Digressions&Impressions

On Nagel vs Neurath on the Icy Slopes

While Frank and his institute began to lose favor at the Rockefeller Foundation, increasing numbers of his colleagues in philosophy – besides Kallen and Hook, who had led the way in the 1940s – began to voice public sympathy with some of the dominant anticommunist themes and postures of Cold War intellectual life. With regard to the questions of whether and how science (or scientific philosophy) should treat matters regarding values – political, social, or ethical – two general stances were popular. The first called for disengagement and held that philosophy of science should abstain from such debates. The second held that philosophy of science should at least support the view that absolute moral and social values exist, if not also the partisan view that those of the West are plainly superior to those of the Soviets. In the early 1950s, Ernest Nagel and Martin Gardner touched on these issues and illustrated these stances.
In an obituary for Felix Kaufman, Ernest Nagel referred to “the acute issue whether questions concerning public and individual moral values are capable of being settled scientifically and objectively.” Nagel approved of Kaufman’s position, one that gingerly sided with Frank’s: If our “rules of application” of moral terms are made explicit, Nagel explained, then “moral questions are as much capable of objective decision as are questions in physics and biology” (Nagel 1950, 467). When it came to politics and the problem of international peace and stability, however, Nagel was skeptical that science or philosophy had any relevance. Reviewing a book by Lewis Mumford, in which he insisted that we must “learn to participate in a durable universal communion with the whole of humanity,” Nagel wrote in the anti-Stalinist New Leader,

It is extremely doubtful whether such a reorientation by every human being constitutes a necessary condition for the disappearance of the cold and hot wars that [Mumford] so fervently desires. There are surely respectable grounds for believing that this objective could be realized within the framework of current political organizations of men and on the basis of current moral ideals.

Nagel did not accept Mumford’s utopianism because history did not support it:

As for Mr. Mumford’s suggestions concerning the establishment of intimate neighborhood organizations, of public-work groups, or of international travel fellowships, there is little evidence to show that such devices have contributed much in the past to the creation of a world community, or that without fundamental political and economic reorganization on both a national and international level they are likely to do so in the future. (Nagel 1951, 22)

Nagel agreed with Kaufman and others (including Frank, Morris, and Dewey) that science and philosophy of science could illuminate values and related questions about human conduct. But with respect to the Cold War, he was resigned to quietism and disengagement. Only political and economic restructuring, and not resources offered by philosophy, could advance a “world community” or reduce international tensions.
Martin Gardner followed Hook and Kallen more closely  than Nagel. When he reviewed books in the New Leader, he urged philosophers of science take up the cause of anticommunism and not retreat to any apolitical posture. George A. Reisch (2005) How the Cold War Transformed Philosophy of Science: To the Icy Slopes of Logic (CUP) 310-311

I was a PhD student when I first encountered Reisch's brilliant, archive based demolition of conventional wisdom and self-serving myths in the profession often based on anecdotal evidence or hearsay. His book (hereafter Icy Slopes) is one of the towering achievements of various intellectual communities (HPS, HOPOS, History of Analytic, 'the fans of Carnap' etc.) that I have self-identified with over a good chunk of my professional life. Unlike much history of analytic philosophy it unsettles comforting narratives. And like many much nobler souls, I applauded and continue to admire his intellectual rehabilitation of those that gave analytic philosophy its original moral and political purpose. 

So, with an intro like that, you know I am going to disagree with Reisch.

Icy Slopes has a rise and fall structure. The passage I quoted is part of one of the key turning points in the book where the unity of science program (initiated by Neurath and left Vienna, and then still promoted by Frank) collapses. And while Nagel is not portrayed as himself a villain, he is treated as an exemplar ("illustrated these stance") of a certain retreat from more ambitious and nobler aspirations. I think Reisch misunderstands Nagel's position, and the misunderstanding is illustrative of a wider problem of his conceptual framework. But at the end of this post I am also going to suggest that in a certain sense Reisch's misunderstanding is not surprising given the way the profession evolved.

So, Reisch is absolutely right that Nagel is an anti-communist (recall; and here on his reviews of Cornforth); see also his withering 1939 review in The Nation of Haldane's The Marxist Philosophy and the Sciences which Nagel republished in Logic Without Metaphysics (1956; hereafter LWM). But it is worth noting, first, that Nagel was not just an anti-Communist. His polemical targets in this period, also include folk who may now classify as sympathetic to the right: these include not just an assortment of neo-Thomists and the obscurantism popular in Nazi Germany, but also harshly negative reviews of the realist theorist Hans Morgenthau's Scientific Man vs Power Politics (in The Yale Law Journal 1952) and (recall) the rhetorically devastating review in (1952) The Journal of Philosophy  of Hayek's The Counter-Revolution of Science both re-published in LWM.

So, even in his anti-communism Nagel did not give fellow critics of Marxism a pass. In that sense Nagel does not fit the stereotype of a cold war intellectual. This gets me to second point. Reisch is, thus, a bit misleading to imply that anticommunist themes and postures shaped Nagel's thought or that he conformed to them. His philosophical politics, if we may call them that, were not centered on anti-marxism as such. Rather his anti-marxism is an expression of his underlying views about philosophy and political life.

As I have discussed (here and here), Nagel's political stance in the period echoes the liberal ideal of rule by discussion. In Nagel's case his political position was ground in and mirrors his metaphysics and, in particular his philosophy of science. While science contains differential expertise and skill, Nagel views it as an open-ended conversation responsive to public criticism and reasons and, thereby, self-corrective. Crucially for Nagel over time the norms and criteria change, even improve. Not unlike say, Polanyi, he treats this idealized conception as a useful model for a political life worth having. This is a progressive conception of the possibilities of political life that, in practice, recognizes many improvements are the consequence of local, accumulating incremental change.+

Nagel indicates something like the image I have just sketched in many places, including, I should note in the rather moving (!) memorial to Felix Kaufman quoted by Reisch! Here is a passage I have in mind:

[Kaufman] was impressed by the fact that empirical science is a potentially ceaseless process of evaluating propositions in the light of changing evidence for them. And very early in his own intellectual development he recognized as an essential feature of scientific method the tacit use of the principle he came to call the principle of permanent control, according to which no proposition dealing with matters of fact is beyond criticism and revision. He employed this principle, which admirably expresses the self-corrective nature of modern science, in his trenchant criticism of atomistic sensationalism, of philosophies of self-evidence, and of dogmatic claims to finality. The principle was indeed an expression of an essential trait of the man himself: of his native abhorrence of authoritarianism, of his readiness to receive as well as to give criticism, and of his disciplined humility in recognizing the difficulties that stand in the way of obtaining warranted knowledge...(466)

And while the point of Nagel's description is to characterize Kaufman's position, and he sometimes disagrees with Kaufman, this part, as Reisch recognizes, he clearly endorses ("admirably expresses...") I know of no reason to assume Nagel ever gave up on this. 

Now it is worth noting that Nagel's view of science was less naive then it may seem now. At the time science was much smaller and the government's role in shaping science underdeveloped. In addition, while Nagel emphasizes the self-corrective nature of science (and may seem to buy into a kind of efficient market in ideas hypothesis), he is not unaware, and calls attention to the fact, that correcting for social and political bias, even in the natural sciences, is often not easy in practice (as he notes in well known material excerpted from The Structure of Science (and circulating as "the value-oriented bias of social inquiry.")* 

Okay, with that out of the way, let's look at Reisch's analysis. It is notable the for Reisch the two (anti-communist) alternatives to the unity of science are (i) quietism or disengagement; (ii) extolling of objective superiority of the West and its values. This leaves a lot of alternatives unexplored. 

For, the point of establishing the possibility of, and reflecting on the form of the of, objective values is not, in Nagel's hands, intended as an uncritical vindication of the status quo. They also provide the possibility of standard of criticism from which one can evaluate (and now I use one of Nagel's favorite words) the "warrant" of one's beliefs. That is to say, the question of, say, demarcation between science and not science is supposed to be able to be mirrored in social and political sphere. And, say, unlike conservative anti-communists, Nagel (and again he praises Kaufman this) firmly believes that tacit rules and commitments of our social practices, including science and induction, must be made explicit and articulated in order to allow for rational evaluation, justification, and correction/improvement.

Now if we turn to Reisch's treatment of Nagel's review of Mumford, something odd happens. A casual reading of Reisch may leave one with the impression that Mumford is a spokesperson for the kind of utopian thought we might associate with the unity of science movement. But Mumford's approach is, and now I quote Nagel himself, more consistent with being a "spokesman for one phase of intellectual and moral liberalism." (LWM 409) Mumford (who is fascinating) is as non-marxist as Nagel is (as Nagel recognizes LWM 410). And Nagel leaves no doubt he is quite familiar with Mumford's wider views.

And if we look at the passage that Reisch quotes in light of that, it is clear that they are criticized not for being politically wrong-headed. But rather that they do not help solve the (recall) transition problem from here to there and so are, at bottom, a "counsel of dark despair," (LWM 411). Such transition problems do not merely beset marxists, but as Nagel clearly recognizes, liberal thinkers too (something recently emphasized (recall) by Scott Scheall). If one's solution to the transition problem requires "wholesale and simultaneous victory over all problems" (LWM 411) then by Nagel's lights (and i am inclined to agree) that's no solution at all. (One can agree to that without always caving to demands of feasibility!) That is to say, a political philosophy that cares about changing the world is bound to confront the transition problem and take it seriously. This fits with Nagel's larger views on (recall) the "demands of responsible discourse."

For, more subtly, Nagel recognizes, thus, that a certain kind of revolutionary stance is not political at all, but a species of "moralizing exhortations." And, in fact, Nagel is explicit that the quetism is Mumford's: "He therefore proposes....the cultivation of the personal withdrawal and solitary reflection." (LWM 412, he then goes on to quote Mumford). So, the very criticism Reisch lodges against Nagel is Nagel's criticism of Mumford!

Now, Reisch and the friends of revolution, inspired by Neurath, need not agree with Nagel's incrementalist position, which embraces a "realizable process of indefinite approximation" (LWM 413, quoting Plato approvingly!) And I recognize that for some Nagel's charge that Mumford does not offer a "serious contribution to the solution of our present social dilemmas" may be unfair or unduly 'realistic' in a conservative way. But Nagel's willingness to work "within the framework of current political organizations...and on the basis of current moral ideals" (414) it is not a withdrawal from the world or a quietest, complacent stance even in international affairs.** (In 1951 the United Nations had been founded, and the Bretton Woods structure was being developed.) It is neither glorification of the American way nor antipolitical.

There is an important issue that ends up obscured by the limitations of Reisch's particular criticism (and may help explain them). One may well wonder, fairly, how Nagel's approach to philosophy of science, and the professionalism that it helped promote, may contribute to the solution of social dilemmas. And from the other end, the philosophical profession Nagel helped shape and bequeath does seem a fair target for Reisch's criticism. We may still wonder how Nagel's "convictions of a people confident that a bold but disciplined intelligence is still a creative power in the world" is actualized in a life of philosophy of science... 

 

Ernest Nagel's Meta-philosophy of science (and politics)

Scientific knowledge does not depend on the possession of  an esoteric capacity 
for grasping  the necessary structure of some superior reality, nor does it require 
modes of warranting beliefs which are discontinuous with operations of thought, 
identifiable and effective in the ordinary affairs of human life. The achievements 
of science are the products of a cooperative social enterprise, which has refined 
and extended skills encountered in the meanest employments of the human 
intellect. The principles of human reason, far from representing the immutable 
traits of all possible being, are socially cultivated standards of competent 
intellectual workmanship. The life of reason as embodied in the community of 
scientific effort is thus a pattern of life that generates an autonomous yet 
controlling ideal. That  ideal requires disciplined dedication without servitude to
 any ultimate authority, imposes responsibility for performance upon individual 
judgment but demands  responsiveness to the criticism of others, and calls for  
adherence to a tradition of workmanship without commitment to any system of 
dogma. To many commentators, the ideals realized in the enterprise of science are
 also the ideals  which areindispensable to the successful operation of any society 
of free men. Many thinkers, indeed, like John Dewey in America, have based their 
hopes for the future of mankind upon the extension of the habits of scientific 
intelligence to every stratum  of communal life and to every  form of social 
organization. --Ernest Nagel (1954), "The Perspectives of Science and the Projects 
of Man," in Sovereign Reason 306

In analytic philosophy fame is transient. For example, Ernest Nagel is now largely forgotten; a few contemporary philosophers of science are familiar with his writings on reductionism and a few others are familiar with his work on function statements. But I doubt he is the subject of many graduate seminars, and I suspect almost none of his papers are taught regularly. That Nagel was once an academic celebrity in the sciences, the leader of what we may call the scientific wing of pragmatism for a generation, influential supervisor, prolific author across many philosophical topics, introducing countless students and their teachers to Gödel’s Proof, and an intellectual polemicist of the first order is now merely a historical footnote.

So, for example, Misak calls "Nagel, perhaps Dewey’s best graduate student" (116) in her influential, The American Pragmatists. She discusses  him as "the paradigm of an analytic, logically inclined, philosopher of science in America," (150). But while she uses some of Nagel's writings throughout her work in discussing other thinkers, and hints that he may be responsible for the synthesis between logical positivism and pragmatism, (163) at bottom she devotes little space to him. 

This neglect is a shame because in addition to being technically among the more competent philosophers of his generation, Nagel was widely read, fiercely sought out opposing views, and always makes clear that philosophical disagreement may matter to the fate of civilization. As regular readers know I think Nagel decisively shaped the character of contemporary analytic philosophy, including its orientation on puzzle solving (even though in some senses we have not been as ambitious as he hoped as we'll show below). But today I assume that in the background. Here I focus on his own sense that in the wake of Einstein the "classical conception of scientific knowledge" (302)* had to be replaced with something new.

For Nagel, science shapes the broader culture in two distinct ways: first, as the fount of technologies and medicines. Second, by challenging established beliefs and intellectual habits (297). And while the former is more emphasized, not the least by scientists in their appeals for resources (296), the latter can also help re-shape (amongst others) our "basic aspirations," our "moral commitments," and "the principles" by which "actions" are evaluated (297).

These facts generate a "threefold task" for philosophy: (i) clarifying the bearing of trends in scientific inquiry upon pervasive conceptions of our place in nature; (ii) of making explicit the intellectual methods by which responsibly held beliefs are achieved; and (iii) of interpreting inherited beliefs and institutions in the light of current additions to knowledge. (297-298) This has an interest implication. For all three involve philosophy of science. And while we tend to think of philosophy of science as a fairly focused enterprise (characterized by (ii)) -- e.g., "the clarification of scientific procedures" (307) and the stress testing of concepts (306) --, for Nagel, philosophy of science becomes first philosophy: "the boundaries of the philosophy of science are in fact the boundaries of philosophy itself." (298) And while (i) speaks to what we may call the existentialist call to philosophy, the third (iii) involves, more dangerously, philosophical activity with the life of her (imagined) community. 

The quoted passage at the top of this post leaves it unclear to what degree Nagel agrees with his mentor, Dewey, that society needs to be modelled on the post-classical conception of science (which is fallabilist, embraces the significance of what is now known as Duhem-Quine, and which is aware of possible incommensurabilities). Nagel rejects the cult of genius and thinks that science is a collaborative enterprise, which involves the systematic and socially sustained refinement of ordinary cognitive processes. So, rather than seeing a scientific society as a means toward a technocratic elite, the idea that society needs to be modeled on science is thought be democratic in a non-trivial sense.

Crucially, on this model science is not a method of silencing others behind an authoritative consensus, but a mechanism by which one becomes responsive to reasons and participates in a social division of labor that is in some sense self-legislating. And so, this helps generate (through its spokesperson in philosophy of science, and in lived experience) an "attitude, at once critical and experimental, toward the perennial as well as the current issues of human life." (307)

And, in fact, Nagel is explicit that philosophy of science so understood, is "a champion of the central values of liberal civilization." And while in the early 1950s the survival of liberal civilization was more secure than it had been in a generation, it is clear that securing such a civilization is one of Nagel's main aims in all his polemics (recall; and here). And, of course, by providing freedom of thought, liberal civilization is conducive to a humane philosophy of science (understood as first philosophy):

The basis for a general outlook on the place of man in nature is supplied by detailed knowledge of the structure of things supplied by the special sciences— an outlook that contemporary philosophy of science has helped to articulate and defend. In the perspective of that outlook, the human creature is not an autonomous empire in the vast entanglement of events and forces constituting the human environment. Nevertheless, no antecedent limits can be set to the power of scientific reason to acquire theoretical mastery over natural and social processes. Every doctrine which pretends to set such limits contains within itself the seeds of intolerance and repression. Moreover, in the perspective of that scientifically grounded outlook, human aspirations are expressions of impulses and needs which, whether these be native or acquired, constitute the ultimate point of reference for every justifiable moral judgment. The adequacy of such aspirations must therefore be evaluated in terms of the structures of human capacities and the order of human preferences. Accordingly, though the forces of nature may one day extinguish the human scene, those forces do not define valid human ideals, and they do not provide the measure of human achievement. But an indispensable condition for the just definition and the realization of those ideals is the employment and extension of the method of intelligence embodied in the scientific enterprise. A judicious confidence in the power of reason to ennoble the human estate may seem shallow to an age in which, despite the dominant position in it of scientific technology, there is a growing and pervasive distrust of the operations of free intelligence. It may indeed be the case that the temper of mind essential to the exercise of such intelligence has no immediate social future. But the cultivation of that intellectual temper is a fundamental condition for every liberal civilization. By making manifest the nature of scientific reason and the grounds for a continued confidence in it, contemporary philosophy of science has been a servant of men’s noblest and most relevant ideals. (307-8)

It would be tempting to give Nagel the last word. But it is notable that he assigns to philosophy of science also (iv) the task of making manifest "the grounds for...continued confidence" in the nature of scientific reason. The future orientation of this task is quite striking. What could merit that?** So, lurking in Nagel's "naturalist" program is the presupposition that either skepticism can be defeated or, well, that we can be persuaded to it ignore its challenge. 

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.

Toward a Political-Moral Epistemology of Computational (process) Reliabilism for AI

Now, one way to reinterpret Goldman’s [process] reliabilism in the context of computer simulations is to say that researchers are justified in believing the results of their simulations because there is a reliable process that produces, most of the time, true beliefs about such results. We can now reinterpret computational reliabilism in the following terms:

(CR) if S’s believing p at t results from m, then S’s belief in p at t is justified. where S is a cognitive agent, p is any truth-valued proposition related to the results of a computer simulation, t is any given time, and m is a reliable computer simulation.

Let us note that the formulation of process reliabilism remains largely unmodified by computational reliabilism, as it is evidenced in (CR). An important – and rather obvious – difference, however, is that process reliabilism is no longer a general account for any p and m, but rather specified for computational undertakings. In this respect, computational reliabilism takes that p is a truth-valued proposition related to the results of a computer simulation. These could be particular, such as ‘the results show that republicans have won,’ ‘the results suggest an increase of temperature in the Arctic as predicted by theory’, and ‘the results are consistent with experimental results,’ among others. Alternatively, they could also be general such as ‘the results are correct of the target system’, ‘the results are valid with respect to the researcher’s corpus of knowledge’, and ‘the results are accurate for their intended use.’ Naturally, the reliable process m is identified with the computer simulation (see Sect. 3.2 for further differences with process reliabilism).
We can now assimilate Goldman’s process reliabilism into our analysis of computational reliabilism: researchers are justified in believing the results of their simulations when there is a reliable process (i.e., the computer simulation) that yields, most of the time, trustworthy results. More formally, the probability that the next set of results of a reliable computer simulation is trustworthy is greater than the probability that the next set of results is trustworthy given that the first set was produced by an unreliable process by mere luck (Durán 2014).-- Durán, Juan M., and Nico Formanek (2018)" Grounds for trust: Essential epistemic opacity and computational reliabilism." Minds and Machines 28.4, 654. [HT Federica Russo]*

In what follows I take Durán & Formanek's computational reliabilism (about simulations) as a baseline. This work draws on some of the best work in recent epistemology and philosophy of science, and offers a compelling framework for how to think of the strictly epistemic grounds for trust for computer simulations. In what follows I  use 'algorithmic mediation' where they use 'simulations.'1

But since the pathbreaking work by Heather Douglas, we also know these epistemic grounds are conceived rather narrowly. So, here I explore how their approach might be enhanced if one takes seriously (to use a jargon from philosophy of science) inductive risk in AI. In particular, I reflect on  the conceptualization of how one should internalize social consequences into the design and programming process of algorithms as they figure in contemporary AI.+ And in particular discussions over machine learning algorithms.

In this post I sketch three areas where this baseline must be enhanced, of which the latter two are specific to machine learning AI, when one takes seriously social consequences for the grounds of trust. The idea is to work toward Ethical Computational (process) Reliabilism (ECR).

First, when one looks at the application of a reliable process outside the lab into social context, one is not just interested in its reliability for a specific task. One would also like to know what the effects are of failure. In particular, one would like to know something about the distribution of possible or likely harms on different kinds of populations, especially if these populations have different kinds of vulnerabilities (and appetites for risk). 

So, for example, something may function reliably as designed with industry beating low failure rates; yet when it breaks, all too rarely, the artifact may still be especially dangerous for kids. Or, some safety gears work swimmingly on average male subjects, less so on average female subjects. Some medicines interact badly with pre-existing conditions in subsets of the population. Now, in many cases the harms that follow from such selective or asymmetric vulnerabilities can be internalized in the design and testing process (and often this is mandated legally or by in-house risk assessment).

Selective or asymmetric vulnerabilities can map onto socially salient issues when they map onto political or morally salient demographics. In the previous paragraph I mentioned children and sex differences. A lot of public discussion of machine learning in AI has (quite naturally) focused on its reinforcement of racial and economic injustice(s). 

How to characterize what counts as as an asymmetric vulnerability is not so easy especially because many of the ethically or politically most salient harms may only become asymmetric due to causally intersectional effects. In addition, some asymmetric harms may be due to the fact that a truthful p reinforces or entrenches a socially bad status quo. For many purposes one may wish to distinguish among such selective vulnerabilities, but here I lump together as an especially important set of unfair outcomes. So, to begin formulating a possible framework:

(ECR) if S’s believing p at t results from m, then S’s belief in p at t is justified. where S is a cognitive agent, p is any truth-valued proposition related to the results of an AI, t is any given time, and m is a reliable algorithmic mediation without generating asymmetric harms to vulnerable populations.

Here 'reliable' already presupposes an ordinary use of the reliable algorithm in an assigned task. One thing that follows from this is that in order to generate an ECR, its sources -- again following,  Durán & Formanek, 1) verification and validation methods; 2) robustness analysis; 3) feedback from trial runs and (prior) implementations; 4) expert knowledge -- must also be made to seek out and track asymmetric vulnerabilities. While this clearly makes initial R&D more expensive, it may reduce litigation costs and social harms (including withdrawal fo the product) downstream. 

Second, algorithmic mediation may generate both unintended and unforeseeable outcomes. Here, too, there are many subtleties.  Some unintended consequences may just be a matter of negligence. And these can be simply assimilated to (ECR).  Morally, legally, and politically one may  be held accountable for those if there are harms in use. As it happens because of concerns over opacity and traceability, algorithmic mediation does generate some special concerns over accountability which are especially salient in the context of informational asymmetries and asymmetric vulnerabilities (with players that range from huge transnational economic agents to dispersed individuals). I return to this before long. But it is morally and politically a very important issue.

Other consequences may be unforeseeable in detail, or their tokens unknown, even though the outcome pattern (or outcome type) may be quite predictable after a while. For example, algorithmic mediation has made financial markets move (i) at much higher speeds and (ii) has also increased the likelihood of mini and maxi flash crashes. The first (i) was entirely predictable (and desired), but the (evolution of) exact speed(s) and volume of market transactions may have been unknowable in advance. And that it would generate new kinds of financial transactions was also known, even if the exact strategies were not. By contrast, it's possible that (ii) was initially unexpected. But by now any given mini-crash may be surprising or unpredictable, but that they occur is foreseeable and so they become a 'new normal.'

That is to say, unforeseeable tokens can occur in foreseeable outcome patterns/types. If an outcome patterns has possible tokens with asymmetric vulnerabilities, these patterns should, all things being equal, be avoided and ought to be internalized in ECR. (Of course, sometimes one can compensate for downside risks, etc.) So, I propose the following modification:

(ECR) if S’s believing p at t results from m, then S’s belief in p at t is justified. where S is a cognitive agent, p is any truth-valued proposition related to the results of an AI, t is any given time, and m is a reliable algorithmic mediation without (intentionally) generating foreseeable asymmetric harm patterns to vulnerable populations.

Obviously, this leaves the prevention, accountability, and remedy of some unforeseeable asymmetric harm patterns outside (ECR). And that is for another occasion.

Third, one instance of an asymmetric vulnerability is the reinforcement of a bad status quo. For many it seems intuitive that p (which is true after all) is ethically neutral, so that p reinforces a bad status quo can elicit shrugs from many. But we should resist the shrug. Classic examples can be found, I learned this via Kristie Dotson, in the way in which crime data is presented and thereby stigmatize vulnerable demographics or promote policies which have side-effects that also harm vulnerable demographics asymmetrically. I return to such harms before long.

But a crucial feature of algorithmic mediation is (as Mittelstadt et al (2016) note) that it can affect how our social reality is conceptualized, and becomes actionable in ways that are utterly unexpected (including a reinforcement of a bad status quo). So, algorithmic mediation can generate  consequences that are not just unintended, but also unforeseeable in principle because they are transformative (Mittelstadt et al also use this terminology going back to Floridi (2014).**

Here, too, the fact that an algorithmic meditation is transformative may be intended and foreseeable. And it is possible that some of the higher order outcome-patterns including the asymmetric vulnerabilities can be predicted. And that is assimilable to (ECR). But some transformations are (ahh) transformative. And these require special treatment. To be continued.

On Adam Tooze, Ideology and Liberal Historiography

If Jamie Dimon quipped that to make sense of Dodd-Frank one needed the services of a lawyer and a psychiatrist, the same is no doubt true of the business of deciphering Donald Trump. But historians can contribute too.¹⁸ Trump offered to a bewildered present a throwback to an earlier era. Born in 1946, the same year as Bill Clinton, Trump would take office at seventy years of age, recycling a rancid version of the baby boomer narrative, which in the 1990s had still seemed fresh. Trump’s racial attitudes reflected the animosities of the era of civil rights, desegregation and New York in the 1970s. His boorish manners and sexism echoed the Manhattan party scene of the 1980s, when bond traders toasted one another as “big swinging dicks.” The sense of national crisis that drove his campaign was a reflux not so much of the recent past as of the first moment when modern Americans felt the world changing around them—the late 1970s and early 1980s. The trauma of defeat in Vietnam, America’s urban crisis and angry Japan bashing—thirty years on Trump was still harping on those fears, but now transposed onto new enemies: China, Islam and undocumented Latino immigrants.--Adam Tooze (2018) Crashed: How a Decade of Financial Crises Changed the World, 569.

One of the non-trivial sub-texts of Tooze's argument is the complacency of the Democrats and even the whole 'political class' in the face of the rise (and eventual) triumph of Trump. I don't want to claim that Tooze always or even often treats his targets as complacent, but it is a recurring theme in the book.* Now, interestingly, on Tooze's presentation, the complacent uniformly believe not just in technocratic skill, but also that by and large the present system is unavoidable (interestingly this flattens any difference between "the determinism of Thatcher and Merkel,"  and that this system is in some world-historical sense progressive ('progress' is repeated frequently in the book).

Tooze is very critical of the political class. Even so, as Perry Anderson notes (83ff.) in his by now (justly) famous review essay, Tooze also admires the complacents's skill which consist in, in part, in managing, so it seems, the irrationality that democracies naturally generate without undoing the liberal status quo (free trade, markets, glocal integragration, etc.) My interest in mentioning this is not Anderson's who is engaged in a polemic from the Marxist left against Tooze's self-proclaimed left-liberalism (p. 21 Crashed; p. 53 in Anderson).+

In fact, I quote the passage at the top of the post  because it is a place where Tooze alerts the reader to his own intellectual perspective (qua historian). And the way I read him, Tooze treats Trump as a kind of expression of atavistic sentiments and outlooks (e.g., 'throwback;' 'recycling;' 'reflux.' etc.) That is to say, just like the complacents, Tooze embraces, despite his sense of contingency, a notion of progress (in his terms an arc of  history) in which some ideas are modern and modernizing and others are old-fashioned or outright primitive. (Because of his progressive Marxism, Anderson can't really object to Tooze's stance here.)

To be sure, Tooze is doing multiple things in the quoted passage: first he explains, briefly, where Trump's ideas come from given Trump's social milieu and personal trajectory.  Second, he is conveying the sense of bewilderment of an establishment of having to confront these ideas and the manner in which they are performed anew. Third, he himself is judging these ideas/performance ("boorish," etc.) I don't think any of these three features are objectionable, and Tooze is transparent about them.

But, fourth, because Tooze treats these ideas as atavistic, he can't quite explain their nature (or their appeal) qua historian (or engaged citizen). So, while throughout the book, Tooze understands the grounds of popular dissatisfaction would with the status quo (the complacents), we're now left with a mystery why the progressive left could rarely capitalize on this dissatisfaction. For, and again Anderson is good on this, why did not ordinary citizens embrace the progressive Keynesian solutions Tooze advocates at the ballot box?

And, in my view, part of the problem for Tooze -- and this emblematic for much liberal thought in the broad sense -- is that he views the fears and desires Trump represents so well as atavistic to be overcome rather than as permanent possibilities of the human condition.** Now the moment one says this one draws the suspicion one is willing to slide into a normative defense of Trump and author authoritarian tendencies. (I have learned that to cast even minimum doubt on the reality of say, moral progress of humanity, is to invite opprobrium.)

But  the fact that those whose business is not to be bewildered were, as Tooze correctly notes, "bewildered" by Trump's rise (and Brexit, etc.) shows that there is a defect in the self-conception of contemporary liberalism. That's different from suggesting that there is a defect in contemporary liberalism that causes the rise of Trump and other authoritarian tendencies (that may be true, too). Since I care about liberalism's survival I think this is a defect in our self-conception worth addressing. The moment one respects what is taken as atavistic as a permanent possibility, one might govern and campaign (and communicate, and theorize, etc.) in a different fashion. For, even recognizing a possibility as a possibility allows one to better prevent it from happening and not treat it as mystery. It is also a start of a better explanation of why it did happen.

That is to say, I think one of the problems in contemporary liberalism is the unstated assumption that certain forms of progress are permanent and that once achieved no back-sliding is genuinely possible. But an ideology that takes its own victories for granted (in being justified and secure), after the fact, really is an ideology.*** What I admire in liberalism is a willingness to learn from experience and self-correct; it strikes me this is a feature within liberalism we need to change, or, better yet, revert back to the wisdom of Judith Shklar and Karl Popper (here; here; here) and be willing to expect the worst.

 

On Kitcher (vs Schliesser) on Synthetic Philosophy

Dewey’s vision is one of “synthetic philosophy.” The philosopher’s reading is not limited to the writings of contemporary specialists in their home sub-branch of the academic profession, but should draw widely from many areas—as the greatest figures in the history of the subject, from Plato to Rawls, have all done. The first task is to understand the questions posed for us by the circumstances of our lives and by the conditions of contemporary inquiry, and that task is unlikely to be undertaken by those whose attention is confined to what a handful of specialists have said about some technical issue. Digging deeply into a particular place can have large and important rewards—but before you choose where to dig (and, perhaps, after you have started but before the treasure has been found), you would do well to reflect on why that specific location is so promising. The fact that generations of predecessors have been excavating in the neighborhood is not a conclusive reason—indeed, you might well reflect on the record of successes they have, or have not, achieved.--Philip Kitcher (2012), Preludes to Pragmatism, ("Introduction"), p xi

By synthetic philosophy I mean a style of philosophy that brings together insights, knowledge, and arguments from the special sciences with the aim to offer a coherent account of complex systems and connect these to a wider culture, policy, or other philosophical projects (or both). In particular, the synthetic glue tends to be a general theory or theories. When I first started blogging about synthetic philosophy, Max Hayward noted that Philip Kitcher had also used  the term in a lovely, polemical interview (reproduced here). Since I made it clear I traced my use back to Spencer, I did not think more of it. Synthetic philosophy is a potentially powerful meme, and it is no surprise it would infect both Kitcher and me. After my Biology & Philosophy (hereafter B&P) piece went viral, more people mentioned Kitcher's use in somewhat accusatory tones. And I realized he had been using the phrase for a while. So, recently (in a forthcoming Dutch piece), I decided to think through our contrasting uses.+ 

Before I get to the differences, I want to emphasize that one can be a synthetic philosopher in his (Deweyan) sense and my own at the same time. So, there is a clear family resemblance between the two notions, and sometimes -- for the Darwinian synthetic philosophers this is not trivial -- one can trace the same lineage of thinkers and problems back. In particular, both approaches feed on the impulse that within philosophy we should be reading widely and aspire "to know a lot of different things and to forge useful synthetic perspectives."

First, for Kitcher, "synthetic philosophy" ought to "replace" analytic philosophy.  I don't think that necessary at all; they can be reciprocal and mutually supportive enterprises. As I noted in B&P, synthetic philosophy can provide analytic philosophy with the data/input that enter into its arguments. Analytic philosophy can provide a home to and train would-be synthetic philosophers and, more important, be a place where would-be-synthesizing theories are tried out and developed. The previous two sentences do justice to the recent history.

I do not mean to deny that the ambitions of synthetic philosophy are a response to the cognitive division of labor and specialization within the sciences that contrast to those characteristic of analytic philosophy response. To simplify: analytic philosophy has both tried to become scientific and a specialist discipline, and it has tried (in Neurath's orchestration and Quine's regimentation) to propose being steward of a specialist language that might unify the sciences (and metaphysics). Synthetic philosophers aspire to play a more a more activist role in the sciences (and beyond).

Second, Kitcher strongly identifies synthetic philosophy with "naturalism." (That's in the next paragraph after the one I quoted above.) But as I clarified (or conceded) in response to a fine post by Martin Lenz, I don't think that necessary. All that is required is that there is a body of knowledge that can be taught, shared, and criticized (in neighboring fields) by others. This is, in fact, compatible with anti-naturalism, and the kinds of projects one can find in continental philosophy, what the Germans call geisteswissenschaft, and theology. None of this is far removed from what Kitcher writes about embracing the 'methodological canons of science of the day'; but he is adamant it rules out all kinds of beasts (e.g., Aristotelian essences, a priori truths, etc.) to be found in ''the philosophical zoo." And my kind of synthetic philosopher probably shares a lot of Kitcher's suspicions, but nows that its tasks are too urgent and difficult to rule out, a fortiori, useful tools and approaches.

Third, and most important, I think the synthetic 'glue' in synthetic philosophy is provided by one or more generalist theories (e.g., Darwinism, Bayesianism, information theory, game theory, actor network theory, etc.) Such theories provide a form of unification and explanation, and frame the kind of puzzles and research questions that can be pursued within the special sciences, a new synthetic special science, and/or philosophy (and public policy). They also provided a species of discipline (for reasons Timothy Williamson has emphasized in a different context). Kitcher quite clearly rejects the very idea (he calls it a "myth") of philosophical "expertise" (219, and elsewhere).* 

Why do I emphasize this? Kitcher's version of synthetic philosophy risks trapping philosophy in the great man of genius syndrome, where the synthetic glue is an act of creative synthesis; it is no coincidence that Kitcher quotes Dewey's gloss on "great innovators" (xi-xv; 172) confronting the problems of their own times approvingly. This is distinct from mastering skills and the (sometimes fine) drudgery of working through and becoming comfortable of a theory within the give and take of academic discussion (as Liam Kofi Bright would say, #noheroes). I am the first to admit that understood like this I am more likely to be (somewhat self-aggrandizing and delusional) synthetic philosopher in Kitcher's sense  and not my own (and that Kitcher clearly could be quite excellent as a synthetic philosopher in my sense)! The synthetic philosopher as I understand her responds to the division of labor among the sciences, by standing within and among them.

To be sure, I do not mean to deny the possibility of intellectual freedom in the synthesizing process. Nor do I mean to reject the hybrid and sometimes eclectic nature of the cosmopolitan, synthetic philosopher (who rejects nativism in all forms). Like Kitcher, I value the possibilities of roaming free exploring desert landscapes and the rich aquatic life below.

A reader sympathetic to Kitcher, and impatient with mine, may note that Kitcher's reasons for rejecting philosophical expertise are ethical. Kitcher's motives are egalitarian, and he wants to block a cognitive hierarchy in which philosophers are superior in the joint ethical practice we must engage in. Regular readers, familiar with my views on methodological analytic egalitarianism, would expect me to agree. But for Kitcher ethical practice (on the value axis) is orthogonal to synthetic philosophy which operates on a "knowledge axis" (xxv; 214ff). By contrast, fourth, I do not see the axes as opposed, and more important, synthetic philosophy can operate at sites where knowledge and values intersect in complex fashion (and draw on expertise in ethics and metaethics and other value-apt sciences). How to do so responsibly is no simple matter and Kitcher himself has important things to say on the topic.

Fifth, there is a more subtle distinction (connected to the third) between Kitcher's approach and my own. The way I understand synthetic philosophy, it is really about the construction of a scientific image that may influence the development of the special sciences, philosophy, public policy, or the manifest image, and in turn be influenced by these. Integrating the various scientific and manifest images into a single picture, let's call it integrative world-making, probably goes beyond synthetic philosophy. I don't deny that some synthetic philosophers -- think of Dennett's use of the universal acid of Darwinism -- may aspire to such a project. In Kitcher's hands integrative world-making (systematic philosophy) is constitutive of synthetic philosophy: "Setting aside any further ventures in philosophical midwifery, societies and individuals continue to need an integrated picture of nature that combines the contributions of different areas of inquiry, and different fields of investigation can be assisted by thinkers whose more synthetic perspective can alert them to missed opportunities and provide them with needed clarification." (215)

Since I like the sentiment in these last words, I think in practice, this may be merely a matter of emphasis. But the way I understand a synthetic philosopher, she may, in principle, more comfortable with a dappled, disunified, and pluralist world  than Kitcher's integrative world-maker.  While I personally feel the pull of a Jamesian reconciliation between the scientific and manifest image, I do not see why this should be the mission of synthetic philosophy. Kitcher (and Kitcher's Dewey) also deny that this should be the main project. But they do continue to embrace integrative world-making.

Finally, for Kitcher synthetic philosophy belongs to pragmatist philosophy. And pragmatism in its beautiful, broad variety does involve non-trivial philosophical commitments about the purpose(s) of philosophy and the nature of  meaning and truth. I don't see why synthetic philosophy must involve commitment to pragmatism understood in a doctrinal sense even if I grant that synthetic philosophy of all stripes can thrive in such a variety. 

Quine, Spencer, Neurath, Regimentation & Imperialism

Each elimination of obscure constructions or notions that we manage to achieve, by paraphrase into more lucid elements, is a clarification of the conceptual scheme of science. The same motives that impel scientists to seek ever simpler and clearer theories adequate to the subject matter of their special sciences are motives for simplification and clarification of the broader framework shared by all the sciences. Here the objective is called philosophical, because of the breadth of the framework concerned; but the motivation is the same. The quest of a simplest, clearest overall pattern of canonical notation is not to be distinguished from a quest of ultimate categories, a limning of the most general traits of reality.--Quine Word and Object, 146-147

Quine's philosophy of science is now primarily understood in terms of his naturalism which makes the distinction between science and philosophy, or metaphysics, blurry. But as is, I assume, well known, he also embraced a distinct task for the philosopher: to paraphrase and thereby regiment scientific language into a simpler and clarified language. This task turns philosophy into a mopping up operation behind the research frontiers of science, but better a job than no job. And while Quine keeps his distance from Neurath's preference for orchestration of the unified language of science, it is notable that Quine thinks there is a possible regimented language of the broader framework shared by all the sciences. This part of Quine   ("ultimate categories, a limning of the most general traits of reality") is beloved by post-Lewisian metaphysicians with their fondness for fundamental languages at the end of scientific development.

Quine devotes the whole of chapter 5 of Word and Object to regimentation, so I do not mean to suggest my previous paragraph is suitable paraphrase for the idea. In particular, there are interesting questions to be asked about the nature of the regimented (logical) language, the relationship between the regimented languages of the special sciences and the broader framework language, and what the further point of a regimented language is. Later in the book, Quine casually notes that regimentation in the way he understands it has already a glorious history: "how powerful this combination is has been borne out by extensive logical regimentations of parts of science, especially mathematics, at the hands of Frege, Peano, and their successors."* (208) Quine's use of 'successors' is a bit cheeky, because back in 1936, in a little known essay, "A theory of classes presupposing no canons of type," Quine himself had deployed a "principle of regimentation" he had contributed to this glorious history.

My interest today is  in the word, ''regimentation,' used to describe this philosophical task. In 1957, Quine had used 'regimentation' in the sense intended here in "The scope and language of science," (BJPS, 8:29: 10). And as Peter Hylton and Gary Kemp note, he gives a nice sense of the point of regimentation there: "To some degree…the scientist can enhance objectivity and diminish the interference of language, by his very choice of language. And we concerned to distill the essence of scientific discourse, can profitably purify the language of science beyond what might reasonably be urged upon the practicing scientist." (7) In so far it is profitable, I leave aside here to what degree the return on investment is worth the opportunity costs.

But Quine does not comment on his choice for using this word. An unscientific search in scholar.google, suggested that the idea of regimenting language was not altogether uncommon in the half century before Quine (as the reference to his 1936 paper suggests). To the best of knowledge, however, Quine invented the coinage of regimentation in the sense under discussion.

I was reminded to reflect on this by a late essay by Herbert Spencer, "Regimentation," which appeared in his collection Facts and Comments (1902).+ Spencer himself knew that the "volume herewith issued I can say with certainty will be my last," and so it can be understood as a kind of literary last testament. As Davies notes, the tenor of the collection is fiercely anti-imperialist. (For an excellent treatment of Spencer's views, which dispels many common myths about him.) And the main argument of the essays I have read so far involves the ways in which imperialism corrupts and infects all of social life, including language, which takes on militaristic tropes and metaphors.** For, Spencer "regimentation is another aspect of that general retrogression shown in growing imperialism and accompanying re-barbarization. "

And like a contemporary cultural critic, Spencer notes that the widespread use of 'regimentation,' and not just it,  indicates the "graduated subordination which we see in an army, characterizes a militant society at large more and more as militancy increases." One of the subtexts of the essay is the co-constitution of imperialism, militancy, and social planning. And echoing Mill, Spencer deplores the way homogeneous society is the product of nation-building at home: "we see that even in a single generation great strides have been taken towards a regimental organization for moulding children after an approved pattern." In particular, regimentation involves the subordination of spiritual ends toward society's directed use of our bodies.++ One of Spencer most striking claims is that such regimentation is directed and developed by city governments in their provision of public goods and services: "each of these is, as we see, like a military administration is having ranks subordinate one to another."

As an aside, the broader thrust of Spencer's essays serve as a record of the retreat of liberalism even in Victorian England at the end of the nineteenth century. This was echoed by a later generation of liberals confronted by its near collapse in the 1930s. I think Spencer's essays can also help explain why liberalisms' ideas lost their vitality in this period. But that's for another occasion.

Let me tie the two threads together. First, Quine's use of 'regimentation' is associated with concepts like purification, clarification, and distillation. These, we may say, chemical metaphors of refinement tend to avoid the militant network of associations that Spencer had diagnosed. This alone is peculiar. 

Here's a speculative thought. There had been a brief, polemical debate about the tendency of the tropes and metaphors used within (let's call it) scientific philosophy. Neurath, when (in 1946) explaining his fondness for, and responding to Kallen's criticism of, his "pet word," 'orchestration,' had gone out of the way to distance it from various totalitarian and programmatic connotations. Instead, for Neurath, orchestration should be understood as a "democracy of cooperation." Kallen had charged the unity of science movement with "imperialistic" designs. The whole exchange is worth reading.*** My present point is that there was self-conscious reflection on the way the network of words used for specialized terminology within scientific philosophy may be associated with wider political trends. 

I have already reflected with the help of Greg Frost-Arnold on Quine's understanding of clarity. So, let me leave that aside here. Regimentation is a form of disciplining (a word notably absent from Quine's idiolect). And, it is worth asking who is, thereby, disciplined. It turns out both science and philosophy. The main benefit of regimentation is more disciplined metaphysics which can take place, as it were, in the regimented language. So, regimentation is a kind of self-disciplining of philosophical activity.

But Quine-ian regimentation is also, a species of disciplining of science; not, I hasten to add, to pursue "practical language reform," (as a Marxist might) or to tell scientists how to do science. Quine thinks, however, that no language is fully "neutral" and that language shapes scientific activity. And a purified language may well improve actual science, indirectly, by avoiding ambiguity and other sources of interferences within science. One important effect of this is that scientists do not have (ahh) the last word on their own output. (Quine here echoes Toland, Berkeley, Hume, in response to Newton.) Their authority is, thus, partially subordinated to philosophy.+++ And, while I would hesitate to call this an 'imperial philosophy,' in this sense, 'regimentation,' seems an apt choice.

Models, Politics, Populations, and Public Science in times of Pandemic (V); On Marc Lipsitch and Synthetic Science

I share Medawar’s pragmatic vision of scientific reasoning. Scientists must resist the temptation to excessive skepticism: the kind that says no evidence is ever quite good enough. Instead they should keep their eyes open for any kind of information that can help them solve problems. Deciding, on principle, to reject some kinds of information outright, or to consider only particular kinds of studies, is counterproductive. Instead of succumbing to what Medawar calls “habitual disbelief,” the scientist should pursue all possible inputs that can sharpen one’s understanding, test one’s preconceptions, suggest novel hypotheses, and identify previously unrecognized inconsistencies and limitations in one’s view of a problem....If the COVID-19 crisis has revealed two “competing” ways of thinking in distinct scientific traditions, it is not between two philosophies of science or two philosophies of evidence so much as between two philosophies of action....

This is not to deny that there are different and valuable perspectives on epidemiology. Like any other field, there are many specialties and subspecialties. They have different methods for how they study the world, how they analyze data, and how they filter new information. No one person can keep up with the flood of scientific information in even one field, and specialization is necessary for progress: different scientists need to use different approaches given their skills, interests, and resources. But specialization should not lead to sects—in this case, a group of scientists who accept only certain kinds of evidence and too rigidly adhere to a philosophy of non-interventionism....

Infectious disease epidemiologists must embrace diverse forms of evidence by the very nature of their subject. We study a wide range of questions: how and under what conditions infectious diseases are transmitted, how pathogens change genetically as they spread among populations and across regions, how those changes affect our health, and how our immune systems protect us and, sometimes, make us vulnerable to severe illness when immune responses get out of control. We also seek to understand what kinds of control measures are most effective in limiting transmission. To understand these issues for even one type of disease—say, coronavirus diseases—requires drawing on a wide range of methodologies and disciplines....

The upshot is that, done well, epidemiology synthesizes many branches of science using many methods, approaches, and forms of evidence. No one can be expert in all of these specialties, and few can even be conversant in all of them. But a scientist should be open to learning about all of these kinds of evidence and more.--Marc Lipsitch (May 12, 2020) "Good Science is Good Science" The Boston Review. [HT Matt Lord]

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

Sometimes I notice the English heritage plaque commemorating Sir Peter Medawar on Downshire Hill, and I am reminded of the joys of reading the belles lettres of urbane scientists who, through their attention to detail, and highly specialized knowledge, open gates to inviting, untrodden paths meandering in nature's library and our location in it. The essays collected in Medawar's The Uniqueness of the Individual are among the greatest in the genre.  

Unlike Kuhn's image of scientists as unreflective puzzle solvers, Medawar developed a strong interest in philosophy, even a philosophy of science that emphasized the creativity of working science.* Because his philosophy of science seems to have been developed under the influence of a philosopher (Popper) I find this work less stimulating.** Although I take a professional interest in the way in which philosophical ideas play a role in (what I call) the image(s) of science -- (recall these posts on Hume, Williamson, Spinoza), within science and public pronouncements of scientists (and aggregators or what Polanyi called 'influentials'). An image of science is a list of characteristics that function as short-hand for representing science when these characteristics are used in debates where one side (or more) appeals to the (epistemic) authority of science to settle debate.

The passages quoted from Lipsitch, an epidemiologist, are a response to an essay by Jonathan Fuller itself a rare philosophical intervention in unfolding scientific and public policy controversy (which I discussed in the fourth post in this series). Lipsitch gives us a glimpse, from under the hood, of a species of (what in honor of Jacob Stegenga I call) policy apt, fast science. This species involves the synthesis of many different kinds of sciences and different kinds of evidence. And Lipsitch self-consciously presents his discipline as synthetic in this fashion.

As it happens, despite the increasing significance of ecology and climate science, there is not more sustained reflection, philosophical or popular, on synthetic disciplines in science.+ In addition to the integration of specialist knowledge from the special sciences it requires non-trivial judgments about the ways these hang together. When these connecting links are subject to experiment and statistical techniques, the integration may itself be a scientific discipline; when these links involve leaps of the imagination or conceptual creativity the integration may be more likened to philosophy (what I call synthetic philosophy). Of course, the dividing line between science and philosophy may be blurry or a matter of sociology of the academy.

When there is time to integrate, one aims to develop the techniques that allow one to stress test the links (models, techniques, experiments, concepts, etc.) constitutive of a synthetic science. But as Lipsitch describes, when such time is absent, a lot of the integration occurs almost heroically on the fly.++ This is so even if the underlying (special) sciences are robust and well understood. Lipsitch describes how good judgment across multiple forms of evidence is required. Presumably this also involves comparing features of COVID-19 to better understood diseases and mechanisms.

It follows from Lipsitch's account of the practice of infectious disease epidemiologists that their synthetic science involves lots of judgment calls that can generate plausible disagreements among even the most informed experts. This matters in a policy context because it means that one is not in the realm of constrained choice -- characteristic of applied ethics and decision theory ; in constrained choice areas, and as Ellsberg taught the intermediary contexts (characterized by constaints and uncertainty), experts can have considerable understanding of possible consequences of policy and one may even evaluate them in light of ends one may wish to pursue or avoid. 

But in the context of fast, synthetic science that aims to be policy apt -- "decisions are urgent and must be made with the evidence we’ve got" -- we should expect lots of controversy: legitimate second guessing of both the judgments internal to infectious disease epidemiology as well as the judgments about how to pursue policy (even if there were consensus about ends/values). From that vantage point it is to be welcomed that Lipsitch engages in critical dialogue with learned critics.

But there is a more important moral lurking here. As I noticed last week (recall this post on Ernest Nagel), it is a deeply ingrained way of thinking about the role of science in public policy that it supplies the decision-makers and public with authoritative, univocal answers to decisions. And when the scientific community does not spontaneously supplies such unicity, the impulse has been to create mechanisms of consensus generation (through surveys, panels, boards, special issues, etc.). The impulse is natural both (a) when one fears that diversity of voices will make space for political arbitrage or even to allow slick operators to sideline  knowledge altogether or (b) thinks that univocalness allows for the benefits of social coordination to materialize.  But premature (that is not stress-tested) consensus generation is also fraught with risks because it also means the authority of a particular synthetic science rests on fragile foundations.*** Thanks to Eric Winsberg I have been alert to the (reputational) risks to science as such that accompany this.

So, the moral is this. Lipsitch rightly calls for a philosophy of action in the context of fast, policy apt synthetic science. In my view, informed by Medawar's embrace of human diversity within (imaginative) science tempered by responsibility and obligations, such action will be better constrained once science and society learn to be comfortable with an image of science that permits listening to and speaking with multiplicity of informed voices. 

 

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.+