On Dull and Lifeless philosophy, and peer review.

Liam Kofi Bright alerted me to a lovely blog post (here) by Alex Douglas (St. Andrews), “Pure, Ornate, and Grotesque Philosophy Literary styles in philosophy.” The core idea of the post is to suggest, after appropriating a three-fold typology of styles, that style is not just a manner of doing philosophy, but something also expressive of a philosophy. And then to offer an affirmation of the grotesque. Especially if you are like me — with a natural aversion to the grotesque — it’s worth reading, because it’s very nicely done (in part because Douglas himself brings a restrained sensibility toward his defense of the grotesque).*

What follows is a response to something that is not wholly incidental to Douglas’ main point, but also relatively minor to his fundamental argument. It appears in the stage-setting process. I quote:

During my doctorate, exposure to peer-reviewed journal articles nearly put me off the whole idea of philosophy…I was disheartened to discover a whole industry churning out reams of (largely unread) philosophical literature so dull and lifeless it would shame the most pedantic monk of the darkest age…

What they think they are doing is scientific research, for which peer review has become the stamp of authenticity. Peer review came about with a Cold War project designed to maximise the financial return on public investment in scientific research, by imposing a quality-control mechanism. Since the result in the natural and social sciences has been failure and a waste of money, you might wonder why the humanities would want to follow the same path. But the move makes sense if the point is to ape the look and feel of what the earthly powers have declared “scientific”. This also makes sense of what has inevitably resulted: a dark fulguration of jargon and formalisation, of “papers” with “abstracts” presenting the “results” of “research”.

Before I continue, I like learning the term, ‘fulguration.’

Okay, I share in Douglas’ disheartenment. But unlike him (I am in my 50s), I have made my peace with the existence of the publication path he rejects. If peer review were to disappear, I wouldn’t mourn it, and I probably would adapt myself pragmatically to whatever succeeds it. My true thoughts are found at digressions, even if they carry the impression of my professional culture.

Be that as it may, authenticity (or its stamp) is not something scientific philosophers (I am something of an expert on the topic [see here]), or scientists aspire to. So, something is off in Douglas’ analysis.

The history of peer review is also much older than Douglas implies. (Since he is a world-class scholar of seventeenth century natural philosophy, I am a bit surprised by his claim here.) The first instance of peer review that I am aware of —note the phrasing — is De Volder’s review of Christiaan Huygens’ reports on the workings of his pendulum clocks for finding longitude at sea voyages in the late 1680s. De Volder was a professor at Leiden University now primarily known for his correspondence with Leibniz. Interestingly enough, the recipient of De Volder’s review was the Director of the Dutch East India company. They had an obvious interest in making sure that any further investment in Huygens’ clocks was based on sound empirical evidence and technical expertise. De Volder’s report is incredibly astute in locating the weak spots in Huygens’ evidential arguments. The issue intersects with Huygens’ empirical argument against the existence of universal gravity, so Hudde and De Volder were also in a good position to convey the strengthts and weaknesses in the controversy with Newton. (I have written up the details here with George E. Smith.)

Interestingly enough, the person in charge of the Dutch East India company, Johannes Hudde, was one of the few people in the world who would not have needed to rely on De Volder’s review. And, presumably, De Volder was asked not just because of his own technical skills (which arguably were inferior to Huygens and Hudde when it came to the mathematics), but primarily because of his impartiality and, perhaps (connected to this), his university affiliation (Leiden University).

I am not, of course, claiming that this is the first instance of peer review in the salient sense (journal publication). In addition, there existed peer reviewers who would evaluate prize claims for clocks to find longitude at sea for a host of European governments during the period.

My suspicion is that Hudde (also a mayor of Amsterdam) — a rather busy man at the time [with complex political situation as a Dutch Stadtholder was busy conquering the British Isles] — sent it out to De Volder expecting to get a positive report back and so impartial support for further investment in Huygens’ clocks. Hudde and Huygens had known each other since each other’s student days.

As my wording implies, I am hedging my bets on what will count as the first version of peer review in our sense. Through Wikipedia, I learned that peer review in journal publishing is often traced back to 1731 and the Medical Essays and Observations, published by the The Royal Society of Edinburgh. But from the wording that is usually quoted I infer it’s possible that the function of the review process then was really to increase readership among the learned of the published works. And most scholars recognize that imdustrial-size peer review as practiced now — as a kind of gold standard of science — is relatively new. So Douglas can defend his position if he wants.

Douglas links to and endorses a recent argument by Adam Mastroianni that I also had read in which standard peer review as practiced during the last half century is a kind of large-scale social experiment that shows peer review a giant waste of epistemic time. Let me stipulate that I agree with the thought that if peer review exists in order to impose quality control and/or even to be part of a larger mechanism ‘to maximise the financial return on public investment in scientific research’ it is likely to be a costly failure.

As regular readers know, I think peer review exists to produce prestige in the Humanities. I also think this is true in the natural and social sciences, but the more important function in those areas is to create a signal that can be used by others to explain and justify decisions in hiring and promotion. (Basically, I am using ‘humanities’ as a proxy for all fields where there is gigantic oversupply of PhDs relative to TT/permanent positions.)

But the underlying reason or function for the existence of peer review is the fact that due to the advanced division of cognitive labor, hyper-specialization (a term I borrow from Lija Millgram’s Great Endarkenment), even very educated people are in a terrible position to make judgments of quality in areas of specialization removed from theirs. And as distance increases from a particular field, people basically lack the ability to make informed decisions on the quality of other people’s work. So, they look for proxies (metrics, perceived character, PhD granting institutions, letters of recommendation, etc.). As these proxies become worthless or as mistrust about them increases, peer review increases in significance. After all, the concept of a ‘peer’ presupposes a kind of epistemic parity.

Now, let me distinguish between generalist journals and sub-field specific journals. Interestingly enough, there are a few generalist journals that deal with all of science (Science, Nature, PNAS, etc.). But most journals are related to a particular science or subject, and are either generalist or sub-field specific. Somewhat oddly, when you reflect on it, it’s the generalist journals that provide most prestige or highest impact hiring signal. I return to this below.

Now, in a specialized journal one usually expects the editor to be a true specialist in the field. (Think of the role of Bob Goodin, former editor of JPP.) In fact, specialist journals with specialist editors really do not need peer review along the three key dimensions of quality control, significance, and originality. They may draw on peer review in order to create the appearance of procedural impartiality or to draw on other people’s expertise to handle the implied workflow (because editors are rather busy). But there is no intrinsic epistemic reason for its existence in a specialist field specific journal. (Oddly, these are often called ‘niche.’)

In fact, until fairly recently even generalist philosophy journals were entrusted to the care of cherished professional figures, whose judgments of quality were (purportedly) widely respected. (Philosophers of a certain age can give you lots more examples of editors --beloved, safe pair of hands— who became closely associated with a particular journal.) This became to be seen as an anachronism as area of specialization became thought important.

Of course, many of the most prestigious journals are generalist in character. This is so for the original learned journals of the Republicof Letters (some still exist), and, of course, so for many journals in the sciences and philosophy. My claim is that once the division of cognitive labor becomes a bit advanced, editors of generalist journals are generally in a terrible position to evaluate submissions along the three key dimensions involved: quality control, significance, and originality. They literally need others to help them out. (I don’t mean to suggest they couldn’t figure this out with enough time and energy, but time is precisely what they are economizing on).

So, even if they were to read all the potentially accepted submissions themselves, editors must farm out quality control to those with the time and interest to keep abreast of the factors salient to the three dimensions. Of course, lurking here are higher dimension problems: how to identify competent and reliable peer reviewers, etc.

So, peer review creates value in the credit economy of a given discipline. Or to be precise, journals create a measure of value. And, in fact, this helps explain the peculiarity I mentioned above. In generalist journals somebody (the editor) has made a decision about relative general significance (to a field, a discipline, or to science/medicine more generally). Of course, grant agencies may amplify these decisions. Publication in a specialist journal does not create any usable signal about relative significance.

Nothing I have said suggests that publication in a peer reviewed journal is an epistemically valuable signal in, say, the context of discovery. Peer review is ordinarily not a mechanism to establish intrinsic merit in the real world. (After all, referees are very busy too with lots of commitments; editors need to take their review on trust, etc.) But it allows a narrative to be created when hiring and allocative decisions need to be justified after the fact, especially in the context where nobody really speaks authoritatively about quality and significance. (My argument echoes Ted Porter’s famous argument on the rise of accounting.) This feature is actually my only reason for rejecting the abolition of pre-publication review as advocated by Liam Kofi Bright and Remko Heessen.

And, if my suspicion is right — that the advanced division of labor, hyper-specialization, undermines our ability to make regular and reliable qualitative judgments in real time about work done outside our own area of specialization — no such relative authoritative judgments are truly possible in the physical/real world. If you have the luxury to ignore such justificatory (ah) processes, you will shrug your shoulders about the merits of peer review.

Let me wrap up. I doubt peer review necessitates the production of dull and lifeless philosophy, unless one assumes (as I do not) that only the grotesque is apt at doing justice to vital thought. But that’s to say not just that nominalism is true, but that our art must be as singular as what one conveys. Douglas has not even started to make the case for any of this.

There is, of course, an art involved in preventing the vitality to be sucked out of one’s prose during the peer review process. (Hint: editors can be an ally.) And while I wouldn’t want to subsist on a primary diet of peer reviewed journal articles alone, they can be nourishing when digested selectively and with discernment.

---

* Douglas quotes Bagehot as follows: “grotesque art does just the contrary. It takes the type, so to say, in difficulties. It gives a representation of it in its minimum development, amid the circumstances least favourable to it, just while it is struggling with obstacles, just where it is encumbered with incongruities. It deals, to use the language of science, not with normal types but with abnormal specimens; to use the language of old philosophy, not with what Nature is striving to be, but with what by some lapse she has happened to become.”