AUSTRIAN ECONOMICS AND COMPATIBILIST FREEDOM

The present paper probes the relation between the metaphysics of human freedom and the Rothbardian branch of Austrian economics. It transpires that Rothbard and his followers embrace metaphysical libertarianism, which holds that free will is incompatible with determinism and that the thesis of determinism is false as pertaining to human action. However, as we demonstrate, their economics with its reliance on value scales requires for its tenability compatibilist freedom. Moreover, we attempt to show that the notion of value scales (or preferences) postulated by them implies that value scales are determinative of choices people make. We contend that it is for this reason that the said Austrians should jettison their metaphysical libertarianism.


Notes
Under metaphysical libertarianism, for agent S to perform an act x freely implies that at the very 1.
least the agent could have refrained from performing x, everything else equal. In other words, just to resort to the possible worlds semantics, according to metaphysical libertarians, an act x is performed freely by the agent S at time t only if there is a possible world where everything is the same up until t and S does something other than x at t.
Of course, 'can' in this context must be given the incompatibilist reading, that is the ability to do 2.
otherwise, everything else equal, which contrasts with the so-called motivational 'can', with the latter denoting the ability to do otherwise if only one had chosen otherwise. On a classic analysis of different senses of 'can', see: Austin (1961). On the ought-implies-can principle, see: e.g. Otsuka (1998); Kramer (2006); Graham (2011).
Most certainly, compatibilism about moral responsibility and determinism is not a novel view 3.
but it traces back to ancient times. Much more recently, but still before Frankfurt, Strawson (1962) powerfully argued that our responsibility practices undercut the whole debate between libertarian free will and determinism. But if so, there is no obstacle to have meaningful responsibility practices under determinism. This Strawsonian view with a Humean twist was brilliantly defended by Russell (1995). Mele (1995) purports to offer an account of moral responsibility that would satisfy both compatibilists and incompatibilists. More recently, compatibilism about moral responsibility and determinism was argued for by e.g. Bok (1998), Fischer and Ravizza (1998), McKenna (1998), Beebee and Mele (2002), Dennett (2003), Fischer (2006), Mele (2006), Moore (2020). However, as mentioned above, compatibilism about moral responsibility and determinism is still open to criticism. For a dissenting (i.e. incompatibilist) view, see: e.g. van Inwagen (1983), Kane (1996), Copp (1997), O'Connor (2000), Pereboom (2005). Incidentally, it should be borne in mind that although Mele's works cited herein defend the One of the reviewers drew our attention to the possibility that, after all, 'choice' might be just "a 5.
technical term not intended to involve metaphysical libertarian freedom (MLF) in the first place". As a next step, the reviewer invites us to "onsider the case of an agent who maximizes her utility", adding that "er 'choices' do not seem free in the relevant MLF sense". Not wanting to precipitate things at this expository stage of our paper, we just want to mention that we are going to argue in subsequent parts of the paper that the Austrian idea of choice precisely necessitates the compatibilist rather than libertarian metaphysics of free will. However, to make the strongest charge out of the reviewer's remark, we should probably take it to imply that, technically speaking, choice can be understood in abstraction from any metaphysics of its freedom, which points to a sort instrumentalist reading of the concept of choice. Granted, this is a feasible (methodological) position. However, Austrians adhere to so-called causal realism, which seeks to identify causal laws underlying market phenomena. The Austrian causal-realist approach was commenced by Menger himself ( 2007). In the words of Salerno (2010, 3): "He concluded that all product prices, rents, wage rates, and interest rates were interrelated and were the systematic outcome of the value judgments of individual consumers who chose between concrete units of different goods according to their subjective values or "marginal utilities"". That is to say, according to Menger, it is real human valuations that causally explain the emergent market phenomena. Given this, merely instrumentalist reading of choice would not do for Austrians. For more on causal-realist approach, see Salerno (2007).
Note that if we indeed succeeded in demonstrating that Austrian economics requires the 6.
adoption of compatibilist metaphysics of freedom, we would, as an interesting corollary, make a considerable step towards the unification of sciences. For then, we would not only employ the univocal sort of freedom (i.e. compatibilist one) both in moral philosophy and in economics but we would also thus render the said two disciplines coherent with a scientific deterministic world-view.
This argument is actually strengthened by Rothbard (2011b, 6) in a later passage wherein he 7.
claims that he proves the existence of MLF indirectly, that is by assuming the truth of determinism arguendo, which apparently leads to a contradiction: "If we are determined in the ideas we accept, then X, the determinist, is determined to believe in determinism, while Y, the believer in free will, is also determined to believe in his own doctrine. Since man's mind is, according to determinism, not free to think and come to conclusions about reality, it is absurd for X to try to convince Y or anyone else of the truth of determinism. In short, the determinist must rely, for the spread of his ideas, on the nondetermined, free-will choices of others, on their free will to adopt or reject ideas (footnote deleted). In the same way, the various brands of determinists-behaviorists, positivists, Marxists, and so on-implicitly claim special exemption for themselves from their own determined systems (footnote deleted). But if a man cannot affirm a proposition without employing its negation, he is not only caught in an inextricable self-contradiction; he is conceding to the negation the status of an axiom." At this point we would like to offer a suggestion to the effect that Rothbard's commitment to 8.
MLF is to some extent an outcome of his aprioristic style of reasoning. It is noteworthy that in his Mantle of Science, Rothbard almost exclusively argues for MLF by resorting to a priori reasoning. Most notably, as observed in the footnote above, Rothbard posits that an attempt to convince others of the truth of determinism enmeshes the determinist in a performative contradiction. For, it is the act of attempting to convince others that apparently presupposes their free will (in the libertarian sense). Hence, allegedly, what the determinist preaches is at odds with what her preaching presupposes (i.e. the audience's ability to freely (in the MLF sense) "adopt or reject ideas". Yet, as we claim, the act of trying to convince others of some beliefs is reconcilable with compatibilist metaphysics. After all, to try to convince one's audience of a certain claim is to give them a reason to believe it. Presumably, it should not matter whether one's interlocutor was ultimately determined to appreciate the strength of the reason provided. However, what does matter is that the interlocutor changes her mind via reasons. Moreover, we contend that the Rothbardian belief of the interlocutor's being free "to adopt or reject ideas" can be best explained in terms of uncertainty (or the present ignorance of future facts) rather than determinism. That is, the convincing party does not know whether her interlocutor is going to be effectively convinced or not. If the former somehow knew that the latter is going to reject determinism, she would never engage in the act of persuasion in the first place. Therefore, if anything, it is the persuading party's ignorance of an outcome rather than the audience's MLF that the act of persuasion presupposes. Rothbard (2011b, 7) makes a similarly conceptual point when professing that "if our ideas are determined, then we have no way of freely revising our judgments and of learning truth". It is now the concept of learning that seemingly presupposes MLF on the part of learners. But suppose person P learned the truth of 2 + 2 = 4 and P was determined to accept (and not reject) the soundness of this proposition. Would not we be prone to saying that P learned that 2 + 2 = 4 even though he was not free (in MLF sense) to reject this proposition? Certainly, Rothbard might argue that this act of P could not amount to learning since the presupposition (i.e. P's MLF) was not met. Fair enough, by stipulation, P's coming to accept the truth of 2 + 2 = 4 would not be the act of learning in the Rothbardian sense but it would clearly be one in the ordinary sense. We can reconstruct another Rothbardian argument for MLF along the following lines. Says our author: "On the formal fact that man uses means to attain ends we ground the science of praxeology, or economics; psychology is the study of how and why man chooses the contents of his ends", while adding that "f men are like stones, if they are not purposive beings and do not strive for ends, then there is no economics, no psychology" (Rothbard 2011b, 4). Apparently, the Rothbardian reasoning assumes the form of modus tollens. That is, if human agents are determined, then the disciplines of praxeology, economics or psychology are impossible. But the said disciplines are possible (or even impossible to deny without running into a performative contradiction). Therefore, human agents are not determined, which was to be demonstrated. However, what we must take heed of is that this argument is question-begging. That is, since, by Rothbardian lights, the existence of praxeology or economics presupposes MLF, how does he know that praxeology is applicable in the first place. After all, whether human agents are endowed with MLF or not is a point at issue. Hence, to reason from the existence of praxeology to the existence of MLF is to beg the question. For, according to Rothbard, praxeology is possible (or is applicable) only when determinism is ruled out but whether determinism is true or not is something yet to be established. And the presumed fact that the possibility of praxeology requires MLF cannot in and of itself establish the truth of MLF, for when MLF is in doubt, so is praxeology. The same criticism applies to Rothbard's a priori arguments from the concepts of persuasion or of learning, which seemingly serve to establish the truth of MLF. Even if the concept of learning and persuading were (contrary to fact, as we believe) to imply MLF, we could not non-question-beggingly infer MLF from the said concepts. After all, if we do not know just yet whether MLF holds in the actual world, we cannot know whether the concepts of learning and persuading are ever exemplified. Therefore, the Rothbardian a priori conceptual reasoning seems to get it backwards. Now a word of caution is advisable at this point regarding classifying Rothbard's reasoning as aprioristic. As rightly pointed out to us by an anonymous referee of this journal, Rothbard presents his position as an alternative to the alleged Kantian apriorism of Mises and thinks that the truth of the fundamental axioms of his system can be assessed by observation, especially by introspection. However, at the same time Rothbard (2011, 108-109) claims that "this type of 'empiricism' is so out of step with modern empiricism that I may just as well continue to call it a priori for present purposes. For (1) it is a law of reality that is not conceivably falsifiable." And elsewhere (Rothbard 2011b, 6) he contends that "if a man cannot affirm a proposition without employing its negation, he is not only caught in an inextricable self-contradiction; he is conceding to the negation the status of an axiom." It is therefore important to distinguish between, on the one hand, Kantian apriorism-sometimes problematically ascribed to Mises and his understanding of the action axiom-which claims, in a nutshell, that Formen der Anschauung (space and time) and categories of Verstand are logical conditions of possibility of any experience (even an introspective one) and knowledge (even an unfalsifiable one) whatsoever and that they must be assumed in order for the latter to even occur and, on the other hand, Aristotelian-Thomistic apriorism-subscribed to by Rothbard-according to which the first principles of theoretical and practical knowledge, although grasped by our nous or intellectus by way of insight into the previous experience, are aprioristic in the sense of being per se nota, self-evident or underived from any prior proposition and thus capable of being justified only via dialectical arguments demonstrating that any negation of these principles presupposes their truth. Rothbard's reasoning is aprioristic only in this second, less radical, sense of the word. On the relation between these two kinds of apriorism as well as on each of them in separation see, inter alia, Veatch (1965, 239-263), Timmons (1997, 1-13), Copleston (1955), Finnis (1998, 86-90).
However, there are Austrians who deny that man never chooses under indifference. See: 9.
To use the parlance of mainstream economics, we shall henceforth refer to the Rothbardian 10.
view on the relation between choices and preference as the evidential view. Just to reiterate slightly, on this view, value scales (or preferences) exist prior to actual choices and, epistemologically speaking, we can infer the former from the latter, whereas, genetically speaking, it is the former that guide the latter. For an excellent elaboration of the evidential view, see: Hausmann (2012, 88-103).
the spirit of Austrian economics with its commitment to the idea that people's actions trace (are indicative of) their respective most valued ends. Jeffrey (1983, 225) captures the idea of this sort of evaluations very sharply indeed: "I am concerned with preference all things considered, so that one can prefer buying a Datsun to buying a Porsche even though one prefers the Porsche qua fast (e.g., since one prefers the Datsun qua cheap, and takes that desideratum to outweigh speed under the circumstances). Pref = preference tout court = preference on the balance".
That is to say, it may well be true that were S to be a single man, he would buy white shirts 12.
instead of blue ones.
However, there are dissenting views within the Austrian camp having it that it is the relation of 14.
weak preference that is fundamental to action, with strict preference being only derivative (see: e.g. Machaj 2007;O'Neill 2010). That is, subject S strictly prefers x to y iff he weakly prefers x to y and he does not weakly prefer y to x. Mind you, if the actor weakly prefers x to y, it is by definition possible that he is indifferent between x to y. And if it is weak preference that governs actions, it is conceptually possible for man to act (or choose) under indifference.
We picked up Rothbard's exposition for we believe it is probably the most pellucid. However, 15.
as mentioned above, the view that man acts only on strict preference is the prevailing view among Austrians. Its most prominent proponents also include Mises ( 1998); Block (1980;1999;) Hoppe (2005.
As noted by one of the referees, there are other possible value scales that S's behaviour in W* 16.
may be evidentiary of. Indeed, from the fact that S spends his first hour on walking and his subsequent hour on resting we cannot infer with apodictic certainty how S framed his choice. It might as well be the case that the value scale underlying S's behaviour was the following: (1) first walking and then resting; (2) first resting and then walking; (3) doing a salto. Still, various ways in which S can frame his choice have interesting ramifications. First, given the framing envisaged by the referee, options 2 and 3 on the proposed value scale are not demonstrated, whereas on the grounds of the value scale we suggested in the body of the text, options 1 and 2 are demonstrated and only option 3 remains undemonstrated. The reviewer's framing, by contrast, is agnostic about the content of option 2 and 3 as there is no behaviour on the part of S that would satisfy either preference (desire) 2 or preference (desire) 3. Second, from our framing it follows that S preferred walking to resting, which, in turn, does not follow from the framing proposed by the referee, for if S had had only one hour to economize, he might have preferred resting to walking even though he in fact preferred first walking and resting to resting and then walking. Finally, our way of framing choices resembles Rothbard's simply because we mainly discuss this author's position. In this regard, it seems preferable to alternative framings.
At least in the ex ante sense. In Austrian economics, the actor's actual choice, as compared to 17.
other possible choices, benefits him most in expectation. This is, of course, not to deny that it is only contingently true that it also benefits him most ex post. The human actor is fallible: he may misjudge probabilities, fail to appreciate relevant causal relations or simply may not be imaginative or knowledgeable enough to envisage alternative courses of action. Given all this, it might as well be the case that he would have been better off ex post had he chosen otherwise.
It is important to note yet again that, strictly speaking, preferences alone do not determine 18.
choices. Contemporary philosophy of action has it that preferences (or desires) can issue in an action only when coupled with beliefs, as maintained by e.g. Goldman (1970) or Davidson (1980, 3-19). Some other philosophers (see e.g. Moore 1993, 113-165) hold that we need to postulate some more immediate causes, such as, say, volitions, that eventually bring about actions. And it is common sense: if a given actor's highest valued end is to marry a particular lady but he deems this end unattainable, he would not be acting on this unsatisfied preference. As Mises ( 1998, 14) put it: "But to make a man act, uneasiness and the image of a more satisfactory state alone are not sufficient. A third condition is required: the expectation that purposeful behavior has the power to remove or at least to alleviate the felt uneasiness. In the absence of this condition no action is feasible". Therefore, it should be borne in mind that the sentence 'preferences determine choices' is not literally true for unsatisfied preferences alone are insufficient for an action to be taken. Accompanying beliefs of the proper content are necessary. Technically speaking, if one most wants the state of affair S to obtain, one will do nothing to bring it about unless one also believes that he can employ some means M that will make S obtain. It is only then that one starts employing M, which is another way (however convoluted) way of saying that one starts acting.
A very interesting debate concerning the stability of preferences ensued between Block and 19.
Barnett (2012), who explicitly reject the idea, and Hudik (2012). Slightly ironically, the former authors attempted to debunk the alleged rational requirement of transitive preferences as their main target. However, as demonstrated by Hudik (2012, 459-460), if the assumption of transitivity of preferences is dropped, then the observer cannot distinguish between the agent's having unstable or intransitive preferences. Consequently, if Block and Barnett want to remain unyielding about the apparent instability of preferences as explanatory of "all observed choice inconsistency", they should rather embrace their transitivity rather than reject it. Hudik (2012, 460-461) argues that when it comes to "market-level phenomena" such as an increased demand for a certain commodity or service, preference changes constitute sort of last-resort explanations. Instead, he submits that such facts might be better explained by an increase in real income or by a lower price of the said commodity or service. Given this, the assumption of stable preferences appears to be quite reasonable.
We can easily generalize the above point. To avoid explanatory regressus ad infinitum, Austrian 20. must assume that there exists such an n-order preference that determines a choice of a particular n-1-order value scale, which in turn determines a choice of a particular n-2-order value scale etc. until we reach the level of a first-order value scale, which directly determines an action towards the satisfaction of a particular highest valued material end.
After all, we seriously take the libertarian claim that literally everything else is equal in W* but S 21.
acts in W* in a different way from the one in which he acts in W.
The problem of luck primarily arises in moral philosophy. Indeed, there are thinkers who 22.
believe that luck rules out responsibility in the basic-desert sense completely (e.g. Levy 2011; Caruso 2019), the position labelled hard luck. Mele (2006), on the other hand, offers probably the most comprehensive review of the problem to be found. For the classical exposition of luck and its taxonomy, see: Nagel (1979).
An anonymous reviewer entertained a very interesting possibility of higher-order preferences 23.
being "the outcome of an indeterminate process that is not a choice process but still has some stable characteristics, so that it is not pure luck." The thought is that although first-order value scales would function deterministically (i.e. given the actor's beliefs and the first-order value scale at a certain time, he or she would inevitably make a definite choice), the actor's higherorder preferences (or value scales) would be governed only by propensities and would thus be indeterministic. And if so, then in the end actual choices made by economic agents would not be fully determined. When taken seriously (i.e. on the realist reading thereof), propensities are such dispositions of an object that account for its differential behaviour (i.e. different relative frequencies of certain outcomes). For instance, if a coin has a propensity 0.7 for landing heads uppermost, this very property is responsible for the coin landing (more or less) 70% of times heads uppermost in the long run, as opposed to landing (more or less) 30% of times tails uppermost, with everything else being equal (for an elaboration of the propensity interpretation of probability statements see: Mackie (1973, 179-187)). Now, by analogy, if our actor has a 80% propensity for Value Scale 1 and 20% propensity for Value Scale 2 , this ipso facto predicts that in the long run he or she will be guided by V 1 four times as often as by V 2 . However, if so, then at least in the long run, the actor is determined to be guided by the ratio between these two value scales. To wit, if the actor's propensity under consideration is something which generates the stipulated ratio (i.e. 4:1) of first-order value scales, then the actor is not free to choose to be guided by the said two value scales at a different ratio. To conclude, propensity argument seems to salvage a hint of indeterminism, while still facing determinism in the long run. However, rescuing indeterminism in this manner still runs into our luck challenge. To illustrate the point, let us stick to our previous example of the actor being guided by V 1 and V 2 at the ratio of 4:1. Suppose further that in the actual world (W 1 ), the actor is guided over time by the following series of the two value scales: {V 1 , V 1 , V 1 , V 1 , V 2 , …}. Moreover, since the actor is presumed to be endowed with the above-defined propensity, there is a merely possible world W 2 , in which the same actor is guided over time by the following series of the two value scales: {V 1 , V 1 , V 2 , V 1 , V 1 , …}. Clearly then, the 4:1 ratio alone allows for some variability within series.
Specifically, in the envisaged worlds, the two value scales are distributed differently over time. But then again, since everything is the same about W 1 and W 2 , the differential distributions of V 1 and V 2 cannot be explained by the propensity (or anything else, for that matter). Rather, they are lucky. However, according to an anonymous referee, one worry still remains, for we do not conclusively rule out a possibility of "a creative construction of a value scale" in the first place since our example involving propensities still assumes pre-existing value scales and a choice between them. By contrast, the referee invites us to imagine "free creation of value scales". To address this objection, we offer the following two points. First, it seems to us that the most charitable reading of the referee's suggestion is to take it to be an allusion to contra-causal freedom in the form of agent causation (for the major problem haunting agent causation see footnote 37). For, in the absence of value scales of a higher order to choose from, it appears as if the only option left open is that it is the agent herself (as a cause) that "freely" creates a firstorder value scale which, in turn, guides her actions. However, as already mentioned, agent causation-as opposed to event causation-does not fare well among the contemporary views on the metaphysics of free will. Second, we believe that since the scenario suggested by the referee involves no choice, it is automatically beyond the remit of Austrian economics. After all, Mises ( 1988, 3) viewed "modern subjectivist economics" as "a general theory of human choice." Thus, praxeology aims at illuminating human choice. And hence, Mises would be prone to regarding "free creation of value scales" not involving a choice behaviour as an ultimate given, something that praxeology can only take for granted but cannot be a theory of.
One of the reviewers pointed out that this value scale does not reflect the intended 24.
indifference since apparently every entry on a value scale contains several options which the agent is indifferent between and not only those entries with an "or". The reviewer indeed made a valid point and so we take no issue with it. Granted, each and every option on a value scale might be rendered in such a way as to include those aspects of an action that the agent is indifferent between. For instance, our option 2 in v 4 might be rendered in the following way: 2.
Playing bridge indoors or outdoors. For the Actor in question might as well be indifferent between various circumstantial aspects of the game. The reason why we do not describe options in this manner is that we explicitly mark indifference (via "or") only when it is vital for our argument. Therefore, as it stands, v 4 signals that it is both continuing to watch the baseball game and going for a drive that S would strictly prefer to playing bridge, with the question of whether there are some ways of playing it or some circumstantial elements of the game that S is indifferent between being left open Again, see footnote 14. 25.
action. Quite the contrary. Every action necessarily signifies a choice, and every choice signifies a definite preference. Action specifically implies the contrary of indifference. The indifference concept is a particularly unfortunate example of the psychologizing error . If a person is really indifferent between two alternatives, then he cannot and will not choose between them." As already observed, Austrians are barred from saying that S could have chosen to play chess, 27.
everything else equal for-at the very least-S's underlying preferences must have changed too.
After all, he would be ex hypothesi indifferent between indefinitely many options, which seems 28.
to be the only way to grant to him the possibility of behaving differentially across worlds, everything else equal.
It is indeed puzzling that MLF-oriented Austrians would ever want to have this variety of will. 29.
After all, would they really want the sort of control (or the lack thereof) whereby their respective differential cross-world behaviors (e.g. whether they play cards in W and go swimming in W*) is just a matter of luck and not of their preferences or anything else for that matter? Upon reflection, this sort of control looks like a travesty of control.
It is worth noting in passing that it was also Menger that was most probably a determinist too. 30.
Austrian Economics and Compatibilist Freedom https://science-direct.net/austrian-economics-and-compatibilist-freedom/ Published by ElSevier BV Copyright © 2022 (PDF Created by Saudi MEP) Page: 11 Consider the following citation: "All things are subject to the law of cause and effect. This great principle knows no exception, and we would search in vain in the realm of experience for an example to the contrary." (Menger 2007, 51) A very persuasive construal of Mises' philosophy along determinist (compatibilist) lines can be 31.
The idea is that the truth of determinism does not entail perfect knowledge. Especially, and 32.
crucially, the former does not in and of itself entail present knowledge of future actions. And since the relation of entailment between the two does not hold, it is conceivable to have (radical) uncertainty coupled with determinism. That is, even if all the events were to be causally determined, this very fact would still not guarantee that future events (esp. actions) can be knowable now. Actually, Morgenstern (1928, 96) advanced a stronger thesis, while trying to demonstrate that predictions in social sciences are necessarily flawed. For, if a social scientist is to publicly announce her prediction, the prediction will fail once it affects the information set on which economic agents act. This finding only strengthens our point that metaphysical determinism does not immediately translate into epistemic determinism. For, if epistemic determinism is ruled out on independent grounds, then the possibility of ending up with metaphysical determinism and epistemic indeterminism is all the more viable. This is explicitly acknowledged by Gray ( 1984, 9), who has it that "Hayek's 'compatibilist 33. standpoint in respect of freedom of the will-his belief that the causal determination of human actions is fully compatible with ascribing responsibility to human agents for what they do-is analogous with his stance on the mind-body question." For a classic exposition of the way Hayek tackles the mind-body problem, see Hayek (1952). In fact, Hayek's skepticism as to the possibility of the mind explaining itself constitutes yet another reason why he may be validly described as a metaphysical determinist (compatibilist) and an epistemological indeterminist. Again, as Gray ( 1984, 9) puts it, "Hayek is concerned to deny any ultimate dualism in metaphysics or ontology, while at the same time insisting that a dualism in our practical thought and in scientific method is unavoidable for us." Hayek's compatibilist stance falls into what is recently labelled "cheap compatibilism", the idea 34.
that punitive measures as justified by forward-looking deterrence-related utilitarian considerations are compatible with determinism. What is "cheap" about this sort of compatibilism is that responsibility assigned is not desert-based. (Moore 2020, 208) That is to say, on the grounds of cheap compatibilism, responsibility assignments do not reflect the offender's moral desert and is thus not backward-looking. Rather, the institution of punishment is justified instrumentally as it is supposed to deter potential offenders and is therefore, by contrast, forward-looking.
Compare Rothbard's (2011b, 8-10)  Mantle of Science. Moreover, it is worth noting that the Rothbardian (1982 2002. Ethics of liberty. New York and London: New York University Press." href="https://link.springer.com/article/10.1007/s10838-023-09640-x#ref-CR58" id="ref-linksection-d91280806e2354">2002]) theory of punishment also contrasts sharply with that of Hayek's. For Rothbard, the rationale of responsibility assignment has nothing to with the deterrence of future crimes. Rather, ascriptions of responsibility are motivated by backwardlooking considerations. Specifically, Rothbard's theory of punishment is retributive. That is to say, the reason why punishment is inflicted is that the punished person culpably committed a prior wrong. Or still in other words, according to Rothbard, the institution of punishment is desert-based rather than informed by forward-looking considerations such as, say, minimizing future crimes.
One cannot fail to see that these remarks are indicative of the author's belief in MLF. 36.
For a defense of agent causal accounts of free will, see e.g. Clarke (1993), O'Connor (2000. It 37. must be heeded, though, that agent causation is rather severely criticised these days. For a critique of this doctrine, see e.g. Moore (2020, 61), where the author contends that postulating "the "agent-causation" supposedly distinctive of persons bringing about changes in the world through their actions" is a desperate move which "could say little about the nature of this sui generis kind of causation, other than it was not to be confused with ordinary causation and it is the kind of causation persons uniquely originate." As such, most apparently, agent-causation is of little explanatory value and makes our ontology unnecessarily expensive.
At some other place, Kirzner (2000, 58-59) laments that "the Paretian device of accounting for 38. market outcomes" eliminates "ny freedom of the consumer to choose a market basket in a way that might permit him to err, to exercise imagination concerning, say future price changes" and it is for that reason that Kirzner believes we need (libertarian) freedom or-which is pretty much the same thing-Lachmannian (1977) "autonomy of the human mind". But then again, our compatibilism is unscathed by this criticism. After all, we submit that it is a combination of the actor's fallible knowledge and preferences that determine his or her action. Hence, on our grounds, it is unproblematic to envisage, say, a speculative act wherein the actor overestimates some future demand for a given commodity and makes purchases which only ex post prove to amount to an entrepreneurial error. To put it simply, since the actor was acting-among other things-on a false belief, he ultimately erred. It should be clear to see that the explanation of entrepreneurial errors does not need to resort to "uncaused" agential causes.