In the early 1990s, walking through the panel section of a physics meeting, a paper on statistical distribution of energy levels in a crystal caught my attention. Well, the topic isn't important here, don't worry, the topic is different. I asked the colleague who was presenting the panel about this statistic and the answer was that he had done the calculations on energy levels, he did not know how to interpret them and the person who could answer my question was the other author, who was not present. I met this colleague some time later and learned from him about this level statistics, which I ended up using in some works already in this century. But that wasn't why I never forgot the episode, but because of the inescapable question: what did it mean to be the author of a work without having authority over all of its content? (It must be said that the two in this "case" became good scientists).
I remember here that author and authority are words with the same origin. At the time, the discomfort that the issue aroused was alleviated by the memory of a practice internalized in the scientific community to which he belonged. Important experiments are carried out by different groups using sophisticated samples obtained in a few laboratories. The use of these samples configures a common type of collaboration and the articles resulting from these collaborations always have the producer of the samples as the author. Part of the community's paradigm is the clear perception of the specific role of that author in the list of authors. His authority is over the sample and not the resulting research itself. So, initially: what is an author? Looking at science from databases, an author is anyone who signs an indexed article.
Today the majority of articles are multi-authored, many with ten, fifteen or more authors in various areas of natural sciences. In the databases, they all automatically receive an additional citation when the article in question is cited. But what is the role of each of the authors? Does the function (contribution) of a given author deserve the citation that he automatically receives? The problem has started to be partially addressed by some magazines, which ask for a description of each author's role in the published article. But things got more complicated. Jason Osborne and Abigail Holland address the question in an article with the interesting title “What is authorship and what should it be?” [I], starting with the enunciation of two extremes. On the one hand, they mention an article reporting an international clinical study with more than 900 authors. Relatively few, next to the other mention: the almost 3000 authors who signed an article on the detection of the Higgs boson (the so-called god particle). On the other hand, they resemble the inventory of impossibly productive authors with 32 or more articles per year: one article every week and a half. My old postdoctoral supervisor is a Nobel laureate and, looking through his CV the other day, I discovered that his record was (at the turn of the century) to sign 10 scientific articles per year, even though he had an entire department at his disposal (not just administratively, but academically). In an ethnography by memory exercise I can say that he had authority over the construction of the content, the articulation and consistency of the various parts of that content and, finally, its textualization: he signed it and was in fact the author with authority over the article as a all. I suspect the possibility of authors with these characteristics achieving annual productivity much higher than that. The perception is that academic evaluation guidelines encourage this hyper-authorship, as well as bringing authors forward in time. 50 years ago it was expected that a physicist would begin to be an author at the end of his doctorate. At the turn of the century it became almost mandatory for the master's student to sign a scientific article. Today it is common for undergraduate students to be authors. What does such precocity entail?
"What is an author?" is the name of a conference by Michel Foucault from 1969, but the French philosopher was not considering scientific authorship, but in recent decades it has become increasingly relevant. A symptom of this relevance, in addition to the questions I pose above, is the end of the title of the article by Osborne and Holland, which I mention above: “a survey of prominent guidelines for determining authorship in scientific publications”. From a more dense point of view, I recommend the book “Scientific authorship – credit and intellectual property in science”, organized by Mario Biagioli and Peter Galison [II]. The book is divided into three parts: the emergence of authorship, the limits of authorship and the fragmentation of authorship. This announces what, when you think about it, would not be unexpected: if the scientific article changed over time, why not the very meaning of its authorship? Emergence refers to the delimitation, still in the modern period, of who could be considered a scientific author, that is, a scientist. The limits of authorship refer to more contemporary issues such as the scientific author's relationship with other non-academic actors. The simplest example, although not as thought-provoking, is the conflict between the scientist's interest in publishing and the requirement for secrecy due to intellectual property issues of those who finance the research. Another is the relationship between the canon of scientific knowledge and native knowledge. The fragmentation of authorship is largely dedicated to multi-authorship, which I dealt with at the beginning of this column. One of the chapters, by Peter Galison, is free access: “The collective author” [III]. This collective author is the “collaboration” itself, which involves, for example, the almost 3000 authors of the work mentioned by Osborne and Holland. This immense list of authors is like a petition that validates the claim of knowledge obtained through the “collaboration” declared in the article. It is part of an institutionalization of science characteristic of a small club of institutions such as CERN (European Center for Nuclear Research), whose history is for another time. For now, the questions raised by Galison in the construction of his arguments and which apply to less numerous multi-authorships are of interest. Again: in a multi-authorship, what is the role of each author? If each author has a different role, who has an overview of the work and can account for the consistency of the article and the knowledge it claims (like my former supervisor in the example above)? In other words, of all the authors who take credit equally in the citations recorded in the databases, who actually builds an apperception about the article they subscribe to? Apperception is a fundamental ingredient in the training and practice of scientists and cannot be accelerated.
Last question given the way things are going: what science are we doing and what scientists are we training?
[I] 'What is authority, and what should it be? A survey of prominent guidelines for determining authorship in scientific publications. Jason W. Osborne and Abigail Holland, Practical Assessment, Research & Evaluation, Vol 14, No 15
[II] Scientific Authorship: Credit and Intellectual Property in Science, edited by Mario Biagioli and Peter Galison. Routledge (Taylor and Francis Group), 2003
[III] https://galison.scholar.harvard.edu/publications/collective-author
