🍎your Scholarly Digest 1st May, 2025

Academia essentials hand-picked fortnightly for the mindful scholar

Was this newsletter forwarded to you? Sign up to get it in your inbox every week.
Know someone who will enjoy The Scholarly Letter? Forward it to them.

Hi Scholar,

This is the 25th Letter we’ve sent in 2025. To mark this (mini)-feat, we thought we’d share why we decided to call this project The Scholarly Letter. Besides, we’ve explained elsewhere why we chose the word ‘scholarly’, so it only seems fitting to tell you why we chose ‘letter’.

It wasn’t random. A while ago, we came across the Republic of Letters - a virtual, global community of intellectuals in the 17th and 18th centuries who kept the lifeblood of ideas flowing through one key medium: the letter. Long before the internet, scholars, scientists, and philosophers debated, criticised, and shared their thinking by post. This metaphysical republic of ideas was built entirely on letters: lots of them.

That spirit animates The Scholarly Letter. What we send aren’t letters in the traditional sense: we know they’re emails. But the intention is the same: to explore the ideas, debates, and questions that matter to scholars. And maybe, we preserve a little of the slowness and intentionality of the post, in contrast to the evanescence of tweets and algorithmic scrolls.

We’re not quite Voltaire with his 15,000 letters but this is our 25th, and who knows? Maybe one day we’ll get there. Maybe one day we’ll even send out physical ones.

BRAIN FOOD

Has Critique Run Out of Steam?

Bruno Latour’s academic career was defined by critique and a sharp attention to what he saw as a “lack of scientific certainty inherent in the construction of facts”. This idea, that facts are not discovered but constructed by interactions between human and non-human actors, is central to Actor-Network-Theory. This theory, which Latour is most well-known for, was aimed at exposing how “objective facts” are capable of hiding the assumptions and ideologies embedded within the processes that make science, research and knowledge possible. In other words, the goal was to challenge the invincibility of objectivity. 

Yes, yes, last week’s brain food was also about (strong) objectivity, we know, but this week the focus is different. 

This week’s Brain Food is an essay Latour wrote titled “Why Has Critique Run out of Steam?”. In it, he reflects on and worries about the consequences of his role in developing critical theory. The essence of his argument is that over time, critique has become too similar to deconstruction; a new kind of critique is needed.   

Initially, critique aimed to expose the bias and power embedded in what were blindly accepted facts: in other words, the problem was too much trust in facts that were actually ideological. Latour worries that critique is now eating itself. Factual statements - like the reality of climate change or vaccine safety - can now be challenged using the same tools of power analysis and social construction that once successfully debunked harmful ideologies masqueraded as facts. These were the very tools used to debunk so-called “scientific” justifications of racism, which claimed biological differences in intelligence or morality.

Today, the problem is too much distrust in facts that are actually sound. 

Latour argues that we ended up here because those who developed critical theory, including himself, have been overly focused on ‘debunking’: see something that claims to be factual and then reveal it to be actually shaped by power, ideology or social factors. This way of critiquing is all about subtracting false claims to reality, focused on tearing down, never on building.   

There is a need, therefore, for critique which does not just focus on what we critique, but why and how we do so. Latour challenges us to take more responsibility for the consequences of our critique, to shift the focus of our critique from matters of fact, to matters of concern. It’s a challenge that aligns with the vision of the Scholar we laid out in The Scholar Manifesto, as a publication who does not critique to destroy, but to nurture and support the conditions under which what we critique matters. 

NEWS

Lost in Translation

When AI is used unsupervised to produce research articles, it has a tendency to leave “fingerprints”. These examples of fingerprints range from the tortured - “big data” changed to “colossal information” - to the obvious - “as an AI model, I do not have opinions”. The latest fingerprint to have caught the attention of research integrity watchers is a fictional experimental technique named “vegetative electron microscopy” (“scanning electron microscopy” however, does indeed exist). 

Now, we could spend time spelling out the consequences of nonsense phrases like “vegetative electron microscopy” contaminating the datasets used to train LLMs (bad). But this particular fingerprint gets the News slot this week because of the story behind how it came to exist in the literature. 

Unlike most others, this fingerprint was not created by researchers as a result of AI use to produce papers themselves. 

Initially, it was suggested that an error was made when a paper published in 1959 was digitized. In this instance, the Optical Character Recognition technology, which extracts text from images, may have mistakenly read text in adjacent columns as part of a single column of text (click here for an image of the paper in question). After this error, anyone using AI to generate research articles without checking the output may have included the nonsense phrase in their paper. 

However, examples of this phrase appeared in articles authored by Iranian researchers as early as 2019 - three years before widespread AI use began following ChatGPT’s public release. Researchers from Iran have a bad rap when it comes to research integrity and are often associated with data fabrication and more recently unsupervised AI use. However, users on Reddit pointed out the Persian words for “vegetative” and “scan” are extremely similar:

The phrase probably originated when an earlier paper containing the term for “scan” was automatically translated into English by Google Scholar: an error during the translation is probably what caused the term “vegetative” to be inserted into the paper. This mistranslated article was subsequently used in the training data for GPT3 and persists in more up-to-date models like GPT4o meaning it's probably permanently embedded into AI training datasets. 

RESOURCE

Slightly Rude Notes on Writing

Writing is a skill that's not so easy to teach, especially not through a structured, linear sequence of steps in a course. Yet it's a skill that can be learned. If you ask us, the best way to learn how to write isn't by reading instructions about writing - aka teaching materials, but by reading slightly unhinged reflections on the act itself. 28 Slightly Rude Notes on Writing is one of those pieces: sharp, irreverent, and oddly inspiring for anyone aspiring to get better at writing. Here are two notes from it which we think might resonate with you: 

OPPORTUNITIES

Funded PhDs, Postdocs and Academic Job Openings

• Postdoctoral and Faculty Positions @ Pennsylvania State University, USA: Postdocs: click here / Faculty positions: click here

• PhD and Postdoctoral Positions @ TU Delft University, Netherlands: PhD positions click here / Postdocs click here

• Postdoctoral Positions and Fellowship Opportunities @ Kings College London, UK: Postdocs: click here / Fellowships: click here

• PhD Positions @ Coventry University, UK: click here

KEEPING IT REAL

So Long, and Thanks for All the Delays

Rather than thanking his family for their support, his colleagues for their thoughts or his publisher for their patience, Anthony Standen thanked the Long Island Railroad for all the time it gave him to write his book, ‘Science is a Sacred Cow’.   

We care about what you think and would love to hear from you. Hit reply or drop a comment and tell us what you like (or don't) about The Scholarly Letter.