We Need to Rethink First-Author Culture
The undue importance attached to first or "lead" authorship is one of the more absurd features of modern academic culture. In theoretical high-energy physics, this obsession has largely been abandoned because intellectual contributions are often distributed across a collaboration, and author order is a poor proxy for who did what. A paper is not a rat race in which the first name gets the prize.
Even in fields that do not, or only loosely, follow alphabetical author order, first authorship is often partly a bureaucratic convention. PhD students may be given first authorship because institutional or career requirements demand "first-author papers" for a thesis, degree, or future postdoctoral applications, even when such requirements are not explicitly stated. In the typical setup involving a PhD student, a postdoc, and faculty, the chances of a postdoc getting a first-author paper can therefore be greatly diminished. Their contribution may nevertheless be more important to the success of the work than that of the nominal first author. Yet simplistic metrics reduce everything to a position in the author list. It is especially strange when a postdoc gets less credit despite shaping the project, developing the numerics, and mentoring junior researchers.
This seems particularly problematic in India, where schemes such as the National Post-Doctoral Fellowship (NPDF) appear to place undue importance on first authorship. In fields such as general relativity, quantum information, computational physics, and theoretical high-energy physics, being a second author during a PhD can represent substantial, sustained, and genuinely independent intellectual work. The position in the author list tells you little about the depth of that contribution.
Academics on selection committees increasingly seem unable to evaluate contributions beyond their expertise and instead fall back on paper counts, h-index, and numbers of first-author or single-author papers. Perhaps the sheer number of applicants makes their job difficult. But these practices create perverse incentives and encourage gaming of bibliometrics through citation cartels and salami slicing.
More importantly, the battle for the first-author slot discourages collaboration, first among postdocs at the same career stage, and then between postdocs and PhD students. A postdoc may think twice before collaborating with another postdoc, or taking on a novice PhD student, if sharing the work could mean sacrificing a first-author paper that matters for their career.
That undermines mentorship, collaboration, and the transfer of expertise. It is a disaster waiting to happen in my narrow domain of theoretical gravitational physics, and in several disciplines, it has arguably already happened.
N.B. I leave you with a proposed solution: use the principle of superposition and make everyone "first author". And there is even a LaTeX package for this!
Erik D. Demaine and Martin L. Demaine, Every Author as First Author, arXiv:2304.01393