If one scans the publicity pages of university websites, or the science reporting that stems from the same source, one finds an unending series of articles of the type “study finds X is linked to Y”, where X is some aspect of everyday life, and Y is some supposedly desirable or undesirable outcome.
The claims range from the blatantly obvious (countless variants of “poverty leads to ill health”), through the inscrutable (many diet and cancer studies), to the hopelessly confounded, but politically convenient (“climate change leads to [every bad thing you can possibly think of]”). One very rarely sees a claimed result that is both striking and seriously convincing.
The claims range from the blatantly obvious, through the inscrutable , to the hopelessly confounded.
Moreover “linked to” are weasel words. To the layperson, they appear to state that a causal link has been established, and publicists expect exactly this association to be made, particularly as they know few will read beyond the headline and the first paragraph. Those who do will find, almost invariably, that the researchers are not claiming a causal link, but merely stating that they have established a statistical correlation, deemed “significant” (this is a term of art) by some agreed disciplinary standard, that may serve as a prompt for “further research”. Has any grant-dependent scientist ever concluded from his results that he has been going down a blind alley, and further research is not needed?
Most of this vaunted research is garbage. It either affirms things for which there was no need to carry out research, or it adds a dot or so to an ever-expanding landscape of blurry “links”, where no chain of causality is ever firmly established, and so no sharper understanding of the world is gained. A lot of the research is also garbage in the more technical sense that its producers have falsified their data, monkeyed with their statistical techniques, or deluded themselves into extracting a signal from noise. There have been attempts for decades to curb these practices, and that has perhaps led to less very bad work being done at what are meant to be very good universities. But it’s by no means certain that even this has been achieved, and if there were fundamental zeal for reform, most of these “linked to” studies would not be published at all, still less trumpeted. Yet their number grows significantly every year.
If there were fundamental zeal for reform, most of these studies would not be published at all.
Why then does this circus continue, and continue to expand? For a start, there are simply a huge number of universities, most of them with their cohorts of scientists and social scientists who are meant to be practising and publishing research. “Linked to” studies are then one of the easiest sorts of research to produce, and their potential number is infinite. In any field, with a minimal input of imagination, and any dataset one can lay one’s hands on, a link between some X and some Y can be established. That the links are rarely particularly clear is all to the good: the research program remains open indefinitely.
Still one might hope universities would at least be embarrassed about this research, instead of making it a central lever of their publicity efforts. Alas, in their very banality and stupidity, “linked to” studies are the form of academic work that often most easily gains traction with the wider public. They meet the expectation that universities be doing something “useful”—“curing cancer” is the tiresome paradigmatic example here. Not much cancer has been cured of late, but “linking” olive oil to heart disease, or TikTok scrolling to poor sleep, are feeble entries in the same genre of “useful” research.
Behind this lies a denuded view both of natural philosophy and of the good life. Scientific enquiry is reduced, as a practice, to the hunt for statistical correlations, with any structured thought on the causes of things pushed into the background. And the purpose of such enquiry corresponds to an understanding of what we live for that places us barely above cattle at a trough: comfort, “well-being”, muscle tone, gut health, simply being alive a few more years.
Universities are often, and often justly, reproached for reducing much of their research to left-inflected activism. But even this disaster is secondary to their adopting a model of what constitutes the good that reduces it to the useful. Cicero taught the opposite, that the utile (“useful”) could not be other than the honestum (“honourable”) … That’s all very well, you say, but are there any studies showing that reading Cicero is linked to higher average lifetime earnings and better gut health?
