Portal de Periódicos da CAPES

The Decline of the Academic Auteur

2020; Elsevier BV; Volume: 2; Issue: 3 Linguagem: Inglês

10.1016/j.matt.2020.01.025

2590-2393

Steve Cranford,

Academic Publishing and Open Access

In early October last year (2019), the acclaimed filmmaker Martin Scorsese was interviewed by Empire magazine, where he claimed that Marvel films do not qualify as cinema, and they are more akin to "theme parks". This resulted in a social media firestorm and an extended op-ed piece in The New York Times (NYT) in November, where Scorsese doubled down on his stance—mostly on his own definition of "cinema". As an amateur film buff and Marvel fan, but also a professional editor and publisher, the NYT piece and online discussions resulted in a subconscious mish-mashing of the state of film and academic publishing in my mind—for better or for worse. While one industry is focused on subjective entertainment and the other focused on objective science (hopefully, it's obvious which one is which), they both ultimately have a similar challenge—disseminating works (film and research articles) to the widest possible audience. This, however, is very much the top-down goal of the industry, not the primary motivation of the creators—the individuals who produce the work. Ideally, scientists enter their field due to innate curiosity about the workings of nature and the universe or with some ultimate societal problem or challenge to solve. Publications and citations are a byproduct of scientific output. Likewise, filmmakers typically want to tell stories and express ideas through images on the screen, extracting an emotional response from the audience. Awards and box office receipts should not be the primary motivation. Is this too idealistic a perspective to hold today? Can there be a balance between motives? Scorsese believes the best film, or *ahem* "cinema", arises from "auteur theory". The French word auteur literally translates to the English "author". Within the context of cinema, the word auteur is used to describe a director who exerts a high level of control across all aspects of a film to reflect their own artistic vision. Auteur directors generally have a distinctive style across their filmography and often fill other roles besides directing, including writing, editing, and even acting. It is frequently argued that the best filmmakers are those who write and direct their own works and have a unique, personal vision. Think of Steven Spielberg, Stanley Kubrick, Francis Ford Coppola, Christopher Nolan, Quentin Tarantino, John Carpenter, Wes Anderson, or Spike Lee—they all have such of unique style all onto themselves. You know their names because their films are successful, award-winning, and influential. But can you rank them by box office gross? The academic equivalent is the independent and successful professor. Running a small group in an elite institution, they publish seminal works regularly, are well-respected, world-renowned, well-funded, and given the academic freedom to pursue interests of their choice—the academic auteur. The ideal academic auteur is as fictional as The Incredible Hulk. The prevailing structures of academic career advancement mean the biggest rewards often follow the flashiest work, not the best. Like movies with the most advanced special effects, all journals—from Cell, Nature, Science, and Matter to smaller niche society titles—love articles that result in the largest net readership. Increasing the number of readers increases the number of citations, and in our film/paper analogy, citations are the box office gross. A journal's impact factor, while stretched over an (arbitrary) 2-year span, is about as important as a film's opening weekend gross. Impact factor says something about the popularity of the work, but absolutely nothing about the quality. Now, a quality paper can become highly cited because it is good science—or because it is eye-catching. The Academy Award for best film has been given to popular movies (although a rarity). Likewise, the Nobel Prize is not given out to those with highest number of citations. Like a film's box office gross, citations influence the science that scientists do. I went on SCOPUS and did a little investigating. Rather than a field-specific journal, I choose science as a representative publication. I think we can all agree it is a reputable publication—akin to the Universal/Paramount/Warner Bros./Disney major studios. While publishing since 1880, I chose 1976 as my sample year (just for fun, the year Scorsese released Taxi Driver). I then checked the top 20 most heavily cited papers. Six papers had only one author, eight papers had two authors, four papers had three authors, a single paper had four authors, and one outlier had five more authors (a high 36). Considering the common practice of progeny plus PI as co-authors, the "auteur" PI's (single or two-authors) accounted for 70% of the top 20 articles. Neglecting the outlier (36), the average authors per paper was exactly 2.0. Flashforward to the "paper-by-committee" era—from the same process of the 20 most highly cited Science articles in 2018 resulted in only four articles with six or fewer co-authors (only a single paper with two authors, and no single-author papers). Another six articles had between 10 and 20 authors, while seven had between 20 and 50 authors. One paper listed 71 co-authors. For those keeping track, that's only eighteen total papers. The final two outliers in this case had 200 and 571 listed authors respectively. Ignoring those two extremes, the average number of authors (for the 18 papers) is about 22 (rounding up for complete authors). While far from a complete statistical analysis, the two selected years tell a compelling story. In this issue of Matter, for example, there are 10 research articles. The number of authors is: 4, 5, 5, 8, 8, 10, 11, 12, 15, and 20. The average number of authors is 9.8. Five of the articles have at least 10 co-authors. Reflective of the times, there are no single or co-authored papers. The minimum number (4) is the output of a theoretical/computational study (by Haji-Akbari and colleagues), and such papers are known to require less personnel than experimental efforts. What does this tell us exactly? Nothing about the science itself. Good science can clearly be done by an individual or a consortium. But it does suggest how science is performed and disseminated. Clearly, there is a trend from an individual—the classic portrayal of a stodgy professor in a lab—to science by committee. The growing size of collaborations is science has been evolving in recent years. Combining resources can result in more rigorous, complete, and groundbreaking science. It is difficult for a single research group to provide deep expertise in materials synthesis combined with computational simulation and in operando characterization. Big science problems require ever increasing effort (and associated funding). Even more fundamental problems are tackled by consortia, resulting in the previously mentioned Science publications with hundreds of co-authors. Every contribution counts. This has also led to more cross-disciplinary efforts, where one field can draw expertise from another, resulting in novel findings and applications. Transdisciplinary trends necessitate a more open and accepting groupthink mentality, enabling the best ideas from two (or more) specialties to produce a new idea. Modern science takes a village. It's a beautiful sentiment. Scorsese laments that true cinema is "about characters—the complexity of people and their contradictory and sometimes paradoxical natures, the way they can hurt one another and love one another and suddenly come face to face with themselves." I'm sorry, but while Travis Bickle and Bill the Butcher have nuance, Thanos and Tony Stark are far from one-dimensional characters, and Steve Rogers is more than America's ass. Similarly, in science, sometimes the best original ideas are produced by committee, emerging from a set of small contributions. While one can consider a film's auteur as the creative driver, it somewhat overlooks the small army of personnel necessary to make a movie. According to the internet movie database (IMDB), the film Goodfellas has credits for 288 individuals, from executive producer to craft services. Academic studies with numerous contributors typical only give one credit: co-author. Perhaps academia should delineate contributions beyond first author, corresponding author, and those in between. But that is an argument for another day. Another feature of group-think science is that is effectively mitigates risk. When Scorsese reflects on the risk-averse "nature of modern film franchises: market-researched, audience-tested, vetted, modified, revetted and remodified until they're ready for consumption," he could also be referring to a modern science manuscript – funded-research, audience-tested, reviewed, revised, re-reviewed and re-revised until ready for publication. While the accolades for an important study is proudly shared, the disappointment of a less-impactful work, or, even worse, so-called "incremental" findings, can be diluted across multiple individuals. Just like in finance, if you hedge your bets and diversify your research portfolio, the continuous gains of successful projects can easily absorb the losses of a few side projects. This results in the highly networked scientist/academic, involved in numerous multi-university grants and centers, seemingly presented work at every conference across diverse disciplines, and publishing across a spectrum of journals and fields. A modern Renaissance man is the hub of academic connections, not a recluse polymath behind a library of obscure texts. Gone are the days where an acclaimed researcher can be a rogue thinker, taking long walks on campus grounds to gather thoughts for the next single-authored breakthrough. It happens, but it is the exception to the new status quo. This is not to stifle independent thought. Risk is a good thing in science. Bold, original ideas drive fields forward. There must be a balance between leading and following, iconoclastic and cooperative, original and derivative, etc. But what about the success? As of the writing of this work, according to Box Office Mojo, Scorsese's 24 feature films have earned a lifetime gross of over $1.9 billion. The Marvel Cinematic Universe (MCU), in comparison, has released 23 films from 2008 (Iron Man) through 2019 (Spider-Man: Far from Home). The penultimate film Avengers: Endgame grossed $2.5 billion. In total, the MCU is the highest-grossing film franchise of all time, having grossed over $22.5 billion at the global box office. That's 12 times Mr. Scorsese's gross. If box office sales were equated to citations or H-index, Scorsese is still at the assistant professor level. He would have difficulty getting tenure at a top research institution. Lucky for Scorsese (and assistant professors), there is more to film (and research) than box office gross (and citations). I was just making a comparison on a single metric. Of course, Scorsese first made a name for himself in the 70's, when the top grossing film of the year (Rocky) made about $120 million. Major films today exceed that amount in opening weekends. Taxi Driver, for those who are interested, placed 15th that year and grossed approximately $30 million upon release (in 1976 dollars; adjusted, the gross is on the order of $120M). Times were different in film, just like times are different in academia. One can't thrive in academics as an independent creative, producing a seminal paper every few years. Even if it is The Irishman. It's possible, but not encouraged. In Scorsese's NYT piece, he states "Cinema is an art form that brings you the unexpected. In superhero movies, nothing is at risk…" There is a clear parallel between the academic auteur (cinema) and research-by-committee (superhero movies). From a very cynical outlook, one tells a single-minded story with focus and passion, while the other assembles a plot with known box-office-performance components, lacking emotion beyond the merchandising opportunities. The newest generation of researchers must balance risk with reward, contribute across fields, and understand their own value. A small role in a larger scheme is no less worthy than an autonomous study. Solid work is always valued and recognized, but sometimes, great works require collective minds, connecting stories beyond an individual. There's plenty of room for popcorn blockbusters and Oscar-winning darlings, both equally valid forms of film. While the current state of research is explicitly collaborative, resulting in the decline of the academic auteur, I'd rather live in a world with teams of academic superheroes.