“Liberalism and / or Socialism?”
The Wrong Question.
NOTE: I originally wrote an earlier version of this text for a conference on the topic “Liberalism and / or Socialism: Tensions, Exchanges, and Convergences from the 19th Century to Today,” held at the University of Lorraine, October 21-23, 2021. I would like to thank the conference organizers and participants for their helpful comments. Any remaining errors, whether of commission or omission, are my own, of course.
Introduction
Political questions are typically framed in normative terms, in terms of the political actions that we (or our political representatives) “ought” to take or, alternatively, in terms of the political philosophies that “should” inform our political actions. “Should we be liberals or socialists, or should we (somehow) combine liberalism and socialism?” Indeed, political philosophy itself is typically conceived as a branch of moral philosophy, i.e., as that branch of philosophy that considers normative questions as they arise in specifically political contexts.
Notice, however, that the process of reasoning about political decisions is quite different from the one that typically operates in individual decision-making contexts. When we make decisions for ourselves, as individuals, we rarely, if ever, reason on normative grounds alone. We do not reason in terms of what we ought to do without prior consideration, albeit perhaps only un- or sub-conscious consideration, of what we know enough to do. Recent research in epistemology and psychology suggests that individual reasoning always proceeds from consideration (perhaps only un- or sub-conscious consideration) of what the decision-maker knows enough to do to consideration of what the decision-maker ought to do.
The question that motivates this research concerns the origins and determinants of individual decision-makers’ incentive structures. The preference rankings from which a decision-maker chooses always include only a limited range of courses of action, never every action, that could be chosen in the given decision context. What circumstances determine whether a course of action is included as an option and where it is ranked relative to other options in a decision-maker’s incentive structure? Scheall (2019) and Scheall and Crutchfield (2021a) offer several philosophical arguments for the logical priority of the epistemic, the thesis that the nature and extent of a decision-maker’s relevant ignorance regarding various courses of action determine which actions count as options, and where options are ranked relative to each other in the decision-maker’s incentive structure. Of course, non-epistemic considerations, such as the moral, prudential, or pecuniary value of an option relative to other options, also figure in decision-making. However, if the thesis of the logical priority of the epistemic is sound, then such normative considerations impinge on options only after (logically and temporally) the decision-maker’s relevant ignorance has first culled and sorted courses of action into a cognitively tractable preference ranking. Crutchfield, et al (2023) test the logical priority of the epistemic empirically and find the thesis supported by the experimental evidence. Similarly, the experiments of Kirfel and Lagnado (2021) and Kirfel and Phillips (2021, 2022) show that various judgments are affected by decision-makers’ epistemic circumstances.
The arguments and evidence from this literature suggest that we typically do not take on obligations or even treat as options courses of action that we do not know enough to do. Simply put, ignorance limits options and, therefore, obligations.
In political contexts, however, we tend to reason on the basis of purported obligations that, given the nature and extent of their own ignorance, may not even be options for the policymakers tasked with political decision-making. Crutchfield and Scheall (2019) apply the logical priority of the epistemic to the problem of the obligations of surrogate decision-makers, while Scheall (2019, 2020), and Crutchfield and Scheall (2021b) consider the obligations of policymakers, who constitute a particular class of surrogate decision-makers, in the light of the logical priority of the epistemic. Political questions are typically posed and debates around such questions emerge with little, if any, prior consideration of a question that, if there is anything to these epistemological and psychological developments, is, logically speaking, more fundamental: “What can we effectively achieve through political action? What goals fall within and what goals lie outside the scope of political action?” The reasoning process that operates in individual decision-making contexts – and, importantly, serves to protect the individual decision-maker against disastrous choices – is short-circuited in political environments, often to the detriment of the consequences of the decisions taken.[1]
Because we pose and argue about normative political questions without first getting the descriptive facts straight, we often embark on political projects that have little hope of success. More precisely, because we do not first clarify what policymakers know enough to do and what they are too ignorant to do, because we have little idea what political action can and cannot achieve on the basis of policymakers’ existing knowledge and learning capacities, we sometimes pursue goals through political action that cannot be realized unless spontaneous forces intervene to compensate for the consequences of policymaker ignorance (Scheall 2019, 2020). Such spontaneity is necessary to realize policy goals that policymakers are too ignorant to realize on their own.
Anyone who accepts a principle like ought implies can or any of its logically weaker variants, such as the principle that ought presupposes or is a conversational implication of can, is committed to rejecting “ought” claims that assert obligations to do things that cannot be done. The logic of such principles is that nothing that cannot be done can be an obligation.[2] Thus, anyone who accepts such a principle is committed to rejecting political debate that runs purely in normative terms without prior consideration of what can and cannot be done, given existing knowledge and relevant spontaneous forces. Since most, if not all, people accept some such principle, most, if not all, people are implicitly committed to rejecting the traditional – purely normative – form of political debate. That many people nevertheless engage in such debate reveals a significant inconsistency in how these people think about and assign obligations to policymakers.
If the question “Liberalism and / or Socialism?” should be interpreted as the normative question, “Should we be liberals or socialists, or should we (somehow) combine liberalism and socialism?” then it is the wrong – or, more exactly, a premature – question to ask. Before we ask whether we ought to be liberals or socialists, or (somehow) combine the two, we need to get clear on which, if any, of these political systems we can do.
The Epistemic Argument against Socialism (and Other Forms of Interventionist Economic Policymaking)
As is well-known, Ludwig von Mises ([1920] 1935; [1922] 2009) and F. A. Hayek ([1935a] 1997; [1935b] 1997; [1940] 1997; [1945] 2014) leveled formidable, perhaps even devastating, arguments against the epistemic feasibility of centrally-planned socialism. Any attempt to plan and control a modern industrialized economy from the center would fail to coordinate economic activity as well as the price system coordinates economic activity in market economies, because, simply put, the policymakers tasked with deliberately coordinating the economy from the center would inevitably lack some of the knowledge required of effective central coordination.
In Mises’ ([1920] 1935) original German-language socialist calculation argument, because market prices would cease to exist in an economy where the means of production had been fully socialized, the centralized administrators of a socialist system would necessarily lack knowledge of the profits and losses expected to follow from various production decisions. Without this knowledge, socialist administrators could never deliberately coordinate production with consumption as well as a system of freely-adjusting prices spontaneously equilibrates supply and demand in market economies. The effort to deliberately produce to satisfy consumer demand would inevitably lead to shortages of some goods and surpluses of others, which is to say, to economic disequilibrium.
The relevant terms shifted (multiple times) in the years between Mises’ original contribution and Hayek’s various forays into the debate. In particular, various responses and attempted solutions were put forward in reaction to Mises’ argument (see, e.g., Taylor 1929; Dickinson 1933, 1939; Lange and Taylor 1939). In replying to these arguments, Hayek extended and further generalized Mises’ argument (Scheall 2020). Whereas Mises had assumed a fully socialized economy isolated from all market-determined prices, Hayek showed that even if these assumptions were relaxed – even if, say, elements of market competition were introduced – central planning would nevertheless fail to achieve the performance of the price system as an equilibrating mechanism. Because policymakers could never know at any given time the various arrays of prices across the economy, or even within a single market, that would tend to promote rather than prevent equilibrium, efforts to deliberately plan the economy to this end would underwhelm relative to the results of unplanned markets.
Indeed, Hayek’s ([1931, 1935] 2012; [1933] 2012) famous business cycle theory (and its various elaborations) can be understood as a further extension and generalization of the same basic epistemic argument (Scheall 2015). In earlier versions of the theory, it is the knowledge of bankers – be they central or private bankers – that is crucial to avoiding or moderating effects of the cycle. Because bankers cannot know the array(s) of interest rates that will promote a tendency toward equilibrium between saving and investment, attempts to promote this tendency by manipulating the money supply or advancing public works, though they may appear to stimulate the economy in the short-run, will set in motion a series of events tending to promote disequilibrium in the long(er) run (Hayek [1933] 2012, 131). In later versions of the theory, particularly the sketch of the “epistemic theory of industrial fluctuations” (Scheall 2015) that Hayek ([1975] 2014) offered in his 1974 Nobel Prize lecture, furtively titled “The Pretence of Knowledge,” it is the knowledge – or, more exactly, the ignorance – of economic policymakers that is disequilibrating. The (Keynesian) toolkit of economic policymaking is inadequate to the epistemic requirements of effective countercyclical demand management. Efforts to apply Keynesian means to the countercyclical end, Hayek argued, tend to prevent rather than promote the tendency toward economic equilibrium. Even in an otherwise liberal context of market competition, for epistemic reasons concerning their ignorance of required causal knowledge, policymakers would fail to effectively promote equilibrium through deliberate policies aimed at that end. This goal requires the significant intervention of spontaneous forces, if it is to be realized, despite economic policymakers’ ignorance of much of the knowledge required to realize it deliberately.
The Epistemic Argument against Liberalism
Although Hayek extended Mises’ original socialist calculation argument beyond the narrow (and unrealistic) context of a fully socialized economy completely isolated from all market-determined prices and further generalized the argument by applying it to less socialistic, more liberal, policymaking contexts, he did not generalize the argument as far as it can be generalized. In particular, he did not consider the epistemic difficulties that policymakers confront in either liberalizing a relatively illiberal economy or maintaining the existing liberality of an already liberal economy. He seemed to assume, without argument or evidence, that knowledge problems of the sort that he and Mises had raised for socialists and Keynesians would emerge only in the context of activist policymaking, and that such problems were either negligible, or easily resolved, where the relevant policymaking goal was either further liberalization or maintenance of a given level of liberality.
However, I have recently argued elsewhere (Scheall 2020) that this latter assumption is groundless.[3]Policymaker ignorance can prevent liberalization or maintenance of a given level of liberality just as it can hinder central planning and Keynesian demand management. The same basic argument that Mises and Hayek advanced against would-be central planners and countercyclical demand managers can be leveled against the would-be liberalizing policymaker or defender of existing liberties.
All policies, even those meant to maintain the status quo, aim to realize some state(s) of affairs. It is always an open question – whether the intended state of affairs is a socialist paradise, a liberal utopia, or maintenance of the status quo – whether policymakers possess (or can acquire) the knowledge required to realize the intended state of affairs (Scheall 2020). Whatever the goal(s) of policy, it should never just be assumed a priori that policymakers are adequately epistemically equipped. Mises and Hayek pointed out that defenders of economic interventionism had made precisely this illegitimate a priori assumption concerning the adequacy of the epistemic capacities of central planners and Keynesian demand managers. However, Mises and Hayek made precisely the same illegitimate a priori assumption concerning the adequacy of the epistemic capacities of both liberalizing policymakers and protectors of prevailing liberties (Scheall 2020).
As is evident in the manifest failures of many efforts to “nation-build” liberal societies upon the remnants of comparatively illiberal ones, liberalization can proceed in better and worse ways, and there is no reason to assume a priori that policymakers know enough to choose the better and reject the worse ways. Similarly, as is evident in many past instances of the de-liberalization of relatively liberal societies, not to mention the apparent disintegration of liberal democracies currently in progress, maintaining a given level of liberality is not necessarily simple, and there is no reason to assume a priori that policymakers know enough to stem the de-liberalizing tide (Scheall 2020).
In short, except perhaps in the simplest and most banal contexts, policymaking is always epistemically burdensome. What’s more, it is always an empirical question, not one to be assumed away a priori, whether and to what extent policymakers’ existing knowledge and learning capacities are adequate to the realization of some policy goal (and, relatedly, whether and to what extent spontaneous forces must intervene if the goal is to be realized despite relevant policymaker ignorance). This is true, again, whether the goal is socialism, liberalism, or some combination thereof (Scheall 2020). That is, we can – and one might think, should – get some empirical purchase on the policy goals that policymaker knowledge makes possible and the goals that policymaker ignorance forecloses.
More to the present point, it might be wise to have this knowledge – i.e., knowledge of what deliberate policymaking can and cannot achieve – in hand before engaging in normative reflection on policies that ought to be pursued. After all, this is how reasoning proceeds in everyday life, in individual decision-making contexts, from (perhaps only un- or sub-conscious) consideration of what can be done to consideration of what ought to be done. Except in political contexts (more carefully, except in surrogate decision-making contexts), we typically do not reason or act on the basis of purported obligations without prior consideration of our knowledge, ignorance, and capacity for future learning. This reasoning process would seem to serve an obvious evolutionary function in promoting the survival of individuals who (and, therefore, of species whose members) engage in it. That reasoning in political decision-making contexts does not proceed in this fashion, that such reasoning skips the crucial step of determining what the relevant actors’, i.e., policymakers’, knowledge permits and what their ignorance forecloses (or, more exactly, given the possibility of spontaneous realization of a goal, what their ignorance makes more complicated), would seem to be an essential element of any explanation of why policymaking fails as often as it does and why constituents are so frequently dissatisfied with their political representatives. If we wish to avoid political disaster or even just political disappointment, a greater focus on policymaker ignorance and its consequences is warranted.
The Logical Priority of the Epistemic
According to the thesis of the logical priority of the epistemic, the nature and extent of our ignorance (that and how) with respect to various courses of action – our epistemic burdens – serve to determine our incentive structures (Scheall 2019; Scheall and Crutchfield 2021). Courses of action that seem to bear impossibly heavy epistemic burdens – which is to say, courses of action with regard to which a decision-maker is seemingly irremediably ignorant – typically do not appear in the decision-maker’s incentive structure. They are not even options for the decision-maker. More generally, as compared to options that seem less epistemically burdensome, courses of action that appear to bear relatively heavy epistemic burdens tend to be systematically discounted in a decision-maker’s incentive structure. In short, epistemic considerations are logically prior to other normative (i.e., moral, prudential, and pecuniary) considerations in human decision-making, at least, that is, outside of political (surrogate decision-making) contexts. One cannot have obligations without first doing some epistemic work—at least, again, outside of political contexts.
In a series of single- and co-authored papers, I have offered several arguments and pieces of evidence in support of this thesis (see, esp. Scheall 2019; Scheall 2020; Scheall and Crutchfield 2021; Crutchfield, et al 2023). First, there is a simple introspective argument: when we reflect on our own reasoning and decision-making processes it is clear that courses of action about which we are irremediably ignorant do not figure and that courses of action about which we are relatively ignorant figure less than those about which we are relatively knowledgeable. Second, there is a more complicated argument about what the word “can” must mean in principles like ought implies can and its logically weaker variants, if these principles are to be practically useful. The only meaning of “can” that makes such principles useful for the purposes of distinguishing one’s potential obligations from one’s non-obligations is deliberately can. But, to say that someone “deliberately can” do something is just to say that the person’s epistemic capacities are adequate with respect to it. If this is right, then obligations are determined (in the first instance, if not fully) by epistemic considerations (Scheall 2019; Scheall 2020; Scheall and Crutchfield 2021). A bit more carefully, as discussed in Footnote 2, unless there is some non-epistemic sense of the word “can,” the thesis of the logical priority of the epistemic follows (Crutchfield, et al 2023). Third, there is the obvious evolutionary advantage that members of species which reason in accordance with the logical priority of the epistemic possess over species the members of which obligate themselves without prior reflection on their relevant epistemic burdens. One is reminded of W. V. O. Quine’s (1969, p. 126) famous quip that “Creatures inveterately wrong in their inductions have a pathetic but praiseworthy tendency to die before reproducing their kind.” Something similar can be said of creatures who commit themselves to courses of action that are relatively more likely to fail for reasons concerning the creatures’ relevant ignorance. Creatures with a tendency to obligate themselves and each other without prior consideration of relevant epistemic burdens would also have a pathetic tendency to die before reproducing their kind. Fourth, several collaborators and I have conducted a series of psychological experiments that seem to support (or, more exactly, do not falsify) the thesis of the logical priority of the epistemic. These studies have shown, for example, that experimental subjects responses to the famous “trolley problem” depend on their knowledge and ignorance of relevant circumstances (Crutchfield, et al 2023).
So, when we decide as individuals, we reason from our knowledge to our obligations and not on the basis of our purported obligations alone. However, when we (or our political representatives) make decisions concerning collective action, we rarely, if ever, start from consideration of the limits that political ignorance places on such action. If we did, our political selves might be less like those pathetic creatures who, because they obligate themselves to pursue epistemically problematic, if not impossible, results, die out before reproducing their kind.
Hayekian Political Epistemology
There are at least two ways that the nature and extent of relevant policymaker ignorance might be scientifically analyzed and its real-world consequences, potentially at least, mitigated. These methods are not mutually exclusive and might be fruitfully combined. Both methods are based on the work of F. A. Hayek; together they constitute what I call “Hayekian political epistemology” (Scheall 2020).
The epistemic-mechanistic approach would investigate, both empirically and theoretically, the possibility of mechanisms for communicating relevant knowledge from constituents concerning their preferred political ends and the means for their realization to policymakers, as well as relevant knowledge from policymakers regarding the nature and extent of their knowledge to constituents (Scheall 2020). An effective epistemic mechanism would inform policymakers of their constituents’ policy demands and how to realize related policy goals; it would ensure that policymakers know what constituents want (or where their interests lie) and how to satisfy these demands (or how to promote these interests). At the same time, an effective epistemic mechanism would inform constituents of their policymakers’ epistemic limitations; it would serve to ensure that constituents do not demand policies the epistemic requirements of which exceed the epistemic capacities of policymakers. An effective epistemic mechanism would promote a tendency toward a kind of equilibrium (or, in Hayek’s [1976] terms, a kind of order), a balancing of the supply of epistemically-surmountable policies with the demand for them. The epistemic-mechanistic approach would serve to mitigate the problem of policymaker ignorance by improving the relevant knowledge of both policymakers and constituents so that members of both groups could make better, more actionable, plans of political action.
On the other hand, the constitutional approach to Hayekian political epistemology would take political knowledge as given and aim to mitigate the consequences of policymaker ignorance by limiting political action to policy ends with respect to which policymaker knowledge was adequate to make a positive contribution (and would leave the realization of all other goals to spontaneity). The constitutional approach would investigate the existing knowledge and learning capacities of policymakers and the knowledge required to realize various policy ends (Scheall 2020). More exactly, the constitutional approach would inquire into the theoretical and empirical knowledge required to realize some policy goal and compare these with the theories and data available to policymakers. The constitutional approach would, in other words, try to gain some empirical purchase on policymakers’ relevant epistemic burdens (and, thus, on the nature and extent of the spontaneous forces required to realize goals in the presence of relevant policymaker ignorance). We could then constitutionally proscribe political action to goals that policymakers’ epistemic capacities were adequate to help realize. We could never make policymakers truly omniscient and omnipotent, which would be the only full-fledged solution to the problem of policymaker ignorance, but we might make them functionally omniscient and omnipotent by limiting their range of political motion to goals they are knowledgeable enough to help realize (Scheall 2020).
The Implications of the Logical Priority of the Epistemic for the Question “Liberalism and / or Socialism?”
If we were to reason in political decision-making contexts the way that we reason in individual decision-making contexts, the results of (something like) Hayekian political epistemology would figure in our normative deliberations about the policies we ought to pursue. We would no longer engage in political debates without an understanding of the limits of what political action can achieve, as determined by the nature and extent of policymaker ignorance.
It is important to be careful here. I do not pretend that Hayekian political-epistemological analysis is likely to provide us with complete and infallible knowledge about policymaker knowledge. However, such analysis would provide us with a better understanding than we currently possess of the limits of effective political action and, thus, would place us in a better position than we currently occupy to avoid political disappointment.
It is also important to note what does not follow from the current argument. It is no part of the argument that policymakers are obligated to reject policies with regard to which they are relevantly ignorant. To draw this inference is to confuse the logic of the argument. If policymakers are irremediably ignorant with respect to policy P, then there can be no obligation for them to pursue P; however, this does not mean that an obligation to pursue not-P emerges in its place.
Similarly, it is no part of the argument that policymakers must always wait for the results of (potentially time-consuming) Hayekian political-epistemological analysis before taking a political decision. “Snap” political decisions must sometimes be made. This does not mean, however, that snap political decisions must be made in complete ignorance of the nature and extent, or of the effects likely to follow from a snap political decision made in the presence, of relevant policymaker ignorance. In other words, just because snap political decisions must sometimes be made is no reason to delude ourselves about the wisdom upon which they are based or the consequences likely to follow from epistemically-inadequate snap political decisions.
If we were to reason in political decision-making contexts in accordance with the logical priority of the epistemic, we would not consider the “Liberalism and / or Socialism?” question without something like the results of a Hayekian political-epistemological analysis of the prospects for liberalism, socialism, and for political systems in between, given policymakers’ knowledge and learning capacities. To ask this question in the absence of such an analysis is to place the normative cart before the epistemic horse that in fact drives decision-making (Scheall 2019). It is to ask a question without all of the data in hand required to answer it rationally. It is to ask the question prematurely.
References
Besch, Thomas. 2011. Factualism, normativism and the bounds of normativity. Dialogue 50 (2): 347-365.
Cooper, Neil. 1966. Some presuppositions of moral judgments. Mind 75 (297): 45-57.
Crutchfield, P. and S. Scheall. 2019. Epistemic burdens and the incentives of surrogate decision-makers. Medicine, Health Care, and Philosophy 22: 613–621.
Crutchfield, P., S. Scheall, M. Rzeszutek, C. Cardoso Sao Mateus, amd H. Brown. 2023. Ignorance and moral judgment: Testing the logical priority of the epistemic. Consciousness and Cognition 108 (published online 30 January 2023).
Dickinson, Henry. 1933. Price formation in a socialist community. Economic Journal 43 (170): 237–250.
Dickinson, Henry. 1939. Economics of socialism. Oxford: Oxford University Press.
Driver, Julia. 2011. Promising too much. In Promises and agreements, ed. Hanoch Scheinman, 183-197. Oxford: Oxford University Press.
Hampshire, Stuart. 1951. Symposium: Freedom of the will. Aristotelian Society Supplementary Volume 25, 161-178.
Hare, Richard. 1951. Symposium: Freedom of the will. Aristotelian Society Supplementary Volume 25: 201-216.
Hare, Richard. 1963. Freedom and reason. Oxford: Clarendon Press.
Hayek, F. A. 1976. Law, Legislation and Liberty, Volume 2: The Mirage of Social Justice. Chicago: University of Chicago Press.
Hayek, F. A. (1935a) 1997. The nature and history of the problem. In The collected works of F. A. Hayek, volume 10: socialism and war, ed. Bruce Caldwell, 53–78. Chicago: University of Chicago Press.
Hayek, F. A. (1935b) 1997. The present state of the debate. In The collected works of F. A. Hayek, volume 10: socialism and war, ed. Bruce Caldwell, 79–116. Chicago: University of Chicago Press.
Hayek, F. A. (1940) 1997. Socialist calculation: The competitive ‘solution’.” In The collected works of F. A. Hayek, volume 10: socialism and war, ed. Bruce Caldwell, 117–140. Chicago: University of Chicago Press.
Hayek, F. A. (1931, 1935) 2012. Prices and production. In The collected works of F. A. Hayek, volume 7, business cycles, part I, ed. Hansjörg Klausinger, 167–283. Chicago: University of Chicago Press.
Hayek, F. A. (1933) 2012. Monetary theory and the trade cycle. In The collected works of F. A. Hayek, volume 7, business cycles, Part I, ed. Hansjörg Klausinger, 47–165. Chicago: University of Chicago Press.
Hayek, F. A. (1945) 2014. “The use of knowledge in society.” In The collected works of F. A. Hayek, volume 15, the market and other orders, ed. Bruce Caldwell, 93–104. Chicago: University of Chicago Press.
Hayek, F. A. (1975) 2014. “The pretence of knowledge.” In The collected works of F. A. Hayek, volume 15, the market and other orders, ed. Bruce Caldwell, 362–372. Chicago: University of Chicago Press.
Kirfel, Lara and David Lagnado. 2021. Causation by ignorance. Proceedings of the Annual Meeting of the Cognitive Science Society 43: 966-972.
Kirfel, Lara and Jonathan Phillips. 2021. The impact of ignorance beyond causation: an experimental meta-analysis. Proceedings of the Annual Meeting of the Cognitive Science Society 43: 1595-1601.
Kirfel, Lara and Jonathan Phillips. 2022. The pervasive impact of ignorance. https://phillab.host.dartmouth.edu/downloads/Pervasive_Ignorance_FullPaper.pdf. Accessed 20 November 2022.
Lange, Oskar and Fred Taylor. 1938. On the economic theory of socialism, ed. B. E. Lipincott. Minneapolis: University of Minnesota Press.
Littlejohn, Clayton. 2009. ‘Ought’, ‘can’, and practical reasons. American Philosophical Quarterly 46 (4): 363-373.
Martin, Wayne. 2009. Ought but cannot. Proceedings of the Aristotelian Society 109 (2): 103-128.
Mises, Ludwig von. (1920) 1935. Economic calculation in the socialist commonwealth. In Collectivist economic planning, ed. F. A. Hayek, 87-130. London: Routledge.
Mises, Ludwig von. (1922) 2009. Socialism: an economic and sociological analysis. Auburn, AL: Ludwig von Mises Institute.
Oppenheim, Felix. 1987. National interest, rationality, and morality. Political theory 15 (3):369-389.
Quine, W. V. O. 1969. Natural kinds. In Ontological relativity and other essays, eds. Jaegwon Kim and Ernest Sosa, 114-138. New York: Columbia University Press.
Saka, Paul. 2000. Ought does not imply can. American Philosophical Quarterly 37 (2): 93-105.
Scheall, Scott. 2015. Hayek’s epistemic theory of industrial fluctuations. History of Economic Ideas 23 (1): 101–122.
Scheall, Scott. 2019. Ignorance and the incentive structure confronting policymakers. Cosmos + Taxis: Studies in Emergent Order and Organization 7 (1–2): 39–51.
Scheall, Scott. 2020. F. A. Hayek and the epistemology of politics: the curious task of economics. London: Routledge.
Scheall, Scott and Parker Crutchfield. 2021a. The priority of the epistemic. Episteme, 18 (4): 726-737.
Scheall, Scott and Parker Crutchfield. 2021b. A case study in the problem of policymaker ignorance: political responses to COVID-19. Cosmos + Taxis: Studies in Emergent Order and Organization 9 (5-6): 18-28.
Sinnott-Armstrong, Walter. 1984. ‘Ought’ conversationally implies ‘can’. The Philosophical Review 93: 249-261.
Taylor, Fred. 1929. The guidance of production in a socialist state. American Economic Review 19: 1–8.
Vallentyne, Peter. 1989. Two types of moral dilemmas. Erkenntnis 30: 301-318.
Vogelstein, Eric. 2012. Subjective reasons. Ethical Theory and Moral Practice 15: 239-257.
[1] Technically, such “short-circuiting” occurs in all surrogate decision-making contexts, where decision-makers decide on behalf (and, ostensibly, in the interests) of some other person(s) (Crutchfield and Scheall 2019). Policymakers are surrogate decision-makers wherever their decisions are expected to promote the interests of constituents. Given that, in modern political contexts, even in autocratic dictatorships, it is more or less universally assumed that policymakers (qua policymakers) should act to promote their constituents’ interests, policymakers (qua policymakers) are more or less universally surrogate decision-makers.
[2] This statement requires some qualification. As might be expected, there is considerable debate in the philosophical literature concerning the exact nature of the relationship that obtains between ought and can. If ought implies can, then an especially logically strong relationship – logical implication – obtains between the two terms. If ought implies can, then nothing that a person cannot do is among their potential obligations. Anyone committed to ought implies can is thus, as indicated in the text, committed to limiting a person’s potential obligations to things they can do. However, some philosophers argue that the relationship between ought and can is logically weaker than implication. Some argue that ought (merely) presupposes can (Hampshire 1951; Hare 1951, 1963; Cooper 1966; Martin 2009; Besch 2011; Driver 2011), in which case a person’s potential obligations presuppose only things that the person can do; a course of action that presupposes something that the person cannot do cannot be among the person’s potential obligations. Other philosophers argue that ought is a conversational implication of can (Sinnott-Armstrong 1984; Oppenheim 1987; Vallentyne 1989; Saka 2000; Littlejohn 2009; and Vogelstein 2012). If this is right, then the relation between ought and can is a matter of conversational pragmatics rather than of pure logic. A potential obligation cannot be interpreted, according to prevailing conversational conventions, to imply something that the actor cannot do; a course of action is among a person’s potential obligations if, given prevailing conversational standards, it is commonly construed to be among the things the person can do. There are other, even logically weaker, candidates for the relationship betweenought and can. For instance, perhaps ought (merely) makes plausible can, in which case a course of action is among a person’s obligations if (and to the extent that) it is plausible that the person can perform the action. If this is the case, then the set of a person’s potential obligations is “fuzzy”: whether a course of action is included in the set of a person’s potential obligations is not a binary matter, but a function of the degree of its plausibility (Scheall and Crutchfield 2021).
Nothing in the argument of the present paper hinges on a particular relationship obtaining between ought and can. In particular, the argument does not require that an especially strong relationship obtain between ought and can. This is because the meaning of the word “can” is inherently epistemic. Unless there is a non-epistemic sense of the word “can,” i.e., a sense that does not make the things a person can do at all dependent on their knowledge, which is implausible, then any of the popular candidates for the relationship between ought and can entails the thesis of the logical priority of the epistemic (Crutchfield, et al 2023).
[3] Policymaking might be relatively epistemically simpler in more liberal environments, but it does not follow that realizing and sustaining such environments must also be epistemically simple (Scheall 2020, 88).