[realradicalwash]

Great article. A snippet from the conclusion that I found particularly interesting: "History of science all too often highlights "superstar" scientists who have illustrious careers, overcoming great obstacles and resistance to new ideas, leading to a public vindication, sometimes followed by a Nobel Prize. Miescher’s story is different. There was no immediate recognition nor ultimate vindication in his case, yet his discovery proved fundamental for the whole of modern biology. Excavating stories such as his from the past, and incorporating them into the history of genetics, helps us portray the process of science more realistically."

[dekhn]

The writeup in Mayr's book "The Growth of Biological Thought" is really enjoyable. Miescher started by working with pus collected from hospital patients, but quickly moved to Rhine salmon "because their sperm is filled with the substance". His work was so surprising that his advisor Hoppe-Seylor (a well-known biologist for many other things) redid all his experiments. It's actually one of the great examples of protein bias- they couldn't understand why the molecule they purified didn't contain sulfur (DNA doesnt' have sulfur, while many proteins do).

[teorema]

The problem is now the superstardom narrative has grown in importance in academics, regardless of validity. It's not just something we project into the past anymore, it's something that is needed to project into the future, to survive at the most basic levels. Everyone is the next W&C (ignoring the complexity of the truth around that narrative), until they have tenure or are pushed out altogether. In this way false narratives about the past become impossible goals for the future, which leads to false claims in the present.

[realradicalwash]

being in academia myself, this hits home so hard :/ the field picks some rockstars, which imv is to a good extent based on charisma, and then goes with that.

one of 'our' rockstars has pushed a certain hypothesis for more than 10 years. it is plausible on a conceptual level, but I have yet to see systematic, large-scale evidence for it. it would not be entirely trivial, but also not that hard to test.

here is what happened: that person has fallen from grace due to some troublesome accusations. those accusations did not stick up in court btw. the case was made public 2017, the court case was closed this year. still that person's reputation is tarnished, so is their rockstar aura.

here comes the kicker: just this year, the first study comes out, presenting results that that hypothesis does not hold up, certainly in some contexts. i know for a fact that at least one more team has had similar results a few years ago - but when they first presented the results informally, they felt pressured to shelf the project.

[teorema]

I'm not sure how much my cynicism is just me or my immediate experiences over the years, but I don't really trust the academic system anymore in terms of credit or even as much in terms of validity. Or rather I do, but in some abstract sense, in that if I see a meta-analysis with some mechanism of dealing with publication bias (formal modeling of bias, or preregistration) I might believe something, but I don't attach any credit to the body of work beyond those who as a group in toto were involved in it.

Too much credit is given to trappings of progress or something of that sort; conversely, people are punished too much for things that are irrelevant, and both seem to be weighted capriciously.

I see similar viewpoints expressed repeatedly in various outlets, but little seems to change.