Digressions&Impressions

On Foucault on 17 January 1979: On Dynamics of Liberty or Method and Enemies (V)

Undertaking the history of regimes of veridiction—and not the history of truth, the history of error, or the history of ideology, etcetera—obviously means abandoning once again that well-known critique of European rationality and its excesses, which has been constantly taken up in various forms since the beginning of the nineteenth century. From romanticism to the Frankfurt School, what has always been called into question and challenged has been rationality with the weight of power supposedly peculiar to it. Now the critique* of knowledge I would propose does not in fact consist in denouncing what is continually—I was going to say monotonously—oppressive under reason, for after all, believe me, insanity (déraison) is just as oppressive. Nor would this political critique of knowledge consist in flushing out the presumption of power in every truth affirmed, for again, believe me, there is just as much abuse of power in the lie or error. The critique I propose consists in determining under what conditions and with what effects a veridiction is exercised, that is to say, once again, a type of formulation falling under particular rules of verification and falsification. Michel Foucault, 17 January 1979, lecture 2  The Birth of Biopolitics. translated by Graham Burchell, 35-36

One strain easy to miss in the fascinating ambitions of lecture 2, is Foucault's polemic with the Frankfurt school. That is to say, I missed it when I taught last year it only to be taken by surprise (looking ahead) in lecture 5. His main criticism above seems to be that ideology critique lacks originality (it goes back to romanticism) and is tired old hat ("etcetera;" "once again;" "well- known;" "constantly" "always"). The more subtle criticism, which is a consequentialist argument from authority ("believe me") is is that if the effect of ideology critique is a regime of unreason or error, it is no better. That is to say, and this has been noted by critics of Foucault, there is a kind of status quo bias built into Foucault's methodology; it is not sufficient to show that the status quo is problematic if you don't have a path toward a better outcome. 

It would be tempting now to turn to Foucault's methodology, but, as it happens, the polemic with Frankfurt is not finished yet. A few pages (or since these are lectures, minutes) later he claims that:

We should keep in mind that heterogeneity is never a principle of exclusion; it never prevents coexistence, conjunction, or connection. And it is precisely in this case, in this kind of analysis, that we emphasize, and must emphasize a non-dialectical logic if want to avoid being simplistic. For what is dialectical logic? Dialectical logic puts to work contradictory terms within the homogeneous. I suggest replacing this dialectical logic with what I would call a strategic logic. A logic of strategy does not stress contradictory terms within a homogeneity that promises their resolution in a unity. The function of strategic logic is to establish the possible connections between disparate terms which remain disparate. The logic of strategy is the logic of connections between the heterogeneous and not the logic of the homogenization of the contradictory. So let’s reject the logic of the dialectic and try to see—this is what I will try to show in these lectures—the connections which succeeded in holding together and conjoining the fundamental axiomatic of the rights of man and the utilitarian calculus of the independence of the governed. (42-43)

The heterogeneity he has in mind (recall) is liberalism's two traditions/concepts of liberty -- the contractualist-republican and utilitarian ones. And the significance of his claim, to liberalism, and this anticipates structurally something Jacob Levy has recently emphasized (recall;and recall) in a slightly different fashion, is the creative tension, a dynamics of liberty, within liberalism of these non-identical yet coexisting conceptualizations of liberty. And strikingly, Foucault wants to adopt a logic of inquiry that does justice to and can convey precisely these dynamics, "the connections which succeeded in holding together and conjoining the fundamental axiomatic of the rights of man and the utilitarian calculus of the independence of the governed. " As I signaled earlier in the week, by adopting what he calls strategic logic, Foucault's rewriting of the tradition is simultaneously a means of extending it because it facilitates the expression of this creative dynamic. 

At this point it is unclear whether Foucault thinks that by so doing one gives liberalism a new lease on life or one helps to push it over the precipice. But he is explicit that dialectical logic (which I take him to associate with the Frankfurt school, especially Horkheimer and Marcuse [both mentioned later in the lecture series)) is both simplistic and incapable of respecting the creativity lodged within the liberal tradition. De facto he is saying that the unity they offer is not worth having.

I could stop here, but it is worth noting what liberalism's rule of reason he is willing to apply strategic logic to is. He writes:

Governmental reason in its modern form, in the form established at the beginning of the eighteenth century with the fundamental characteristic of a search for the principle of its self-limitation, is a reason that functions in terms of interest. But this is no longer the interest of an entirely self-referring state which only seeks its own growth, wealth, population, and power, as was the state of raison d’État. In the principle to which governmental reason must conform, interest is now interests, a complex interplay between individual and collective interests, between social utility and economic profit, between the equilibrium of the market and the regime of public authorities, between basic rights and the independence of the governed. Government, at any rate, government in this new governmental reason, is something that works with interests....In its new regime, government is basically no longer to be exercised over subjects and other things subjected through these subjects. Government is now to be exercised over what we could call the phenomenal republic of interests. (44...46)

The rule of reason is thus the rule of interests, and this is, quite clearly, itself in constant motion a moving target. And in part it is a moving target because as interests are both shaped and pursued, perceived needs and how these perceptions are constituted evolve. So, we  may say that in this phenomenal republic there is lodged a further complex dynamics that has, in turn, a complex interplay with the dynamics of liberty. 

These complex, multi-faceted dynamics come together in the market which is (recall) a stupendous machine that helps coordinate the pursuit of our needs. I put it like this because here in lecture 2, Foucault has already signaled, without mentioning his name, that Hayek is not an aberration from liberalism but central to the tradition. But the lecture ends on alerting to the reader to the significance of James Buchanan's research program (e.g. Calculus of Consent):

The fundamental question of liberalism is: What is the utility value of government and all actions of government in a society where exchange determines the true value of things? this question encapsulates the fundamental questions raised by liberalism. With this question liberalism posed the fundamental question of government, which is whether all the political, economic, and other forms which have been contrasted with liberalism can really avoid this question and avoid formulating this question of the utility of a government in a regime where exchange determines the value of things. (46)

This is, of course, precisely the research agenda of public choice theory (Buchanan, Tullock), which goes unnamed in the Birth of Biopolitics (and perhaps everywhere in Foucault's work). And, of course, if we look at Buchanan's research project closely it mixes the two traditions noted by Foucault. The contractualist-republican element is (recall) exhibited in the significance to Buchanan of the constitutional moment; this is the part of Buchanan where he is very close to Rawls.  But within the constitutional order Buchanan inherits (what Foucault calls) the radical tradition and applies in Foucaults terms the utility value of government and all actions of government in a society where exchange determines the true value of things.* That is to say, Foucault helps explain why the Virginia school is, itself, an expression of the central problematics of liberalism. 

 

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.+

 

On Ellsberg's defense of Knightian Uncertainty, Arrow, and Ambiguity (II)

In most of this literature, the term ‘ambiguity’ has been treated as a synonym for what Knight called ‘uncertainty’ namely the fact that relative likelihoods are not characterized by well-defined numerical probabilities. The standard method of dealing with ambiguity in decision theory is to endow the decisionmaker with multiple priors as in Gilboa and Schmeidler (1989). This approach may be combined with a variety of preference models, notably including the maxmin model of Gilboa and Schmeidler (1989) and the smooth model of Klibanoff et al. (2005).
For a non-specialist this is puzzling; there is no obvious link to the ordinary meaning (or meanings)1 of ambiguity as a characteristic of propositions with more than one interpretation. In its normal usage, ambiguity is a linguistic concept, but in decision theory it is typically treated as a property of preferences.
The now-standard usage is quite different from that in Ellsberg’s (1961) original article. Ellsberg treated ambiguity, not as a property of preferences or relative likelihoods, but as a property of the information on which judgements of relative likelihoods might be based.

“Responses from confessed violators [of the expected utility (EU) axioms] indicate that the difference is not to be found in terms of the two factors commonly used to determine a choice situation, the relative desirability of the possible payoffs and the relative likelihood of the events affecting them, but in a third dimension of the problem of choice: the nature of one’s information concerning the relative likelihood of events. What is at issue might be called the ambiguity of this information, a quality depending on the amount, type, reliability and ‘unanimity’ of information, and giving rise to one’s degree of ‘confidence’ in an estimate of relative likelihoods.”

---Simon Grant, Ani Guerdjikova, John Quiggin "Ambiguity and awareness: a coherent multiple priors model." (17 April, 2020) [quoting p. 657 of Ellsberg's (1961) "Risk, Ambiguity, and the Savage Axiom]

In Yesterday's post I noted that while writing in defense of Knight, Ellsberg (1961) innovatively identified Knightian uncertainty with a class of decisions that can be specified precisely as empirical violations of the (Ramsey-)Savage axioms.* In particular, in such instances (when the formalization is violated) there is "no way to infer meaningful [subjective] probabilities for those events from their choices, and theories which purported to describe their uncertainty in terms of probabilities would be quite inapplicable in that area." (646; [parentheses added].) This identification was an innovation made possible and intelligible by formalization.  

As I remarked, Ellsberg noted that it is possible to treat the very same formalization as a [A] normative standard for a certain kind of ideal/idealized agent or [B] as a device to make predictions about agents. Ellsberg recognizes that Savage was (primarily) tempted by the normative interpretation whereas Ellsberg was primarily focused on the empirical interpretation. In particular, he wanted to identify classes of choice circumstances in which the possible rationality (669) of one's decision could not be articulated in or explained by reference to the Savage axioms. 

The Savage formalization allows the attribution to agents of precise subjective probabilities about distributions in the world. That is, in certain choice circumstances, the formalism allows for an interpretation of behavior that reveals if not inner states than at least structures that are compatible with such inner states. (It may be useful to compare this with the Samuelsonian strategy of revealed preference.)

With that in place, let's turn to ambiguity. As Grant/Guerdjikova/Quiggin note, in much of decision theory 'knightian uncertainty' is treated as synonymous with 'ambiguity' and has come to mean the absence of well-defined quantitative probabilities.+ And as they correctly note this is not quite what Ellsberg had in mind (and about that more in a second). Before I get to Ellsberg and ambiguity, it is worth noting that the absence of well-defined quantitative probabilities is indeed what Knight meant by uncertainty in the contrast between risk and uncertainty.

However, what the literature means by the absence of well-defined quantitative probability is (and now I quote Gilboa and Schmeidler (1989)) circumstances in which one "considers a set of priors as possible." (emphasis in original). It is important that this set is limited. Now, this is not quite what Knight, generally, intended to mean  by the contrast between risk and uncertainty.  In paradigmatic cases of uncertainty for Knight (and Keynes), we really have no right to make claims about the priors let alone treating them as well defined possibilities.

As an aside, this is not to deny that, as Richard Pettigrew noted to me, that "there are situations in which your evidence and/or current stage of scientific theorising is not going to deliver a unique probability assignment over the states you consider" but in which one has reasons to limit one's subjective probabilities to a set of possible distributions. And if forced to choose one may well come to think of your own (again quoting Pettigrew) as "your best subjective probabilities and use those while knowing that it would be equally compatible with your evidence to adopt an alternative set of probabilities."** If you and the other involved are experts in more or less the same situation, we can call this situation 'reasonable expert disagreement.' In fact, Savage himself noted (without having to contemplate violations of his axioms) that that “we must be prepared to find reasoning inadequate to bring about complete agreement.”  (1954 [1972]:7; see also 3, 67ff.) This is not trivial in the historical context, as I have noted elsewhere, in which the very idea of science involves the arrival of expert consensus. 

Now, as Grant/Guerdjikova/Quiggin note, Ellsberg thought ambiguity something else: namely it is a property of the "nature of one’s information."  And throughout the paper, Ellsberg goes on to deny that this feature "cannot be conveniently described in terms of a sample distribution." (659) Rather the point of using ''ambiguity" is to try to describe circumstances "when there are questions of reliability and relevance of information, and particularly where there is conflicting opinion and evidence." (659) So, one measure , but it is only one such measure, that one's information is ambiguous in Ellsberg's sense is the kind of reasonable expert disagreement [that is, situations one knows an expert's subjective probability about a distribution is compatible with another expert adopting an alternative set of probabilities.] But it is only one such measure because ambiguity itself cannot (to repeat) be reduced to and articulated in one distribution or another. As Ellsberg noted (660), Savage (57-60), too, had tried to characterize the nature of such ambiguity; Savage had explicitly acknowledged that formal theories known to him lacked resources to do so in satisfying ways.

Now, so for Ellsberg ambiguity is a "subjective variable" (660) about the nature of one's information. But the circumstances under which it arises can be characterized objectively. The proxy he uses to characterize these circumstances are violations of the Savage axioms can be predicted to occur and, perhaps, thought (somewhat) reasonable.

But that is not all of it (and this connects to Gilboa/Schmeidler).*** In circumstances of ambiguity one has, according to Ellsberg, some subjective grounds to "eliminate" some "possible distributions," but not all such possibilities. (661) For, Gilboa/Schmeidler (and the Bayesian literature) that is treated as the occasion to look for solutions to decision problems in terms of sets of priors as opposed to unique priors. In so doing, they end up echoing what I have called the displacement strategy of Knightian uncertainty. 

It is worth noting that there is an interesting difference of emphasis between trying to characterize an optimal decision strategy (in the manner of Gilboa/Schmeidler), say because one must choose or wish to guide policy-makers, and between wishing to promote understanding of how to think about or study one's information about possible distributions. (I have jokingly called the latter modal economics.) Ellsberg is clearly less interested in the former, and more interested in the latter.  Ambiguity in Ellsberg's sense does influence choice (664); yet Ellsberg is interested in finding out when and how, and to create "testable propositions" about such states [669], not to tell you what strategies to use in choice.

That is to say, Knight thought of uncertainty in terms of not knowing the (range and nature of) possible distributions. Ellsberg thinks we can develop rigorous thought  about what a theorist taught me to call intermediate circumstances in which we have some information about the possibilities. Somewhat surprisingly, according to Ellsberg -- and here he refers also to Chipman's experiments++ -- this thought has to be developed not from normative first principles, but rather experimentally ((655) within what in reviewing Donald Davidson, Suppes, and Siegel, he calls (in 1957) "the new field of experimental economics;" (1009). Of course, Ellsberg soon got preoccupied by different concerns. But that's really for another occasion.

So, Ellsberg has been remembered as generating a "paradox" in subjective expected utility theory. (This seems to be due to Harry V. Roberts.) This way of thinking about him says more about the economists' attitude toward their own past than Ellsberg. We have seen that alone among the responses to Knight that are still acknowledged by contemporary mathematical economists, Ellsberg wrote to vindicate Knight.+++ But he did so by using the canons of rational decision-making to specify when some circumstances of true uncertainty might occur. We could call them Ellsberg circumstances. These Ellsberg circumstances are characterized by constrained decision options in which information about world has the character of what he called ambiguity. In Ellsberg circumstances one grasps, we might say, the possible contours of the underlying distributions. Whether, in Ellsberg circumstances, it might pay to take bets on these contours is a tough challenge (one may well be in context of what he called "treacherous possibilities." (666)) He then opened the door not to better decision rules, but rather to the epistemology of information, which uniquely he thought ought to be explored experimentally.  

 

 

 

On the Crisis of Liberalism (VII): Pandemics and the Strong State

[I]f some day a legislature, having exhausted all other ways of improving mankind, should forbid the scoring of baseball games, it might still be possible to play some sort of game in which the umpire decided according to his own sense of fair play how long the game should last, when each team should go to bat, and who should be regarded as the winner. If that game were reported in the newspapers it would consist of a record of the umpire's decisions, plus the reporter's impression of the hoots and cheers of the crowd, plus at best a vague account of how certain men, who had no specified position on the field moved around for a few hours on an unmarked piece of sod. The more you try to imagine the logic of so absurd a predicament, the more clear it becomes that for the purposes of newsgathering, (let alone the purposes of playing the game) it is impossible to do much without an apparatus and rules for naming, scoring, recording. Because that machinery is far from perfect, the umpire's life is often a distracted one....

Wherever there is a good machinery of record, the modern news service works with great precision. There is one on the stock exchange, and the news of price movements is flashed over tickers with dependable accuracy. There is a machinery for election returns, and when the counting and tabulating are well done, the result of a national election is usually known on the night of the election. In civilized communities deaths, births, marriages and divorces are recorded, and are known accurately except where there is concealment or neglect. The machinery exists for some, and only some, aspects of industry and government, in varying degrees of precision for securities, money and staples, bank clearances, realty transactions, wage scales. It exists for imports and exports because they pass through a custom house and can be directly recorded. It exists in nothing like the same degree for internal trade, and especially for trade over the counter.

It will be found, I think, that there is a very direct relation between the certainty of news and the system of record. If you call to mind the topics which form the principal indictment by reformers against the press, you find they are subjects in which the newspaper occupies the position of the umpire in the unscored baseball game. All news about states of mind is of this character: so are all descriptions of personalities, of sincerity, aspiration, motive, intention, of mass feeling, of national feeling, of public opinion, the policies of foreign governments. So is much news about what is going to happen. So are questions turning on private profit, private income, wages, working conditions, the efficiency of labor, educational opportunity, unemployment, monotony, health, discrimination, unfairness, restraint of trade, waste, "backward peoples," conservatism, imperialism, radicalism, liberty, honor, righteousness. All involve data that are at best spasmodically recorded. The data may be hidden because of a censorship or a tradition of privacy, they may not exist because nobody thinks record important, because he thinks it red tape, or because nobody has yet invented an objective system of measurement. Then the news on these subjects is bound to be debatable, when it is not wholly neglected. The events which are not scored are reported either as personal and conventional opinions, or they are not news. They do not take shape until somebody protests, or somebody investigates, or somebody publicly, in the etymological meaning of the word, makes an issue of them.--Walter Lippmann (1921) Public Opinion, Chapter 23.

The second wave of liberalism (recall) has ended and liberalism may not survive. Some readers will say, good riddance, and to you I say, I hope you do better. This is the seventh post (recall here (I);  here (II); here (III); here (IV); here (V); here (VI) in an open-ended series (see also here, here, here,  here; and here). Critics of liberalism remind their readers during the pandemic that we care for more than "market logic" (Lisa Herzog in the New Statesman) or that there is more to the economy and living than "exchange value" and "how false our beliefs about markets are" (Simon Mair in The Conversation). Indeed, given the intellectual implosion of Richard Epstein this week, one may be forgiven into thinking that a variety of classical liberalism is truly on (ahh) a ventilator. But see these two fine essays by Nick Cowen. here and here; the second one reminds fans of the Non-Aggression Principle  that "natural rights are conceived to protect life as well as liberty and property." 

Friends and critics of markets have been much impressed (recall) by Hayek's argument that markets are impressive epistemic machines. But as is well known, Hayek thinks that the state and evolving common law provides the framework for the market and, at bottom, act as (to use David Schmidtz's phrasing) its impartial umpire. Hayek's friends, the ordoliberals (Ordos), went further (recall here) and claimed that quite a bit of state capacity, expertise, and even a proper ruling ethos is needed for the state to act as an umpire and to help the market function properly. 

More than any other liberal thinker, but drawing on ideas visible in Adam Smith, Walter Lippmann recognized that for markets, the media, and public opinion to function, the state is needed not just as impartial rule-setter and arbiter, but as a generator and storehouse of reliable and usable information. Conveniently he uses the umpire analogy for this. As the quoted passage reveals, Lippmann didn't think that only the state could be a reliable source of information (the stock-market is a private association).  It's compatible with his view that for-profit enterprise can also do so.

A core capacity of the modern state is to be a machinery of record. This is, of course, also recognized by those like Foucault and James C. Scott who argue that legibility of the population is a key feature of state-building and statecraft. This is a feature and not a bug of liberalism. Because (and to combine Lippmann and Hayek) in virtue of having a machinery of record, the market machinery can do its fallible work. That means that one of the distinctive mind-sets and dispositions of the liberal state is to gather data in order to discharge its proper functions. And, as Adam Smith already noted, to "prevent" the "spread" of a most loathsome "disease" deserves the "most serious attention of government" (WN 5.i.f.60.) 

Of course, all states collect information. The distinctly liberal state does so in a rule-following fashion and respecting certain limitations which include the possibility of publicly reconstructing the data-chain and so being contestable.* One reason why modern (neural-networky) AI is causing so much trouble for liberal thinkers is that while it is rule governed its internal workings are also opaque and so difficult to contest.

We see this in the attitude of the German government which started building up the capacity to test early in the pandemic. Germany conducts "50,000 coronavirus tests a day." This is partly due, as the Financial Times reports, to its decentralized testing regime. (The US's reliance on the CDC makes it vulnerable to mismanagement, underfunding or simple errors of judgment.) But partly this is also due to the fact that German liberalism, as represented by the ORDOs, has not treated well-directed state capacity itself with suspicion.

No doubt even well functioning machinery of record can be abused or captured by sinister interests. But, going forward it would be useful for liberals to remember Lippmann's insight that maintaining the Weberian state as a machinery of record with the dispositions to act in the public's interest, remains a vital element of a liberalism worth having.

 

On Aspiration (II)

In all the liberal and ingenious arts, in painting, in poetry, in music, in eloquence, in philosophy, the great artist feels always the real imperfection of his own best works, and is more sensible than any man how much they fall short of that ideal perfection of which he has formed some conception, which he imitates as well as he can, but which he despairs of ever equalling. It is the inferior artist only, who is ever perfectly satisfied with his own performances. He has little conception of this ideal perfection, about which he has little employed his thoughts; and it is chiefly to the works of other artists, of, perhaps, a still lower order, that he deigns to compare his own works. Boileau, the great French poet (in some of his works, perhaps not inferior to the greatest poet of the same kind, either ancient or modern), used to say, that no great man was ever completely satisfied with his own works.--Adam Smith The Theory of Moral Sentiments, 6.3.26, p. 248

Last week, I noted that Agnes Callard's inspiring Aspiration: The Agency of Becoming left some intertwined problems because her analysis seems to require that the normativity of aspiration is not grounded in the intentions of the aspiring agent. (I called this the problem of aspirational normativity.) More subtly if such normativity exists at all it seems to have to be grounded in a counterfactual self that (prospectively) has a very limited identity with the aspiring agent. (I called this the problem of counterfactual strangers.) The two problems are distinct, but in what follows I sketch a proposal that may resolve them simultaneously.

Counterfactual agents are a staple of impartial spectator theories, and play different roles in them. For present purposes, I am going to stipulate that in Adam Smith's theory any impartial spectator is always a counterfactual agent with complex identity profiles.* The ways of being of such agents are complicated, but I am going to skip over that issue right now. Here I am primarily interested in one feature: namely the role of an impartial spectator theory by which the norms of our actions would be judged; these norms "express" in the felicitous phrase of Samuel Fleischacker, "the feelings of an impartial spectator." These norms express the feelings of an impartial spectator because in Smith's theory, it's the impartial spectator that sets meta-ethical standards ("that precise and distinct measure can be found nowhere but in the sympathetic feelings of the impartial and well-informed
spectator" TMS 7.2.1. 49, p. 294).**

Impartial spectator theories do not fit easily into the current taste for monistic meta-ethical systems, so I accept (for the sake of argument) that they have considerable weaknesses in meta-ethics. But as it happens they are especially suitable to explaining the normativity of aspiration. Because whatever normativity aspiration has it inherits from (i) the worthiness of the activity it is aspiringing toward, and (ii) the aptness of the process by which aspiration is fulfilled. Both of these involve the kind of contextual judgments that are a strength of Smithian impartial spectator theories. 

More subtly, while prospectively the aspirant lacks -- to adopt Callard's terminology -- the intrinsic rational ground to justify her aspiration, she can point to exemplars who extrinsically inspire or guide her path; some of these (exemplars) may well act as mentors along the way. Callard has little to say about exemplars, but mentors play a key role in her account.+  

As an aside, I do not mean to suggest that (i-ii) exhaust the relevant conditions; perhaps the impartial spectator would also make contextually salient judgments about (iii) the fitness of the aspirant for the aspiring activity; and (iv) the suitability of the adopted exemplars (and mentors). 

Okay, as it happens Smith's moral philosophy is especially fruitful when it comes to reflecting on the normative grounding and standard-setting of exemplars within practices it is worthy to aspire to. I do not mean to suggest Smith's account explains especially well why particular practices are worth aspiring to (and if Callard is right such an account can only become available to the aspirant after the aspiring activity has been completed for the greater part), but rather if we accept that such practices exist and that they are worthy of aspiration, Smith offers (even if we note his gendered biases/examples) a useful account of how his impartial spectator account can function and help articulate the standards within such worthy-of-aspiration practices.  In fact, Smith thinks we have a good enough grip on the existence and tropes surrounding the reality of standards within such worthy-of-aspiration practices that he uses these, as he does in the larger context of the quoted passage at the top of this post, to illuminate his meta-ethical account of the role of standards in moral life.

So, here's my suggestion, in order to tackle the problem of aspirational normativity and the problem of counterfactual strangers, we need to introduce an impartial spectator theory. This theory posits a counterfactual agent with which one shares partial identity through which one can endorse the exemplary standards of an aspiring-worthy-activity. This counterfactual agent with which one shares partial identity will be the source of normativity for one's proleptic reasons to aspire to become like, in certain crucial ways, the exemplar that is guiding one's aspiring activity. 

If one is unfamiliar with Smithian impartial spectator theories, the previous paragraph will probably seem outright mysterious. (Even if one is uninterested in reading my book on Adam Smith, one could do worse with one's time than read The Theory of Moral Sentiments.) But if one is familiar with these, the previous paragraph should provide the discerning reader with a how-possible sketch to explain the normative ground of proleptic reasons. Of course, lurking in here is also a problem of teleology: how can proleptic reasons gain a justifiable grip on an agent?

Now it is tempting to answer that question by saying 'all of us have access to the judgments of the impartial spectator' at any given time (see Adam Smith's TMS). But that answer may seem to skip over the particularly teleological nature of the problem. So, let me grant that before that answer can seem fully compelling, I need to offer an account of naturalistic teleology. I leave that to a future post.

 

 

On Chicago Economics (Becker vs Kirzner), Foucault, and Darwinism (via Daniel Dennett & David Haig)

Becker, for example—the most radical of the American neoliberals, if you like—says...that the object of economic analysis can be extended...and that economic laws and economic analysis can perfectly well be applied to non-rational conduct, that is to say, to conduct which does not seek at all, or, at any rate, not only to optimize the allocation of scarce resources to a determinate end. Becker says: Basically, economic analysis can perfectly well find its points of anchorage and effectiveness if an individual’s conduct answers to the single clause that the conduct in question reacts to reality in a nonrandom way. That is to say, any conduct which responds systematically to modifications in the variables of the environment, in other words, any conduct, as Becker says, which “accepts reality,” must be susceptible to economic analysis. Homo oeconomicus is someone who accepts reality. Rational conduct is any conduct which is sensitive to modifications in the variables of the environment and which responds to this in a non-random way, in a systematic way, and economics can therefore be defined as the science of the systematic nature of responses to environmental variables.
This is a colossal definition, which obviously economists are far from endorsing, but it has a certain interest. It has a practical interest, if you like, inasmuch as if you define the object of economic analysis as the set of systematic responses to the variables of the environment, then you can see the possibility of integrating within economics a set of techniques, those called behavioral techniques, which are currently in fashion in the United States. You find these methods in their purest, most rigorous, strictest or aberrant forms, as you wish, in Skinner, and precisely they do not consist in analyzing the meaning of different kinds of conduct, but simply in seeing how, through mechanisms of reinforcement, a given play of stimuli entail responses whose systematic nature can be observed and on the basis of which other variables of behavior can be introduced. In fact, all these behavioral techniques show how psychology understood in these terms can enter the definition of economics given by Becker....
...Becker’s definition, which, again, although it is not recognized by the average economist, or even by the majority of them, nonetheless, despite its isolated character, enables us to highlight a paradox, because homo oeconomicus as he appears in the eighteenth century—I will come back to this shortly—basically functions as what could be called an intangible element with regard to the exercise of power. Homo oeconomicus is someone who pursues his own interest, and whose interest is such that it converges spontaneously with the interest of others. From the point of view of a theory of government, homo oeconomicus is the person who must be let alone. With regard to homo oeconomicus, one must laisser-faire; he is the subject or object of laissez -faire. And now, in Becker’s definition which I have just given, homo oeconomicus, that is to say, the person who accepts reality or who responds systematically to modifications in the variables of the environment, appears precisely as someone manageable, someone who responds systematically to systematic modifications artificially introduced into the environment. Homo oeconomicus is someone who is eminently governable. From being the intangible partner of laissez-faire, homo oeconomicus now becomes the correlate of a governmentality which will act on the environment and systematically modify its variables.--Michel Foucault, The Birth of Biopolitics, 28 March 1979, translated by Graham Burchell,  269-271

While it is not the main topic of today's digression, it is worth noting how incredibly prescient Foucault is in grasping that what is now known as nudging is an eminently foreseeable, internal development of Chicago economics. In context that point is even clearer in light of the concluding remarks of the previous lecture (of March 21), which focuses on the possibility of "an environmental type of intervention" to produce or stabilize homo oeconomicus which in the Becker/Stigler version of Chicago just is MaxU.

Now, Foucault rightly notes that Becker actually relies on a relatively thin notion of agency which really amounts to responding 'systematically to modifications in the variables of the environment.' Incentives matter, after all. Let's call this the modern economist's conception of agency.  

As an aside, as I have noted before Foucault actually understates the situation here. Becker's own position develops an idea from Alchian (and noted by Arrow) that even 'agents' who behave stochastically, will, if they respond to the environment -- and there is some selective winnowing -- approximate the behavior of MaxU, or, as Milton Friedman also emphasized, can be treated as if they do so. 

Okay, with that in place, let's turn to Daniel Dennett's treatment of the Warden's Rule: "if it can happen, it will happen. (See Freedom Evolves, 2003, 160-61. The version I discuss below is reproduced from Dennett's preface to David Haig's excellent From Darwin to Derrida (forthcoming); recall this post by Haig.)

The Warden's Rule...is an improvement on “Murphy’s Law,” that whatever can go wrong will go wrong. It is an improvement because there are real circumstances where a version of it is true, not merely an amusing expression of pessimism. It is supposedly the maxim of every prison warden. Its rationale is that prisons are full of prisoners with time on their hands and a measure of patience and an inquisitive and competitive nature. They will exhaustively search every aspect of the system in which they are incarcerated looking for a way out, or a way of improving their condition, however arduous the path.  How clever are the prisoners?  Some are brilliant, others just dogged; if they share their discoveries it doesn’t much matter who gets credit for the innovations. Instead of testing them for intelligence, test their environments for opportunities. (In Elbow Room (1984), I described the existence of a bag of jewels in a trash can a few feet away as a bare opportunity—if only I had known about it, I could have been rich, but I had no information at all that would lead me to check that trash can for treasure.) According to the Warden’s Rule, even bare opportunities must be counted, since prisoners have nothing better to do than hunt for them. No trash cans go untested in a prison.

Bare opportunities abound in the world of evolution, and it doesn’t take intelligence to stumble across them; it just takes lots of time for variations to arise and get tried. If there is information in an evolutionary environment, a difference that could make a difference to something that could detect or just respond to that difference, it will make a difference (generally—this is a rule of thumb), a difference that in turn can be amplified by reproduction, making a bigger difference and, in a recursive process, making the detection of that difference more likely in the future, and so forth.  Yes, this is just another way of explaining how natural selection works, but it works this way all the way down, so that even simple molecular structures can be usefully seen to be like those persistent prisoners, ever delving, ever seeking an advantage that they can exploit to further their own “interests”.  The prisoners have a single overarching goal: Escape!  These agents have a different goal: Reproduce!--Daniel Dennett.

The warden rule prompts a useful contrast with the modern economist's conception of agency. What the warden's rule suggest is that rather than treating agents as responding systematically to modifications in the variables of the environment, agents are better thought of as responding to systematic possibilities in the environment. The difference isn't primarily about the shift from incentives (a) to information (after all, the variables in the environment can be treated as information). Nor is the difference (b) that the warden rule stipulates away search and opportunity costs (something an economist would not do)--the warden rule effectively assumes that prisoners have infinite time (and patience, etc.). I don't mean to deny that (b) is important. That evolution takes place in time, often extended periods of time, is non-trivial. Rather it is, primarily about the shift from tracking features of the environment (c) to searching and exploiting possibilities in the environment. In particular, the warden rule (d) tacitly recognizes the agency of the agent in her shaping the environment in ways that cannot be effectively foreseen. I view (c)-(d) as connected. 

Now, I could stop here.* As Foucault notes, but does not explore, Becker's analysis received a response by Kirzner in 1962 (the leader of the Austrian school of economics in exile). As I noted a few weeks ago, Kirzner was very interested in entrepreneurship. This matters for my present purposes because (d) just describes entrepreneurship (or attributes it to organisms). So, today, I am going to assume Kirzner's commitment to some analogue of (d). But I am interested in his relationship toward (c). To get to that we must note that Kirzner conceived of market processes as essentially about "the systematic way in which plan revisions are made as a consequence of the disappointment of earlier plans." (381; Kirzner is describing buying and selling plans.) In other context, it would be tempting to talk about updating here. But we can discern here that Kirzner is articulating features of (c). 

The connection between (c) and the intentional stance is also made explicit by Kirzner: "only by assuming that buyers purposefully seek to achieve given goals can we predict that their thwarted market plans of yesterday will lead to their systematically offering more attractive choices to sellers today." (382) Of course, Kirzner may well think that the intentional stance is more than a mere stance, but as he goes to acknowledge one need not assume such purposeful rationality.+ He recognizes that market institutions as a system may induce behavior that can be treated as purposeful. It is worth asking if he was already aware of Vernon Smith's striking experimental results (about that some other time).

Now, one may think that I have not said sufficient amount to motivate that Kirzner is really articulating a version of (c). But the relationship between search (which is his main focus in context), the intentional stance, and the role of possibility in the environment, is pretty clear in his thought: "on the assumption of rationality, a market demand curve is seen as a set of hypothetical equilibrium positions." (I agree it is not identical to the Warden Rule, but we're not far removed either.) 

Okay, let me zoom out again. Foucault, correctly noted that one way to understand the difference between classical liberalism (e.g., Adam Smith) and many branches of twentieth century neoliberalism, is that the former treated economic analysis in terms of exchange and the latter in terms of competition. Both of these are fundamentally about processes. Foucault treats Becker (with a nod to Robbins) as something radically new: now "the starting point and general frame of reference for economic analysis should be the way in which individuals allocate these scarce means to alternative ends." (223, 14 March 1979). And with the ends given (by experts or politicians, tradition, society, etc.) the individual can be nudged into being good at such allocations.

As Kirzner discerned, Becker's approach is not at all about process. By contrast he thinks, "it is primarily upon the systematic revisions of disappointed plans that the market process depends." (384) He thinks this feature of purposeful behavior generates stable outcome patterns associated with market processes. As is well known, Becker's approach was vastly more influential in economics and facilitated the manner in which it became the imperial discipline in other social sciences. So, Foucault was prescient to ignore Kirzner and focus on Becker.

Let me wrap up. I understand Vernon Smith's contribution to economics in part to teach us all that the micro-foundations of economics are unknown. (Let's not even begin with macro.) However, in other part, some institutions and norms generate remarkably stable outcome patterns without us really understanding why. With Dennett and David Haig re-articulating Darwinian theory, and, in particular, supplying a new set of ideas about natural teleology and social functions (recall Haig here; and me on Dennett here); and about which more soon), a Darwinian process oriented approach (not necessary Kirzner's) is ripe for a major relaunch in the human sciences, also in economics. 

 

On Milton Friedman and Hayek on Consensus

I venture the judgment, however, that currently in the Western world, and especially in the United States, differences about economic policy among disinterested citizens derive predominant from different predictions about the economic consequences of taking action - differences that in principle can be eliminated by the progress of positive economics - rather than from fundamental differences in basic values, differences about which men can ultimately only fight.--Milton Friedman (1953) "The Methodology of Positive economics" in Essays in Positive Economics, p. 5

Friedman relies here on [A] a tacit distinction between basic values and not-so-basic values. When there is (fundamental) disagreement over these society returns, we can say, to the state of nature. So, not unlike one of his Chicago mentors, Henry Simons, he relies on the idea (recall) that states presuppose considerable consensus. (I have noted the somewhat disastrous afterlife of this idea in Chicago economics elsewhere; e.g.  here and here.) Another, no less consequential, distinction he relies on [B]  is that between disinterested and (ahh) self-interested citizen. Presumably the disinterested citizen advocates the public or common interest, whereas the self-interested citizen promotes a self/sinister interest. 

This means that in a functioning political unity -- i.e., one with considerable background consensus -- if we encounter public disagreement, there are for Friedman three possible sources of controversy: (i) disagreement over likely outcomes of policy; (ii) disagreement over which no-so-basic value to prioritize [in the context of major agreement]; (iii) people advocating their own interest as opposed to public interest. In principle, and this is the point of Friedman's essay, (i) can be eliminated as positive (that is empirical) economics progresses. This suggests, and again this is no surprise, that for Friedman a mature positive economics will be characterized by consensus over the relationship between policy and outcomes. In addition, in principle, (ii) can be eliminated as normative economics progresses. Friedman does not spend much time on this, and it is unclear if he agrees with Robbons who does not expect such convergence in normative economics, or with later more optimistic professional philosophers.

An important consequence of Friedman's stance is that it entails that if, one agrees with him and understands oneself as a policy-scientist (and Friedman often does), one can model public policy debates with representative agents who exhibit public spirit. For these will all agree on the common good. For those who think of 'Chicago' as a bastion of self-interest this is surprising. It is worth noting this because it makes clear that the subsequent public choice (Buchanan/Tullock) and rent-seeking (George Stigler, etc.) revolution (by his Chicago trained contemporaries) also impacts Friedman's own program. 

Another important consequence of Friedman's stance is that one cannot assume that public spirited representative agents will agree across space and time, especially as they become culturally more distant. This has obvious consequences  when one aims to do comparative policy work.

I could end here, but I want to call attention to a feature in Hayek's thought (ca 1945). Consider the closing two sentences, in particular, from the following passage from a justly famous essay:

The price system is just one of those formations which man has learned to use (though he is still very far from having learned to make the best use of it) after he had stumbled upon it without understanding it. Through it not only a division of labor but also a coordinated utilization of resources based on an equally divided knowledge has become possible. The people who like to deride any suggestion that this may be so usually distort the argument by insinuating that it asserts that by some miracle just that sort of system has spontaneously grown up which is best suited to modern civilization. It is the other way round: man has been able to develop that division of labor on which our civilization is based because he happened to stumble upon a method which made it possible. Had he not done so he might still have developed some other, altogether different, type of civilization, something like the "state" of the termite ants, or some other altogether unimaginable type. All that we can say is that nobody has yet succeeded in designing an alternative system in which certain features of the existing one can be preserved which are dear even to those who most violently assail it-such as particularly the extent to which the individual can choose his pursuits and consequently freely use his own knowledge and skill. F Hayek (1945 "The Use of Knowledge in Society," 528

In the paper, Hayek treats the price system as a natural machine with a clear purpose: "We must look at the price system as such a mechanism for communicating information if we want to understand its real function." (526) This is compatible with the thought (a) that the price system has many derivative functions and (b) that the real function can be undermined or obstructed. (For Hayek's argument, (b) is very important.) And one can easily see, that the gap between Hayek and Ordoliberalism is fairly small, if Hayek were to grant the Ordos that (c) one of the key functions of the state is not just not to interfere with, but to maintain actively the conditions that enable the proper functioning of the price system.

Now, one may well wonder, why Hayek assumes that there is one and only one "real function" of the price system. I think he helps himself to this claim without argument. But let's stipulate he is right about this, and that it's methodologically kosher to attribute functions to social systems at all. Even so, one may well wonder what function the communicating of information serves. In this paper he is rather silent about this more fundamental end. And here it is clear that whatever it is, not unlike Friedman later, Hayek relatives this more fundamental function to "our civilization."

And somewhat surprisingly, not unlike Simons and Friedman, Hayek also assumes considerable agreement over what Friedman calls basic values.  This is clear from his response tho those who reject the price system as privileged machine. To them he de facto says, 'we agree about some of these fundamental ends -- which may call in short-hand individual 'freedom' and 'autonomy' -- and any replacement of the price-system will have to secure these, too.' My interest here is not whether this is an accurate claim about either the status of these ends in our civilization nor the extent to which illiberal critics of the price-system would endorse these. (I am, it should be clear, skeptical of both.) But rather that Hayek, too, ends up suggesting that if there is any disagreement over the functioning of the price system, it will not be due to disagreement over fundamental ends. And this suggests, that if there is disagreement, the other side(s) is/are engaged in special pleading. 

Not for the first time I am struck by the lack of space for genuine pluralism over ends in twentieth century liberalism. As regular readers know I think this plays no small role in the predicament it finds itself today.

 

Kant, Neo-Liberalism, and Migration

This a libertarian must believe, for war is the great threat to his kind of society. There is simply no democratic answer to the problem of external defense, save indefinite extension of federalism, first, into a predominantly powerful supranational federation, and then gradually into inclusive world organiza­tion of all nations capable of responsible participation. Here the important next steps must be taken in the field of com­mercial policies; and the next conspicuous institutional innova­tion will be an international court with compulsory jurisdiction, albeit only among some Western democracies at the start. Libertarian democracy can survive without world order but not without secular movement toward such order.--H. Simons (1948) Economic Policy for a Free Society: A Credo, pp. 15-16 

First, a terminological point. Simons' use of 'libertarian' is a bit confusing. It means, for him, in "the English-Continental sense" liberal (given that in the US context 'liberalism' became associated with the New Deal). In contemporary discourse he would be considered a bleeding heart libertarian. Either way, my interest in Simons (recall) is that he helped shape the Chicago school of economics in crucial fashion.  And that he is one of the founding thinkers of the constitutive schools of twentieth century neo-liberalism, which was at the heart of what I call the second wave of liberalism (roughly 1945-2008).

So, I was a bit surprised to see such a clear statement of the  Kantian program (recall) for perpetual peace. Kant thought that growing regional federated commercial liberal democracies (which he called 'republics' because the term 'democracy' was still reserved for direct rule by the people) were the key to world peace. (The idea that liberal democracies don't fight with each other can also be traced back to Kant and Addison.) 

Simons is clear that such supranational federations -- the idea anticipates the EU -- would be committed to internal free-trade and, ideally, to external free trade. A free trade regime will generate international policy coordination and international laws. That is to say, Simons thinks that international institutions will develop functionally out of a joint commitment to free trade among like-minded peoples. I put it like this to do help explain his relativizing the scheme to 'Western Democracies.'

The point is not that Simons thinks non Western peoples are incapable of democratic federation. Rather, Simons thinks any such democratic, liberal and commercial supranational federation presuppose a"moral consensus." So, shared values are presupposed in any stable government. (This idea has a powerful afterlife in the ideas of his students, Milton Friedman and George Stigler.) While tacitly drawing on Max Weber (Simons' Chicago colleague (recall) Frank Knight had introduced Weber to US readers), he puts it as follows: 

The basic function of central government is to sustain domestic peace. Internal order is prerequisite to external defense and, of course, is the essence of world federation. The good central government will represent a monopoly of violence; it must sustain that monopoly against both its constituent political units and all extra-governmental bodies. It must promote all kinds of peaceful intercourse, intellectual and commercial. It must articulate the prevailing moral consensus and promote enlargement of that consensus by organized, free discussion and legislative-judicial experimentation. (16)*

That is to say, one of the goals of the liberal state (itself ideally a federated state) is to promote moral consensus by being the site of reasoned debate, and by engaging like-minded states to create the international infrastructure that will facilitate more trade and more moral consensus, including a legal regime.

Now, Simons is clear that these are ideal types. The real world is messy and imperfect. And he explicitly acknowledges that liberalism must be possible without it existing everywhere. But he does think movement toward such a Kantian order is required to keep faith in liberal progress. The EU has internalized this idea in its ordinary self-conception. The fact that Brexit has become truly thinkable will be an interesting test case for the survival of this idea (and the EU).

That liberalism must be possible without it existing everywhere is not obvious. Because Simons is clear that the very possibility of war is a threat to a liberal state. For it requires considerable preparedness; and such preparedness involves planning and subordinating individual freedoms to collective survival. The idea is worked out at greater length in an earlier text by Von Mises (1927; unmentioned by Simons), who is clearly indebted to Kant (whom he cites in relevant context a few times in the text):

One would think that after the experience of the [first] World War the realization of the necessity of perpetual peace would have become increasingly common. However, it is still not appreciated that everlasting peace can be achieved only by putting the liberal program into effect generally and holding to it constantly and consistently. "Liberal Foreign Policy" in Liberalism: in the Classical tradition (110-111)

This would be a natural place to end. 

But Mises also complicates the situation. One may well wonder, and critics of Liberalism have wondered, why if Liberalism promotes peace the first wave of liberalism ended in the disaster of first world war.+ Mises himself suggests that it is the rise of nationalism and zero-sum colonial competition that is to blame.+ Mises is quite capable of divorcing colonial/imperial projects from liberalism. But the situation is not so easy when it comes to liberalism's relationship to nationalism:

As long as nations were ruled by monarchical despots, the idea of adjusting the boundaries of the state to coincide with the boundaries between nationalities could not find acceptance. If a potentate desired to incorporate a province into his realm, he cared little whether the inhabitants—the subjects—agreed to a change of rulers or not. The only consideration that was regarded as relevant was whether the available military forces were sufficient to conquer and hold the territory in question. One justified one's conduct publicly by the more or less artificial construction of a legal claim. The nationality of the inhabitants of the area concerned was not taken into account at all.
It was with the rise of liberalism that the question of how the boundaries of states are to be drawn first became a problem independent of military, historical, and legal considerations. Liberalism, which founds the state on the will of the majority of the people living in a certain territory, disallows all military considerations that were formerly decisive in defining the boundaries of the state. It rejects the right of conquest. It cannot understand how people can speak of "strategic frontiers" and finds entirely incomprehensible the demand that a piece of land be incorporated into one's own state in order to possess a glacis. Liberalism does not acknowledge the historical right of a prince to inherit a province.

To simplify: while nations pre-exist liberalism, liberalism makes nationalism politically feasible. (It is worth comparing (recall) Mises with Herzl.) As Mises points out (on the next page) the key mechanism is the plebescite or referendum. Moreover, where national identities must co-exist, and mutual trust is low, the democratic impulse of liberalism also makes nationalism a permanent political problem. Mises is quite frank about this. For Mises cultural and ethnic pluralism is a source of tension. 

Mises' solution to the problems of a multi-national democracy is to double-down and to suggest minimizing government functions to bare (liberal) necessities (rule of law, defense of property rights, and some coordination functions, etc.) such that gaining power becomes of little value to people of ill will. As regular readers know I don't think Mises, nor any of his modern libertarian heirs, can explain how in the eternal presence of rent-seeking, one can attain such a nightwatchman state without a massive change of popular opinion (and the political dis-empowerment of those that benefit from spoils and rents who are by definition very well organized politically). So I view this as unstable. 

But, where liberalism generates homogeneous nation states, and has what Simons will call moral consensus, Mises allows that public provision of public goods (like education) need not cause internal strife. I think this helps predict and explain why in relatively homogeneous liberal democracies welfare states are less controversial (than in the more pluralistic ones). In homogeneous states the moral consensus can be broader. But if they engage in free trade and maintain a liberal economy, these tend to be wealthy and growing and so become, as Mises emphasizes, very attractive to immigrants and, thereby, become multicultural/national. So, by Mises' own lights -- he is for open borders [Simons is more reserved about this]-- this is a recipe for more internal tension and so turns out not to be stable either.**

I have gone on long enough. But the problem is that the early neo-liberals did not develop a genuinely pluralist conception of moral and political life; and so their commitment to open borders and democratic liberalism are a source of tension.     

 

The road to serfdom before Hayek (Knight, Lippmann, and a note on Weber today)

The author insists earnestly and at length that a regime of collectivism will inevitably be a dictatorship, a tyranny over the whole cultural and personal life of man as well as over economic activity, however defined. The turning-point from liberalism as laisser faire to liberalism as collectivism, miscalled liberalism, is set at about 1870.

The current phase of the movement toward collectivism in the United States, as regards the structure and process of government, is the active development of pressure groups under the wing of the government itself, together with the displacement of the activities of legislature and courts by administrative organs and processes. The original liberal movement was, more than any other one thing, a struggle for an effective legal order. Our self-styled liberals of today are doing their utmost to get away from legality. Ostensibly they would have the administration operate under a general legislative grant of power. But it is obviously a short step to a condition in which the legislative mandate will be under the control of the administration of the day itself-and then a shorter, relatively immaterial step to the adjournment "without day" of all lawmaking by elected representatives as a superfluous formality or an impediment to action. This means the end of all popular control of government, all "responsibility." Only an omnipotent and irresponsible state can deal with or suppress the organized pressure groups-or all but one of them, which one would become the state. All this seems to the reviewer as clearly as it does to Mr. Lippmann not only historically inevitable but actually necessary if there is to be any political order at all in a country where the leading economic and (perhaps other) special interests have been taught to think that it is the business of government to fight their battles against "business" and have been organized, or encouraged or allowed to organize, to get "justice" for themselves, or their "rights" -or simply what they want. [p. 865]...

Perhaps even more "glaring" is the absence of any real argument on the question as to why a collectivist regime must be a despotism. Even in a book of this sort, intended for a wide audience, it is somewhat jarring to have it simply asserted as the proclamation of an "inexorable law" (or a plain dogma), not merely that collectivism means tyranny but apparently that it is impossible, or at least theoretically incompatible with social life above the level of savagery. We read: "The really inexorable law of modern society is the law of the industrial revolution, that nations must practise the division of labor in wide markets or sink into squalor and servitude." ...... And again, "There is no way of practising the division of labor, and of harvesting the fruits of it, except in a social order which preserves and strives to perfect the freedom of the market" (pp. 206 and 207).--Frank Knight  (1938) reviewing Walter Lippmann's The Good Society in the Journal of Political Economy. 867 [Ht Ross Emmett]

Frank Knight, who is now primarily known as author of Risk, Uncertainty, and Profit and the Godfather of Chicago economics, was also one of the first, prominent Weberians Stateside. (Among other things he published a translation of Weber's Wirtschaftsgeschichte (General Economic History)—as Ross Emmett (the greatest Knight scholar) taught me --,the first book by Weber to appear in English in 1927.) I return to Weber below.

Knight is clear that by 1938, by then then in the second FDR administration, the sense of the state of crisis in liberalism had not been abated. The idea of crisis frames the review; and as he puts it near the end of his review, "It is not too much to say that the history of modern democratic systems leaves the issue unsettled as to whether representative machinery of any kind, or government based in any way on popular discussion, can meet such a crisis and survive." (271) On the very page cited by Knight above (207), Lippmann himself speaks of "the debacle of nineteenth-century liberalism."  According to Lippmann (this is not in Knight), liberals started to loose the battle of ideas from 1848 onward, and subsequently to loose political battles, too.* It was not at all obvious, subsequent the first world war, the Russian revolution, and the great depression, that liberalism would survive the century. It is no surprise that much of the most arresting writing of the period is so illiberal in character. 

In the review, Knight treats liberalism as primarily characterized by three features: the rule of law (an "effective legal order"), representative democracy, and government by "popular discussion." Contemporary readers may suspect that "the rule of law" is primarily about property rights, but Lippmann is explicit that he thinks of something more expansive with an "effective legal order;" that is, it involves commitment to the idea that "individuals have equal and reciprocal rights and duties." (The Good Society, 267) In fact, Lippmann is extremely critical of late nineteenth century liberal jurisprudence which, disguising the role of power and utility in generating property rights, had turned property rights into a universal law (e.g Good Society, 258). I mention this because both Lippmann (and Knight) are notable among liberals of the crisis-ridden age for their continued adherence to democratic norms.** 

Knight's response to Lippmann's book is notable because he calls attention to (what we may call) the road to serfdom before Hayek thesis he attributes to Lippmann. I am not just making a joke. Hayek read The Good Society (he participated in a famous colloquium on the book in 1938 in Paris). And as scholars have noted he engaged with Lippmann's writings in the period prior to the Road to Serfdom (1944). Lippmann almost wrote the preface to the Road to Serfdom. And Hayek almost added a "postscript to The Road to Serfdom, written in 1948," in which Hayek noted that "although he had written the book chiefly on the basis of his knowledge of Britain and Germany, he was aware that American writers had made similar points about the situation in the United States, including ‘‘such well known books as Mr Walter Lippmann’s Good Society, which of course I knew and to which I ought to have acknowledged greater indebtedness.’’"+  

I take the 'road to serfdom thesis' to entail that once one switches to an extensively planned economy ('collectivism" in Knight's and Lippmann's terminology), the needs of planning and stability will ensure that a dictatorship will follow. I read Knight as attributing a version of this thesis to Lippmann. [There are passages that suggest Hayek may have flirted with the idea that even the welfare state would inevitably generate dictatorship (as Andrew Farrant and Edward McPhail argue), but for present purposes I remain agnostic on that more expansive version of thesis.]

What's striking about Knight's interpretation of Lippmann is that the road to serfdom is taken to be immanent in the New Deal policy of organizing interest/pressure groups. In Lippmann's book, however, pressure groups precede the New Deal by "sixty years."His main example of the role of pressure groups in what he calls "gradual collectivism" is the tariff of the Hawley-Smoot Bill in 1930. It is the case, however, that Lippmann claims that "the one certain thing is that in a democratic society the granting of some privileges must be followed by the granting of more privileges." (116) This suggests an open-ended process to ever more collectivism.

Even so, the thrust of Knight's review is peculiar because after thanking Mises and Hayek, Lippmann writes in his acknowledgments of his indebtedness to "John Maynard Keynes, who has done so much to demonstrate to the free peoples that the modern economy can be regulated without dictatorship." (viii) That is to say, from the start of his book Lippmann thinks democracies can find ways to prevent the road to serfdom. 

In fact, the very pages in The Good Society that Knight cites and quotes also claim "the renascence of liberalism may be regarded as assured." And Lippmann thinks this is the case because he thinks (i) something like liberalism is the natural accompaniment of an economy with advanced division of labor (206); (ii) he thinks collectivism, "in practice" (that is peace-time) is impossible and leads to a decline in living standards, terror ("brutality" 2o6). In addition, (iii) he thinks the collectivist states become "prey" and risk "conquest." (207)

Somewhat striking, and ironically, Lippmann's confidence that collectivism in practice is impossible (ii) is grounded in the apparent retreats from communism by Lenin and Stalin (205-8). Knight clearly suspects (867), however, that Lippmann's confidence is influenced not just by these historical developments in Russia, but also, and primarily, by Mises' contributions to the socialist calculation debate ,which Lippmann had discussed earlier in the Good Society and had praised (94ff; in fact, Lipmann's accompanying footnote treats Hayek and Weber as fellow contributors to this argument. (It is the only mention of Weber  and one of two mentions of Hayek in text of the The Good Society.) 

Knight uses the occasion of the review to lampoon Mises's argument ("wishful thinking"). The core point of Knight is that Mises argument is entirely too abstract, and that science lacks the expertise to make any claims about what is possible or not in reality: "No one knows at all definitely what is politically possible....There is practically no science in terms of which any fairly certain prediction can be made as to how a government will operate under any given conditions." (868) While I do not want to imply that Hayek read the review, what is striking about Hayek's great breakthrough in the social calculation debate in "the use of knowledge in society" is to turn away from questions of possibility, and, instead, to turn the attention to the nature of the "a process by which knowledge is constantly communicated and acquired" (Hayek 1945: 530).  [We might say the debate went from metaphysics/modality to epistemology.]

Knight, by contrast, thinks there is no reason to be sure that collectivism is impossible. So, in this respect he is much more pessimistic than Lippmann. And he is confident that full blown collectivism must devolve into dictatorship (868-869). So, Knight clearly holds a point of no-return version of the road-to-serfdom. However, he does not think that point has been reached yet or that reaching that point is especially likely.

So, here's my hypothesis. The road to serfdom thesis was if not inspired by Lippmann, at least prompted, in part, by him. But Lippmann did not hold the thesis; it is articulated by Knight in his review of Lippmann and (mistakenly) ascribed to Lippmann. Knight, however, thinks there is nothing inevitable about the thesis because he thinks the future is still very much open. I cannot prove that Hayek read Knight's review of Lippmann. (Knight was later a somewhat ambivalent referee for The University of Chicago Press of Road to Serfdom.) But Knight articulated several major challenges to liberal self-reflection in his review of Lippmann. And some of Hayek's major contributions to the liberal self-imagine, for good and ill, can be fruitfully interpreted as responses to that review. 

This Digression is somewhat long. But even so, I promised a comment on Knight's Weberianism. It can be discerned in the following passage:

It might have been thought self-evident, if anything in the field of politics can be called such, that the first essential function and task of government is to preserve unthreatened its own monopoly of political power, and that this means prevention of the development of dangerous power groups outside itself.

Knight has turned Weber's "monopoly of legitimate violence/force" (in Politics as a Vocation) into a "monopoly of power." The substitution may be thought innocent given that one may reasonably hold that power equals possible violence. But in Knight's hand the state does not tolerate what we may call ([recall]) following Bagg) concentrated power. And so one of the tasks of the state is to prevent the growth of concentrated power and doing so without devolving in fascism.