Liberal Epistemic Relativism and the Rejection of Science
A Cautionary Tale in Reaping What You Sow
This will be a two-part essay, with this first part setting out the case that the liberal embrace of epistemic relativism has been instrumental to eroding trust in science and academic institutions. I wanted to explore why this shift from positivism to relativism has occurred, as a cultural phenomenon, but this essay turned out long enough, so the whys and wherefores of the cultural shift of priorities in science and academia will be explored in a follow-up essay to this in, I expect, about a week or so.
A recent social media post by Dr Jess Steier, a prominent science communicator, provided the motivating spark for this essay. However, many of the issues I’ll articulate have been occupying room in the back of my mind for some time, as will likely become apparent. The original post concerned the wild and utterly baseless assertions that COVID vaccines have “killed thousands”, a common claim in the anti-vaccine realm, which itself now skews heavily to the MAGA/MAHA Right (although anti-vaccine conspiracies skew in both political directions). Dr Steier wrote:
“How did we end up in such different realities? ...is there a fundamental difference in how we’re evaluating the same information? I struggle to reconcile these dramatically different conclusions drawn from what should be observable reality. We’re experiencing a profound crisis of trust rather than simply a disagreement about data interpretation. Many people no longer trust traditional sources of health information–scientists, doctors, public health officials–to prioritise public welfare. And without that trust, evidence can't bridge our divide.”
The post itself was written in a manner that communicated genuine inquiry, and Dr Steier has since expanded her thinking in a measured Substack essay. For this, the author deserves credit, as her tone and reasoning stood in stark contrast to most of the public-facing realm of “science communication” - Americanised to “SciComms” - where the politicised overtones of similar posts drip with the seething condescension and aloof arrogance that typifies the manner in which American liberals1 speak about the political (read: educational) divide splitting their nation.
The comments section underneath the original post was, however, riddled with the kind of lofty self-tribute that evokes the slogan popular among American liberals that, “We Believe...Science is Real...”. Ironically, the slogan betrays Scientism as a belief system, rather than any commitment to, or understanding of, what in fact makes science “real”: positivism, objectivity, and a commitment to truth.
Yet when it comes to the interface between academia and science, and important socio-cultural issues, liberals have spent three decades denigrating and violating these very epistemic principles. Science cannot be “real” when moral and scientific relativism are the defining epistemic framework of the contemporary liberal information ecosystem, academia in particular. Liberals do not get to choose to be defenders of science when it comes to climate change or vaccines while also, for example, rejecting basic facts about biology and championing, or acquiescing to, the profoundly unethical and pseudo-scientific “affirmative care mode” for gender dysphoric youth, or turning the cheek on spurious retractions of papers on fatal police shootings because they don’t like how the Wall Street Journal is interpreting the findings.
And yet the almost-exclusively liberal world of science and academia appears to be confused as to how and why we ended up with such mistrust and, indeed, dismissal of scientific research and academic institutions. Predictably, the usual tropes have emerged, emphasising the apparent mental and moral defects of Red America, the same pattern that followed the November election. I thought the November election might serve as an inflexion point for liberals, an opportunity to reflect on the glaring defects in their political movement. But liberal intellectual culture precludes reflection, preferring to march on with a self-aggrandising assumption of moral and intellectual superiority.
The same pattern is repeating itself over the months subsequent to the election with regard to science and academic research. The Trump administration is pursuing a dangerous game of championing pseudo-scientific policies, promoting incompetent charlatans into positions of influence for public health, and swinging a wrecking ball at research and academic institutions. That this is an unprecedented full-frontal assault on science and academia is beyond question. But it is all too convenient for liberals to now claim a self-righteous position as defenders of the values that make science and academic research function - truth, objectivity, and reason - when liberals have presided over a decades-long assault on these values in academia.
That the liberal assault—characterised by vindictive intolerance, censorship of research and ideas deemed Verboten, policing of language and thought, and the postmodernisation of science—has been more insidious, and comes from within academic institutions, does not make it any less damaging to the credibility of, and trust in, science and academia. Liberals can only claim a high ground in defence of science if they can demonstrate that the core principles of the scientific method’s epistemic framework come first, before any politically partisan interpretation of an issue. The problem is that liberals can’t; they ceded the high ground by consistently demonstrating that their supposed commitment to these principles extends only insofar as these principles align with their political views.
This reflects a wider trend that has repeated itself in politics and cultural life for over a decade. Liberals choose a hill to die on, almost always related to questions of race, gender, sexuality, or other identity parameters. Rather than defend the hill with empirical rigour, they resort to relativist, subjectivist standards that mire the issue in question in absurdity, contradiction, and incoherence. In doing so, they abandon the hill, giving away the high ground they might otherwise have had on the issue. The Right realises no one is holding the high ground, and sees its opportunity to send in the bulldozers and flatten the hill. Both sides end up in the mud, shouting into the void. Yet liberals remain utterly convinced they’re still on high ground, looking down on everyone else, because none of them ever look outside their circle. Instead, they cloister together, heads buried, affirming amongst themselves that there is no mud, that they are uniquely positioned on high ground. Meanwhile, the bulldozers move on to the next abandoned hill.
The proverbial hill in this analogy—the high ground—is epistemic. The foundations of the rigour in the sciences, our academic institutions, and indeed our legal systems, are grounded in the specific ontology and epistemology of realism and empiricism, respectively. Other ontological and epistemic frameworks exist, but none exhibit the utility of realism and positivism for advancing knowledge, testing theories and ideas, and understanding the world we inhabit.
Subjectivity may be useful to add a human voice and perspective to knowledge, but subjectivism contains an inherent trapdoor that is all too easily fallen into; the relativist fallacy, where some characteristic of the identity of an individual speaker, or a group, is taken to imply the truth of their arguments or claims. In a positivist framework, knowledge is independent of any such characteristics; to use the common trope, E=mc2, whether you want to believe it or not, whether you’re Black, Brown, or a stale Pale Male.
The meaning of ceding the high ground is the liberal embrace of relativism and the long march of moral and epistemic relativism through our academic institutions. It is beyond contest that academia is a sphere of public life overwhelmingly dominated by Left-leaning liberals (to such an extent that an unintentionally hilarious recent article in The Atlantic talked of "DEI for conservative academics”). This results in a cloistered, self-reverential ecosystem in which heads are buried from other contesting perspectives and ideas, and from the real world itself. Uncoupling ideas and ideologies from the need for contact with reality is a recipe for bad ideas and absurd ideologies to develop and disseminate. And when academic ideologies are institutionalised, they filter from the dusty corridors of the Ivory Tower into wider society in the heads of graduates, and into the cultural discourse.
Did science and academia walk itself into this “profound crisis of trust”? I want to circle back to this concept of “SciComms” because the culture in this corner of the internet provides some answers to its questions in this respect. As cultural products of academic institutions, the public-facing SciComms community provide a barometer for the culture of those institutions, which are taken out into the wider discourse. And as an interface between science and the public, SciComms provides a stress test of the degree to which the aforementioned principles are adhered. It is a test that I think this world fails.
“SciComms” or PolitiCISedComms?
If the term “SciComms” is news to you, we should define what this concept describes. Broadly speaking, “science communication” involves the presentation and dissemination of scientific concepts and research to the public. This may be directed towards other academics within a specific domain of science, other scientists in non-related fields, or, more commonly, the lay public. SciComms encompasses various media, including social media platforms, podcasts, and ancillary publications, such as Nature and the Scientific American magazines. As a disclosure, I am part of this world, co-hosting a nutrition science podcast while maintaining a modestly-sized social media profile. Thus, what follows here is not intended to be mere finger-pointing; I recognise that I have been part of this culture and how it operates.
Over the past year, accelerated in earnest by the response of the social media SciComms community since the U.S. election, I’ve become increasingly uncomfortable with the barely-veiled politicised nature of SciComms rhetoric, dressing contested issues up as objective “truth”. Much of what I’m writing here stems from my reflection on this awkward interface between science, SciComms, and the politicised “public square” of social media. The SciComms world leverages the perceived authority of science to portray contested social, cultural and political questions as matters of settled fact according to the liberal position on such issues. TheScience™ thus becomes the legitimator of the liberal worldview, from immigration to gender to the war in Gaza. Consider the following graph, which was shared widely on social media by prominent SciComms accounts as some “proof” that Republican voters acted out of misinformation.
Every one of these questions was based on highly technical points that were arguably framed precisely to make Republican voters look “misinformed”, particularly given the substantial issue of liberal bias in social studies. For example, the bottom statement played on a temporally acute technicality in the wider context of four years of historic levels of immigration, most of it illegal, and the “lowest level in the last few years” was a specious distinction in the context of levels of border crossings that were orders of magnitudes greater than at any point in the Bush, Obama, and Trump 1.0 administrations, which you can see in the figure from The Economist below. The same was true for the question on inflation, a deceptive manipulation of the reality of the surge in inflation during the Biden administration, which negated much of the GDP growth in that period, perpetuating economic pain in large sections of society. The framing of the questions was so obviously disingenuous, but a good example of how American liberals, reinforced by the SciComms community, play on data to explain away and delegitimise the realities people that they don’t like experience on the ground.
By trying to catch people out on technical details, it made it appear that anyone losing ground in the economy, or concerned with historic levels of illegal immigration, or concerned with violent crime in cities, was somehow “misinformed” and disentitled to that view. Which, of course, was the entire purpose of the exercise: to strip those voters of their agency and frame the election as another liberal purity test. “We lost, but we’re still right about everything.” The rigour and framing of the survey were irrelevant; all that mattered was that the answers confirmed a preconceived view of Trump/Republican voters, and sharing it widely would be met with clapping hands and fire emojis in the DMs. “Isn’t it so good to be self-evidently correct on every issue? And look—TheScience™ says we’re right, too. I can sleep at night knowing that the problem in our politics is that everyone but me is stupid, because We Believe Science is Real!”
In the context of the politically-charged nature of the Trump administration’s assault on science and academia, one might be tempted to defend liberal-leaning politicised science and academia by suggesting that this is a “false equivalence” or that “all science is political”. Both of these positions are specious. In the first instance, it is not a false equivalence to focus on liberal politicisation of science and academia while Trump’s assault rages on, because the Trump administration’s assault is itself a response to the perception on the Right that science and academia have been compromised by liberal bias. Reactionary politics need something to react to. The question is whether there are seeds of truth to the causes of the reaction, not conveniently taking the reaction itself as grounds to dismiss that question.
In the second instance, “all science is political”, a view popularised in the SciComms community (Nature and Scientific American being two cringeworthy examples), is naïve because it fundamentally misconceives the basis upon which the integrity of science and its institutions are built: impartiality, transparency, and accountability. Our legal systems and political institutions are grounded on the same fundamental principles. Democracies function on the expectation that institutions of state, which include academia, exercise impartiality, conduct themselves transparently, serve the public over partisanship, and may be held accountable. For science and academia to have any intellectual or moral high ground requires adherence to these principles to be demonstrated. If political bias and partisanship can be demonstrated instead, trust in science and these institutions is eroded. And it is hard to win back.
To win that trust back, however, requires liberals to take a cold, hard look at the epistemic environment they have cultivated, and see whether the epistemic walk matches the lofty scientific talk. Yet there is a distinct absence of one characteristic crucial to the integrity of science and its dissemination: humility. The lack of humility isn’t unique to SciComms; humility is almost entirely absent from contemporary liberals, who have instead declared themselves paragons of virtue and built a politics around Moral Purity. The culture among liberals is thus: “We’re right...even when we’re wrong. And yes, mistakes were made...but not by us.” Humility is a prerequisite to self-reflection, and little self-reflection is evident among contemporary liberals.
And despite the self-reverential manner in which SciComms positions itself, speaking to the values of openness, dialogue, and objectivity, when I’ve raised the topic of liberal bias with fellow liberals in the SciComms realm, with few exceptions, it has been met with the typical trajectory of conversations about issues within the liberal sphere: Deny, Equivocate, Excuse, Deny. It is met with either summary dismissal, “...but the Right tho...”, or outright hostility.
The “Two Cultures”
The problem isn’t that SciComms is riddled with dishonest actors. On the contrary, I genuinely think, and indeed most science communicators display, honesty and genuine intent with what they disseminate to the public. Rather, the problem is threefold; the first being the nature of the relationship between communicator and audience in the context of the liberal information ecosystem; the second being the clash of science and cultural commentary; the third being epistemic relativism and the glaring absence of addressing pseudoscience and bad academic practice from within ones own tent.
Regarding the first issue, when a science communicator develops trust with their audience, their audience leans on them as a trustworthy source of information on all manner of topics. This may come from a well-intended place, but it has the effect of positioning the communicator, whether they intend it or not, as somewhat omnipotent. This parasocial dynamic also engenders a feedback loop between the interests and political persuasion of the audience (i.e., overwhelmingly Left-leaning liberals on platforms like Instagram or BlueAnon) and the content covered—or not—by the communicator. The atmosphere thus has the appearance of openness and discourse, but lying in wait are all the toxic, censorious traits of the liberal information ecosystem. The result is a form of self-censorship where, to take the gender medicine example below, no prominent science communicator could ever post anything critical about this pseudoscientific discipline; they would immediately become “problematic” and rouse Moral Outrage for having transgressed against this most Sacred Cause of the Progressive Left.
The second problem goes back to C.P. Snow’s 1959 book, The Two Cultures, in which he articulated the polar split in Western intellectual life between the arts and humanities, and the sciences: “Between the two, a gulf of mutual incomprehension...but most of all lack of understanding.” Snow saw literary intellectuals as viewing scientists as “ignorant specialists” with little grasp of the core matter of the humanities and, consequently, “unaware of man's condition.” Conversely, scientists viewed literary intellectuals as “totally lacking in foresight” in whom “their own ignorance and their own specialisation is startling.” I am a product of both cultures—my undergraduate degree was in history and English literature, followed by a law degree, and then an MSc and PhD in nutrition science—and sympathise both with the degree of truth in Snow’s observations, and with his lamentation that this polar split deprives us of enriched understanding that could come from more cross-pollination.
Snow’s polar concept holds some merit; scientists have very deep domain-specific scientific knowledge, but, with a few exceptions, a narrow range. So when SciComms steps into the role of broad socio-cultural-political commentary, but without any real depth of understanding for such topics and armed instead with a few selective quantified metrics, it can be an awkward, uninformed fit. This was on display after last November’s election, when the SciComms world seemed to think it could understand the election through some crude indicators, with little to no historical, sociological, or political context. As if GDP can capture the sheer scale of inequality and the psychological impact of status disaffection; as if employment figures negate the deep resentment and shattering of community life in the Rust Belt. Bear in mind that the same indicators told liberals that Brexit wouldn’t happen, Trump 1.0 wouldn’t happen, and Harris would Walz into the White House. Data in the absence of context is all too easily ignorance dressed up as foresight.
Yet the more that liberals position science as part of their worldview, the more alienating the entire enterprise becomes. A new “two cultures” has emerged, not between spheres of intellectual life, but between educational status and its strong ties to political leanings. College-educated (especially White) women skew overwhelmingly liberal and Democrat, while (especially White) men without a college degree skew overwhelmingly Right and Republican. I would estimate that ~8 out of 10 major SciComms accounts are White, American, highly educated women, most of whom can barely conceal their contempt for the Other Side of America that shares their pigment but lacks their level of education.
In this new “two cultures”, one educated and liberal and the other uneducated and reactionary, liberals imbue science with moral qualities and moral conviction reflecting their social status; if liberals could just explain to the Plebs what TheData™ says about their life, rather than what their day to day experience tells them, the Plebs would see the light, and the self-evident truths of the liberal worldview would be revealed to them. Hallelujah! The reality is that the sense of a vertical imposition of edicts from a self-appointed intellectual elite drives anyone less educated away. To be fair, I’ve had a decent education, and this kind of arrogant, didactic liberal scientific-moralising makes my skin crawl, too.
The Worm at the Core
But the third issue is the most substantive for the loss of trust in science and academia, which is the fact that for these institutions to hold legitimacy, they need to demonstrate that they exercise impartiality, transparency, bipartisanship, and accountability. The entire purpose of the scientific method is to account for and attempt to minimise human cognitive biases and fallibility, and this equally applies to political leanings. This means that in practice, liberals must be seen to be calling out liberal bullshit and pseudoscience where they see it. Instead, the SciComms world only defends these principles when it is the political Right/MAHA that breaches them.
As the most egregious example, consider that not a single prominent SciComms platform—not one—has held to account the biggest medical scandal of the past 50 years, namely the “affirmative care model” for gender dysphoria.2 A model that constitutes a violation of the most basic tenets of medical and scientific ethics, that a modicum of adherence to ethical principles and fundamentals of evidence evaluation would reveal, has somehow either escaped comment by most of the SciComms community or has been actively aided and abetted by members thereof. Revealing the depths of the unethical and pseudoscientific basis of this model has been left to investigative journalists, such as Hannah Barnes and Jesse Singal, because academics cowed from their ethical and scientific duty.
One of the world’s most prestigious medical and scientific journals, the New England Journal of Medicine (NEJM), has a worryingly consistent run of form in throwing its weight behind this pseudoscientific model, revealing some troubling conflicts of interest along the way. It has published several primary papers in which the claims of the paper are simply unsupported by the data therein, which in turn become central to the claim that the model is “settled science”; the Journal of the American Medical Association (JAMA) has shown the same course.3 In one example that beggars belief for publication in a journal like NEJM, the authors altered their hypothesis from their preregistered protocol and selectively reported only two out of eight preregistered outcomes, excluding the outcomes of most interest to the prevailing theories underpinning the “affirmative care model”. NEJM then recently published an editorial that amounts to a hit-piece that contains not just opinion, but outright falsehoods and misrepresentations about the methodology and findings of the U.K.’s Cass Review.
The implication is clear: if the social cause to which the research relates is a liberal shibboleth, then studies that would not otherwise have been published in a rigorous journal get a hall pass. The obvious, uncomfortable question then begs: if NEJM and JAMA are willing to abandon their basic commitments to scientific integrity to favour a political cause dressed up as a medical model, can you trust the research published on other medical and scientific questions? What about vaccine research? Or statins? I personally don’t think that the “gender medicine” bias necessarily implicates all other areas of research published in these journals, but I certainly have sympathy with how it looks, and how it looks is utterly corrosive to the trust and perceived integrity of science. Just imagine that any of this came from the Right; the SciComms community would be relentless posting and publishing, positioning themselves as front-line troops on the barricades in defence of science and truth. Yet when the rot comes from within the liberal tent, they are nowhere to be found. Epistemic relativism at its worst.
This is political bias, nothing more, and it pervades this entire sordid episode for science and medicine. When it is revealed that authorities like the American Academy of Pediatrics, American Psychological Association and the Endocrine Society are engaging “policy-based evidence making” underscored by circular referencing to non-evidence-based “best practice” guidelines generated by World Professional Association for Transgender Health, is there any public call out of this manufactured, false consensus? When Carole Hooven, an evolutionary biologist, is forced out of Harvard for stating basic facts about biology, or the research of Michael Bailey or Lisa Littman is retracted for little justifiable reason other than the findings are considered Verboten to the model's unfalsifiable premises, is the SciComms world leaping into the public fray to decry academic censorship and defend academic freedom and transparency? When Kathleen Stock faced death threats and required police protection at the University of Sussex before being forced into early retirement, was the SciComms community manning the barricades in defence of academic integrity? The answer to each of these questions is a resounding “No”.
When a paper in the prestigious Proceedings of the National Academy of Science found no evidence of racial disparity in fatal police shootings (it found that the fatally shot individual was more likely to correspond to the race of the shooting officer), it was retracted because of the potential for “misinterpretation and partisan political use of a scientific article after its publication.” The paper contained no methodological or statistical errors, nor was it based on fabricated data, which would warrant a retraction. The primary “error” related to their choice of language in the significance statement that accompanied the paper. But in reading the justification for retraction, the motivation is plain: how the Right-wing press interpreted the findings. In other words, the findings were subjugated to the social cause to which the study related, judged through a politicised lens. The SciComms community took no issue with this against academic censorship.
In fact, when it comes to several issues concerning liberal social justice shibboleths, SciComms outlets like Nature and the Scientific American have not merely been passive bystanders, but actively aid and abet the dissemination of relativist, pseudoscientific views. Again, there is a simple test to apply here: imagine if conservative or Right-wing journals and periodicals engaged in the very same behaviour? It would feed the We-Believe-in-Science-Outrage Complex for weeks.
Perhaps the most public example of overtly politicised bias in the liberal world of academia came during COVID-19 lockdowns when, with the surge of Black Lives Matters protests, public health academics and officials U-turned on their heels with statements like, “the public health risks of not protesting to demand an end to systemic racism greatly exceed the harms of the virus." If social causes exceeded the harms of the virus, why not protests against closing the economy, a major harm that arguably also exceeded the harms of the virus? All that distinguishes the two is value-laden judgment calls. The U-turn undermined public trust in the objectivity and integrity of science and public health officials by demonstrating that lockdowns always contained trade-offs between viral spread and social factors, i.e., judgment calls that did not come down to evidence. The justifications for the U-turn were largely specious references to non-evidential matters that reframed the debate away from COVID-19: the interface of scientific and moral relativism.
I’ve painted a brief picture with the foregoing examples; now imagine how it looks when those examples are fleshed out and fully ventilated, particularly by Right-wing Shock Jocks. It reeks of ideologically compromised institutions, ideologically compromised journals, and sketchy medical authorities. It appears like a fundamentally untrustworthy system that is compromised by the overwhelming political leanings of the people running these very institutions and systems. Yet there is a systematic failure among liberals to get to grips with the pathology under their own roof. And the Trump administrations wild, incoherent and damaging attacks on science and academia now provide the perfect excuse for liberals to close their eyes and pretend there is nothing to see, that there was never anything to see, that could explain why so many to the Right of centre have come to reject science and academia as politically compromised.
Science, Academia, and Institutional Failure
As a final point for this first part, but a crucial point nonetheless, consider that the loss of trust in academic institutions does not exist in a vacuum. The sense of mistrust in science and academic institutions comes, temporally, after the other institutions that used to protect citizens not only failed to safeguard them, but took an active role in the destruction of the post-World War II social contract, from banks and financial institutions, to educational institutions, to the institutions of state.
To whole swathes of American society, the swathes that liberals scorn and denigrate, life for the past two to three decades has consisted of off-shored jobs, home foreclosures, roadside bombs in Iraq, opioids and despair. And liberals just rub their noses in all of it, while wondering why there is a breakdown of trust in institutions, particularly those considered part of the amorphous “elite” and dominated by liberals—those institutions now under attack by the Trump administration.
As it relates to science and academia, there is culpability here that liberals need to face. To quote a recent David Brooks essay in The Atlantic:
“The left really did purge conservatives from universities and other cultural power centres. The left really did valorize a “meritocratic” caste system that privileged the children of the affluent and screwed the working class. The left really did pontificate to their moral inferiors on everything from gender to the environment. The left really did create a stifling orthodoxy that stamped out dissent. If you tell half the country that their voices don't matter, then the voiceless are going to flip over the table.”
And the deeper reality that underpins the politicised bias of the liberal-dominated worlds of science and academia is that a cultural shift, which was initially confined to the humanities, eventually metastasised into science. This epistemic Cultural Revolution subverted the principles upon which the scientific method is predicated with a political ideology masquerading as an epistemic framework. We’ll examine that premise in the follow-up to this essay.
For this essay, the term “liberal” is used in its generic American sense of “left of centre”, without necessarily delineating between more traditional centre-Left liberals or the far Progressive Left. However, while almost all of the harebrained ideological nonsense referred to in this essay emanates from the Progressive Left, the failure of “common sense” liberals to stand up to this has resulted in the lunatics running the Leftist asylum. So, implicating “liberals” in the broadest sense of the word is intended and, as will hopefully become clear in the essay, justified.
It should be noted that this is now an exclusively American pathology. Any countries that have conducted independent systematic reviews of the evidence for “gender medicine”, including the U.K., Finland, Sweden, and Norway, have all arrived at the same conclusion: that the evidence is too weak to support the model, in particular, the early medicalisation of dysphoria. It is not coincidental that these countries all have public health care systems that are grounded in a duty of care. America’s health care system is grounded in a duty of profit, which renders dubious both the refusal of American medical authorities to conduct an independent systematic review, and suspicious the collusion of various bodies and journals like NEJM in manufacturing a false consensus regarding the “affirmative care model”, and in attacking, with falsehoods, the U.K.’s systematic review. American liberals frame this as a matter of gender dysphoric youth benefiting from the model; the real beneficiaries of a system of lifelong medicalisation in the privatised U.S. healthcare system are clinicians.