Coercion, voluntary exchange, and the Austrian School of Economics

Abstract

In this paper we analyse the concept of coerced exchange (and partly of voluntary exchange inasmuch as the absence of coercion is its necessary condition), which is of utmost importance to economic theory in general and to the Austrian School of Economics in particular. The subject matter literature normally assumes that a coerced action occurs under threat. Threats in turn can be studied from the perspective of speech act theory, which is concerned with the speaker’s intentions. Ultimately, our goal is to provide a descriptive (i.e. non-moralized) definitions of threat and coercion, based on the analysis of the coercer’s intentions. If successful, we would be in a position to present such an account of coerced and voluntary exchanges that is compatible with both speech act theory and the Austrian methodology. Although we focus on the Austrian School of Economics, we believe that our investigations might impact on economic theory in general. We also criticize a rights-based account of coercion employed in the research practice of some neo-Austrians and based on the libertarian ethic of property rights.

1 Introduction

A well-known contention holds that when a voluntary exchange takes place, both parties benefit (at least ex ante). This is in fact the key proposition presumed to be true both within the Austrian and the neoclassical paradigm. It is also assumed in the most recognized economic theories and models such as the model of perfect competition or general equilibrium theory. It also figured explicitly in Vilfredo Pareto’s welfare economics.

Pareto is famous for his rule of efficiency, according to which economic changes are efficient (Pareto-superior exchanges take place) whenever there is at least one person who benefits (is better off) and no one who loses (is worse off) (see, for instance: Wight, 2015, p. 81). Thanks to that understanding of efficiency, Pareto could thus escape the difficulties with interpersonal comparisons of utility. In consequence, it is at least intuitively plausible that voluntary exchanges are efficient in Pareto’s sense and that in the case of coerced exchange one party benefits at the expense of the other,¹ thus violating the Paretian rule of efficiency. These considerations tell us what the results (or expected results) of voluntary exchanges are. However, this still leaves the concept of voluntary exchange unanalysed.

Before proceeding, a few systematizing remarks are due at this point. In this paper we adopt the following assumptions concerning relations between the following notions (those premises are usually accepted in the literature dealing with the problem of coercion, see: Anderson, 2011; Lamond, 2020). First, coercion renders actions (and transactions) involuntary. Therefore, it is an analysis of the concept of coercion that is the focal point of our essay. Second, coercion occurs only in interpersonal relations, which implies that one cannot be coerced by entities that are not persons (or at least we do not deal with such types of coercion). Third, coercion is a notion that is different from the one of violence: in the case of coercion one party resorts to a threat² to get someone to do something (coercion takes place only when the victim surrenders, i.e. when the threat is successful), while violence can occur without a threat at all (one party does not necessarily want another to do something, but makes him worse off directly, e.g. in case of battery or robbery).

As threat (the sine qua non condition of coercion) can be classified as a speech act (wordless communications can be treated as speech acts too, see: Green, 2020), and because the speech act theory is essentially concerned with the analysis of the speaker’s intentions, the idea of applying the said theory to investigating the coercer’s intentions seems especially interesting and apparently promising. Even though the concept of coercion was not originally formulated on the grounds of speech act theory, we believe that the latter may well serve to revise the former.³

The main goal of this paper is to provide such a definition of coercion that is compatible with the foundations of the methodology of the Austrian School of Economics. The rationale for the present research is that the notions (however clearly defined) of voluntary exchange and coercion as employed by Austrians are inadequate, with the main indictment being that they run counter to Austrian methodology. Although this paper aims at revising the concepts of coercion and voluntariness specifically within the Austrian School of Economics, the considerations presented herein might impact on the economic theory in general. Our thesis is that the intention-based concept of coercion (Lamond, 1996, 2000, 2020) best fits the Austrian methodology. Another problem to solve is the relation between the Austrian economics and libertarian ethic. Even though the Austrians contend that “Austrian economics and libertarianism, although often found together, are distinct both in principle and practice” (Egger, 1979, p. 119), we argue that they are not always distinct in practice. In fact, we submit that they are bound together in the very methodological apparatus employed by some well-known Austrians (Block, 1995, 2000; Block & Gordon, 1985; Herbener, 1997, 2008; Hoppe, [1988] 2010, [1993] 2006, 1994; Huerta de Soto, [1992] 2010; Hülsmann, 2004, 2008; Rothbard, [1970] 2009b). Moreover, we believe that our findings will help to make “a further step in the consistent application of subjectivism” (Hayek, 1952, p. 31) in economic analysis, which seems desirable from the point of view of the Austrian School.

The reasons for choosing the Austrian School are as follows. First, it is characterized by quite consistent and sophisticated methodology, which has been appreciated by the influential philosopher of economics, Uskali Mäki, who claims that the Austrian School is “one of the most self-reflective traditions in economics”.⁴ Second, the notions in question are especially important to this intellectual current since a great number of its theories rely on them; for example, monopoly theory, welfare theory, monetary theory, theory of public goods, theory of interventionism, theory of entrepreneurship or comparative analysis of economic systems (see e.g.: Hoppe, [1988] 2010; Huerta de Soto, [1992] 2010; Hülsmann, 2008; Kirzner, 1973; Rothbard, [1962 ]2009a, [1970] 2009b; Wiśniewski, 2018).

Even if there were some attempts to deal with the problem of coerced and voluntary exchanges in the economic literature (e.g. Christiano, 2016; Dawson, 2015; Hausman & McPherson, 2006; Johnson, 2015; Klein, 2007; Munger, 2011; Musgrave, 1939; Preston, 1984), none of them proposed a solution that was similar to ours (at least to the best of our knowledge).

This paper proceeds as follows. "Foundations of the Austrian methodology and theory of exchange" section presents the most important methodological postulates of the Austrian School of Economics and the foundations of its theory of exchange in order to understand what criteria the concept of coercion should meet to be consistent with the Austrian methodology. In "Problem of coercion in philosophy" section, we take a look at those concepts of coercion that occur in the contemporary literature on the subject. Next, in "Rights-based concept of coercion in neo-Austrian economics" section, we analyse theories of voluntary and coerced exchange that are being ventured in the research practice of the representatives of the Austrian School, while showing that some of them use these concepts (i.e. voluntary and coerced exchange) in a way that smuggles the libertarian ethic of property rights, thus making them inconsistent with the methodological postulates put forward by Austrians. In "Speech act theory and intention-based account of coercion" section, we offer a solution which seems compatible both with the Austrian methodology and speech act theory at the same time. The last section contains a short summary.

2 Foundations of the Austrian methodology and theory of exchange

In the eyes of Austrians, economics is a science of human action (praxeology). The central proposition of the science is the so called action axiom, which holds that an individual acts in order to attain certain ends by employing definite means. Among the most fundamental methodological postulates of the Austrian School are methodological individualism and methodological subjectivism (Martin, 2015).

Methodological individualism postulates a certain norm of explanations in sciences concerned with human action. Specifically, what would count as “satisfactory explanations” in the said sciences would be the ones that account for “the change of social institutions, social processes, social facts, and social structures” by citing “the interplay of individual intended human actions” (Linsbichler, 2017, p. 17).⁵ Some traces of this principle can be already found in Carl Menger, who contended that “economic phenomena theoretically are reduced ultimately to individual economic efforts or to their simplest constituent elements, and are thus explained” (Menger, [1963] 1985, pp. 90–91). It was the author of the term methodological individualism, Joseph Schumpeter himself, who already stressed that this methodological principle should be distinguished from both ethical and political individualism (and thus should not be associated e.g. with egoism or liberalism, respectively). Methodological individualism neither necessarily presupposes ontological individualism (Schumpeter, [1909] 1980). It rather emphasizes that “everything social must in some way be recognizable in the action of the individual” (Mises, [1933] 2011, p. 39). The same has been restated by Friedrich von Hayek and Ludwig Lachmann. Hayek, in his description of “the individualist and»compositive« method of the social sciences”, stresses that explanations of complex patterns of social order ultimately refer to beliefs held by acting individuals and to their interactions (Hayek, 1952, pp. 36–43). Lachmann suggests that explanations in social sciences rely on tracing social phenomena back to intentions and plans of certain acting individuals (Lachmann, 1971, p. 20). Finally, it is worth emphasising that according to many Austrians methodological individualism is concerned with explaining social facts as unintended results of intended individuals’ actions (see: Menger, [1963] 1985; Hayek, 1952; Linsbichler, 2021; Neck, 2022).

Methodological subjectivism, in its turn, holds that social scientists should not be concerned with the (external) physical reality, but rather with acting individuals’ mental states. As Austrians claim, there are no objective economic values inhering in material objects. Economic value is rather relative to human needs, desires or preferences. Moreover, Austrians emphasize that only ordinal scales of values are warranted in economic analysis and no interpersonal, intertemporal or quantitative comparisons of utility are valid. This subjective value theory, originally put forward by Menger ([1976] 2007), became the cornerstone of the Austrian School. But there is more to methodological subjectivism than just that. This principle can be treated as the extension of the Mengerian theory of value. It was Mises who consistently expanded subjectivism (see: White, 1984; Lavoie, 1991; Mises, [1933] 2011). As Mises put it: “Praxeological reality is not the physical universe, but man’s conscious reaction to the given state of this universe. Economics is not about things and tangible material objects; it is about men, their meanings and actions.” (Mises, [1949] 1998, pp. 92–93). Mises also maintained that it is methodological subjectivism that allows economics to be an objective and value-free science (Mises, [1949] 1998, p. 21). Hayek in turn famously claimed that the data of social sciences comprises what people believe and think, and not what exists out there independently of subjective beliefs and valuations. To that effect, he stated what follows: “That the objects of economic activity cannot be defined in objective terms but only with reference to a human purpose goes without saying. Neither a»commodity« or an»economic good,« nor»food« or»money,« can be defined in physical terms but only in terms of views people hold about things” (Hayek, 1952, p. 31). Or, as Don Lavoie expressed it: “It is not directly the way the world is which matters to economics, but rather the way it appears to acting agents” (Lavoie, 1991, p. 476).

Some Austrians went even further. According to so called radical subjectivism, with its originator, Ludwig Lachmann, in the first place, methodological subjectivism is strongly connected with the method of Verstehen (i.e. interpreting other people’s actions by understanding the meaning they ascribe to them), as put forward by Max Weber (Lachmann, 1971). Lavoie even suggests that economics, as a human science, is concerned with understanding rather than explanation (Lavoie, 1991, p. 473). This orientation is also called the hermeneutical turn in the Austrian economics.

To sum up, Austrians consistently claim that economics is fundamentally about intentions and meanings that people attach to things. Subjectivists hold that economic goods and economic values exist only in the minds of economic actors. Therefore, we could say that methodological subjectivism is based on the belief that social and economic phenomena are relational (and not independent), i.e. they exist in relation to acting individuals’ mental states. It means that economics deals with the mental and not with the physical aspect of reality.

Next, there is a value freedom postulate (Wertfreiheit). According to this principle, no value judgements are acceptable within (positive, non-normative) economic theory. As Austrians contend, value freedom postulate has it that “personal value judgements are in no way expressed in the substantive content of the science” (Kirzner, 1994, p. 313). It is not permissible for the economist qua economist (and not ethicist) to judge whether a change in state of affairs is good or not. In order to assess that, one needs to assume some standard of what is desirable and what is not. As we believe, in more general terms, value freedom means that the descriptive and explanatory power of economic theory should be independent of value judgements or ethical assumptions.⁶

Now, we can turn to the foundations of the Austrian theory of exchange. The fundamental question here concerns the motives of human action. As Mises states: “Acting man is eager to substitute a more satisfactory state of affairs for a less satisfactory. His mind imagines conditions which suit him better, and his action aims at bringing about this desired state. The incentive that impels a man to act is always some uneasiness” (Mises [1949] 1998, p. 13).

As we believe, the above quotation points to a plausible conjecture that the Misesian concept of “uneasiness” might be identified with an unsatisfied desire (or preference). If so, then we could say that this general motive which incites an individual to perform a given action is an unsatisfied preference.⁷ As Mises clarifies, to take an action, an individual must be also convinced that his goal is achievable (ibid., p. 14) and that the benefit from the action will be greater than its opportunity cost, which can be identified with the second-best alternative which has to be foregone (ibid., p. 97).⁸ Therefore, we can say that an individual always undertakes the most, at least at the moment of choice, subjectively desirable action.⁹ In other words, he acts in a certain way because he prefers it to acting in a different way or not acting at all. Attaining ends is in turn to be identified with satisfying desires (or preferences). Whenever an individual attains a certain end, his preference satisfaction increases, and when a fewer of his ends are being attained, his preference satisfaction decreases (see: Rothbard, [1962] 2009a, p. 18).¹⁰

Since every action is “an attempt to substitute a more satisfactory state of affairs for a less satisfactory one”, it is always a kind of exchange (Mises, [1949] 1998, p. 97).¹¹ According to Mises, when an exchange is made without cooperation with other individuals (that is without expectation that another individual will perform an action satisfying some preference of the acting individual), it is an autistic exchange.¹² On the other hand, when an individual expects cooperation from another, it is an interpersonal exchange (ibid., pp. 195–196). While the former type of exchange is not a subject of interest to catallactics (the theory of market exchange) at all, the latter is its very core. It is worth noticing that a consequence of this distinction is that autistic exchanges occur not only when there is no interaction between individuals, but also when there is an interaction without cooperation. For instance, giving a present (when one does not expect receiving anything from the other party in the future) is classified as an autistic exchange (ibid.). Consequently, we should also thus conceive of actions involving violence (but not coercion), such as battery, theft or robbery.

It seems important to mention that Rothbard categorizes actions slightly differently. According to him, gifts as well as murder or robbery are to be put under the rubric of interpersonal exchanges (Rothbard, [1962] 2009a, pp. 79–94). We, however, do not deal with such actions here since the very structure of coercive proposals requires assuming a scenario wherein the proposer gets the recipient to do something for him, with the latter expecting undesirable counteraction in case of his non-compliance (like, for instance, in the case of the following proposal: “if you do not give me your wallet, I will hit you”). Thus, we take the Misesian criterion of cooperation for granted in defining interpersonal exchanges.

For our investigations the distinction between two types of interpersonal exchange is crucial. Mises calls the first of them an exchange based on contractual bonds and the second an exchange based on hegemonic bonds. As he maintains, in the case of contractual bonds “the logical relation between the cooperating individuals is symmetrical” (Mises, [1949] 1998, p. 196). There is no director and subordinate. Each party to the exchange has the same relation to the other, and each party receives “definite quantities of goods and services of a definite quality” (ibid., p. 197). The main difference between contractual bonds and hegemonic bonds is that in the case of the latter “the scope in which the choices of the individuals determine the course of events” is radically smaller. Mises writes: “In choosing subjection in a hegemonic body a man neither gives nor receives anything that is definite” (ibid.). This means that in such a situation, the only choice of a subordinate individual is subordination. He, unlike the director, does not realize his goals (his main purpose is to act as the director demands). It might be said that motives of his action are external to him. The person who is mainly acting here is the one who gives the orders. The relation between individuals is asymmetrical then (ibid., pp. 196–197).

Mises talks about physical violence and the threat thereof as possible motives for choosing obedience. At the same time, however, he claims that the family is a hegemonic structure too, so he does not seem to contend that these motives (physical violence and the threat thereof) are the only ones that might underlie a hegemonic bond (ibid., p. 197). In the end, it would be suspicious to claim that only preferences related to physical reality matter in economics, since the science of human action is concerned with all types of goals and preferences. It is also worth remembering that according to Mises “no physical violence and compulsion can possibly force a man against his will to remain in the status of the ward of a hegemonic order” (ibid.). Quite the contrary, the only result of violence or the threat thereof is that “subjection as a rule is considered more desirable than rebellion” (ibid.). Hence, the hegemonic bond occurs only when an individual chooses to succumb to subordination, and he does so because he prefers the consequences of obedience to the consequences of disobedience. If an individual opposes the subordination, hegemonic bond does not occur (ibid.).

It is worth emphasizing that Mises does not talk about voluntariness and coercion here. Indeed, since he recognizes family as a hegemonic structure, we could suppose that according to him individuals can engage in hegemonic structures voluntarily. In other words, it seems that according to Mises, coercion is not necessary for hegemonic bond to occur. The only important thing is that the hegemon wants the counterparty to realize his goals and gives the orders which the counterparty realizes, regardless of the fact whether the latter does it voluntarily or not.

It is also Rothbard ([1962] 2009a, pp. 82–84) who resorts to the term of hegemonic bonds. However, in this author’s view, exchanges based on hegemonic bonds are always coerced. In other words, he, contrary to Mises, identifies exchanges based on hegemonic bonds with coerced exchanges, thus claiming that here, as opposed to voluntary exchanges (which he identifies with exchanges based on contractual bonds) one party gains at the expense of the other (Rothbard, [1962] 2009a, p. 84). As he states, coercion (or intervention, with both terms being used interchangeably here) evokes negative “direct effects on utility” (it decreases preference satisfaction of an individual) (ibid., pp. 878–885). To sum up, Mises’s conception of hegemonic bonds cuts across the Rothbardian distinction between voluntary and coercive exchanges. By contrast, in Rothbard’s view, the concept of hegemonic bonds is co-extensive with the concept of coerced exchanges. However, this differential conceptualization of hegemonic bonds by Mises and Rothbard seems to amount to nothing more than a merely classificatory (or verbal) issue. Still, since we are primarily interested in the problem of coercion and voluntariness, we are supposed to talk about coerced and voluntary exchanges.

Finally, our investigations should not ignore considerations on coercion put forward by Hayek. In his Constitution of Liberty he dedicates the whole chapter to the issue of coercion. Hayek explicitly accepts the assumptions mentioned by us in the introduction: that coercion occurs only between acting individuals; that there is a difference between coercion and violence; and that coercion implies “the threat of inflicting harm” (Hayek, [1960] 1978, pp. 133–135). Moreover, Hayek notices that a coerced person—in spite of the fact that he or she is made to serve “another man’s will”—acts (contrary to victims of violence, who do not act).¹³

Interestingly, Hayek distinguishes between coercion and power, and his conception of power seems to correspond to the Misesian conception of hegemonic bonds. Hayek admits that in the case of power one can direct other people without threatening them. For instance, a director of an enterprise tells other people what to do in order to achieve some common goal, but he does not pose a threat at them. Thus, power can be exercised voluntarily, i.e. without coercion (still, it might be coercive too). It seems then that exchanges based on power can be identified with exchanges based on hegemonic bonds (in Mises’s terminology).¹⁴

Despite its advantages (mentioned above) it is necessary to notice that Hayek’s position on coercion has been criticized from many different points of view for its flaws and inconsistencies (see: Hamowy, 1971; Hoppe, 1994; Paul, 1980; Rothbard, [1982] 1998; Viner, 1961). For Hayek indeed provides us with problematic criteria of coerced actions. First, he contends that coercion can occur when “the services of a particular person” are crucial to one’s existence.¹⁵ Unfortunately, this seems to be a very strange criterion for a subjectivist to adopt. Certainly, this condition is purely objective, since it does not refer to mental states (beliefs or value judgments) of acting individuals, but to their “existence” as such. Fortunately, Hayek also presents a subjectivist criterion, i.e. that coercion can occur when what is under threat is something that someone considers the most valuable (Hayek, 1960, 1978).

Second, Hayek claims that for coercion to occur, another necessary condition must be met. That is, there must be no other people who can satisfy one’s desire. To put it in a different way, if there are other people who can satisfy one’s particular desire, then coercion cannot occur under such circumstances. This is because Hayek regards competition between sellers as a sufficient condition for voluntary exchanges to take place.¹⁶ Accordingly, he states:

“A monopolist could exercise true coercion, however, if he were, say, the owner of a spring in an oasis. Let us say that other persons settled there on the assumption that water would always be available at a reasonable price and then found, perhaps because a second spring dried up, that they had no choice but to do whatever the owner of the spring demanded of them if they were to survive: here would be a clear case of coercion.” (Hayek, [1960] 1978, p. 136)

It seems that there are at least two problems with this position. First, if we reject the objective criterion of survival (or existence) and just stick to subjective desires, then every monopolist could coerce other people. For instance, imagine that there is only one university in the particular area. The university authorities inform a student: “pay us $2,000 in tuition or we will expel you from the university.” Since the student has no choice but to pay (given that he or she wants to graduate in higher education), Hayek’s position predicts that he or she was coerced (perhaps the distinction between warnings and threats could help here, but, unfortunately, Hayek apparently does not take it into consideration—we elaborate on the distinction in the next sections). The second problem with the criterion of competition is that there could be more than one “coercer” at the same time, which predicts that no “true coercion” would obtain in that case. For instance, let us imagine that P₁ comes to Q and says: “if you pay me $3,000, I will be defending you, but if not, I will kill you”. Then P₂ comes to Q and says: “if you pay me $2,000, I will be defending you, but if not, I will kill you”. Does P₁ or P₂ coerce Q? Since there is a competition between P₁ and P₂, Hayek would have to admit that there is no coercion here. But let us imagine that P₂ drops out of the picture, everything else equal. In this case, the Hayekian position predicts that P₁ coerces Q even though in both scenarios P₁ performs exactly the same action. Mutatis mutandis, if P₁ were to drop out of the picture, suddenly P₂ would start counting as a coercer, which seems equally problematic for the same reason (for other comments on the issue of coercion, monopoly and competition in Hayek see also: Paul, 1980, pp. 41–42).¹⁷

Third, in his discussion of coercion Hayek uses a problematic concept of “reasonable price”. This, however, seems incompatible with the Austrian methodology. Thus, Hamowy rhetorically asks: “at what price does the contract alter its nature and become an instance of»coercion«?” (Hamowy, 1971, p. 354).

Fourth, it seems that there are also problems with the criterion of avoidability, which Hayek presents to suggest that we “need never be coerced”, if only we know in advance that in a particular situation (that can be avoided) we would be “coerced”. Unfortunately, as Hamowy observes:

“It follows from this that if Mr. X warns me that he is going to kill me if I buy anything from Mr. Y, and if the products available from Mr. Y are also available elsewhere (probably from Mr. X), such action on the part of Mr. X is non-coercive! Avoidability of the action is sufficient, according to this criterion, to set up a situation theoretically identical to one in which a threat does not occur at all.” (Hamowy, 1971, p. 356)

Given the above problems with the Hayekian criteria of coercion, it seems that we should look for different solutions. Incidentally, the Hayekian conception of coercion has been criticized also by Rothbard ([1982] 1998) and his followers (e.g. Hoppe, 1994).

Taghizadegan and Otto (2015, p. 300) in their turn seem to rightly claim, referring to Hayek, that: “Human actions that start with offers that can be rejected free of cost by the counterparty belong […] to the area of catallaxy”. Indeed, it seems that catallactics is to be identified with the area of voluntary exchanges. But the term “free of cost” needs to be more precise, since, as we have seen, every action (and thus every type of exchange) entails an opportunity cost.

As we will try to show later, according to Rothbard and his followers, coercion basically reduces to the aggressive use of physical violence¹⁸ or the threat thereof, and proposals can be recognized as coercive threats only if they announce a (property) right violation as the counteraction in the case of the victim’s non-compliance. Moreover, (property) rights are usually treated by those Austrians as rationally justified natural rights and as given exogenously in their economic theory. As we will argue later, this position is inconsistent with the remainder of the Austrian methodology. Now, however, we would like to turn to the sketchy analysis of philosophical approaches to the concept of coercion, showing their inconsistency with the Austrian methodological postulates presented in this section.

3 Problem of coercion in philosophy

Coercion is a concept that is widely studied in ethics as well as in legal and political philosophy. The problem of coercion has been already stated in Aristotle’s Nicomachean Ethics. The philosopher from Stagira claimed that coercion makes an action involuntary, and therefore reduces moral responsibility of the coerced person (Aristotle, 1984, pp. 1110a–1111b).¹⁹ Although Aristotle predicted the existence of different types of coercion, at least already from the times of Thomas Aquinas, it is usually assumed that coercion occurs only in relations between people, i.e. when one person forces another to do something (Anderson, 2011).²⁰

The new standard of discussion about coercion in the contemporary philosophy was set by Robert Nozick thanks to his essay Coercion (1969). Anderson (2011) notices that there are a few types of approaches to the said problem. Nozick can be perceived as the precursor of at least two of them, both of which can be classified as so called baseline approaches.

In this kind of analysis, it is usually assumed that threats coerce, while offers do not, and what distinguishes threats from offers is that “compared to the baseline, threats worsen the coercee's situation, while offers do not” (Anderson, 2011). More precisely, threatening party by the very issue of the threat renders the performance of an action that the threatened party would otherwise like to take (while the threatening party would not like the threatened party to take) less attractive for them. For instance, in a paradigm case of threat, where P comes to Q and says: “Your money or your life”, Q’s refusal to give P the money appears less desirable for Q than giving it. In other words, the alternative action (preservation of money) appears less desirable for Q than acting as P demands (giving him money), due to the P’s announced counteraction.

There are two kinds of baselines that Nozick uses in his seminal paper. According to him, one baseline can be identified with the “normal or natural or expected course of events” (Nozick, 1969, p. 447). In his analysis, Nozick presents a thought experiment in which he suggests that the normal course of events is what is predictable based on prior experience. In his example, P usually sells drugs to Q, but once he refuses to do so unless Q beats up a certain person (then P will even give Q these drugs for free). That would mean that P threatens (and if he does it successfully, then coerces) Q. On the other hand, if P meets Q for the first time, then his proposal would be treated as an offer (Nozick, 1969, p. 448).

As Nozick observes, there are some problems with this criterion. One is that it is relative to our sense of “normal or expected course of events”. In his other thought experiment, Nozick presents a situation in which Q is going to drown in the lake, and P proposes to save him if Q promises to do something for P in the future. As Nozick states: “Whether P’s saying that he will save Q if and only if Q makes the promise is an offer to save Q or a threat not to save Q depends upon what the normal or expected course of events is” (Nozick, 1969, p. 450).

Another problem with this criterion might be that it is overinclusive, for it allows for warnings to count as threats. For instance, P comes to Q and says: “In the near future, I am going to build an ugly building in your neighbourhood. But I can do that elsewhere, if you pay me for the transportation”. As P’s plans are not normal or expected course of events for Q, then Q’s alternative is worse compared to his baseline. In consequence, even if P’s intent was not to coerce Q to pay him, but just to warn Q and give him some alternative, we would have to classify P’s proposal as a threat.²¹

Moreover, this criterion takes into account both speaker’s intentions and recipient’s understanding of a situation. Undoubtedly, this is justified in the analysis of coercion, wherein the necessity of establishing whether a threat is successful or not is widely acknowledged, but as long as we are concerned with the problem of threat as such, we should focus solely on speaker’s beliefs and intentions.²² Only when we formulate a plausible concept of threat can we go on to probe it further.

The second baseline which Nozick proposes is a normative (moral) baseline. He comes to the conclusion that the “normally expected course of events” baseline is sometimes insufficient and then he formulates another thought experiment. He imagines a situation in which a slave owner beats his slave every morning for no reason, but once he proposes not to beat the slave if the latter does something special for him. If the previous criterion were to be applied, this proposal would be treated as an offer: compared to a “normally expected course of events” the slave is better off having the opportunity to avoid being beaten. However, as Nozick contends, since slavery and battery are themselves immoral, such a proposal should be treated as a threat. It is because no one has a right to enslave and beat another person—at least in the absence of any prior contracts to that effect. In other words, by receiving this proposal the slave is rendered worse off compared to the situation in which his moral rights are respected (Nozick, 1969, p. 450).

Ultimately, the selection of a baseline, Nozick contends, depends on the preferences of the recipient: if he prefers a “normally expected course of events” to the consequences of the refusal of the proposal, then he is confronted with a threat (even though “morally expected course of events” is different from the normally expected one; let us imagine here a thief who is usually unpunished for his crimes: the punishment is “morally expected” for him, even though it is not “normally expected”). If the recipient prefers “morally expected course of events” to the consequences of the refusal of the proposal, then he also faces a threat (even though “normally expected course of events” is different from the morally expected one) (Nozick, 1969, pp. 450–451).

The rights-based view of coercion and voluntariness is shared by Wertheimer (1987). There are a few problems with this approach. First, the most general problem with this rights-based account is that it does not conceptually allow for saying that the legal system coerces people. For instance, legal systems usually provide that the penalty for theft is imprisonment. In this situation, we could not say that law enforcement agents coerce the thief (when caught and convicted) to go to prison, for the thief does not have a right not to go to prison. The only possible answer here would be then that the thief goes to prison voluntarily. Unfortunately, this sounds odd and counterintuitive, to say the least.

Second, as a general argument against rights-based positions as such, we must note that different rights are recognized by different moral or legal systems. And since there is more than one ethical (or legal) system in the world, this ipso facto implies that what is coercive relative to one system is not necessarily coercive relative to another.²³ However, if Austrian economists subscribed to this approach (actually, some of them do, and we will critically analyse their position in the next section), they would have to choose a definite moral system as a basis for their economic theory. Unfortunately, it would still collide with their postulate of value freedom. Moreover, as our agenda is to elaborate a non-relative concept of coerced exchanges, the relativization of the concept of coercion (and voluntariness) to a given ethical or legal system would be most damaging to our project in the first place. After all, we are interested in coerced exchanges as such, and not in coerced exchanges relative to an ethical (or legal) system. That is to say, our ambition is to lay out such an account of coerced exchanges that can be applied across ethical (or legal) systems. Another problem with the rights-based approach on the grounds of the Austrian methodology will be presented in the next section.

A significant contribution to the philosophical debate on voluntariness was made also by Olsaretti (1998). In her article, she criticizes rights-based conceptions of coercion and voluntariness. Her solution, however, is normative too. In her view, there is a need to resort to so called acceptable alternatives. This concept, however, requires some objective (and normative) standard of what is “acceptable”. Unfortunately, this seems inconsistent with both the Austrian subjectivism and value freedom postulate.

Yet another baseline approach has been presented by Zimmerman (1981). According to him, both threats and offers can be coercive and there is no sharp difference between them. In order to assess whether a proposal (whether an offer or threat) is coercive or not, we have to establish if the proposal maker actively prevents the recipient from being in a situation in which the latter has better alternatives to choose from. If the proposal maker makes it impossible for the recipient to be in a better position, then he makes him worse off compared to the baseline, which is the absence of his interference. In consequence, if the recipient accepts the offer, coercion occurs. In his example, Zimmerman imagines a situation in which an island inhabitant receives a work offer from the only other person on this island. Therefore, accepting this offer is the only way for him to survive. In consequence, he accepts the offer. Was he coerced to do so? Zimmerman would say that it depends whether the proposal maker made it impossible for the recipient to receive other offers (for instance, by making it impossible for him to escape from the island).

The problem with this approach is that it resigns from the commonly accepted difference between (coercive) threats and (non-coercive) offers. There is, however, a possibility of preserving the distinction, introducing the notion of exploitation in the case of Zimmerman’s coercive offers. In the above scenario with the proposal maker actively preventing the island inhabitant from having better alternatives, we could validly speak of exploitation. Admittedly, the proposal maker could resort to coercion to make it impossible for the recipient to have better alternatives, but at the moment of formulating it, this proposal as such would be classified as an exploitative (but not coercive) offer. Therefore, as usually assumed, coercion would occur only in situations where threats are posed (see: Anderson, 2011). Unfortunately, as Zimmerman does not provide us with an independent definition of threat, these considerations seem useless for the purpose of our investigations.

Now we can turn to so called non-baseline approaches. Feinberg (1986) argues that coercive proposals generate so much pressure on the recipient’s will that the latter is unfree to make his choice. This account requires employing some method of pressure calculation and comparison between the degrees of pressure that proposals exert on the recipient’s will. Unfortunately, this position allows for tempting offers being coercive too as long as they create very strong pressure. This seems to be the first problem. The second is that pressure is not a praxeological concept. Defining pressure in a formal and non-psychological way is most probably impossible. Even if we somehow tried to express pressure in terms of utility (or welfare), we need to remember that Austrian economics bans interpersonal, intertemporal and quantitative comparisons of utility (one can only say that one prefers X to Y, but he cannot say how much). Hence, appealing to calculations of pressure seems inconsistent with this methodology.

Another interesting non-baseline position is the one which focuses on the intentions of the proposal maker (Lamond, 1996, 2000, 2020). As we believe, this is the most appropriate solution to the problem of coercion on the grounds of the Austrian methodology. In our opinion, this account meets all the Misesian criteria, i.e. it refers only to these analytical categories which constitute the essence of the Austrian theory of action: intentions (goals, aims, ends), desires (preferences), and beliefs (expectations), and is thus compatible with the Austrian methodology.²⁴ In the next section, however, we will show that some contemporary Austrians use a moral, rights-based baseline, when defining coercion and voluntariness. More specifically, they found these notions upon the libertarian ethic of property rights.

4 Rights-based concept of coercion in neo-Austrian economics

Rothbard and his followers consistently contend that an exchange is mutually beneficial whenever it is voluntarily undertaken by both parties.²⁵ In his recent paper, it is Rothbard’s follower, Block, who holds what follows: “The only way we can deduce benefits for sure is from voluntary interaction” (Block, 2019, p. 106). From the works of Rothbard ([1982] 1998, [1973] 2006, [1962] 2009a, [1970] 2009b) we can infer in turn that by coercion he means a physical violation of property rights or the threat thereof. Moreover, it seems to be based on his libertarian ethic of property rights. That would mean, however, that he uses rights-based concept of coercion. As we will try to show, this position seems incompatible with the Austrian methodology.

Commenting on Rothbard, High (1985) notices that the notions of voluntariness and coercion rely on certain ethical assumptions. He argues that this is the reason why economics needs to employ ethics in its theoretical apparatus.²⁶ In the economic discourse, the ethical entanglement of these concepts was also noticed by Hausman and McPherson (2006, pp. 36–37). Since ethics is traditionally considered a normative discipline, such understanding of these notions could threaten the Austrian value freedom postulate.

The problem of ethical entanglement of Rothbardian economic theory was noticed by Terrell (1999) in his comment on the Austrian critique of Coase theorem, and by Gunning (2000) in his critique of Cordato ([1992] 2007). These authors, however, do not indicate precisely where the problem really lies. On the other hand, Cordato ([1992] 2007, pp. 41–42) rightly recognizes that the Rothbardian concepts of voluntariness and coercion are based on the libertarian ethic of property rights, and it is for this reason that his concept of economic efficiency is normatively entangled. Cordato’s point, however, seems to be underestimated in the literature. For instance, Gordon (1993) in his comment on Cordato ([1992] 2007) notices the problem, but he disregards its gravity and dismisses it in just a few sentences.

In the recent research it has become clear that the notions of threat, voluntariness, coercion or even freedom, as used by deontological libertarians such as Rothbard, rely upon physical violation or non-violation of property rights (Dominiak & Wysocki, 2016; Dominiak, 2017, 2018). It seems that Austro-libertarians²⁷ are oblivious of the ethical entanglement of the concepts in question since they use them in their economic theory which they regard as value-free (Block, 2005; Rothbard, [1976] 2011). This problem is especially conspicuous in the Austrian debate on blackmail and extortion. Block and Gordon (1985) distinguish between efficient and inefficient economic exchanges, based on the criterion of physical violation of property rights (where property rights are treated as naturally given). In their view, because blackmail (as they define it) is a kind of threat which does not announce the violation of property rights, an exchange between the blackmailer and blackmailee is voluntary and economically efficient.²⁸ On the other hand, extortion involves a threat of property rights violation, and hence it makes a resultant exchange economically inefficient.²⁹

It is worth noticing here that Hayek held a different view on that issue. He did not regard physical violence as a necessary condition for coercion to occur. He emphasized that “the threat of physical force is not the only way in which coercion can be exercised” (Hayek, [1960] 1978, p. 135). Moreover, as he put it: “True coercion occurs when […] the knower of an evil secret blackmails his victim” (ibid., pp. 137–138).

Nonetheless, according to Rothbard and his followers, only illegitimate proposals (i.e. those that announce the physical violation of property rights) count as threats and can render exchanges involuntary. Consequently, we should say that proposals that do not violate property rights are not threats at all. Instead, they should be treated as offers. In his defence of Rothbard’s welfare economics, Herbener (2008) takes a similar position:

“A voluntary exchange occurs when neither trader uses or threatens violence against the property of the other. If the two persons trade the ownership of property without aggressive violence, then the exchange is voluntary. Given their natural ownership of property, each person chooses an alternative he prefers more than the non-interaction alternative. Both traders benefit. If one person violently aggresses against the property of the other person, then the exchange is involuntary. Given their natural ownership of property, the aggressor chooses an alternative that he prefers more than the non-interaction alternative and the victim is forced to choose an alternative that he prefers less than the non-interaction alternative. The aggressor benefits and the victim loses.” (Herbener, 2008, p. 61)

One might argue, however, that, given this quote, it is not clear whether Herbener has in mind the legal concept of natural ownership (or property right, as defined in the libertarian concept of natural law), or rather the sociological (or economic) one. The latter, as defined by Mises ([1922] 1962), refers to the fact of physical possession of a good, not the right to possess it.³⁰ This sociological phenomenon cannot be identified with the legal concept of ownership. Indeed, the sociological interpretation of the passage quoted above is very likely, given the footnote Herbener wrote in his other paper (Herbener, 1997, p. 99).

Unfortunately, neither of these interpretations satisfies the standards set by praxeology. First, note that the argument presented by Block and Gordon (1985) can have ethical significance, but it is difficult to understand the mental leap into economics made therein. As economics talks about acting individuals’ mental states, the mere reference to physical infringement of property rights appears insufficient at best and utterly misleading at worst. Second, even if we mitigate this position by reference to the sociological (or economic) concept of ownership, as Herbener (1997) does, it is still unclear why only the physical possession of goods should matter here, since the Austrian economic analysis examines all types of desires, not just those related to physical aspect of reality (as outlined in "Foundations of the Austrian methodology and theory of exchange" section). Actually, both interpretations lead us to counterintuitive conclusions and are inconsistent with the Austrian methodological principles. To illustrate this thesis, let us consider the following thought experiment (for a similar thought experiment and the critique thereof, see our discussion with Jakub Bożydar Wiśniewski: Wiśniewski, 2019; Wysocki & Megger, 2019, 2020; and our previous works: Megger, 2021; Wysocki, 2021):

• Case (1) P comes to Q and says: pay me $1000 or I will (illegitimately) imprison you. As Q prefers losing $1000 to being imprisoned and he does not see any other reasonable options to avoid the threat, he decides to pay $1000 to P. And there are no further consequences.

• Case (2) P comes to Q and says: pay me $1000 or I will destroy your reputation. As Q prefers losing $1000 to losing his reputation and he does not see any other reasonable options to avoid the threat, he decides to pay $1000 to P. And there are no further consequences.

The question is why the above contrasting scenarios demonstrate the inadequacy of the rights-based approach to economic efficiency? It is because they both have the same formal structure and—by stipulation—the same final outcome. However, the assessment (in terms of voluntariness and efficiency) of the outcome is for Rothbard and his followers different in each case: case (1) would be treated as an example of coerced and thus economically inefficient exchange; and case (2) would be treated as an example of voluntary and thus economically efficient exchange. Intuitively, this conclusion is at least suspicious.

Of course, one might say that the formal structure of our argument is correct, but the material is not. For why should we contend that these results are identical? One could argue that in each case Q bought different services. In the case (1) he bought his freedom and in the case (2) he bought the gossip’s silence. Indeed, it is impossible to deny it.

It should therefore be clarified that what matters here are the formal similarities between these scenarios (in terms of their mere reference to analytic categories of praxeology), and this should be the most important consideration here since both Mises and Rothbard contend that praxeology is after all a formal science (see: Mises, [1949] 1998, pp. 32–36; Rothbard, [1962] 2009a, p. 73). Thanks to that, our argument meets all the essential conditions of the Austrian methodology. We can present the formal similarities between our scenarios as follows:

• (1) P effectively makes Q expect that if Q refuses to accede to P’s demand, P will frustrate Q’s preference either for being free or for having unspoiled reputation, respectively (Q believes P’s promised action is going to frustrate P’s particular preference; that is, for being free or for having good reputation, respectively);

• (2) Q prefers losing $1000 to P’s counteraction in the case of Q’s refusal to accede to P’s demand;

• (3) the only purpose for which Q pays P is to avoid the execution of P’s counteraction;

• (4) cost incurred by Q is the same ($1000);

• (5) after all, P lets Q go free, and there are no further consequences.

In consequence, we could say that the result is the same in both scenarios, because Q loses the same amount of money, P lets Q go free, and Q acts for the same reasons (only in order to avoid P’s counteraction).

For Rothbard and his followers the concept of coercion is based on the libertarian ethic of property rights. Hence, as we can observe, although these Austrians argue that the economic theory itself is value-free, they somehow fail to see that the notions they use to distinguish between welfare increasing and welfare decreasing exchanges are value-laden.³¹ The reason is that applying such rights-based concepts always requires answers to normative questions like: Which rights are justified?³² Who should have these rights? How to justifiably acquire them?³³ In addition, this position leads us to weird consequences in economics (i.e. it generates counterintuitive results).

Finally, we could consider yet another interpretation of property rights. More precisely, they could be understood not as natural rights, but as a social institution. Thanks to that, rights-based concept of coercion could meet the value freedom condition. However, it is worth noticing here that Austrian economics is clearly consistent with an endogenic perspective on property rights (according to which property rights are subject to economic analysis), and not with the exogenic one (which assumes that property rights are already given) (Leeson, 2012; Rajagopalan & Rizzo, 2019). Hence, property rights can be subject to economic explanations. But if we conceive of property rights as a social institution (and not a moral standard), it does not solve our problems. For institutions emerge through interactions and exchanges between acting individuals. Hence, we could not properly distinguish between voluntary and coerced exchanges before the institution of property rights emerged. But if so, we would not have analytic tools ensuring the explanation of how property rights emerged (via voluntary or coerced actions).

Now, the question is: are there such concepts of threat, coercion and voluntariness that fit the Austrian methodology? Undoubtedly, talking about voluntary transactions as the essence of free market has a strong rhetorical value. However, as we could have seen above, there are serious difficulties in defining the terms in question in a nonnormative and unambiguous way. If Austrians want to stick to declared value freedom together with methodological subjectivism and methodological individualism, they should look for such an account of coercion that is based on the fundamental categories of praxeology. This would mean that whether an action is voluntary or not depends exclusively on specific desires, beliefs and intentions of interacting individuals. An attempt to provide such an account will be made in the next section.

5 Speech act theory and intention-based account of coercion

Undoubtedly, a threat is a speech act. It can be also called an “illocution”. It is because nowadays, “speech act” and “illocution” are terms that are often used interchangeably (see: Green, 2020). According to the precursor of speech act theory, John Austin, “illocution” is that dimension of utterance that contains the speaker’s intention (Austin, 1962). According to Searle and Vanderveken (1985), an illocutionary act is an act which is performed through “a sentence in an appropriate context with certain intentions” (Searle & Vanderveken, 1985, p. 1). Moreover, a broader definition of speech acts allows them to be wordless acts of communication (Green, 2020). In consequence, we could say that an illocution is an action aimed at achieving some goal by means of communication. This means that the speech act theory is concerned with the goals which people want to realize by virtue of communication. Thanks to that, we could perhaps say that speech act theory is, like economics, a part of the general science of human action (praxeology).

There are a few types of speech acts, but for the purpose of our investigations two are especially important; that is, (1) so called commissive illocutions, by which the speaker commits himself to perform certain actions in the future, and (2) directive illocutions, by which the speaker wants the recipient to do something.

Some scholars see far-reaching similarity between threats and promises (Salgueiro, 2010). Both so called elementary threats and elementary promises are kinds of commissive illocutions. Consider, for instance, a sentence: “I will kill you!”. In this case, the speaker communicates to the recipient an intention to kill him. In other words, we might say that the speaker commits himself to killing the recipient.³⁴ Such threats are very similar to elementary promises, such as, say, “I will buy you a car”. In this case, the speaker communicates to the recipient an intention to buy him a car. In other words, we might say that the speaker commits himself to buying the recipient a car. The difference between the two illocutions is that in the case of threat the speaker announces that he will do something unwanted (unwelcome, not preferred) by the recipient (or at least the speaker has such a belief), and in the case of promise the speaker announces that he will do something desired (welcome, preferred) by the recipient (or at least the speaker has such a belief) (Salgueiro, 2010).³⁵ Things become more complicated, when we consider so called conditional threats.

Both conditional threats and conditional promises can be either just commissive or directive-commissive. To understand the difference between them, let us consider the following example: “If they set me free, I will kill you!”. As this threat requires some condition to be fulfilled, it is a conditional threat. However, there is no directive element yet. An example of directive-commissive threat might be: “I will kill you unless you give me your wallet!”. The second part of this sentence contains a directive illocution: the speaker wants the recipient to give up his wallet. In such illocutionary acts, the directive part of the threat is even more important. While the commissive element of the threat is just aimed at making non-compliance less desirable for the recipient, the directive one is crucial to understand what the speaker wants to achieve (see: Salgueiro, 2010).³⁶ Such directive-commissive conditional threats (DCCT) are the proper subject matter in our analysis of coercion.³⁷

As it seems, in the philosophy of coercion there is a position that is consistent with this speech act theory approach. As we believe, it is also compatible with the methodology of the Austrian School of Economics.

According to Lamond (1996), the baseline approaches to the problem of coercion are misguided. Instead, he claims that the analysis of coercion should primarily rely upon investigating the intentions of the proposal maker, and only then on the reaction of the recipient.³⁸ As Lamond observes, we need to distinguish two aspects of coercion:

“The first is that one person aims to make another person do something by threatening her. The second is that the threat succeeds in forcing the other person to comply with whatever is demanded. If A coerced B into doing Y, then A intended to make B do Y, and A succeeded in having B do Y (by threatening to do X should B not do Y).” (Lamond, 2000, p. 52).

In other words, we have to make the distinction between coercive intent (“to coerce”) and coercive effect (“to believe to be coerced”). Coercive intent, in its turn, is what it takes for a threat to be coercive in the first place. That is to say, no speech act would count as a coercive threat unless it is accompanied with a coercive intent. And finally, when such a threat is successful (the recipient acts as the proposal maker wants), the coercive effect takes place.

In his analysis of threats, Lamond distinguishes between simple threats and conditional threats (in the same way as Salgueiro distinguishes between elementary and conditional threats) and calls the latter coercive threats (simple threats cannot coerce) (Lamond, 1996, p. 225).³⁹ According to him:

“There are three features of coercive threats which are crucial to a proper understanding of their nature. The first is that the proposed consequence is unwelcome to the recipient. The second is that the maker of the threat proposes to bring about the unwelcome consequence because the consequence is unwelcome to the recipient. The third is that the maker commits herself to bringing about the consequence if the recipient fails to act as demanded.” (Lamond, 1996, p. 225)

As Lamond claims, to grasp the difference between offers and both (coercive) threats and (conditional) warnings, the first condition is crucial. In order to distinguish threats from warnings, the most important is the second condition. First, however, let us shed light on the compatibility of this approach with the above speech act analysis. In a footnote in his paper, Lamond presents the constitutive conditions of coercive threats more technically:

“P coercively threatens Q when

• (1) P communicates to Q an intention to X unless Q does Y, and

• (2) P intends by (1) to commit herself to X unless Q does Y, and

• (3) P’s doing X would be unwelcome to Q, and P knows why this is so, and

• (4)

  • (a) P’s reason for (1) is to have Q do Y, and

  • (b) P’s reason for committing herself to X unless Q does Y is that X is both unwelcome to Q and more unwelcome to Q than having to do Y, and

  • (c) Q knows both (4)(a) and (4)(b)” (Lamond, 1996, p. 225).

As we see, condition (1) captures the fact that P performs a speech act, i.e. he communicates some intention in order to realize his goal (to influence Q’s behaviour). Condition (2) has it that P’s illocution is commissive. Both conditions (1) and (2) contain a part “unless Q does Y”, which means that P wants Q to do something (this is the directive dimension of P’s proposal) and that P’s threat is conditional. Conditions (3) and (4) explain the rest of the content of the coercive situation: P’s speech act is successful (because Q finds P’s doing X more unwelcome than doing Y), P rightly recognizes Q’s preferences and uses this knowledge to influence Q’s actions, and Q rightly interprets P’s beliefs and intentions. If Q surrenders in this situation, then he performs a coerced (involuntary) action (exchange).

Then Lamond adds that bluffs can be coercive too (because they are aimed at coercive effect), even though they lack some features of real threats:

“The central case allows secondary cases to be understood: thus bluffs involve P lacking the intent in (2) but seeking to have Q believe that (2) and (4)(b) are true; and threat which P hopes will be defied so that she can rationalize doing X involve (4)(a) and (b) being false but P seeking to have Q believe them. Secondary cases thus turn on Q reasonably believing, in the context in which P communicates that she is going to X, either (i) that P has the listed intentions, reasons, and beliefs, or (ii) that P wants Q to so believe.” (Lamond, 1996, p. 225)⁴⁰

Now, we can turn to the distinction between offers, (conditional) warnings and coercive threats. First, let us notice some similarities between these illocutions. All of them serve the speaker to influence the recipient’s behaviour. Moreover, all these illocutions are or can be kinds of conditional proposals. According to the speech act theory, offer in itself is a conditional promise (Searle & Vanderveken, 1985, pp. 195–196). Both warnings and threats can be in turn either elementary or conditional.

As Lamond observes, offers, as opposed to threats, involve a promise to do something that the recipient finds welcome.⁴¹ For instance, if P comes to Q and says: “I can sell you a bread for $3”, then P assumes that Q might find this proposal welcome, but P does not assume that not selling Q a bread will be unwelcome to Q (for P might as well believe that there is a chance Q might want to buy the bread from P).⁴²

How do we know if the proposal will be unwelcome? Ultimately, it depends on the recipient’s desires. If the recipient wants the proposed consequence to occur, then the proposal is welcome. If he prefers the proposed consequence not to occur, then the proposal is unwelcome. Moreover, Lamond admits that it is the recipient’s expectations and knowledge (both of them being subsumable under a general notion of beliefs) about actions the proposal maker commits himself or herself to performing that play an important role here. For instance, suppose Q expects P to do X or knows that P is committed to doing X. Now, if P proposes Q not to do X unless Q does Y, then Q finds P’s proposal unwelcome. On the other hand, if a stranger (someone towards whom Q has no expectations) comes to Q and says that he is not going to do X unless Q does Y, then Q does not find this proposal unwelcome, for he does not have any expectations associated with the stranger (Lamond, 1996, pp. 226–227).

Although the criterion of unwelcomeness suffices to distinguish offers from both threats and warnings, it is not sufficient to distinguish threats from warnings. Why is this so? First let us compare elementary (non-conditional) warnings and elementary threats. Let us imagine a situation in which P comes to his neighbour, Q, and says that he is going to mow his loan on Saturday (to use Lamond’s example). As Q did not expect that, he finds this information unwelcome. As Lamond observers, whether this is a threat or a warning depends on P’s reasons to do that. If P is concerned about Q’s comfort, then he warns Q. On the other hand, if P intends to reciprocate for a raucous all-night party by mowing at that time because Q dislikes it, then P is making a threat (Lamond, 1996, p. 227).

The same applies to conditional warnings and conditional threats. For instance, if the teacher, P, informs his student, Q, that he will have to fail him if Q does not submit his work by a certain date, then P is making a conditional warning. Even though Q finds this proposal unwelcome, P does not intend to fail Q because being failed is unwelcome to Q. It is perhaps simply his duty. In other words, failing Q would not be the purpose of P’s actions (in case Q does not meet the deadline), but simply unintended (even if conscious) consequence of his actions.⁴³

To sum up, only in the case of a coercive threat (DCCT) the proposal maker’s reason for committing himself or herself to performing the counteraction which is unwelcome to the recipient is precisely that this counteraction will be unwelcome to the recipient (it will frustrate the recipient’s preferences). In other words, in cases of offers and warnings, the proposal maker does not have intentions to frustrate the recipient’s preferences if the recipient opts out. Frustrating them is just an unintended consequence of his counteractions. It is just the opposite in the case of threat: here the speaker commits himself to doing something unwelcome to the recipient (if the recipient refuses to accede) precisely because this is unwelcome.

6 Concluding remarks

As we tried to show, there is a concept of threat that is compatible with both speech act theory and the methodology of the Austrian School of Economics. The basic advantage of the Lamond’s concept of threat (or at least according to our interpretation thereof) is that it is free of any moral assumptions. Hence, we are in a position to say that a legal system coerces people to refrain from performing particular actions (Lamond, 2000). Moreover, it allows us to understand the complexity of coercion and to grasp the difference between threats, warnings and offers. Finally, it refers only to purely praxeological categories (such as desires, beliefs and intentions), and neither narrows them down to just physical reality nor assumes any additional categories (such as rights) as exogenously given. That makes it consistent with methodological subjectivism, methodological individualism and the principle of value freedom.

However, the question might still arise of how our theory deals with radical subjectivism. One might add: why shall we be interested in both coercer’s intentions and victim’s beliefs and desires, and not only with the latter? Is not the fact that the victim believes he or she is coerced sufficient for coercion to occur? After all, Hayek observed that facts of the social sciences are what people believe and think (Hayek, 1943, 1952).

As we believe, methodological subjectivism does not deny that mental states of acting individuals exist objectively (even if they can be known only indirectly and are open to misinterpretations) and allows us to analyse intersubjective facts.⁴⁴ Even if such facts are subjective in one sense, they are objective in another. John Searle provides us with a helpful distinction here. As he contends, social facts are ontologically subjective (they exist only in relation to mental states of acting individuals; for instance, pain, as an ontologically subjective entity, cannot exist except as experienced by an individual), but epistemically objective (facts that make statements about them true are independent of the observer’s judgments; for instance, the judgment “John feels pain” is epistemically objective, for what makes it true does not depend on the observer’s attitudes, opinions or judgements) (see: Searle, 1996, pp. 7–9).

In this paper we were interested in the concept of coerced exchange (which is a kind of interpersonal exchange), and not just coercive effect (which might occur independently of the speaker’s intentions, as in the case of misunderstanding). According to the position we presented in our paper, coercion (consisting of coercive intent and coercive effect) is an intersubjective fact. Thus it is not sufficient to look at the beliefs and desires of the recipient of a proposal. Rather, we need to take into account mental states of both individuals—pretty much as in the case of institutions. To illustrate this view on coercion, let us resort to the well-known example provided by Hayek. In his Constitution of Liberty there is a suggestion that a husband can be coerced by his nagging wife (Hayek, [1960] 1978, p. 138). The prediction of our position is as follows. If the wife issues a threat (a proper speech act) and is accompanied with required mental states, then indeed, one can validly speak of a coercive nagging wife. However, the fact that the husband feels coerced (i.e. coercive effect takes place) is insufficient to establish the fact that the husband was indeed coerced into some exchange as the coercive intent on the part of the spouse might as well be missing.

As we believe, these considerations can make an important contribution to the foundations of the Austrian theory of exchange (catallactics). They could provide more precise answers to many questions about the coercive nature of certain institutions (especially of tax system and state’s law) as well as of state interventions. Our results might be also important for an endogenic perspective on property rights in economic theory (according to which property rights should not be assumed as exogenously given in economic analysis). Moreover, they might illuminate entrepreneurship theory and the theory of market coordination. More specifically, it might at least help to understand why Austrians could be justified in their contention that entrepreneurial activity coordinates the market, while coercion discoordinates it. According to Austrians, in the market process, entrepreneurs, through voluntary exchanges, match ends and means of different economic agents (including themselves), thus satisfying their desires. More precisely, if an entrepreneur wants to gain market profit, he or she has to satisfy consumers’ desires. At the same time, Austrians contend that by means of coercion one party satisfies his or her desires at the expense of the other (see: Kirzner, 1973; Huerta de Soto, [1992] 2010, 2009).

Now the question arises: why should our (or Lamond’s) intention-based concept of coercion be particularly suitable for explaining market discoordination? Why should not more permissive concepts of coercion (i.e. the ones that allow for different mental states on the part of the coercer such as knowledge, recklessness or negligence) equally well account for discoordinating moves on the market? It seems to us that the fact that our concept of coercion definitionally requires, secondary cases aside, that a threat posed be a commissive speech act—which implies that the threatening party commits himself or herself to executing the threat in case of the victim’s disobedience—is its virtue rather than its vice. For, it is precisely such a notion of threat that well explains why threats are particularly damaging for market coordination. If, to take another extreme, the concept of threat would allow for threats to be made negligently, such threats would contribute much less to the discoordination of the market process as they would be made as a mere unwitting side effect of the “threatening” party’s action. By contrast, if the threatening party genuinely intends to bring about unwelcome consequences to the victim once the latter does not comply, this ipso facto implies that the former deliberately employs certain means which are aimed at satisfying only his or her preference, while frustrating the victim’s preference. Therefore, it is threats, as understood by us, that would be more likely (than the threats which apparently can be made non-intentionally) to lead to market discoordination.

Moreover, it is worth noticing that the account of coercion developed here sheds some light on the process of price formation on the market. After all, market prices are shaped as a consequence of voluntary exchanges. Moreover, the absence of coercion is a necessary condition of voluntariness of the exchange. And coercion was the main focus of the present paper. Hence, as our account of coercion partly illuminates the concept of voluntary exchange, so does it partly illuminate the process of price formation on the market. We hope our insights will provide an impulse for further fruitful investigations.