Digressions&Impressions

On JPP, Wiley, Prestige Production, and Game Theoretic Modeling

[I will phase out D&I at typepad. This post was first published at:  digressions.impressions.substack here. To receive new posts and support my work, consider becoming a free or paid subscriber at <digressionsimpressions.substack.com>]

As Leiterreports (and here; here) and Dailynous reported, Wiley (a commercial publisher), got rid of Robert Goodin (a prominent political philosopher) as editor of the Journal of Political Philosophy (JPP), which is (one of) the leading journals in political philosophy in good part thanks to Goodin's long stewardship.  It is no surprise that most of Goodin's fellow editors and board resigned in protest. There has been a lovely closing of ranks.

Because I am going to be rather (sternly) critical of the ranks, I want to stress that Wiley's explanation was ridiculously and scandalously vague. More important, the manner in which they fired Goodin also suggests unwarranted editorial interference that actually risks endangering academic freedom. So, resignation was the right thing to do.

But as the resigning editorial board members' comments started to appear in public, it seems that folk in the know believe that the cause of Wiley's decision is that it wanted to increase the number of published papers rather rapidly up to a tenfold more (see here; here). Immediately one started to read on social media that pursuit of profit is undermining quality. (This show up in the threads I have already linked to above, but also in discussions elsewhere.)

However, a major reason why Wiley is nudging many presses to expand journal space is that universities, libraries, and grant agencies have contracted with them to do so in the context of expanding open access. From the perspective of grant agencies and universities this makes a lot of sense: they are funding research that, if of sufficient quality, needs an outlet. (For those who work in a grant system the absence of slots often becomes visible only after acceptance when one discovers that a journal has no more slots available in a given year!) I am not a fan of these contracts because they basically show what happens when well-connected corporations take advantage of well-intentioned policy shifts: they capture the financial rents. Yet, it doesn't follow that open access or expanding journal space (and so increasing acceptance rates) is a mistake. And in what follows I focus on the latter.

For, as is well known, academic philosophy has incredibly low acceptance rates in its flagship journals (often under 5% acceptance rates).* I couldn't find a recent number for JPP, but it seems to be around 5% and I wouldn't be surprised if it were lower. A quick count suggests, JPP publishes 24 articles a year (give or take). Another leading journal, P&PA "has historically published 12-16 articles per year." And it does not get much better at a journal like CRISPP (about 20). [I make the relationship between slots and acceptance rates more precise below.] The situation is not much better in most areas of professional philosophy (but see below).

I have noted in the past that these acceptance rates only make sense, epistemically, if philosophical activity had enormous inductive risk to society. Nobody believes that as Tim Crane argued in an important essay in TLS (back in 2018; see here for ungated version). Tim helpfully suggests that the real explanation is that "publication in one of these journals may no longer be that sole decisive achievement that will get you that job or grant." That's clearly true, but not wholly right not just because I would add promotion to his list. As Anna Stilz, whose comments were highlighted by Justin Weinberg, notes, publication in these journals is a "signalling device that validates the importance of someone’s work when they go up for tenure and promotion."

However, as should be well known, compared to other disciplines, including the sciences, professional philosophy's acceptance rates is especially low (recall Justin Weinber's discussion). And other disciplines also hire and promote people. Thus, I suggest the real function of such very low acceptance rates is the felt need for prestige production in professional philosophy among the circle entrusted to run leading journals.+ Before I get to the attempt to conflate such 'prestige' or 'importance' with 'quality' that insiders often make, it is worth noting that the needs of such signaling are not evidently co-extensive with the pursuit of truth (leaving wisdom aside) or progress in the field. In fact, they clearly can come apart.

For, there are some (quite predictable and also widely discussed) side-effects of the very low acceptance rates: very fine papers need to be shopped around to many journals before they find a home. This has overburdened the supply of referees, and it has put tremendous pressure on editorial process at many journals. We are a discipline that is familiar with moratoria on submission and backlogs from acceptance to publication at quite a few journals. There is also often a huge delay between submission and verdict. And consequently overall a huge delay between research and actual publication for too many papers. Thus, as organized, prestige generation slows down whatever rate of progress one might hope for in the field. (Yes, I am a skeptic about philosophical progress, so I use 'progress'  here relative to a research frontier or relative to how competent judges in a field would describe it.)** 

The problem with this state of affair is not merely epistemic. It also generates huge additional anxiety for people who are or expect to be on the job market and who are up for promotion. (This is also true in other fields, of course.) With acceptance rates so low, they basically face a (weighted) lottery whether their papers are accepted in a time frame needed. This state of affairs probably reinforces all kinds of structural injustice, again not only relative to other disciplines. The fact that many leading journals have, sometimes for decades on end, been entrusted to a 'safe pair of hands' (a distinguished, often male, philosopher) probably increases the bottlenecks and the risks of such structural injustice, but -- like the fact that we see so little experimentation with journal form (and kinds of review) [something I strongly advocate] -- I leave that aside here.

Now, I am not claiming editors are gaming acceptance rates directly. But they do clearly view number of issue/volume slots or pages as a hard constraint (perhaps regulated by an implicit willingness to let publication not be delayed for more than a year or two after official acceptance). But notice that restricting supply of slots is not the same as selecting on quality. If one really selected on quality alone one would allow the supply of slots/pages to be flexible. So, in what follows (and above) I use 'low acceptance rates' as a proxy for limiting the number of publication slots, or selectivity.

Above I noted that prestige generation trade-offs with progress at the low rates of acceptance in professional philosophy. We don't need to model that. (It gets more complicated as acceptance rates rise significantly because there may well be an inflection point where too much ease of publication undermines progress.) But we need some modeling to help understand that prestige generation or selectivity and quality can come apart significantly.

Luckily for me (and the quality of discussion) I was sent a pre-print of a paper, "Academic journals, incentives, and the quality of peer review: a model," by K. J. S. Zollman, J. Garcıa, and T. Handfield conditionally accepted at Philosophy of Science that has a game-theoretic model exploring possible trade offs between aiming for quality and (you guessed it) for low acceptance rates, in contexts where selectivity is taken to be a measure or proxy for quality. The paper should be publicly available fairly soon. But my discussion won’t turn on technical details and its possible lack of robustness. (I call incentive-alert philosophy of science, ‘public choice philosophy of science.’) In their model quality is an effect of peer review which is costly. And somewhat worrisome they conclude that "journals incentivized by selectivity have strong incentives to maintain worse peer review than those incentivized by their quality." I claim this nicely describes well what happens in professional philosophy. 

Now, they note that aiming for low acceptance or selectivity creates an incentive among editors to discourage "self-selection because a paper which is not submitted cannot be rejected. This results in a strange process whereby journals make peer review worse in an attempt to induce bad papers to submit, but maintain sufficiently good peer review to ensure that a large proportion of those bad papers will probably be rejected." And somebody may well object that referee quality at our leading journals is -- despite the boorish incivility and rudeness -- actually qualitatively great. Let's assume there is a lot of correct testimony to this effect.

One can explain such testimony in light of Zollman/Garcia/Handfield model, if we pay attention to the role of desk rejects (something they ignore.) Let's return to JPP; it is especially famous for two interrelated features: it offered quick (even helpful) desk-rejects (often/always by Goodin) to the vast-majority of initial submissions and then stellar review process on the papers sent out to reviewers which help improve the final publications. (I know important scholars who almost only review for JPP.) And this creates (now quoting Zollman et al) "the appearance of high standards."

One may think that JPP has, thus, hit on a fine equilibrium between balancing peer review costs and quality; and that it actually is a good way to aim for selectivity and quality at once. And people do like to point to the quality of the papers appearing in JPP. (In 2021 its impact factor was according the journal's website 1.881--quite respectable for philosophy standards, but not especially noteworthy. Something fans of selectivity tend to ignore.) Fair enough.

But, of course, desk-reject even by an impartial and distinguished and unrushed editor is a very coarse-grained (and very cheap) peer review system. And the Zollman et al. model explains that this does reduce quality. Obviously, this presupposes that among the desk-rejects are sufficiently many papers that with (costly) peer review would have been as good as the papers that JPP did publish. That's obviously true, but that's because among the papers published in very selective peers of JPP and even slightly less selective journals there is very little difference in overall quality. But the nice thing about working with a game-theoretic model like Zollman et al.'s is you don't have to trust my judgments, it's a predictable effect of the incentives -- artificially limiting supply! -- shaping decisions. Of course, the model may not be robust so here I have used it solely to help think through the issues. But notice that when multiple journals aim to be selective it becomes much costlier to provide adequate peer review because referees get overworked (and perhaps bored by seeing the same papers multiple times).

Now, I am not against prestige in professional philosophy or higher education (although one does wonder if excessive selectivity in admissions to top colleges actually is conducive to society's needs). But from the perspective of the epistemic (and social) needs of the discipline and profession, it is a mistake to have excessive selectivity in philosophy journal publication help produce or coincide with that prestige. It slows down research and, as the model suggests, plausibly hurts quality. (It probably does so by reducing intellectual diversity and increasing risk aversion.) 

So, from my perspective, while the publishers should respect editorial independence and not use heavy-handed practices, nudging our leading journals to accept more papers is the right thing to do even if the motive -- profit -- or honoring flawed open access contracts need not be so. We need such a regime change.

Of course, the increase should be gradual and not tenfold at once. The critics of Wiley have a point! As should be clear most arguments that conflate incredible selectivity and quality in defense of the status quo seem bogus and, I say this politely (these are my friends, after all), seem generally self-serving not the discipline's interest.

Tom Hurka suggested that such selectivity allows one to "keep abreast of important new developments and contributions in the field without having to wade through dozens or hundreds of published pieces." This was very true before the age of scholar.google and philpapers. But less so now. I do think we need more outlets like (as Ingrid Robeyns suggested in context) Philosophy Compass that (especially with the disappearance of review chapters in dissertations) give opinionated surveys of recent developments, but that's orthogonal to the present discussion.

Finally, I don't want to claim that increasing, even doubling slots or reducing/halving selectivity (to say 10-20% acceptance rates) solves the profession's social problems. Undoubtedly, if journal publication in particular venues is seen as key, then we cannot prevent an arms race among job candidates and tenure seekers to publish in such journals. And maybe the gains to the referee process will be ephemeral or temporary--only time and experience will tell. But my argument for the epistemic gains from increasing slots remain.

Academic disciplines do need high quality signals for quality in publication for institutional and epistemic reasons; excessive selectivity is a bad mechanism to produce such a signal and it perniciously confers prestige that reinforces all the bad effects of the very steep disciplinary prestige hierarchies. I hope the scholars who nobly and correctly resigned in protest also work to change this bad status quo.

• This first appeared at: <On JPP, Wiley, Prestige Production, and Modeling Quality: (substack.com)> a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.

---

*I am going to ignore the pervasive editorial unaccountability and deviations of best scholarly practices that also show in these journals.

**Of course, the salience of significant work seeps out to well connected insiders, but that just strengthens my argument.

+Some field specific journals within philosophy that actually presuppose considerable technical expertise (in the sciences or ancient languages) have much higher acceptance rates without corresponding loss of prestige. I think this is due to the fact that the prestige is already sufficiently accorded to the expertise (and that editors in these fields can focus more on keeping the discussion going aright).

 

Lenz's Contact Problem with some reflections on Condorcet and Rousseau

Especially early modern explanations seem to construe the mind as something that is tucked away in a body and thus inaccessible by other minds. If this is correct, how could we even begin to think of a way that my thoughts influence yours? Are there ways of transmission or other modes of influence between different minds? I would like to call this the contact problem. Intersubjective explanations, it seems, must specify ways in which one mind can affect another mind. As I will argue, at least some early modern philosophers addressed this problem and provided intersubjective accounts of the mind. What is more, they relied on different models of intersubjectivity. To present three different but crucial explanatory models, I will focus on Spinoza, Locke and Hume: Spinoza will be shown to opt for a metaphysical model, Locke as resorting to a linguistic model and Hume as relying on a medical model that combines assumptions about contagion and sympathy.
Before we take a brief look at these models, let us take a step back and look at the contact problem again. Taking human minds as individual units hidden inside a body suggests that a direct contact between minds is impossible. I can tell you what I think, but in doing so, it’s not my thought that is transmitted. Getting at the thought seems to require an inference on the part of the listener. By contrast, if I endorse a view of the mind that takes mental states to be ingrained in behaviour, such that my mental states are not hidden but part of behavioural patterns, then my mind does seem to be more directly accessible. It seems, then, that the contact problem poses an enormous obstacle for the former but not for the latter view of the mind. On this assumption, intersubjective explanations work well in a behavioural view of the mind but not on what has been called “mentalism”. Martin Lenz (2022) Socializing Minds: Intersubjectivity in Early Modern Philosophy (Oxford University Press), pp, 93-4 (emphases in original)*

I have to admit that when I first started to read Lenz's very entertaining Socializing Minds, I was a bit dubious about the so-called contact problem. It seemed to me, and sometimes still seems to me, an artifact of behaviorialism and some of Ryle's more dubious historiographic moves (which, alas, he was not immune to). But because Lenz's book is so much fun -- it mixes contemporary insights with very erudite and clearly written history of philosophy in refreshing ways easily moving between Avoerrism and memes-- I decided to revise my view. Reflecting on the contact problem is generative and so (by my lights) cannot be a mere artifact (which are sterile).

I have to admit that I don't understand why inference is thought to be a barrier to contact -- the contact problem seems to be rather the no unmediated contact problem --, but obviously for the contact problem to have real bite, by inference here (in this context) is meant something like 'not truth preserving' inference or distorting. And clearly it's true that many if not most early modern philosophers thought that language and other mechanisms of social contact are distorting of reality (and don't do full justice to our ideas of it). Lenz shows that despite this commitment, Spinoza, Locke, and Hume have different kinds of theories in which our minds are, de facto, dependant on other minds, although -- he may disagree with this -- the effect is that in interacting with others the vulnerability to embracing the false remains. (Perhaps, in Locke the vulnerability is also simultaneously the grounds for possibility of reaching some truths. About that some other time.)

Anyway, that the no contact problem seems so much an artificat of behavioralism made me wonder if one couldn't generate it in an alternative framework. This made me consider whether there are circumstances in which no contact is desirable. I immediately thought of Condorcet's Jury theorem which, formally, requires that voters do not communicate with each other and make up their own minds on a decision problem. Without such independence the result that (with rather weak assumptions about human nature) more folk do better than fewer would be unsurprising (and not so unsettling to the elitist and hierarchical mindset).+ Of course, the formation of Rousseau's general will also requires such independence.

Now, while Rousseau's account of the general will has non-trivial formal similarity with Condorcet's jury theorem -- and I would be amazed if it didn't shape Condorcet's thinking --, it is worth signaling that the nature of their independence pushes in different directions: (i) and skating over many interpretive disputes, in Rousseau's general will set up, independence is really designed to create a kind of impartiality not tainted by self-interest (and irrelevant personality features). And this is why this kind of independence is compatible with a kind of representative agent approach (familiar from, and perhaps an ultimate source of, Rawls' original position).

A representative agent approach would undermine the force of the Condorcet jury theorem; in it independence is really about a kind of authenticity of one's judgment. Other minds are corrupting (because they undermine independence). Now, I don't mean to deny that the previous two sentences are things Rousseau himself might say. I am just trying to track a conceptual distinction. (To be sure I think both directions have roots in Spinoza's debts to stoicism--but about that in a future post.) 

That is, the no contact problem is no problem in two contexts: first, one requires or needs authentic judgments that give each of us weight. That is, if one takes democracy seriously. (And lurking here are also protestant ideas about conscience that I have not touched upon above.) Second, one is dealing with a decision problem that is best handled by an impartial agent that represents (the true interests of us) all. That is, if one takes a certain species of objectivity/impartiality seriously. Put like that the two sources of the desirability of no contact seem to be a feature and not a bug of a certain conception of modern political life. And so, to put it as a serious joke, Wittgenstein and Ryle are untimely.

 

On the Marketplace of Ideas: on Mill, Holmes, Kalven, Condorcet and Carl Schmitt

Since the work of Jill Gordon (1997) it ought to be common knowledge that J.S. Mill did not defend the idea that the market place of ideas would inevitably lead to truth. (This fits with my own view that I have expressed over quite a number of digressions: see, for example, here; here,  and here.) Unfortunately, this leaves a bit puzzling how the phrase and idea came to be associated with his name and a kind of (providential) liberal faith. (And depending one's political commitments this is a sign of liberal naivete or well grounded liberal optimism.) Moreover, most sophisticated liberals would never assert that markets always produce optimal outcomes, so lurking here is also a question about the mechanism. (I return to that below.)

I suspect that true scholars are familiar with the fact that Justice Oliver Wendell Holmes's dissent in Abrams v United States (1919) is often thought to originate the idea and closely associated phrase. (As it happens Holmes did re-read Mill's On Liberty that year.) But Holmes wrote, "the best test of truth is the power of the thought to get itself accepted in the competition of the market." This does not actually claim that truth will prevail in the market for ideas (and does not use the phrase 'marketplace of ideas' at all). Holmes' position is more sober than that. I am persuaded by Vincent Blasi (2004) that Holmes never intended the providential claim about the competition of ideas. (Blasi is also the source of my claim that Holmes did read Mill.)

In a footnote (41), Blasi suggests that in 1935, David M. Newbold used the phrase "market place of ideas" first in a letter to the New York Times. But as quoted by Blasi that letter does not connect it to generation of truth (but to 'public opinion.') It's likely that this Newbold was the author of a book called Notes on the Introduction of Equity Jurisdiction Into Maryland:1634-1720. So he was probably a lawyer or law professor. In a Denver Law Review article of 1940, one can read that "Our entire system of free public education is based upon it and out of it has developed the American type of government of public opinion. Our people from the earliest days, unconsciously perhaps, have been demonstrating in every generation their faith in the doctrine that the test of truth is its "ability to get itself accepted" in the marketplace of ideas." This clearly quotes Holmes, but with Newbold's phrase. The first connection between the 'market place of ideas' and 'truth' that I have been able to find with some googling is in Senate testimony in June, 1941, by Mr. Louis G. Caldwell, an attorney representing mutual broadcasting system to the Interstate Commerce committee (here). The context is freedom of speech (and the role radio has in expanding it or not). A decade later the phrase, and its connection to truth is much more widespread. But I have been unable to find anyone expressing the idea that the marketplace of ideas leads to truth, and that somehow Mill had defended it before 1960. 

In 1960, the very eminent law professor, Harry Kalven Jr. did a kind of retrospective in The Chicago Law Review on the Scopes trial to celebrate the (near) centenary of Darwin's Origin of Species. To people who have any association with The University of Chicago, and those who have studied debates on academic freedom, Kalven's name is quite familiar (recall here on the Kalven report). The retrospective is interesting in its own right. But here I just quote a passage: "It is keyed, in the tradition-honored phrases of Milton, Mill, and Holmes, to confidence that truth will not be bested in a fair fight, to competition in the marketplace of ideas." (p. 516) Kalven's own view is not quite this, I think. But that's for another time. The quoted passages does come pretty close to the idea usually associated to Mill. (And unusually for memetics, also provides evidence for the role of Holmes in the process.)* And since Kalven was important and influential it is not unlikely he is the source of connecting Mill to truth prevailing in the marketplace of ideas (which starts to circulate widely in the 1960s).

That's the best I can do on this. I am open to learning from other people's search abilities. I could close here, but I was still left with a puzzle how the idea mistakenly associated with Mill came to be thought of as somehow intrinsic to liberalism. And I think I have a (surprising) answer to that, although it is a bit convoluted. I quote a long paragraph:

De Maistre and Donoso Cortes were incapable of such "organic" thinking. De Maistre showed this by his total lack of understanding of Schelling's philosophy of life; Donoso Cortes was gripped by horror when he was confronted with Hegelianism in Berlin in 1849. Both were diplomats and politicians with much experience and practice and had concluded sufficiently sensible compromises. But a systematic and metaphysical compromise was to them inconceivable. To suspend the decision at the crucial point by denying that there was at all something to be decided upon must have appeared to them to be a strange pantheistic confusion. Liberalism, with its contradictions and compromises, existed for Donoso Cortes only in that short interim period in which it was possible to answer the question "Christ or Barabbas?" with a proposal to adjourn or appoint a commission of investigation. Such a position was not accidental but was based on liberal metaphysics. The bourgeoisie is the class committed to freedom of speech and freedom of the press, and it did not arrive at those freedoms from any kind of arbitrary psychological and economic conditions, from thinking in terms of trade, or the like. It has long been known that the idea of the liberal rights of man stemmed from the North American states. Though Georg Jellinek recently demonstrated the North American origin of those freedoms, the thesis would hardly have surprised the Catholic philosopher of the state (nor, incidentally, would it have surprised Karl Marx, the author of the essay on the Jewish question). Further, the economic postulates of free trade and commerce are, for an examination within the realm of the history of ideas, only derivatives of a metaphysical core. Donoso Cortes in his radical intellectuality saw only the theology of the foe. He did not "theologize" in the least; there were no ambiguous, mystical combinations and analogies, no Orphic oracle. The letters about actual political questions revealed a sober attitude, often frightening and without any sort of illusion or any touch of the quixotic; in his systematic train of thought there was an effort to be concise in the good dogmatic tradition of theology. His intuition into things intellectual was therefore often striking. His definition of the bourgeoisie as a clasa discutidora and his recognition that its religion resides in freedom of speech and of the press are examples. I do not consider this to be the last word on Continental liberalism in its entirety, but it is certainly a most striking observation. In view of the system of a Condorcet, for example, whose typical meaning Wolzendorff, perhaps because of intellectual affinity, recognized and superbly described, one must truly believe that the ideal of political life consists in discussing, not only in the legislative body but also among the entire population, if human society will transform itself into a monstrous club, and if truth will emerge automatically through voting. Donoso Cortes considered continuous discussion a method of circumventing responsibility and of ascribing to freedom of speech and of the press an excessive importance that in the final analysis permits the decision to be evaded. Just as liberalism discusses and negotiates every political detail, so it also wants to dissolve metaphysical truth in a discussion. The essence of liberalism is negotiation, a cautious half measure, in the hope that the definitive dispute, the decisive bloody battle, can be transformed into a parliamentary debate and permit the decision to be suspended forever in an everlasting discussion.--Schmitt (1922/1935) Political Theology, translated by George Schwab, The University of Chicago Press (2005), pp. 61-3 [Originally published as Politische Theologie: Vier Kapitel zur Lehre von der Souveranitat]+

Okay, there is much going on here. Donoso Cortés (1809 – 1853) is a fascinating character who was familiar with the first political group to call themselves 'liberal' and eventually became an ardent opponent of them. (Since Adam Smith describes his system of natural liberty as 'liberal,' I don't think it's anachronistic to claim him also as a liberal--despite learned protestations in Cambridge.) Schmitt admires both Maistre and Donoso Cortés, but as this passage reveals he also sees some of their intellectual limitations.

It is by no means easy to establish in each sentence where Donoso Cortés ends and Schmitt begins. But it's pretty clear that for them liberalism is identified with freedom of speech and the bourgeoise as the disputational/discussing class (clasa discutidora).** (Schmitt almost certainly made a transcription error because Donoso Cortés used the phrase 'las clases discutidoras' in a letter of 1851.**)

But$, strikingly, that freedom of speech will generate truth is then associated with 'the system of Condorcet.' And, in fact, given the significance given to voting in Schmitt's argument, it's pretty clear we have a kind of reference to Condorcet's famous Jury theorem. And, indeed, in the jury theorem the more voters there are (alongside some minimal assumptions) the more likely one is to reach truth. And, over time, as voting population keeps growing, this becomes pretty infallible. Here truth is a kind of outcome of a process, and it is correct to say of it that (give some modest assumptions) "truth will emerge automatically through voting." 

However, I use a 'kind of reference' because the original jury theorem requires no communication among voters. So it's a pretty bad model for thinking about free speech/press.++ It's unclear to me if the conflation is in Wolzendorff or in Schmitt. Part of the historical problem is that Condorcet did also defend freedom of speech (with some attenuations) in a famous fragment from 1776. 

So, here's my suggestion: Schmitt is the source of the idea that liberalism is intrinsically committed to the idea that truth naturally will emerge from free discussion. This idea is associated with Condorcet (who was indeed a proponent of freedom of speech). But it's based on a (perhaps honest) misunderstanding of the jury theorem and the role it plays in Condorcet's thought! 

Schmitt's criticism of the idea he attributes to liberalism is not that it's false or a bad mechanism to discover truth, or somehow naïve. Rather, it's that it's a device to avoid decisive, existential decisions: "The essence of liberalism is negotiation, a cautious half measure, in the hope that the definitive dispute, the decisive bloody battle, can be transformed into a parliamentary debate and permit the decision to be suspended forever in an everlasting discussion." That's actually a sound interpretation of the liberal ideal of rule by discussion, which is all about avoiding fighting (worst possible outcome) if one can continue talking (probably bad, but not so bad).

Schmitt, a jurist himself, was very widely read. So, my hypothesis is that Schmitt's idea about the nature of liberalism as a regime of talk that will produce truth circulated among jurists (perhaps with the reference to Condorcet being dropped once it moved into Anglophone world), and at some point merged with Judge Holmes' idea to produce Kalven's formulation. Okay, that's enough speculation -- I almost wrote 'cheap talk' -- for a day.

 

On Pynchon on 1984

By the time they have left the Ministry of Love, Winston and Julia have entered permanently the condition of doublethink, the anterooms of annihilation, no longer in love but able to hate and love Big Brother at the same time. It is as dark an ending as can be imagined. But strangely, it is not quite the end. We turn the page to find appended what seems to be some kind of critical essay, "The Principles of Newspeak". We remember that at the beginning, we were given the option, by way of a footnote, to turn to the back of the book and read it. Some readers do this, and some don't - we might see it nowadays as an early example of hypertext. Back in 1948, this final section apparently bothered the American Book-of-the-Month Club enough for them to demand that it be cut, along with the chapters quoted from Emmanuel Goldstein's book, as a condition of acceptance by the club. Though he stood to lose at least £40,000 in American sales, Orwell refused to make the changes, telling his agent, "A book is built up as a balanced structure and one cannot simply remove large chunks here and there unless one is ready to recast the whole thing . . . I really cannot allow my work to be mucked about beyond a
certain point, and I doubt whether it even pays in the long run." Three weeks later the BOMC relented, but the question remains, why end a novel as passionate, violent and dark as this one with what appears to be a scholarly appendix?
The answer may lie in simple grammar. From its first sentence, "The Principles of Newspeak" is written consistently in the past tense, as if to suggest some later piece of history, post- 1984 , in which Newspeak has become literally a thing of the past - as if in some way the anonymous author of this piece is by now free to discuss, critically and objectively, the political system of which Newspeak was, in its time, the essence. Moreover, it is our own pre-Newspeak English language that is being used to write the essay. Newspeak was supposed to have become general by 2050, and yet it appears that it did not last that long, let alone triumph, that the ancient humanistic ways of thinking inherent in standard English have persisted, survived, and ultimately prevailed, and that perhaps the social and moral order it speaks for has even, somehow, been restored.--Thomas Pynchon (2003) "Introduction" Nineteen Eighty-Four By George Orwell [HT: Victor Gijsbers]

We know that Pynchon is not the only major novelist that grappled with Orwell's Appendix to Nineteen Eighty-Four; at the end The Handmaid's Tale, Atwood, who according to her own testimony about 1984, "read it again and again," pays homage to it in the "Historical Notes" themselves "Being a partial transcript of the proceedings of the Twelfth Symposium on Gileadean Studies, held as part of the International Historical Association Convention held at the University of Denay, Nunavit, on June 25, 2195." In presenting these historical notes, Atwood, reassures the reader, who may have no taste for (recall) Socratic metaphysics --  "for everything that has come into being destruction is appointed, not even such a fabric as this will abide for all time, but it shall surely be dissolved" (Plato, Republic, 546a)" -- that the "Gilead period," too, will come to an end in some sense. 

Pynchon, who is not exactly known for his optimism, is right to pay attention to the Appendix to 1984, but I am not entirely sure that his hesitant ("perhaps") optimism about what the Appendix signals can withstand scrutiny. His interpretation assumes, first, that the (let's call it) authorial stance of the Appendix has to be taken at face value. That's decidedly odd because 1984 represents us with a world in which just about every text is a forgery of some sorts. (While it would be a mistake to treat them alike, Orwell and Borges were contemporaries, after all.) In fact, much of the action of 1984 consists in the production of forgeries (not just of texts of memories). The production of newspapers by the Ministry of Truth and the torture chambers of Ministry of love have this in common. So, why would the Appendix be any different?

As an aside, the apparent function of the forgeries in Oceania appear to be to produce a totalizing infallibility about everything, not the least the past, of the Party. (No Pope would dream of this!) And what it illustrates is that even if one were to grant (for the sake of amusement) the linguistic idealist, that all is text, what one may call the production values that inform or are exhibited by the production of texts make a huge difference. Linguistic idealism comes in many varieties. If one treats this focus on 'values' as bourgeois, feel free to replace 'production values' with 'means of production' in the previous sentence. In so far as Orwell is a true socialist (and an author), this attention to the conditions of textual production need not surprise. 

But let's stipulate that we need to take the Appendix at face value. You may say, 'What else can we do, after all?' It's true, as Pynchon emphasizes, that the Appendix treats Newspeak as a thing of the past. Given that Pynchon explicitly understands Newspeak as the essence of Oceania, and in his argument tacitly relies on the metaphysical principle, I'll paraphrase Spinoza (E2D2), that when its essence is taken away the entity ceases to be, Pynchon infers from the demise of Newspeak quite reasonably that Oceania must have ended.

Before I get to the merits of Pynchon's argument, a reader may suspect that I will object to Pynchon's metaphysics as an imposition on the world of 1984. Of course, I am assuming you are not the kind of reader who may find my parading of metaphysical jargon quite pedantic. Even so it's worth noting that the core torture scenes in Part III of 1984, emphasize that metaphysics is salient. The torturer, O'Brien, repeatedly, in fact, reminds his victim, Winston Smith, that metaphysics is not his "strong point." So I am not going to object to the form of Pynchon's argument.*

Newspeak is what the philosophers call a final language. In the text -- I have in mind both 1984 and its Appendix -- this is marked by the contrast between the provisional versions (as, I now quote the Appendix, "embodied in the Ninth and Tenth editions of the Newspeak Dictionary") -- and its definitive form "embodied in the Eleventh Edition of the Dictionary."  As we learn from the Appendix and, earlier in the novel, from Syme, who, before he is presumably vaporized, is one of the developers of Newspeak, Newspeak is supposed to be an austere language: its vocabulary culled down, and, through some semantic tools the expressive power of particular words is greatly enhanced, the language's expressive power is curtailed. It's a final language designed to be spoken and used by the Party.**

Now, Syme indeed suggests that the "Revolution will be complete when the language is perfect. Newspeak is Ingsoc and Ingsoc is Newspeak." The narrator of 1984 adds that this is expressed with "a sort of mystical satisfaction." And indeed this idea, the perfection of language, is characteristic of what we may call Enlightenment mysticism (and satirized by Swift). Pynchon's interpretation of 1984 is rooted in Syme's understanding of Newspeak.

The Appendix, which follows Syme's interpretation of Newspeak sometimes literally, omits this mystical claim. I'll quote the Appendix (and then suggest an alternative to Pynchon's interpretation):

The purpose of Newspeak was not only to provide a medium of expression for the world-view and mental habits proper to the devotees of Ingsoc, but to make all other modes of thought impossible. It was intended that when Newspeak had been adopted once and for all and Oldspeak forgotten, a heretical thought--that is, a thought diverging from the principles of Ingsoc--should be literally unthinkable, at least so far as thought is dependent on words. Its vocabulary was so constructed as to give exact and often very subtle expression to every meaning that a Party member could properly wish to express, while excluding all other meanings and also the possibility of arriving at them by indirect methods. This was done partly by the invention of new words, but chiefly by eliminating undesirable words and by stripping such words as remained of unorthodox meanings, and so far as possible of all secondary meanings whatever.

Unlike Syme, the Appendix places greatest emphasis on the fact that the purpose of Newspeak was to make what in 1984 is called 'Thoughtcrime' impossible. Even a Quine-ean couldn't be heretical through indirect methods in Newspeak. We might say, then, that it is regimentation that is the fundamental function of Newspeak as a final language.

Interestingly enough, this is an aside, the author of the Appendix recognizes that once Newspeak is perfected, it would make translation of books and even complex fragments from older languages into Newspeak in a certain way impossible unless they either referred "to some technical processor some very simple everyday actions." (The Appendix offers the Declaration of Independence as an example of how "impossible" would be to render it into Newspeak.) One wishes (recall) our advocates of formal languages would keep this in mind.

Be that as it may, a more natural interpretation of the demise of Newspeak presents itself now.  For, from the point of view of the Party, the elimination of even possible Thoughtcrime would be self-defeating. For Thoughtcrime grounds the repressive apparatus of the Party. The possibility of Thoughtcrimes are -- as much as the conditions of permanent War -- central to the structure of self-discipline of the Party. The elimination of Thoughtcrime as a category would spell doom to the survival of the Party because it would lose its main means of self-control as a ruling, corporate spirit. It would then have to start codifying laws of some sort and bureaucratize, formalize these. This was not a problem in the world of 1984 because the perfection of Newspeak had been projected, as the Appendix notes in its closing sentence, into a more distant future (2050).

So, a more natural political possibility is that the Party abandoned the project of Newspeak because it saw that its perfection would be self-undermining and incompatible with the ideological needs of Ingsoc in power.*** (By contrast, a socialist language planner like Carnap recognizes that Esperanto, for example, is great for a world of public Enlightenment.) And the Appendix reflects the writings of someone who is intimately familiar with the abandonment of Newspeak (not the demise of Oceania).  Admittedly, there are some passages in the Appendix that exhibit a rather free-thinking attitude: the use of the Declaration of Independence as an example, in particular, is rather cheeky. And it forms the best evidence for Pynchon's interpretation. But the more literal reading of the Appendix understands it as a fragment from a reconstruction of the rise and fall of Newspeak as a social enterprise. That is compatible with the survival of Oceania once the Party realized that the mystical attitude toward Newspeak threatened to make it the essence of Oceania.+

 

 

21 March 1979: Foucault Diagnoses American Analytic Philosophy (XXXI)

The second interesting use of these neo-liberal analyses is that the economic grid will or should make it possible to test governmental action, gauge its validity, and to object to activities of the public authorities on the grounds of their abuses, excesses, futility, and wasteful expenditure. In short, the economic grid is not applied in this case in order to understand social processes and make them intelligible; it involves anchoring and justifying a permanent political criticism of political and governmental action. It involves scrutinizing every action of the public authorities in terms of the game of supply and demand, in terms of efficiency with regard to the particular elements of this game, and in terms of the cost of intervention by the public authorities in the field of the market. In short, it involves criticism of the governmentality actually exercised which is not just a political or juridical criticism; it is a market criticism, the cynicism of a market criticism opposed to the action of public authorities. This is not just an empty project or a theorist’s idea. In the United States a permanent exercise of this type of criticism has developed especially in an institution which was not in fact created for this, since it was created before the development of the neo-liberal school, before the development of the Chicago School. This institution is the American Enterprise Institute whose essential function, now, is to measure all public activities in cost-benefit terms, whether these activities be the famous big social programs concerning, for example, education, health, and racial segregation developed by the Kennedy and Johnson administrations in the decade 1960–1970. This type of criticism also involves measuring the activity of the numerous federal agencies established since the New Deal and especially since the end of the Second World War, such as the Food and Health Administration, the Federal Trade Commission, and so on. So, it is criticism in the form of what could be called an “economic positivism”; a permanent criticism of governmental policy.
Seeing the deployment of this type of criticism one cannot help thinking of an analogy, which I will leave as such: the positivist critique of ordinary language. When you consider the way in which the Americans have employed logic, the logical positivism of the Vienna School, in order to apply it to scientific, philosophical, or everyday discourse, you see there too a kind of filtering of every statement whatsoever in terms of contradiction, lack of consistency, nonsense. To some extent we can say that the economic critique the neo-liberals try to apply to governmental policy is also a filtering of every action by the public authorities in terms of contradiction, lack of consistency, and nonsense. The general form of the market becomes an instrument, a tool of discrimination in the debate with the administration. In other words, in classical liberalism the government was called upon to respect the form of the market and laisser-faire. Here, laissez-faire is turned into a do-not-laisser-faire government, in the name of a law of the market which will enable each of its activities to be measured and assessed. Laissez-faire is thus turned round, and the market is no longer a principle of government’s self-limitation; it is a principle turned against it. It is a sort of permanent economic tribunal confronting government.--Michel Foucault, 21 March, 1979,  translated by Graham Burchell, Lecture 10, The Birth of Biopolitics, 246-247 [emphases added]

It's difficult to leave the riches of lecture 9 behind without exhausting all possible angles of comment, but it it is clear that in lecture 10 Foucault starts harvesting the fruits of the first nine lectures. This lecture contains two great set-pieces: (i) the comparison between the neo-liberalism of ORDOs and Chicago-school in part to explore the contrasting analysis of the relationship between market/state and market/society; (ii) the evolution of what we may call law and criminality since the Enlightenment in light of Becker's approach. So, it may strike you as a bit perverse that today's post zones in on an apparent aside. 

So, the emphasized passage is odd because it is by no means obvious that Foucault's audience ("one") will have thought of this particular analogy. As I have noted throughout my comments, Foucault is constantly assuming a kind of Bien pensant marxism in his audience. I don't want to make too much of the grammatical structure of his thought -- and I am not conformatible in the French [On ne peut pas ne pas penser, en voyant s'exercer ce type-là de critique à une analogie que je laisse encore une fois sous la forme d'analogie], but it is notable that the (analogical) thought is inevitable (and that Foucault represents his own thought in such disassociated way, and being so analogical) while Foucault takes responsibility for conveying it in the (analogical) manner he has the thought. 

Before, I get to the substance of Foucault's claim, a bit of context is useful. First, I would not be surprised if the possibility of visiting Berkeley (which starts the following year) was not already on Foucault's horizon. As we know he was visiting California from the mid 70s onward. Second, Foucault would have tracked Derrida's confrontation with analytic philosophy, first in the brilliant, and oddly neglected, (1973) The Archeology of the Frivolous. This book is ostensibly about Condillac, but it is simultaneously (recall this post) a critique of Foucualt's earlier practice of archeology and a clever exploration of the nature of analysis. I would be amazed if Foucault were unfamiliar with it because I think it can help explain some important shifts in Foucault's method(s)/stance(s). And then in the infamous exchange with Searle (Foucault's future colleague) that unfolded through the 1970s.* I assume that Foucault would have been familiar with this 'event'. So, perhaps, for Foucault, a possible personal confrontation with American analytic philosophy would have made the thought inevitable.

Let's return to the emphasized bit in the quoted passage from lecture 10 above. And to avoid misunderstanding, Foucault here is not discussing ordinary language philosophy. Earlier, in Lecture 4 (see p. 76) Foucault had noticed the significance of the movement from Vienna (and Freiburg) of certain intellectual practices to the States.** And this point gets amplified by Foucault's surprising organicist commitment in lecture 8, that when ideas travel from their origin, they are radicalized/amplified in new environments. (We may call such intensification in a new ecological niche a law of memetics!) So, on Foucault's view, we may say that what starts out as possible source of emancipation in Left Vienna, is turned into a form of speech policing Stateside.+ 

And Foucault is explicit that the policing speech is not just directed at ordinary speech, but also at scientific and philosophical speech. The significance of this is that Foucault interprets the (ahh) tools and practices of this branch of analytic philosophy in terms of technologies of control.

As an aside, it is worth noting, and a kind of historical irony, that Foucault here is kind of tracking the general atmosphere of victory of Quine-ean regimentation and, especially, a wider naturalizing program(s) over the Carnapian explanation, which had explicitly limited itself (recall here; and here) to artificial speech. I use 'general atmosphere' because what Foucault is describing (as theoretical virtues and non-virtues "contradiction, lack of consistency, nonsense",) while undoubted familiar in some sense, is surely not Quine-ean regimentation (with its own specialist clarified language)--this is a diffuser form of score-keeping familiar from polemics and classroom instruction.

So, I use 'general atmosphere' for another reason. Because the analogy itself is I think itself intended to point to what happens when certain practices leave the rarified technical discussion in elite journals and seminar rooms, and become a more general mode of thought. For, as a disciplinary practice of deploying (philosophical) technologies of speech control that American analytic philosophy is used as an elucidatory analogy for the fact of what happens when think tanks deploy techniques developed by the Chicago school to try to score and police government policy.

This is not to say one can avoid mentioning Quine. For Quine's (rather juridical) tribunal of experience is now transformed (by way of the magic of analogy) "into permanent economic tribunal confronting government."  That is to say, as the ideas of the Chicago school leave Hyde Park and are institutionalizes by well financed interests, a democratic-republican government is now made to account itself not to the voters, and their representatives, or the press, or supreme court judges, but to an intellectual edifice that deploys (protected) theoretical speech. And this edifice represents itself a strand of the liberal tradition that, as Foucault argued in lecture 9, is itself one of the animating principles of the constitutional order. 

I return to the previous paragraph in subsequent digressions on Birth of Biopolitics. Foucault leaves hanging what to make of his view of American analytic philosophy. It's important to realize that the analogy is not a moralizing criticism. If I understand Foucault's larger meta-philosophical (political) position, he thinks that the development and application of technologies of control have a kind of inevitability to them. And it is not impossible -- but here i speculate -- that he believes that (the philosophical politics of) philosophical speech itself is (given the significance of the law of non-contradiction to its way of being) also, always, constituted by technologies of control. And what may be novel here, is unusual clarity and zest of the Americans in deploying such technologies of speech control.

 

 

 

 

On the Convergence of Economics and Computer Science (with comments on the role of moral reasoning in AI)

We can generalize this idea across our entire model space. Suppose we enumerated all the possible numerical pairs <error, unfairness> achieved by the models we are considering (e.g., SAT cutoffs)....

So each point corresponds to a different model; the x-coordinate of the point is the model’s error, and the y-coordinate is its unfairness score...The key thing to realize is that any model that is not on this boundary is a “bad” model that we should eliminate from consideration, because we can always improve on either its fairness score or its accuracy (or both) without hurting the other measure by moving to a point on this boundary.

The technical name for this boundary is the Pareto frontier or Pareto curve, and it constitutes the set of “reasonable” choices for the trade-off between accuracy and fairness. Pareto frontiers, which are named after the 19th-century Italian economist Vilfredo Pareto, are actually more general than just accuracy-fairness trade-offs, and can be used to quantify the “good” solutions to any optimization problem in which there are multiple competing criteria. One of the most common examples is the “efficient frontier” in portfolio management, which quantifies the trade-off between returns and risk (or volatility) in stock investing.

The Pareto frontier of accuracy and fairness is necessarily silent about which point we should choose along the frontier, because that is a matter of judgment about the relative importance of accuracy and fairness. The Pareto frontier makes our problem as quantitative as possible, but no more so.

The good news is that generally speaking, whenever we have practical algorithms for “standard,” accuracy-only machine learning for a class of models, we also have practical algorithms for tracing out this Pareto frontier. These algorithms will be a little bit more complicated—after all, they must identify a collection of models rather than just a single one—but not by much... 

While the idea of considering cold, quantitative trade-offs between accuracy and fairness might make you uncomfortable, the point is that there is simply no escaping the Pareto frontier. Machine learning engineers and policymakers alike can be ignorant of it or refuse to look at it. But once we pick a decision-making model (which might in fact be a human decision-maker), there are only two possibilities. Either that model is not on the Pareto frontier, in which case it’s a “bad” model (since it could be improved in at least one measure without harm in the other), or it is on the frontier, in which case it implicitly commits to a numerical weighting of the relative importance of error and unfairness. Thinking about fairness in less quantitative ways does nothing to change these realities—it only obscures them.

Making the trade-off between accuracy and fairness quantitative does not remove the importance of human judgment, policy, and ethics—it simply focuses them where they are most crucial and useful, which is in deciding exactly which model on the Pareto frontier is best (in addition to choosing the notion of fairness in the first place, and which group or groups merit protection under it, both of which we discuss shortly). Such decisions should be informed by many factors that cannot be made quantitative, including what the societal goal of protecting a particular group is and what is at stake. Most of us would agree that while both racial bias in the ads users are shown online and racial bias in lending decisions are undesirable, the potential harms to individuals in the latter far exceed those in the former. So in choosing a point on the Pareto frontier for a lending algorithm, we might prefer to err strongly on the side of fairness—for example, insisting that the false rejection rate across different racial groups be very nearly equal, even at the cost of reducing bank profits. We’ll make more mistakes this way—both false rejections of creditworthy applicants and loans granted to parties who will default—but those mistakes will not be disproportionately concentrated in any one racial group. This is the bargain we must accept for strong fairness guarantees. Michael Kearns & Aaron Roth (2019) The Ethical Algorithm: The Science of socially Aware Algorithmic Design, Oxford University Press, 80-84 [the linked page should have their diagram]

A few weeks ago, after I started blogging (here and here) about the ethics of algorithms,* I received a Facebook advertisement-notification from Amazon that Michael Kearns and Aaron Roth (the authors of The Ethical Algorithm) had become Amazon scholars. "What an interesting way," I thought, "to advertise books to me." I clicked on the link and read the corporately sponsored interview with them. Perhaps because my expectations were low , but I was surprised how thoughtful -- "the main thesis of our book, which is that in any particular problem you have to start by thinking carefully about what you want in terms of fairness or privacy or some other social desideratum, and then how you relatively value things like that compared to other things you might care about, such as accuracy" -- they sounded and how relevant to my own project. It was clear theirs was a project about thinking about trade-offs (a favored phrase of theirs) about social values in algorithm design. More important to me, since I had just argued that thinking about the ethics of algorithms seems to replicate the very social problems familiar from thinking about ethics in economics, I wanted to see if my hunch was correct. Two clicks later I ordered the book, and a day later it was delivered.

A few years ago (well back in 2012), I used to joke that as even theoretical economists become more applied, becoming (thanks to increasingly cheap computer power) experts at massaging results out of giant "administrative data-sets,"+ economics ran the risk of being displaced by statisticians and, especially, computer scientists. While then I was largely ignorant of deep learning, it was clear to me that fruitful searches for robust and surprising correlations in data didn't require (as Justin Wolfers then suggested) the restrictive assumptions of economic theory. (Think about it: econometrics and economic theory artificially restrict the search space that can be more fully explored with deep learning machinery.) And indeed deep learning can dispense with economic theory.**

But in reading their interview and their splendid The Ethical Algorithm I realized that back in 2012, I had missed a crucial issue: that algorithmic design is conceived in terms of optimization problems under constraints. Since Lionel Robbins (back in the 1930s) this just is the definition of economics (and it enabled a split between ethics and economics). And once in algorithmic design you are interested in more than predictive accuracy, and so have to deal "with multiple competing criteria," a whole set of mathematically precise diagnostic tools familiar from economics can be imported into algorithmic design (as the passage quoted above suggests). So, I now realize that increasingly computer science and economics will merge (as presumably is happening already in finance-notice the passing mention of "portfolio management" in the block quote above). 

The previous is sufficient for a digression. But I also want to call attention to how moral issues are conceived, and not, in the (ahh) paradigm articulated by Kearns and Roth. And, in particular, I want to call attention to four levels, or sites, where they enter in. So, formally moral issues enter into two sites of algorithmic design: (i) first a choice in setting a goal (such as privacy or fairness) or success criterion for the algorithm. In general, Kearns and Roth view this as a constraint on accuracy that generates a trade-off among accuracy and other possible criteria. Kearns & Roth are very good at explaining that the way one makes precise these moral goals need not be univocal or can be very context sensitive. (So, there are different ways to think about fairness or privacy in algorithmic settings.) 

In turn, these trade-offs can be modeled in terms of pareto frontiers. And this generates the second site: (ii) at the level of a decision about ''which model on the Pareto frontier" to use in practice. Conceptually that's a distinct choice from how to encode 'fairness' into an algorithm, but obviously one can imagine that in practice, in the spirit of experimentation, there are going to be interactions between (i) and (ii) within a general goal-oriented design process. (Recall that an algorithm can be understood, even identified, in terms of the functions and goals it serves--something I have promised to return to.) 

Some time soon, I hope to write some more critical posts about their hope of turning algorithm design into a kind of dream of turning ethics into an optimization problem. (That way of putting it is indebted to Kathleen Creel, a young scholar who has written a gem of a paper on opacity in computational systems.) And one can see how the combination (i) and (ii) lend themselves, despite the situationism in (i), to a form of moral reasoning familiar from (and analogous, if not identical to) utilitarianism. Of course, one can let other values enter into the decision of (ii). 

Now, in their argument (ii) tends to happen in algorithm design/development process within, say, a company. But the choice one makes about which model on the Pareto frontier to inhabit generates (ahh) consequences to wider society. Some of these wider consequences -- laws, regulations, social norms, etc. -- are already internalized as constraints, but some are simply outside the company's mission/attention and not anticipated by the legislature.

The point in the previous paragraph fits their general argument. Because the "core concern" of their book is that "optimizing" on some explicit social goals, predictably and foreseeable, has (even in the relatively short term) unintended and perhaps ex ante unknowable side effects. (188) There is, thus, not just (as Kearns & Roth note) a (iii) a prior moral and political decision to be made whether or the extent to which algorithms are permitted or primarily responsible for decisions in a domain or (as they often describe it) norm enforcement (177). In the book they offer as an example, of the former "automated warfare," (175; 178).  But there is also (IV) a question how society should think about, and perhaps compensate for or mitigate, these entirely unintended and perhaps unknowable side effects that are a foreseeable outcome pattern from embracing algorithmic practices. This matters especially if such outcome patterns (recall) create  asymmetric harm patterns to vulnerable populations.

And, not unlike the economists, the way Kearns & Roth have set up their conceptual scheme (IV) turns out to be delegated to policy and so is not part of the ethical algorithm at all (or the responsibility of the firms that profit from them). To put this in their terms, we (now) know that a system with the very best portfolio management can create unexpected, general externalities. Something similar is now foreseeable in their field. (I am not sure what the equivalent term of true Knightian uncertainty in AI is, but they need it!) So given (IV), it would be good to prevent regressive forms of socialization of risk while privatizing profit.

Le Bon vs Condorcet

If the individuals of a crowd confined themselves to putting in common the ordinary qualities of which each of them has his share, there would merely result the striking of an average, and not, as we have said is actually the case, the creation of new characteristics. How is it that these new characteristics are created? This is what we are now to investigate.
    Different causes determine the appearance of these characteristics peculiar to crowds, and not possessed by isolated individuals. The first is that the individual forming part of a crowd acquires, solely from numerical considerations, a sentiment of invincible power which allows him to yield to instincts which, had he been alone, he would perforce have kept under restraint. He will be the less disposed to check himself from the consideration that, a crowd being anonymous, and in consequence irresponsible, the sentiment of responsibility which always controls individuals disappears entirely.
   The second cause, which is contagion, also intervenes to determine the manifestation in crowds of their special characteristics, and at the same time the trend they are to take. Contagion is a phenomenon of which it is easy to establish the presence, but that it is not easy to explain. It must be classed among those phenomena of a hypnotic order, which we shall shortly study. In a crowd every sentiment and act is contagious, and contagious to such a degree that an individual readily sacrifices his personal interest to the collective interest. This is an aptitude very contrary to his nature, and of which a man is scarcely capable, except when he makes part of a crowd.
   A third cause, and by far the most important, determines in the individuals of a crowd special characteristics which are quite contrary at times to those presented by the isolated individual. I allude to that suggestibility of which, moreover, the contagion mentioned above is neither more nor less than an effect.
   To understand this phenomenon it is necessary to bear in mind certain recent physiological discoveries. We know to-day that by various processes an individual may be brought into such a condition that, having entirely lost his conscious personality, he obeys all the suggestions of the operator who has deprived him of it, and commits acts in utter contradiction with his character and habits. The most careful observations seem to prove that an individual immersed for some length of time in a crowd in action soon finds himself—either in consequence of the magnetic influence given out by the crowd, or from some other cause of which we are ignorant—in a special state, which much resembles the state of fascination in which the hypnotised individual finds himself in the hands of the hypnotiser. The activity of the brain being paralysed in the case of the hypnotised subject, the latter becomes the slave of all the unconscious activities of his spinal cord, which the hypnotiser directs at will. The conscious personality has entirely vanished; will and discernment are lost. All feelings and thoughts are bent in the direction determined by the hypnotiser.
   Such also is approximately the state of the individual forming part of a psychological crowd. He is no longer conscious of his acts. In his case, as in the case of the hypnotised subject, at the same time that certain faculties are destroyed, others may be brought to a high degree of exaltation. Under the influence of a suggestion, he will undertake the accomplishment of certain acts with irresistible impetuosity. This impetuosity is the more irresistible in the case of crowds than in that of the hypnotised subject, from the fact that, the suggestion being the same for all the individuals of the crowd, it gains in strength by reciprocity.
The individualities in the crowd who might possess a personality sufficiently strong to resist the suggestion are too few in number to struggle against the current. At the utmost, they may be able to attempt a diversion by means of different suggestions. It is in this way, for instance, that a happy expression, an image opportunely evoked, have occasionally deterred crowds from the most bloodthirsty acts.
   We see, then, that the disappearance of the conscious personality, the predominance of the unconscious personality, the turning of feelings and ideas in an identical direction by means of suggestion and contagion, the tendency to immediately transform the suggested ideas into acts; these, we see, are the principal characteristics of the individual forming part of a crowd. He is no longer himself, but has become an automaton who has ceased to be guided by his will.
   Moreover, by the mere fact that he forms part of an organised crowd, a man descends several rungs in the ladder of civilisation. Isolated, he may be a cultivated individual; in a crowd, he is a barbarian—that is, a creature acting by instinct. He possesses the spontaneity, the violence, the ferocity, and also the enthusiasm and heroism of primitive beings, whom he further tends to resemble by the facility with which he allows himself to be impressed by words and images—which would be entirely without action on each of the isolated individuals composing the crowd—and to be induced to commit acts contrary to his most obvious interests and his best-known habits. An individual in a crowd is a grain of sand amid other grains of sand, which the wind stirs up at will.
   It is for these reasons that juries are seen to deliver verdicts of which each individual juror would disapprove, that parliamentary assemblies adopt laws and measures of which each of their members would disapprove in his own person.--Gustave Le Bon (1986 [1895] [Psychologie des Foules] The Crowd: A Study of the Popular Mind, pp. 6-8 (translation of second edition published by Dover)

The (1785) Condorcet Jury theorem tells us that with modest assumptions, the rule of more is better and, thereby, vindicates democracy from the verdict of history by all the learned. A century later, near the end of his book, Le Bon  admits that "the parliamentary system represents the ideal of all modern civilised peoples. The system is the expression of the idea, psychologically erroneous, but generally admitted, that a large gathering of men is much more capable than a small number of coming to a wise and independent decision on a given subject." (123) While the official theme of Le Bon's book is to offer an explanation of the functioning of crowds qua crowds, there are two sub-themes in his argument: first, how come the psychologically erroneous idea persists as a civilizational dogma (that is, Le Bon gives us a theory of the rise and fall of civilizations and institutions)?; second, why does the Jury Theorem fail so often in practice, despite the fact that it should function as a self-fulling prophecy given that people widely expect it to succeed?+ Lurking in the second sub-theme is an account of why institutions fail in different places despite the pull of incentives, and why demagogues may succeed.

Le Bon is no friend of democracy. But in this book he has little interest in offering a replacement for democracy because he accepts Tocqueville's claim that in an era of equality men have "an almost limitless confidence in the judgments of the public." In fact, he quite clearly argues that limiting the franchise to a meritocratic sub-set (a la Mill or Brennan) won't improve matters. Because for him once in a crowd, even widely diverging people behave similarly and lose a lot of the characteristics that give them their individuality.* In a crowd "men always tend to the same [low] level.....with regard to social problems... mean are substantially, equally ignorant."

Before I confront the announced subject of this post, I need to be clear about some of his population assumptions. Le Bon works with a contrast between barbarism and civilization. And once in a crowd otherwise civilized people are likely to become barbaric. Despite Le Bon's racism, all kinds of people can become civilized. (Throughout the book, he treats Islam as one of the great civilizations.) Le Bon also works with a racial theory. And it does important work for him because for him races have deep-seated commitments that help explain how people qua people respond differently to stimuli and incentives.

I actually think that what Le Bon calls a 'race,' is (for students) better translated by 'nation.' For, while Le Bon is undoubtedly casually racist (and sexist) in our modern sense (and clearly admires Anglo-Saxons over his own Latins), his race theory turns out to be primarily sociological: that is, a race is constituted by shared history, shared dogmatic commitments, and shared practices (including language). So, Le Bon's 'race' is not far removed from Renan's 'nation.' And crucially the biological stock of a race is never pure according to Le Bon. (This is because for him barbarism, which is always prior to civilization, is always a period of mixing.) And somewhat surprisingly -- given his reactionary mindset -- when explaining crowds no race escapes being barbaric in the crowd.

The brilliance of Condorcet's proof resides, in part, in the fact that (a) he assumes very modest voter competence. All that is needed (recall) is better than chance odds for getting things right. Let's call this average intelligence. Unlike modern students of Condorcet, who recognize that the conditions of (b) independence and (c) lack of communication, may often be violated, Le Bon thinks this (a) is the weak part of Condorcet's proof. For  while Le Bon accepts Condorcet's assumption that in a crowd we are (d) sincere, Le Bon thinks that once we become part of a crowd (and for him juries are also crowds) in our sincerity our competence reduces below such average. We become as it were for him hypnotized automatons. 

To be sure, Le bon thinks that in a crowd we stop acting independently in large part because through the mechanism of contagious sympathy, we end up being suggestible by the same influences. Group diversity, which might give a crowd intellectual strength, evaporates for Le Bon and we become homogeneous, hypnotized automatons in a crowd.  And while the Stoic sage is homogeneous with other Stoic sage in his philosophical virtue, crowds act like one in pursuit of one idol or another.

One may well wonder if the shared feeling of invincibility is really the triggering cause of crowd-like phenomena, and to what degree hypnosis is really the best physiological explanation and explanatory analogy for crowd behavior. I am here not to defend Le Bon. But I do note it is no objection to Le Bon's theory that some crowds get it right or produced fine political outcomes. Le Bon explicitly admits that crowds may well be heroic or help found great civilizations (etc.). If you are fond of great strikes or mass protests, Le Bon is happy to admit that these shape history.

Rather the key point is that friends of Condorcet's jury theorem assume methodological individualism: that being within a jury/crowd baseline competence is not reduced or at least not so reduced as to fall under average. And Le Bon points out this commitment (a) needs to be earned if, as is not wholly implausible, cognitive characteristics are changed when in large company (subject to polarization, group loyalty, sense of justice, etc.).

One way to put this is that in practice, the assumptions of independence and lack of communication secure the modest competence the Jury Theorem relies on. Le Bon's point is recognized by many procedural interventions/norms that prevent a group of deciding individuals to act as crowds.** That is to say, deliberative democrats are, in a way, Le Bon's true heirs. 

 

 

 

Models, Politics, Populations, and Public Science in times of Pandemic (IV); on Ernest Nagel and the role of Philosophy of Science

In the midst of actual and impending disaster, men are inclined to listen to any voice speaking with sufficient authority; and during periods of social crisis, when rational methods of inquiry supply no immediate solutions for pressing problems, spokesmen for institu­tional and philosophic theologies find a ready audience for a sys­tematic disparagement of the achievements of empirical science. Ideas which the advance of knowledge had partially driven under­ground during periods of fair social weather, are then insolently proclaimed as panaceas for public and private ills. The assured methods of scientific control and understanding, because they effect no wholesale resolution of problems and because they yield no con­clusions beyond the possibility of error and correction, are then declared to be unsuitable guides for rational living.
The mounting economic and political tensions of our own age have not failed to produce a literature of this type. From various quarters-from men of science, historians of ideas, as well as out­spoken representatives of theological systems-there has come a flood of criticism of modern science and of the secular naturalism which has accompanied its growth. The criticism has been neither uniform nor consistent. But the common objective of much of it has been the limitation of the authority of science, and the institution of methods other than those of controlled experimentation for dis­covering the natures and values of things. Many recent evaluations of science have thus had an obviously malicious intent; and the pres­ent essay will seek to determine briefly to what extent, in the case of several influential types of philosophies of science, good sense has been sacrificed to such malice.--Ernest Nagel (1943) "Malicious Philosophies of Science" Partisan Review (10), jan-feb, reprinted in Sovereign Reason, (1054), p. 18.*

Today is the fourth in a series of posts (see the first here inspired by Phillipe Lemoine; and, inspired by Stegenga, the second here and third here).

I return (recall) to Nagel's brilliantly titled piece because Nagel is explicitly committed to defending the "authority of science," perhaps even unlimited authority of science in a time when it is under attack. In this piece from which I quoted Nagel does not define science, but inter alia he associates it with "controlled experimentation" and methods that allow "the possibility of error and correction" and "objective measures ((33)when discussing well-being). In addition, he clearly believes that science is often a "painfully slow road" (31), which is why there will be circumstances "where rational methods of inquiry supply no immediate solutions for pressing problems."  This entails that sometimes some questions will have to be answered with appeal to "brute facts" when theoretical resources have bottomed out (29). 

I was reminded of Nagel's piece because of the discussion in Stegenga's essay  "Fast Science and the Philosophy of Science" (first published at Auxiliary Hypothesis BJPS and then Dailynous). In it contemporary anxieties about the role of philosophy of science are explored with a wide variety of leading practitioners. At one point Stegena notes:

Many philosophers of science are concerned with the demise in public trust of science, evident in debates about teaching intelligent design in classrooms or in mistrust of climate science. Conversely, many philosophers of science are concerned about excessive trust in particular domains of science, exemplified by, say, unreliable research on pharmaceuticals.  Of course, it is perfectly consistent to hold that science is the best means to learn about our world, while also criticizing sloppy science. 

One further, underlying anxiety we philosophers of science have is to what degree is the authority of the fast, policy apt sciences worth defending.  As Jonathan Fuller puts it in a very thoughtful and illuminating essay, "The normal process of scientific scrutiny and peer review has given way to a fast track from research offices to media headlines and policy panels." Yet this fast science now lend(s) authority to policies with far-reaching consequences pursued by many governments. So, for example, I sense in my sometime co-author, Eric Winsberg, a concern that doubling down on this fast, policy apt science, will risk discrediting the authority of science down stream.  

Now, Nagel's essay is superb at discrediting the attempts by those who wish to argue for the  limitation of the authority of science. (He engages Gilson, Whitehead, Knight, Blanchard, Maritain, Balfour, etc.) Many of their arguments are familiar, and so his essay remains worth re-reading. And since Nagel is the great champion (within pragmatism) of scientific philosophy, his defense of the authority of science comes as no surprise. But if one happens to be reading Nagel in circumstances not entirely unlike the ones he diagnoses in the first sentence of the quoted passage, one is also left with an unexpected conundrum. Let me explain.

By Nagel's explicit lights, there will be circumstances in which slow science cannot offer "immediate solutions for pressing problems." Simultaneously, Nagel's position presupposes there is no such thing as fast science that does offer reliable solutions for urgent problems. To put this in terms of Kofi Bright and Bradley, "We are going to have to respond to COVID19 in absence of the sort of knowledge we should ideally prefer when making rational decisions of great social import." (I quote them because they represent the most accessible and pertinent example of the stance of a scientific philosopher today.)

That is to say, Nagel does not confront the circumstances in which the authority of science is imperfect (and known to be so), and where appealing to it may risk its further authority when it is needed and can be relied upon (again).  Jonathan Fuller's answer to these circumstances is that even so we must rely on imperfect science, all the while trying to "combine theory with evidence and make use of diverse data while demanding data of increasingly higher quality," but keep a "critical" attitude. Kofi Bright and Bradley advocate disciplined decision making, but with heroic awareness of the "responsibility of hard choices" and the possibility even inevitability of "tragedy." Neither Fuller nor Kofi Bright/Bradley really answer Winsberg's question whether relying on flawed science now is worth the risk if it undermines future credibility.

Now, one may think that 'given that we must choose, we may as well choose while relying on an imperfect instrument instead of no instrument. And if we live another day, and need to rebuild the authority of that instrument, we'll fight that battle then.' Perhaps, Winsberg's concern is overblown. 

Of course, it need not be true that an flawed imperfect** instrument is better than no instrument. Sometimes when imperfect means flawed then imperfect instruments can be worse than no instruments. There are, for example, a small sample of airplane crashes where if the pilot hadn't been so trained up, or reliant, on the instruments good sense could have saved the passengers. And if one knows one is in those circumstances then it is foolish to stick with imperfect tools.

The problem is that Nagel's version of scientific philosophy can't really acknowledge the possibility of the equivalent to good sense in my example in the previous paragraph. That's a problem with his philosophy of science.+

As an aside, it is pretty clear what works when a pandemic is about to develop, and Taiwan, New Zealand, and South Korea have all shown how to execute it in ways that are sensitive to scientific authority and respectful of political and civil liberties. The problem elsewhere is there is no obvious good sense once that moment has passed and so one is stuck with the imperfections of fast, policy apt science.

Let me close, Nagel is very disciplined about recognizing that there occasions when the authority of science also means one has no answers. I think, then, he would urge caution rather than relying on fast science. It would be nice to be able to say in such circumstances, let's gather more evidence first before we act. My own view is that, more often than not, we should be willing to delay until we have authoritative guide in science. But, of course, no action can also carry a cost, and there are times where this stance is also foolish. 

But if we must choose -- be it in the heroic-tragic mode of Kofi Bright and & Bradley or in the critical mode of Fuller (these are sometimes compatible) --, then I don't see why philosophy of science can't allow that one is guided by "mystery;" to pick an example that exercises Nagel, say, with the thought that "God is" her "own cause" (29) and, while choosing, embrace one's fate joyfully with prayer, or without.

Models, Politics, Populations, and Public Science in times of Pandemic (II); on Policy Apt Synthetic Philosophy

This post is inspired by recent discussion on Jacob Stegenga's facebook page, and a draft essay he wrote as a follow up to it; this post is also part of a series of posts I started with yesterday's post (which responded to  issues raised by Phillippe Lemoine's essay). Today's post will largely abstract away from the pandemic.--ES

By drawing on ideas about synthetic philosophy (recall), I mean by 'policy-apt synthetic philosophy’ a style of philosophy that brings together insights, knowledge, and arguments from the special sciences and normative theories with the aim to offer a reasonably coherent account of complex systems and, thereby, to connect these to decisions under uncertainty in policy and/or debate about policy. Policy apt synthetic philosophy is made possible by, and a response to problems subsequent the intellectual division of labor within and among rather esoteric academic disciplines and the need to make decisions under uncertainty that ramify across many social domains. 

Before I explain what policy-apt synthetic philosophy is, it may be useful to contrast it with two other related, familiar intellectual activities that it may draw from but it is not identical to. (In practice there will be lots of blending, and that is okay.) First, in applied ethics decision contexts are narrowed down or constrained such that it is possible to apply a particular principle or decision rule/practice to it.  That is, I am conceiving of applied ethics as an instance of casuistry in a descriptive not pejorative sense. (I have been much influenced by a leading campaigner for abolition of the death penalty, Hugo Bedau's (1997) Making Mortal Choices.) In casuistry cases may be very high stakes (life and death) and may seem intractable because they involve conflicting principles or important trade-offs in values. In the best forms of casuistry, the search for the right principles is a bottom up enterprise, "a multiple triangulation of the region in which the best answer lies, the parameters or boundaries of the region being determined by the relevant ethical principles." Naturally, applied ethics often does more than that confronting constrained and partially standardized choice moments when it, for example, helps develop normative principles for whole processes/fields  of activity or decision rules for the allocation of scarce resources. This will involve the ordering and integration of principles and/or a hierarchy of priorities that figure into such constrained decisions that can be made to fall  under a principle (etc.). When I use 'applied ethics,' I mean this wider sense.

Of course, professional applied ethicists  do a lot more than what I have just called 'applied ethics,' including reconceptualizing what counts as a decision (or, say, a disability), and how we should think about about structural features that may be ignored or invisible if we focus primarily on decisions and not the social institutions/norms/resources (etc.) that constrain and structure our lives. Some important professional applied ethicists even think we should reject the focus on choice altogether. That is fine. That they do a lot more of significance is the intellectual labor that is presupposed in my discussion here.

Second, another intellectual practice that has been developed around constrained choice is decision theory (e.g., practiced in philosophy, economics, game theory, and computer science departments). This intellectual practice is concerned with designing tools to reason in a disciplined fashion in choice settings. In particular, it finds ways of aggregating evidence about the environment and outcomes into a mathematical technology so that a mathematically constrained procedure necessitates a course or range of actions in particular choice circumstances.

Readers who will have noticed my appeal to scarcity in my description of applied ethics, will have already noticed that there can be overlap even identity between decision theory and applied ethics. That is to say, much decision theory is de facto normative. Utilitarians, for example, find it relatively easy to move between applied ethics and decision theory because much of normative decision theory is about maximizing/optimizing expected utility. Focusing on expected utility is not the only way one can do decision theory. But generally important normative principle are already internalized in the mathematical technology or the standard understanding of the decision theoretic tools. Even so, and I return to this, this is just a subset of the normative principles and commitments that are taken as significant in applied ethics.  One way to distinguish decision theory from applied ethics, is that the former tends to be about generic choices, while the latter tends to be domain specific, even though the domain can be vast: so, for example, my own work in the ethics of financial debt covers practices in multiple social organizations that impact the whole economy the natural environment, and the rise and fall of states (etc.)

At this point one may wonder why, if we already have applied ethics and decision theory, policy-apt synthetic philosophy is even needed. For in many areas of applied ethics, in particular, ('synthetic') knowledge of the relevant special sciences and complex social activities are presupposed. There are at least three important reasons that I will present as independent, but that also mutually strengthen each other. First, what applied ethics and decision theory have in common is constrained choice. And, what I assume is, that this feature is internalized in the way of thinking (the modeling, the standard examples, the policy advice, etc.) in the fields that focus on decision theory and applied ethics. And for many contexts that is incredibly apt.

But there are a range of policy circumstances, where by treating a choice as constrained in a particular way one may end up doing more harm than good. Paradigmatically these are going to be circumstances in which the very constraints that make it possible to think of a choice as constrained, will make it impossible to track or take into account a range of effects or values that will be harmed or prevented by the decision. (This is why some applied ethicists dislike the focus on choice!) This is a feature not a bug of disciplined decision making in applied ethics and decision theory. But as is well known, in reality there can be huge interactions among different kinds of choices. And so we need a place where the, as it were,  'outputs' of particular areas of applied ethics and decision theory, are themselves integrated and studied and compared with the 'outputs' of different scientific domiains. 

Second, one may think that decision theory is well situated to do so this itself. So, for example, recently Liam Kofi Bright and David Bradley position decision theory as the activity ot generating not "just a single probability distribution over the outcomes of interest, but a family of them." And in so doing one can help policy makers "to choose actions that can be expected to yield benefits across a very wide range of scenarios." But notice three features, (i) this privileges expected utility (or close analogues); but in many political contexts one may well wish to involve other important values such as various civil rights, respect, autonomy, dignity, and various freedoms (of association, of movement, etc.), etc. In order to think of choices generically, decision theory flattens the value-landscape and in some policy contexts this is unwelcome.*

(ii) Integrating and comparing probability distributions (the outputs of scientific modeling) does not come automatically from heaven. In particular, it involves comparing many different kinds of models that have many ad hoc but socially salient features. Or where the features and practices are entirely standard but have enormous social ramifications. (Recall yesterday's discussion of the significant differences between modeling with homogeneous or heterogeneous agents.) This comparing and integrating of different kind of models and values, requires, when these are not domain specific, the skilled activity of the sort I have called synthetic philosophy. (So, in one sense the synthetic philosopher is the underlaborer to the decision theorist.)

(iii) Not unlike most decision theorists, Kofi Bright and Bradley assume that one can assign probability distributions. There are two problems here: a) what makes something a crisis, often in the face of very new circumstances, are precisely circumstances where one can reasonably doubt one has a grip on any of the relevant underlying distributions, that is, one is faced with true uncertainty. And b) often when we assign probability distributions, there are genuine possibilities one cannot then model at the same time. (That is, this is an instance mentioned above: the very constraints that make it possible to think of a choice as constrained, will make it impossible to track or take into account a range of effects or values that will be harmed or prevented by the decision.) One reason I mention Kofi Bright and Bradley is that they are transparent about the limitations of decision theory.

Third, often in a crisis, the normal operations of science is attenuated even partially suspended. That is, the ordinary stress testing of the concepts, data, and methods over an extended period of time and give and take of scientific debate is compressed. So, even the best informed scientific aggregators (or what Polanyi called 'influentials') can't rely on underlying community consensus. In a crisis, what Jacob Stegenga calls 'fast science'  is really fast. So, policy apt scientific philosophy is needed to do the assimilation, partially in public and in real time, of unfolding sciences so that it, down the road, can be turned into the input for disciplines that work on some constrained choice or a field of study.

One may well think that there is no need for policy apt synthetic philosophy because that's why governments have offices for scientific policy/advice that help coordinate decisions that involve multiple sciences and policy domains. And, indeed, such offices often help prepare the inter-agency and inter scientific crisis meetings that have been in the news. And such offices, in turn, can draw on many kinds of experts including applied ethicists and decision theorists and ad hoc government advisors drawn from a wide variety of scientific aggregators/influentials.  Even so, even in extremely well functioning polities, there is a need for a class of individuals that study the principles of such policy coordination and are not government officials so they can contribute alongside the aggregators and applied ethics folk to public discussion in independent fashion.

Okay, let me close here. Today's post was rather abstract and short on normative and scientific detail, and I have assumed that readers can use their imagination in connecting the dots to the unfolding pandemic. For example, yesterday, I privileged a version of the most affected principle combined with a commitment to protecting the most vulnerable. It's not obvious that should always inform policy apt synthetic philosophy. I also have not given any examples of promising policy apt synthetic philosophy (in the past I have praised Rachel Carson). But more important there are also risks to any intellectual activity  that becomes oriented toward policy. To be continued.+

 

The problem of Arithmetic and True Representation; On Checks and Balances

The founders set themselves apart forever from the naive theorists of democracy. They saw, in Burke's phrase, that the constitution of a state is not a "problem of arithmetic." So they refused to identify the will of the people with the transient plurality of the voters in one constituency. They did not say, for example, that if the whole mass of persons votes once, and if one party has 34 per cent of the votes and the other two have 33 per cent each, the winner in the contest is the true representative of the people. They thought of "the people" as having many dimensions in space, in time, in weight, in quality. They thought, as Burke did, that a society is "a partnership in all science  a partnership in all arts a partnership in every virtue, and in all perfection," and "as the ends of such a partnership cannot be obtained in many generations," a civil society is "a partnership not only between those who are living, but between those who are living, those who are dead, and those who are to be born." The American founders sought to represent this many-sided people and they thought of the people's will as an equilibrium of its many elements. And so in their practical arrangement they sought to make the government as nearly representative as possible of the many facets of the popular will, of the people acting as citizens of local communities, acting as citizens of regions, of states, of the nation, acting with remembrance of the past, acting as they felt at the moment, acting as they would feel after fuller consideration. For they gave no credence to the idea that one periodic count of heads could elicit the real will of a large population. The founders sought to approximate a true representation of the people by providing many different ways of counting heads. For the national government, itself a federation of states with complex forms of representation, they provided a House elected for two years from fairly small constituencies of equal size; a Senate in which one third only was elected every two years from the states that is, from constituencies of varying size; a President, chosen, as they conceived it, by electors from the separate states, and for a term of four years, which did not correspond with that of any one group of the legislators; a judiciary appointed for life after confirmation by the Senate. Thus no two branches of the government were chosen by the same constituency or for the same term of office. They provided that for ordinary laws a majority of both Houses and the President must concur, that a two-thirds majority may prevail over the President, that for treaties two thirds of the Senate must concur with the President. They then provided that all the powers exercised by the legislative and the executive branch were subject to the supreme law of the land, and that their specific acts would be invalid if contrary to the Constitution. They provided that the supreme law of the Constitution could be amended only by a complex vote which would ensure as nearly as possible that the decision had been fully considered, that all men had had a chance to hear the issues debated, and that many more than a mere majority had been convinced. Walter Lippmann (1938) The Good Society, 253-5

Last week, I attended a lecture by the eminent political theorist, Elizabeth Cohen, on her book The Political Value of Time. It made me alert to the passage from Lippmann whom (recall) I was reflecting on. (So today's Digression is a hommage to her work.) We ordinarily think, I think, of checks and balances in terms of separation of powers with the aim to prevent one power from dominating the others. Lippmann is not unaware of that argument and I'll presuppose familiarity with that strain of thinking here. But it's notable that this is not Lippmann's focus in the passage quoted. 

Rather, he introduces the topic, while mentioning Madison, in terms of what he calls "the refinement of the popular will" on which "a progressive society must depend." (260; see also 250). This is an allusion to Madison's claim in Federalist 10, in which Madison explains the difference between a republic and a democracy.* The way Lippmann (re-)interprets Madison the refinement of the popular will is not, primarily, an epistemic discovery procedure (a la Condorcet), but rather a means toward developing a proper (democratic) character. (If you are a Marxist feel free to add that he is defending the interests of the landed, propertied classes.) So that the intentions of the masses/mob are turned into a proper popular will of citizens. To be sure, there is also an epistemic element ("true representation"), but that's not the main emphasis.**

Lippmann's key insight is that this requires not just the slowing down of the process -- to make time for deliberation, and second thoughts, but also, and more important, a means to allow the many dimensions of our lives and interests to be voiced differently (in different districts and in different kinds of representation) and at different times.

I don't think this is Lippmann's most important insight in the book, or that the US constitution is especially good at (generating an "equilibrium" out of) this decomposition designed to generate as many facets of citizens's personality as possible. (Lippmann, too, emphasizes the constitution's imperfections.) But even so, i think there is something important in the idea that a true democracy is one in which the complex, even in some sense paradoxical features of social reality are manifested. 

The previous paragraph may sound highfalutin, but I mean to note that Lippmann helps us convey the nature of true democracy in which the very same citizen can vote for, say, Obama, Trump, and Klobuchar (etc.). And that in so doing, democratic life cannot be reduced to the expression of counting heads at a particular moment, but becomes something extended indefinitely in the future. To be part of democratic process is not to arrive at a final decision, but to make future perspectives on decisions possible.+

I did not put the last point in especially charming matter, and also I came close to turning it into a philosophical exercise. But rather, the underlying (somewhat Protestant) sensibility allows that sometimes the partnership is criminal; but the partnership is meant to exhibit delayed gratification, the welcoming of divergent views and reversals of opinion, and of manifesting (to echo the Platonists on democracy) the wonderful and often chaotic appearing tapestry of peoples and experiences that are characteristic of democracy at its best.