Authorship in Academic Literature

CPDD has transparently outlined authorship-related policies in place for validating authorship.1 In that editorial comment, David J. Greenblatt portrays an insightful historical perspective of authorship in the pharmacological and physiological sciences, showing, using MEDLINE/PubMed data, how the number of authors per paper has increased from about 1.5 in the 1950s to about 6 in 2020.1 Another journal (American Journal of Orthodontics and Dentofacial Orthopedics) also recorded a historical increase in the number of authors per paper, noting that between 1956 and 1965, 87% of papers had only 1 author (4% had >3 authors), but in 1966–2015, <5% of papers were authored by a single author.2 Marder (2022) also noted that until the 1960s, derivative works of PhDs usually only carried the PhD student's name, to reflect a personal endeavor, whereas PhD-derived studies post-1970s tended to also include—at minimum—the PhD supervisor as well.3 Who is a valid author, and is there a limit to the number of authors? There are complexities associated with ascertaining the validity of authorship in multiple-author papers, especially in the light of increasingly complex multidisciplinary and “big data” research, so the intellectual significance of individual authors is a key factor.4 Conversely, the failure to attribute authorship to valid intellectual contributors is unethical,4 while sudden requests to change authorship (adding, removing, indicating nonpermitted inclusion, or exclusion) can be challenging for journal editors to detect and manage.5 Ambiguity and discord often arise because there is subjective interpretation surrounding what constitutes a “substantial contribution.”6 It is also complex to ascertain the validity of authorship from a deontological perspective because different forms of contribution may be incommensurable while the determination of responsibility is intangible.7 What should the order of authors be? CPDD does not control the order of authors and has in place a 1-corresponding-author–1-paper rule.1 As in CPDD,1 a common authorship-related editorial policy in biomedical journals is to leave the choice of decision regarding the order of authorship in the hands of authors.8 In such instances, a document that offers a rationale for that order is rarely if ever submitted. However, when reliable verification steps or documents are missing, this may open up the possibility of authorship abuse because the signal sent is that regulation is lax. In guest authorship, guest authors might be offered a senior (last author) position, or they might be assigned a guest co-corresponding author status.9 What tools are available to assist editors in the verification of an author's identification? Open Researcher and Contributor ID (ORCID) is currently likely the most widely adopted author identification persistent identifier. CPDD does not have a mandatory ORCID policy for submissions.1 Despite the documented strengths of ORCID such as author disambiguation,10 it is not always able to disambiguate fake from real authors, especially if there is insufficient background metadata in ORCID accounts of authors with identical names. Unfortunately, ORCID has been abused by providing a form of support or authorship “validation” tool for authors associated with “paper mills.”11 Another mechanism to hold authors accountable for their contributions in academic papers is CRediT (Contributor Roles Taxonomy),12 but since statements that define the roles of each author are controlled by authors themselves, there is no way that editors or reviewers can independently or accurately verify the veracity of those statements. In the author's opinion, there are thus insufficiently robust tools currently available to determine, with absolute certainty, (1) the valid existence and/or identity of an author or (2) their quantitative or qualitative contribution to an academic paper. Some biomedical journals are making laudable efforts to avoid guest, gift, or ghost authorship,13 but a formidable challenge for the academic community would be to devise robust methods and tools and to develop tangible and verifiable criteria that would be able to assess and confirm the validity of an author's identification and of an author's contribution, that is, to validate authorship. The author declares no conflicts of interest of relevance to this topic.