to extract from the material realm of print and therefore easier to secure. In addition, while the security possibilities of technology are cast as such that this ought not to be so—digital locks ought to make information easier to secure in digital form—it is more difficult to trust a system of which one is ignorant, particularly when the information is sensitive and the extent of a breach potentially substantial. In the popular conscious it is this that makes the digital world appear so vulnerable: an invisible, porous and borderless world with its own values and culture of access and dissemination. According to Bayne (2006) we trust physical documents partly on account of the transparency of their authorship. As discussed below, this is consistent with the eDiaMoND study, which shows how BCU workers depend on authorial information in order to assess and interpret the works on which they depend. Such trust, however, is in the works themselves, and not the systems for securing them.

[9] The “observer” is defined for CalFlora purposes as the person who made the observation for the report, being either the collector of the observed specimen or the author of the article from which the report itself was taken (http://www/calflora.org).

[10] According to Van House, while “[i]n professional communities, the identity of an observer is often sufficient” on the ground that “users know who is reliable[,] CalFlora is wrestling with the problem of open access by both users and contributors who do not necessarily know one another” (2002, 108–9).

[11] The reference here and elsewhere to the BCU practices is to the practices reported by Jirotka et al. (2005) and Hartswood et al. (2002).

[12] There is no suggestion by Jirotka et al. that their concept of biographical familiarity does approximate Van House’s concept of provenance; while Jirotka et al. contextualize their study with reference to the CalFlora study (see 2005, 376) they do not specifically refer to Van House’s concept of provenance.

[13] According to Hartswood et al. (2002, 93), “[m]any English and all Scottish screening centres practice double reading, where each mammogram is examined independently by two readers”.

[14] An example of an annotated screening form is reproduced by Hartswood et al. (2002, 93), who also describe in detail the process by which screening forms are annotated (2002, 93–94).

[15] Hartswood et al. (2002, 94), describing “blinding practices in the majority of six centres that formed the core of our studies” as “entirely discretionary” due to the “public character of the screening reporting form”, which “readily affords the second reader the opportunity to view the first reader’s opinion”. See also Jirotka et al. (2005, 383).

[16] See also Jirotka et al. (2005, 383), describing “the collaborative practices underpinning double reading” as “worthy of particular note”, with second readers able either to “annotate the form, in relation to something they themselves find noteworthy, or to give their opinion on the first reader’s … diagnosis”; Hartswood et al. (2002, 91), describing the reading process as having “informal, collaborative features”; namely, the annotations); and Hartswood (2002, 96), stating that “[r]eaders almost always will annotate the … form when they conclude that there is an abnormality”, and will sometimes also annotate it “with comments about features that they have determined to be benign”. Examples of annotations are given by Hartswood et al. (2002, 96); the first comprising the mark “X” on the form’s breast schematic with the words “NRC” (no real change), and the second showing another mark on the schematic with “?” and “BT” (breast tissue) beside it. As the authors explain, the first annotation “seems intended merely to reinforce” the reader’s opinion, while the second is “to express—and to draw attention to—the first reader’s uncertainty”. The form reproduced by Hartswood et al. (2002, 93) contains the following further annotations by the radiographer and first and second readers, respectively: “Moles. Cyst—R Breast 20yrs ago. HRT for 2 yrs. Pain L breast and arm—GP says muscular” [with markings on schematic]; “new” [with circle on schematic]; “BT I think HRT related” [without schematic markings].

[17] See Jirotka et al. (2005, 389), suggesting as a focus for their paper the observation, which they attribute to Berg and Goorman (1999), that “healthcare workers develop a sense of the trustworthiness of the people (and machines) they work with and evaluate the meaning, status and quality of data in accordance”; Jirotka et al. (2005, 391), stating that “while standardization of artifacts used to record decisions (e.g., the screening form) may give the impression of uniformity of practice, their use and interpretation is tied intimately to local practices”; and Jirotka et al. (2005, 382), describing the biographical familiarity that readers exhibit of “their own and their colleagues” performance”, and the way in which, “[b]y working together, readers come to know their own and each other’s strengths and weaknesses”.

[18] Prohibitions against honorary authorship tend to be explicit in the codes of the increasing number of universities that adopt the International Committee of Medical Journal Editors’ (ICMJE) Uniform Requirements for Manuscripts Submitted to Biomedical Journals: Writing and Editing for Biomedical Publication (http://www.icmje.org/#aboutur) (Vancouver Protocol), which provides that “[a]ll persons designated as authors should qualify for authorship, and all those who qualify should be listed.” Even those universities that adopt the Protocol but leave scope for disciplinary differences will often prohibit false (including honorary) attributions of authorship. An example is Monash University, the Handbook for Doctoral and MPhil Degrees of which includes the following provisions (http://www.mrgs.monash.edu.au/research/doctoral/chapter6b.html): “Although Monash is guided by the Vancouver Protocol on authorship, it is important to remember that each discipline has its own customs and practices for joint or multi authorship … Due recognition of all contributions is part of proper research process. However, honorary authorship (where a person is listed as an author when they have not participated in any substantial way) is unacceptable.”

[19] See for example the University of Oxford’s Academic Integrity in Research: Code of Practice and Procedure (http://www.admin.ox.ac.uk/ps/staff/codes/air.shtml): “3. Misconduct for the purpose of this code means the fabrication, falsification, plagiarism, or deception in proposing, carrying out, or reporting results of research.”

[20] The outcome in many of the cases in which copyright has been denied for de minimis reasons may have been different had the subject matter been presented as artistic works on the ground that invented words and slogans convey no meaning and have no significance independent of the context in which they are used, and are thus better treated as artistic works in recognition of their primarily visual or associative significance.

[21] The conclusion that mammograms and diagnostic reports are copyright works means that they are also the object of copyright. Absent any agreement to the contrary, that copyright would not be owned by their authors, but rather their authors’ employers—meaning presumably the individual units for which they work.

[22] The Vancouver Protocol states as follows: “Increasingly, authorship of multi‐center trials is attributed to a group. All members of the group who are named as authors should fully meet the above criteria for authorship … The order of authorship on the byline should be a joint decision of the co‐authors. Authors should be prepared to explain the order in which authors are listed.” See further Biagioli (2002, 263), quoting from Rennie, Tank, and Emanuel (1997):

If there remains a problem with the language of authorship, it can easily be discarded for the language of contributorship. Such changes in nomenclature however do not involve a change in the nature and importance of authorship itself. A model of authorship oriented around contributors rather than authors is used by the American Medical Association (JAMA), which justifies it as follows: “Because the current system of authorship is idiosyncratic, ambiguous, and predisposed to misuse, we propose in its place a radical change: a new system that is accurate and discloses accountability. We propose the substitution of the word and concept contributor for the word and concept author … Abandoning the concept of author in favour of contributor frees us from the historical and emotional connotations of authorship, and leads us to a concept that is far more in line with the actuality of modern scientific work.”