The Uncited Mechanism

What it is

The category was created from The Molecule of More, by Daniel Lieberman and Michael Long, and two chapters make it unarguable. One chapter tells the reader what the dopamine transporter is, how reuptake works and how cocaine blocks it, and prints a labelled neuron diagram to go with it, while its twenty-three-entry reading list contains nothing on transporters, nothing on reuptake and nothing on cocaine. What it does contain is population genetics, psychiatric epidemiology, a government report and a set of magazine pieces. The next chapter rests its closing argument on reward prediction error, and its seven-entry list contains no Wolfram Schultz and no prediction-error paper of any kind. It does contain a study of green-roof microbreaks and a consulting firm's employee-satisfaction ranking, both of which have numbers in them. Of the journal articles read in full across the sampled lists, the number that are primary neuroscience is nil.

A related habit travels with it. The book credits famous interpreters by name in the prose and relegates first authors to the back matter or omits them. Effort-based choice is credited to a university, postural complementarity to a business school, a drug study to British researchers and a case series to a clinic.

In effect

The category had to be tested against its nearest neighbour before it earned a name. In the unfootnoted thesis, an author footnotes what he borrowed and leaves his own conclusions bare, so the gap runs between borrowed material and original judgement and the unsourced sentences are the ones the author is personally responsible for. Here the gap runs the other way: the demographic and epidemiological trimmings are sourced and the borrowed machinery is left bare, so the unsourced sentences are precisely the ones the authors did not originate and could not have originated. The diagnostic question is whether the bare sentence is the author's conclusion or the mechanism they inherited.

The fair defence is that the neuroscience is presented as settled background of the kind a textbook would not cite either, and much of it genuinely is settled. The trouble is that the same prose also asserts things that are not settled in the identical register, and a reader has no way to tell which sentence is which, because neither kind carries a source.

There is an explanation for the distribution, recorded as an explanation and not as an excuse. The senior author is a clinical psychiatrist and psychiatric researcher rather than a basic-science dopamine neuroscientist, and the apparatus fits that exactly: the psychiatric epidemiology is within his field and is the best-sourced material in the book, and the basic dopamine neurophysiology is outside it and is the least sourced.

What it does not say

It does not say the mechanistic passages are wrong. Largely they are correct, and that is what makes the category worth having. The failure is not that the machinery is wrong but that the reader has no way to find out.

It does not measure whether the owner of an idea is named, which is a different test, and a book can pass that one while failing this.

It does not follow from a small apparatus. It is a claim about where the sources sit, and the remedy is one entry in the right place rather than more entries in general.

---