What could be further removed from scientific research than bodybuilding?

The researcher spends most of his time in a laboratory, pushing his mind to the limit to build knowledge, while the bodybuilder spends most of his time in a gym, pushing his body to the limit to build muscle.

To put it simply, one is all muscle power, while the other is all brain power.

Yet, in this article, I aim to show you that scientific research has undergone, over the past few decades, an evolution similar to that of bodybuilding a little over a century ago.

From strong men to… beautiful men?

It was the turn of the 19th and 20th centuries when a certain Eugen Sandow conceived the idea of developing human musculature in a way that went beyond a mere display of strength, replacing it with an ideal of beauty.

In doing so, he invented bodybuilding.

Initially, however, Eugen Sandow was part of a sporting tradition known as “strongman,” but his approach consisted precisely in moving away from the primary function of muscle—namely, performing a useful movement (walking, running, lifting, throwing, etc.)—in favor of an aesthetic function above all else. This idea came to him from his fascination with the Greek and Roman statues that had filled his childhood in Italy, even though these statues actually depicted athletes in action.

From its inception to the present day, bodybuilding has continued to follow this same vein,1 to the point where some bodybuilders today seem to have difficulty walking normally or performing certain everyday tasks (such as scratching their backs, for example). Their mobility is, in fact, hampered by musculature that serves no other purpose than to be admired. To put it another way, modern bodybuilding perfectly illustrates the supremacy of form over function, of aesthetics over utility.

Before bodybuilding, it was understood that muscle development was a byproduct of sustained physical activity. As a result, “measuring” a muscle (its size, shape, and location) was of little interest, since the best way to assess its “value” was to observe its usefulness in one of the physical activities mentioned earlier.

It should be noted that, initially, Sandow was likely aware that the aesthetic aspect of musculature was linked to actual physical activities, since he drew his inspiration from statues depicting ancient athletes, such as the Discobolus. However, bodybuilding severed this link long ago: the aesthetic measurement of muscle is now the sole criterion for evaluating this discipline. In a sense, measurement has replaced what it was supposed to evaluate: athletic performance.

Predictably, this shift has led to abuses, such as the use of substances harmful to health—like anabolic steroids—for the sole purpose of accelerating and amplifying muscle mass gain. This is one of the well-known consequences of Goodhart’s law, which states that when a measure becomes a target, it ceases to be a good measure.

Or to put it another way, as soon as the method of measurement is known, there are always sly dogs who find shortcuts to optimize the measurement, even if it no longer reflects the original intention.

The nature of science or the science of nature?

Following a similar trend, we have seen in recent decades a shift from scientific publication as a vehicle for disseminating new knowledge to scientific publication as an end in itself. In other words, scientific publication, which was initially merely a means to an end, has replaced the original objective, and now it is the publication itself that takes center stage, while its content is ultimately of secondary importance.

Thus, in many contexts—such as rankings of researchers and academic institutions, securing research funding, obtaining an academic position, or earning a promotion—the number of publications and their citations has become the standard measure of the value of research, to the detriment of its relevance, originality, or utility outside the academic world.

So much so that some promotion or research funding committees no longer even read the content of candidates’ scientific articles, but rely solely on statistical measures such as the h-index, for example. In doing so, a growing portion of the scientific community seems to have forgotten that the best way to evaluate research and the scope of its contributions is still to read the scientific articles that describe it.

In a way, just as a bodybuilder engages in “body building” throughout his “career,” refining the appearance of his most photogenic muscles, today’s researcher engages in “resume building” by ensuring that his publications align with the trendiest topics of the moment. With the rising tide of AI, for example, we’ve seen a flood of research articles spiced with machine learning in just about every field.

Once again, dangerous trends that threaten the future of research are on the rise. Thus, the less disruptive an article is—and conversely, the more it aligns with what the community already thinks and has published, while extensively citing previous research—the more likely it is to be accepted and published. And with the advent of LLMs and their extraordinary writing capabilities—much like anabolic steroids in bodybuilding—we’ve shifted into high gear: not only are papers being written with AI, but we can also use it to summarize them, have them reviewed, and create bibliographies (related work), even though the phenomenon of reference hallucinations is increasingly becoming an obvious issue.

A glimmer of hope?

Given this reality, one might wonder whether scientific research is headed for the graveyard of the “Dead Internet,” just as other types of content once created and consumed by and for humans are said to have already done. As a reminder, according to this theory, most of the content available on the Internet is now exchanged between artificial intelligence-based entities rather than between humans.

Coming back to bodybuilding and taking the analogy a step further, there may still be a glimmer of hope: in response to the rise of a hyper-standardized and soulless approach to fitness, a trend emerged in the early 21st century known as street workout.

Although its popularity exploded starting in the 2020s—partly due to the COVID-19 lockdown—its resurgence in the early 2000s stemmed from a desire to achieve physical feats (often likened to superhuman feats), such as the human flag or the front and back lever, for example, rather than a desire to satisfy a purely aesthetic and narcissistic pleasure.

By analogy, we can hope that our curiosity and our need to develop new ideas—two deeply human traits—will ultimately compel us to thoroughly revisit the current research model, which has itself become hyper-standardized, increasingly reliant on AI, and prioritizing quantity over quality.

  1. No pun intended regarding the often-prominent veins of today’s bodybuilders. ↩︎