Amongst many, though certainly not all, political theorists and economists there is a tendency to believe that in the absence of government, mutually beneficial voluntary economic interactions- and hence property rights- cannot exist, or, if they can, do so only infrequently (see, for instance, Murphy and Nagel (2002); Buchanan (1975); Glaeser et al. (2001); Rand (1967) pp. 329 – 337; Friedman (2002); Epstein (1985) chapter 1; Macpherson (1962)). This view has as its philosophical progenitor Thomas Hobbes, who famously concludes in his masterpiece, Leviathan, that in order to allow for mutually beneficial economic interactions- and thus property rights- a civil authority with the power to create and enforce laws is first necessary. What Hobbes (and by implication most modern political theorists and economists) fails to address adequately is that agents can establish property holdings and facilitate economic transactions in the absence of a government via self-enforcing contracts, particularly given his starting assumptions.
I shall take it as the primary goal of the present work to show that, given the conditions of Hobbes’s state of nature argument, agents are more likely than not to develop mutually beneficial contracts which establish property rights and beneficial economic activity. More precisely, I shall argue that Hobbes’s theory of human nature and the hypothetical state of nature are likely to result in contractual arrangements wherein agents benefit from cooperative interaction, thus nullifying the need for a coercive centralizing authority.
Before we proceed, I should like to offer three clarifications. First, in the analysis property rights are understood as conventions of social interaction. This is not to say that one could not extend the argument to incorporate a metaphysically more robust notion of property rights (ala` John Locke or Murray Rothbard, e.g.); only that for present purposes it is unnecessary to complicate the argument with any more philosophically contentious matters than is necessary.
Second, while my analysis may admit of an anarcho-capitalist interpretation, such an interpretation is not required. Rather, one need only view the conclusion narrowly and see that it consists merely in showing that government is unnecessary in order to establish property rights and mutually beneficial economic activity; the possibility remains that a government of some form may serve as the most efficient means by which to protect property rights and facilitate more complex economic interactions.
Third, though Hobbes authored four works in which he addresses his political philosophy, I shall concern myself only with the Leviathan. The reason for this is simple. Hobbes’s other political treatises, The Elements of Law, Natural and Politic (1640), De Cive (1642), The Elements of Law, Natural and Politic (1651) revised with two supplemental parts: Human Nature, or the Fundamental Elements of Policie and De Corpore Politic, are neither as systematic nor as developed as the Leviathan (1651). Hobbes intended the Leviathan to serve as his political magnum opus and, by all measurements, he succeeded. Therefore, it is within the Leviathan that Hobbes presents his mature political philosophy and it is upon that text we shall concentrate our analysis. So, with one minor exception, I shall glean Hobbes’s entire argument for conflict in the state of nature from the Leviathan.
Throughout the Leviathan Hobbes identifies numerous characteristics which he purports are essential to human beings and which guide their social interactions. To some, Hobbes’s theory of human nature presents a less than accurate general description of man and rather presents a pessimistic portrayal of 17th century English land-owning males, and, as such, is readily dismissible (see, for instance, Macpherson 1962, pp. 32 – 33, 67 – 69), while to others Hobbes’s description of man is almost incontestably true (see, for instance, Gauthier 1969, pp. 14 – 18). However, while one may wish to reject both extremes, in order to proceed to the assessment of Hobbes’s argument, it is first necessary to grant his theory of human nature pro tem, for, if from his initial assumptions Hobbes cannot derive a state of conflict which necessitates a government, or if from his assumptions it can be shown that agents can cooperate in the state of nature, then the Hobbesian argument does not succeed.
For our purposes, then, we shall follow Kavka (1986, pp. 33 – 34) in identifying six primary characteristics which Hobbes employs in his argument for conflict in the state of nature. The six assumptions are as follows:
1. Egoism (L 6.1 – 12; 14.8; 15.16; 27.8; 17.1)
2. Concern for reputation and glory and the influence of various passions (L 11.3; 13.4)
3. Forward-lookingness (L 11.1; 11.24; 17.1)
4. Material scarcity and competition (for finite resources) (L 11.3; 13.3)
5. Equality of physical and mental abilities (L 13.1, 2)
6. Rationality (L 2.10; 3.1 – 12; 5. 1 – 6; 11.24 – 27; 13.4 – 5; 10.2)
Assumptions (2) – (6) are more or less straightforward and require no elaboration. The assumption of egoism, however, is less straightforward and will require comment.
For Hobbes, egoism entails the self-interested pursuit and maximization of subjective preferences. However, as Elster (1979) points out, and contrary to common misperceptions, Hobbesian egoism does not translate into local maximization of preferences. That is, Hobbesian individuals are not necessarily concerned with maximizing preferences in the immediate short run. In fact, as we shall see below in his response to the Foole, Hobbes, to borrow Elster’s term, posits global maximizers, i.e., those who seek to maximize their preferences in the long run.
Admittedly, to some, even when construed in Elster’s limited sense, Hobbesian egoism does not immediately commend itself to acceptance. However, while an in-depth analysis of Hobbes’s egoism is beyond the pale of the present work, the assumption is not unlike the assumption of rational self-interest in neoclassical economics, and, insofar as neoclassical economics provides a useful model through which to understand and predict economic phenomena, one could argue that Hobbesian egoism may also provide a useful model through which to understand and predict political and social interactions. If one nevertheless wishes to insist that Hobbes’s egoism is yet much too unrealistic to serve usefully in even the limited sense of a simplifying theoretical assumption, then we can only request that he grant the assumption so the argument may proceed.
The State of Nature
Beginning, as he does, with his limited psychological assumptions, Hobbes next postulates a state of nature: a social arrangement in which there is no civil authority. To many of his contemporaries Hobbes’s state of nature was taken as an historical description of mankind. As an early critic of Hobbes writes, the state of nature never existed, rather “[t]he primogenious and most natural state of mankind was in Adam before the fall, that is, the state of innocence” (Bramhall 1658, p. 568, excerpted from Martinich (2005), p. 63). However, Hobbes nowhere intends the state of nature as an historical account (though, at the time he believed, albeit incorrectly, that many Native Americans did live in such a state). Rather, as Pasquino (2001) observes, Hobbes introduces the state of nature as a thought experiment (cf. L 13.11).
In conjunction with the psychological assumptions, the state of nature, Hobbes believes, produces a state of affairs that, for its inhabitants, is infamously described as “solitary, poor, nasty, brutish, and shorts” because it would be a condition, as he says memorably in the De Cive, of bellum omnium contra omnes (a war of all against all) (L 13.9).
Conflict in the State of Nature
The analysis of conflict which begins in Chapter 13 consists of a three-part argument, the first of which is the cause of competition:
(P1) Agents are by nature roughly equal in cognitive and physical abilities which engenders the “equality of hope in the attaining of our ends” (Assumption 5)
(P2) There is material scarcity; that is to say, resources are limited in such a way that no two or more agents may satisfy all of their wants (Assumption 4)
(P3) Agents are, primarily, self-interested and seek to maximize their preferences (Assumption 1)
(P4) If agents are roughly equal, self-interested and there is material scarcity, then conflict is inevitable due to competition for resources (Premise)
(C1) Therefore, conflict is inevitable due to competition for resources (From P1 – P4)
Competition for resources entails that the products of an agent’s productive capacities are susceptible to the forcible acquisition of others. For, as Hobbes observes, it seems likely that from the assumption of equality each agent may believe herself able to overpower or outwit her competitor and procure his property (L 13.3). Moreover, from the assumption of egoism, it follows that agents in the state of nature certainly possess the motivation to acquire their competitors’ property via force or subterfuge, and from the fact of material scarcity, violent or coercive interactions will likely occur frequently enough to instill within agents the reasonable fear that at any moment they may lose their property. Undoubtedly, then, agents in the state of nature realize how precarious their property holdings are and are thus conduced to produce only that which is necessary for consumption in the short-term and thus eschew long-term accumulation of goods.
The argument from competition, even if sound, is not sufficient to establish conflict in the state of nature of the sort Hobbes requires, however. It may happen that, due to the desire to avoid death and pain, agents will in sufficient numbers refrain from initiating aggression against their competitors- of course, aggression comes at a cost of personal risk. However, Hobbes proceeds to augment the above argument with another cause of conflict: diffidence.
Hobbes introduces the argument from diffidence as a straightforward extension of the assumptions of egoism, forward-lookingness, and rationality. The primary insight in this portion of the argument is the important realization that initiating hostilities against one’s competitors will, on balance, lead to more preferred outcomes than not initiating. In other words, one can minimize one’s losses or maximize one’s gains by employing an actively aggressive stance in conflict. The argument is as follows:
(P1) Self-interested agents generally seek to minimize losses and/or maximize gains (Assumption 1)
(P2) Agents are apt to recognize the obvious advantages of initiating hostilities (Assumptions 3 & 6)
(P3) If agents are self-interested and aware of the advantages of initiating aggression, then they will, on balance, elect to take an active aggressive stance in the state of nature (Premise from C1)
(C2) Therefore, agents will, on balance, elect to take an active aggressive stance in the state of nature (From 1 – 3)
Diffidence is intended by Hobbes to offset motivations which might counsel one to refrain from violence in the state of nature, namely, the risk of death and pain avoidance. Certainly the reasoning possesses a degree of plausibility, and is moreover entirely consonant with his psychological assumptions. What is more, those who survive in the state of nature are likely to take the offensive or actively defend themselves or learn to do so in short order- particularly given the rationality assumption- since such a strategy is likely to minimize loss, ceteris paribus. However, Hobbes does not need an active and constant state of physical violence and deception. Even if agents choose not to initiate aggression in preemptive attacks, it is sufficient that agents spend an exhaustive amount of time and effort in devising defensive tactics (fortifications, traps, etc.) and thus actively defend their property. The time, effort, and materiel directed toward defensive preparations would prevent production and storage of long-term goods and further prevent commerce.
In fact, Hobbes describes the state of nature as a condition of war which “consisteth not in battle only, or the act of fighting; but in a tract of time, wherein the will to contend by battle sufficiently known” (L 13.8). As Hobbes continues, “In such condition, there is no place for industry; because the fruit thereof is uncertain: and consequently no culture of the earth; no navigation, nor use of the commodities that may be imported by sea; no commodious building; no instruments of moving, and removing such things as require much force; no knowledge of the face of the earth; no account of time; no arts; no letters; no society; and which is worst of all, continual fear, and danger of violent death; and the life of man, solitary, poor, nasty, brutish, and short” (L 13.9). Violence may occur but it is the insecurity of property which ensures the prevalence of discord: “no propriety, no dominion, no mine and thine distinct; but only that to be every man’s that he can get, and for so long as he can keep it” (L 13.13).
Laws of Nature and Contracts
The solution to the conflict in the state of nature is the mutual performance of contracts whose primary condition is that one refrain from violating another’s property and life. Reason advises agents to adhere to contracts since a cooperative social environment is most conducive to their successful survival and enjoyment of property.
In chapter 14, Hobbes introduces his laws of nature, maxims of reason, or, to borrow Kant’s classification, hypothetical imperatives. In short, the laws of nature dictate the means appropriate for attaining one’s desired end(s) (which for Hobbes are subjectively determined, cf. Leviathan chapter six) which in this case is the preservation of one’s life and property. For our purposes only the first three laws of nature have any importance. The first states that when peace is reasonably obtainable, one ought to obtain it, but if peace is not reasonably obtainable, one ought to take any action necessary for the assurance of one’s well-being (L 14.4). The second law, which follows immediately in paragraph five, obliges agents to be “willing, when others are so too, as far forth as for peace and defence of himself he shall think it necessary, to lay down this right to all things; and be contented with so much liberty against other men as he would allow other men against himself.” The third law of nature, as we shall see in Hobbes’s response to the Foole, counsels that one ought to perform one’s agreements made.
The types of agreements which might bring about peace are contracts. Contracts are agreements between two or more agents which transfer a right or rights (or abdicate a right, in this case) from one agent(s) to another (L 14.9, 10). Contracts may be further subdivided into two basic types: covenants and mutual promises. Covenants are contracts wherein one party performs while leaving the other party to perform at a later time (L 14.11). Mutual promises are contracts wherein both agents agree to perform “in time to come” (L 14. 11). In order to facilitate the argument, however, hereafter I will refer to mutual promises as contracts. So, it is advised that the reader take note so as to avoid confusion.
Contracts in the State of Nature
At first blush, resolving the conflict in the state of nature seems a simple matter: Agents need only contract with one another with the purpose of respecting property boundaries. Unfortunately the psychological assumptions are such that, according to Hobbes, though agents might be rational to enter into such contracts, they would not be rational to perform them: “For he that performeth first has no assurance the other will perform after, because the bonds of words are too weak to bridle men’s ambition, avarice, anger, and other passions, without the fear of some coercive power” (L 14.18). The incentive structure is apparently, under Hobbes’s calculation, such that it does not lend itself to initial performance of contracts. As Hobbes says, “bonds, by which men are bound and obliged: bonds that have their strength, not from their own nature (for nothing is more easily broken than a man’s word), but from fear of some evil consequence upon the rupture” (L 14.7).
Prisoner’s Dilemma and Cooperation in the State of Nature
It is as this juncture that we may utilize a simple game theoretic device in order to explicate Hobbes’s reasoning here, a prisoner’s dilemma (PD) scenario. A PD scenario is a two-person, non-zero sum bargaining situation with non-cooperative and cooperative elements and the following normal form payoff matrix:
Left & Right
(The numbers in the matrix are payoffs, with the higher value preferred, for Left and Right, with the left number the payoff for Left and the right number the payoff for Right.)
As we see from the payoff matrix, if Left and Right enter into a contract and, say, Left cooperates, there is good reason for Right to defect. If Right defects while Left cooperates, Right will be able to obtain his optimal outcome: the payoff in the upper-right quadrant. As a rational agent, however, Left recognizes that cooperating induces Right to defect, so she adopts the strategy of defecting. Right employs a similar line of reasoning, and so we have a situation in which the stable outcome is where both defect, the lower-right quadrant. In a single play PD, defection, then, is dominant. That is to say, neither Left nor Right have incentive to adopt a cooperative strategy since defection yields the individually most preferred outcome no matter what the other chooses.
However, in the state of nature agents are merely superficially confronted with a single play PD. If an agent contracts with a competitor for mutual recognition of property, then, as with Left and Right, reason will oblige both agents to defect on the contract. But the state of nature is not a one-time iteration of a PD scenario. Instead, agents continuously interact in a series of PD scenarios; each iteration giving rise to another, indefinitely, in time (with the exception of those on their deathbeds, of course). Furthermore, as Hobbes himself notes in his response to the Foole (more on this below), agents in the state of nature learn from previous social interactions and amend future behaviors appropriately.
If the game is repeated, so as to avoid being subverted by backwards induction, whereas before the strategy of defect dominated cooperation, players are now able to teach their competitors that cooperation is the optimal outcome. Consider a simple example. Left and Right enter into an agreement not to molest the other’s fruit trees (Left grows peaches and Right grows cherries). Let us further suppose that, while undesirable, losing fruit from one tree is low risk (i.e. a non-disastrous outcome since each grows many trees). In the single play PD scenario, the dominant strategy is defect or, as it happens here, to steal fruit. In light of indefinite repetition, though, if in the first trial Left cooperates and refrains from stealing Right’s cherries, but Right yet defects and takes Left’s peaches, it is unlikely that Right will defect in the subsequent iteration and, what is more, it is unlikely that Left will punish Right’s previous act of defection. We may identify two reasons for this conclusion. First, insofar as Right recognizes the potential benefits of cooperation over the long run (thus his entry into the contract), he does not cooperate on the premise that Left may herself be uncooperative. However, since Left has cooperated, Right now has reason to believe that Left is cooperative, and will in the next iteration cooperate. That is, Right is aware that Left, if she is willing to cooperate- and she is per her initial strategy of cooperation-, is aware of the potential benefits of cooperation over the long run and, in all likelihood, will perceive the second iteration as an opportunity for Right to evidence that he has learned his lesson from the first trial. Left’s reasoning here is simple: She knows that losses in a handful of low risk trials is counterbalanced by the gains from indefinite cooperation. In short, by cooperating in the low risk first trial, Left signalled to Right that in the next trial he ought to expect her to cooperate. Come the next trial, it is then extremely likely that both Left and Right will cooperate. In short, Left has revealed to Right that she is prepared to play fair.
Furthermore, Right realizes that, in the long run, the costs of losing cherries to Left via theft and the costs of investing the time and effort into protecting the rest of his property from Left will quickly add up to the point that the costs of continued defection are less than any immediate gains which might be gained from a single defection. One may respond that, though this analysis is true, a scenario like this should not occur since Left is unlikely to cooperate in the first place and thus Right could not never have learned of Left’s willingness to cooperate. But if we recall Hobbes’s assumption of egoism, he does not contend that agents are local maximizers of utility. Rather, he holds that agents are global maximizers; they seek to maximize utility in the long run. Moreover, as Gauthier (1986, p. 114) notes, given the rationality assumption, in the state of nature “we [are] aware of each other as potential co-operators in the production of an increased supply of goods, and this awareness enables us to realize new benefits” which gives agents more incentive to cooperate. Both Left and Right, as noted, are aware of the benefits to be had from cooperation, but they are equally aware that if they prove themselves uncooperative, it is unlikely that other agents will elect to seek mutually beneficial interactions with them (in essence, agents learn from the strategic interactions of other agents; more on this below).
In brief, if the initial risks from cooperation are relatively low and the long term potential benefits great, it is entirely consistent with, and indeed an implication of, Hobbes’s psychological assumptions that Left and Right cooperate in repeated PD scenarios.
Hobbes and the Foole
Prima facie it seems that Hobbes recognizes the benefits of cooperative strategies in iterated PD scenarios. In his famous response to the Foole (15.4 – 8), who maintains that “therefore also to make, or not make; keep, or not keep Covenants, was not against Reason, when it conduced to ones benefit,” he provides an (essentially correct) argument for why one ought not, rationally, to forsake the long term gains from cooperation for the immediate gains of contract violation. To quote in length Hobbes’s central response:
“He therefore that breaketh his Covenant, and consequently declareth that he thinks he may with reason do so, cannot be received into any Society, that unite themselves for Peace and defence, but by the errour of them that receive him; nor when he is received, be retayned in it, without seeing the danger of their errour; which errours a man cannot reasonably reckon upon as the means of his security; and therefore if he be left, or cast out of Society, he perisheth; and if he live in Society, it is by the errours of other men, which he could not foresee, nor reckon upon; and consequently against the reason of his preservation; and so, as all men that contribute not to his destruction, forbear him onely out of ignorance of what is good for themselves” (L 15.5).
Hobbes seems to place the Foole in an awkward position here. The Foole insists that violating a contract is not against reason if the immediate gains therefrom outweigh the immediate gains from keeping the contract. However, Hobbes observes that if the Foole continues to violate contracts, it becomes increasingly improbable that other agents will contract with him and heretofore profitable avenues of benefit cease to present themselves. Thus, it is not unreasonable to adhere to contractual arrangements; indeed, not to adhere is unreasonable.
We may go further and complement Hobbes’s argument by noting that the Foole’s position is dangerously inconsistent. The Foole contends that it is often (if not always) rational to break contracts if the defection produces immediate gains greater than the immediate gains produced via adherence. He also seems obliged to admit that agents enter into contracts in order to obtain a net benefit. Now, if the Foole continues to violate contracts in order to maximize immediate gains, it seems that it is unlikely that his fellow would-be contractors will elect to contract with the Foole, thus robbing him of an opportunity to break it and maximize one’s immediate utility in the process. Hence, if the Foole designs to benefit from defecting on contracts, he will have to be able to enter into contractual agreements. However, if he defects on contracts, he will not be able to enter into contractual agreements. It seems, then, that, at the very least, the Foole cannot, rationally, adopt a general strategy of defection and simultaneously have as an aim the maximization of utility.
It appears that for Hobbes reason obliges cooperation in the state of nature. That is, as we have constructed the situation Hobbes accepts that agents are rationally obliged to cooperate in iterated PD scenarios. Many contemporary commentators concur in this assessment and interpret Hobbes as availing himself of an iterated PD in his response to the Foole (Gauthier (1969) pp. 84, 85; Hampton (1986) p. 76; Ullmann-Margalit (1977) pp. 62 – 73; Taylor (1987) pp. 134 – 136; Kavka (1986) pp 137 – 141). Moreover, some commentators (see, for example, Hampton p. 65) further argue that Hobbes’s response makes first performance in a PD rational. In short, the argument is as follows:
(P1) If it is rational for S to perform if D first performs, then D will reason that he ought rationally to perform first in contracts since S is rational and it is rational for S to perform given D’s performance (Premise)
(P2) If D believes he ought rationally to perform, then he will perform (Assumptions 1 & 6)
(P3) Hobbes’s argument against the Foole shows that it is rational for S to perform if D first performs (Premise)
(C1) Therefore, D will reason that he ought rationally to perform first (From P1 and P3)
(C2) Therefore, D will perform (From C1 and P2)
This argument is not sound, however, since Hobbes’s response to the Foole presumes an initial cooperative play, which means, as Hobbes notes, the situation does not constitute a contract: “For the question is not of promises mutuall, where there is no security of performance on either side; as when there is no Civill Power erected over the parties promising; for such promises are no Covenants: But either where one of the parties has performed already; or where there is a Power to make him performe” (L 15.5 emphasis added). In other words, in Hobbes’s response, unlike in a PD, an agent has already performed his part of the contract. A PD, recall, is a scenario in which the agents are ignorant of the each others’ actions and make their decisions simultaneously; each agent selects his course of action without any knowledge of how the other plays (Raiffa and Luce 1967, p. 94. 95; Harsanyi 1977, p. 276, 277). So, properly, Hobbes’s response to the Foole does not take on the form of an iterated PD scenario because the situation is not, properly, a PD scenario. If we recall, Hobbes distinguishes mutual promises from covenants: covenants are contracts wherein an agent has performed while allowing for the future fulfillment on the part of the other agent, and, mutual promises are contracts wherein both agents intend to fulfill their sides of the agreement at a future date. Moreover, it will be remembered that I intended to use ‘contract’ to refer to mutual promises and that the contracts into which Left and Right entered above were not covenants, rather they were mutual promises: each promised to refrain from taking certain actions against, or with respect to, the other’s property. Therefore, while Hobbes’s argument against the Foole significantly resembles the reasoning behind an iterated PD, it is not an iterated PD argument, and so the argument above, which subsumes as a premise that Hobbes’s response to the Foole makes first performance in contracts rational, does not warrant its conclusion. We may conclude then that while Hobbes considers contract performance rational- it is the subject of his third law of nature- he does not think contract performance is rational in an iterated PD scenario. Though this seems to place Hobbes in an awkward position- in particular given the analysis above which shows that cooperation in an iterated PD is rational and follows from Hobbes’s psychological assumptions- to understand why he may yet maintain that it is not rational to cooperate in a PD, we must introduce the third cause of conflict in the state of nature.
Passions and Conflict in the State of Nature
“For the laws of nature (as justice, equity, modesty, mercy, and (in sum) doing to others, as we would be done to,) of themselves, without terror of some power, to cause them to be observed, are contrary to our natural passions, that carry us to partiality, pride, revenge, and the like” (L 17.2). The extent to which Hobbes relies upon passions to engender conflict in the state of nature is not altogether clear, but we may proceed on the probable assumption that Hobbes intends to base a good deal of conflict on passions and mistaken reason to derive the futility of state of nature contracts which he requires.
In numerous passages Hobbes attributes causes of conflict to the existence of various passions which may serve to pervert rational deliberation and disrupt cooperative interactions (L. 11.6; 14.18, 15.51; 17.1, 2). In particular, he identifies the desire for glory as a primary cause of conflict: “So that in the nature of man, we find three principall causes of quarrel. First, Competition; Secondly, Diffidence; Thirdly, Glory” (L 13.6). Hobbes explicitly provides why the existence of passions proves detrimental to state of nature agreements in chapter 14, paragraph 18:
“If a Covenant be made, wherein neither of the parties performe presently, but trust one another; in the condition of meer Nature, (which is a condition of Warre of every man against every man,) upon any reasonable suspition, it is Voyd; But if there be a common Power set over them bothe, with right and force sufficient to compell performance; it is not Voyd. For he that performeth first, has no assurance the other will performe after; because the bonds of words are too weak to bridle mens ambition, avarice, anger, and other Passions, without the feare of some coerceive Power;”
Though agents may enter into contracts, Hobbes implies that the existence of passions increases the likelihood that one’s fellow contractor will not only defect in the first PD trial, but will also fail to learn to cooperate in the repeated trials, and thus it is not rational to cooperate in repeated PD trials. For instance, to return to the case of Left and Right, this means that even in low risks situations Left may reason that it is sufficiently likely that Right will succumb to irrational passions and not only defect on the first trial, but fail to learn to cooperate in any reasonable amount of iterations thereafter, thus robbing Left (and, of course, Right) of any utility which might be gained from long term cooperation. To compound matters, even if Right were not given to his passions, he ought to reason similarly to Left and defect in the first trial, thus giving Left the impression that Right has defected because of the influence of passions.
Thus constructed, the passions account of conflict is not persuasive. On Hobbes’s own assessment, the distribution of individuals given to passions is very low:
“But this, that men are evill by nature, followes not from this principle; for though the wicked were fewer than the righteous, yet because we cannot distinguish them, there is a necessity of suspecting, heeding, anticipating, subjugating, selfe-defending, ever incident to the most honest, and fairest condition’d” (De Cive 1983, p. xvi).
Of course, as Hobbes notes, it is not necessary that the majority of agents be given to passions in order for the argument from passions to have the appearance of soundness; rather, the mere suspicion that one’s fellow contractor often succumbs to her passions can generate strategies of defection. Nevertheless, that the frequency of agents given to passions (hereafter non-cooperative agents) is relatively low entails that the frequency with which rational agents encounter non-cooperative agents is relatively low. Since the agents will, on average, not encounter non-cooperative agents, certain strategies recommend themselves to would-be cooperative agents which may ameliorate the impediments passions present to keeping contracts.
If we recall our analysis of Left and Right, we will remember that the iterated PD entails that rational agents will learn from previous trials to cooperate. (Indeed, Hobbes himself admits as much in his response to the Foole when he says that when the Foole evidences uncooperative behavior, other agents notice and choose to sever any future cooperative interaction with him.) Hence, it follows that if Left cooperates on the first trial and Right defects, if Right is not given to passion-induced irrationality, then Right will cooperate on subsequent trials (that is, if the trials are repeated indefinitely). What is required of us, then, is that we establish that a rational agent is likely to cooperate in the first trial.
As in the original analysis of Left and Right there remains good reason to expect either Left or Right to cooperate initially. In fact, there is reason to expect that both Left and Right will cooperate initially. In any initial contract, the conditions will outline low risk transactions. As in the case above, losing some fruit in the initial trial of a state of nature contract is far more preferable than losing fruit indefinitely and the utility gained from indefinite cooperation will far outweigh any initial loss which might be experienced in a single trial or in a short initial series of trials. In other words, neither Left nor Right have much to lose in the transaction, but both have much to gain, and a simple cost-benefit analysis will advise cooperation. Moreover, in the case that either Left or Right are given to passions, that they are given to passions will be revealed relatively soon upon subsequent iterations (non-cooperative agents do not learn to cooperate, recall, in even low risk transactions), which brings us to another, perhaps more important insight that makes initial cooperation likely and which me mentioned briefly above: signalling.
Given the passions account, agents in the state of nature are now faced with an information problem, which is to say that agents operate in contractual arrangements without knowing with whom they are dealing: either cooperative agents or non-cooperative agents. Per Hobbes’s argument, rational agents are cognizant of the majority existence of other cooperative agents and the relative minority existence of non-cooperative agents. Now, as previously established, rational agents seek to maximize long term utility via indefinite cooperation and, given the assumption of forward-lookingness, presumably aim to take short run actions which will make indefinite cooperation possible. Faced with this situation, cooperative agents are likely to cooperate in the first trial in which the payoffs are low risk as a means by which to signal to other cooperative agents that they are amenable to cooperation, in much the same way that Left communicates his cooperative intentions to Right by first performance and the Foole communicates his non-cooperative intentions to other agents by his lack of performance. In fact, agents may likely avail themselves of any low risk opportunity to convey their cooperative intentions to others similarly disposed. Through the strategy of signalling in low risk PD, cooperative agents can resolve the problem of identifying other cooperative agents.
Having established the likelihood of initial cooperation on the part of rational agents in low risk agreements in the face of the possibility of interacting with non-cooperative agents, we may safely conclude that the likelihood of long run cooperative behavior in iterated PD is high.
Economic Activity and Cooperation in the State of Nature
We have shown that cooperation in low risk and high reward contracts is rational and thus likely to occur; that is, likely to occur given Hobbes’s initial assumptions. The stability of the cooperative interaction is, I think, more stable than one might at first glance suppose. First, as we noted previously, the social interaction is not a definite series of PD (though it is certainly a finite series), and thus we should not expect subversion via backward induction. Second, cooperative agents will have arrived at a pattern of cooperation, accrued substantial benefit in the process, and likely formed an attitude of risk aversion with respect to any particular contract, thus ensuring continued cooperation. In other words, though an agent may realize that she may gain more utility than usual from any single defection, she also has reason to expect that a defection will induce a defection on the part of her fellow contractor, and, since after any series of iterations of the contract the contract will certainly have delivered heretofore unachievable benefits, it becomes increasingly unlikely that an agent will care to risk losing out on future benefits.
However, recognizing and agreeing not to molest another’s property holdings does not itself translate into economic activity; more is required than mere abstention from property violations. But it is here that the Hobbesian argument presents less of a difficulty. We can, indeed we should, expect agents to take advantage of the opportunities which present themselves in low risk economic activities. In our example above, along with realizing the benefits of not molesting Left’s peaches, Right also realizes the benefits of trading a portion of his cherries for a portion of Left’s peaches. Once low risk economic activity develops, the gains from cooperative interaction increase exponentially. A critic may plausibly accept the argument thus far yet reject its application to riskier cooperative interactions. The criticism, though warranted, is not due to a difficulty or oversight of the analysis. Rather, the analysis sought to show that initial cooperation in the state of nature is not only possible, but likely. More precisely, the analysis sought to show that contractual cooperation entails property rights and introduce the framework through which to realize gains from economic activity. A simple and natural extension of the analysis can show that from low risk contracts, riskier transactions may develop. In brief, if there are greater benefits from riskier transactions (and presumably there are), then each agent will accept a higher degree of uncertainty in the contract, which is to say that agents will, ceteris paribus, cooperate in the face of a greater likelihood of defection. This is not to say, however, that the degree of uncertainty must increase. Rather, through continued interactions in low risk contracts, agents are likely to build a rapport with other agents and thus increase, in a stepwise process, the gravity of the transactions without correspondingly increasing the uncertainty of cooperation in any obvious manner.
Summary and Implications
Thomas Hobbes identifies three primary causes of conflict in the state of nature: competition, diffidence, and glory (L 13.6). Throughout the Leviathan he identifies how each contributes to conflict and why the establishment of a government (in Hobbes’s terms, the Sovereign) is necessary to ameliorate the conflict. But he also notes that “if we could suppose a great multitude of men to consent without a common power to keep them all in awe; we might as well suppose all mankind to do the same; and then there neither would be, nor need to be any civil government, or commonwealth at all; because there would be peace without subjection” (L 17.4). It is precisely that which we have endeavored to show: that, given Hobbes’s psychological assumptions and the state of nature, cooperation amongst rationally self-interested agents is likely to occur, and thus a government is not necessary.
As noted above, it remains as a possible implication of the analysis that a government is the most efficient means by which to facilitate economic activity and recognize property rights. Nevertheless, it also remains a strict implication of the analysis that the prevalent belief that government is necessary for the establishment and maintenance of property rights and the existence of mutually beneficial economic activity is incorrect. The I think unavoidable conclusion is that agents roughly similar to those whom Hobbes conceives, if they are able to extract themselves from the state of nature by engendering a government, are also capable of establishing social rules and norms which govern property holdings and property transactions through low risk contracts. If they are incapable of cooperating in low risk, mutually beneficial contracts, then it is not at all clear how they could cooperate in the creation of a government- the creation of which is certainly a high risk endeavor.
Moreover, if one contends that a particular conception of human agency or human nature makes mutually beneficial cooperation outside of a civil authority highly improbable, it follows that the mutual beneficial cooperation requisite for the creation of a civil authority becomes highly improbable. In a nutshell, those who advance such arguments attempt to have their proverbial cake and eat it too.
If one, as Locke does, wishes to dismiss Hobbes for postulating humans as being unacceptably anti-social, then the argument for cooperation in the state of nature becomes stronger and the likelihood of cooperation greater (1988, IX, 128). Humans are not as anti-social as Hobbes would have us suppose, which serves only as a more forceful impetus to social cooperation in the absence of coercive civil authorities of the sort he would like to engender. The belief that social cooperation cannot occur (or not in enough quantity or not of the proper sort) without a centralizing civil authority is a category mistake: Social cooperation is not political cooperation.
Buchanan, J. (1975) The Limits of Liberty. University of Chicago Press.
Elster, J. (1979) Ulysses and the Sirens. Cambridge University Press.
Epstein, R. (1985) Takings: Private Property and the Power of Eminent Domain. Harvard University Press.
Friedman, M. (2002) Capitalism and Freedom. University of Chicago Press.
Gauthier, D. (1969) The Logic of Leviathan: The Moral and Political Theory of Hobbes. Oxford University Press.
Gauthier, D. (1986) Morals By Agreement. Oxford University Press.
Glaeser, E., S. Johnson, and A. Shleifer, (2001) “Coase versus the
Coasians,” Quarterly Journal of Economics 116: 853 – 899.
Hampton, J. (1986) Hobbes and the Social Contract Tradition. Cambridge University Press.
Harsanyi, J. (1977) Rational Behavior and Bargaining Equilibrium in Games and Social Situations. Cambridge University Press.
Kavka, G. (1986) Hobbesian Moral and Political Theory. Princeton University Press.
Lock, John (1988) Two Treatises of Government. Peter Laslett (ed.) Cambridge University Press.
Luce, R. and H. Raiffa (1967) Games and Decision. Wiley & Sons.
Macpherson, C.B. (1962) The Political Theory of Possessive Individualism: Hobbes to Locke. Oxford
Martinich, A.P. (2005) Hobbes. Routledge.
Murphy, L. and T. Nagel (2002) The Myth of Ownership: Taxes and Justice. Oxford University Press.
Pasquino, P. (2001) “Hobbes, Religion, and Rational Choice: Hobbes’s Two Leviathans and the Fool,” Pacific Philosophical Quarterly 82: 406 – 419
Rand, A. (1967) “The Nature of Government,” in Capitalism: The Unknown Idea. Signet: 329 – 337
Skinner, Q. (2002) Visions of Politics Vol. III: Hobbes and Civil Science. Cambridge University Press.
Taylor, A.E. (1965) “The Ethical Doctrine of Hobbes,” in Hobbes Studies. K.C. Brown (ed.) Basil Blackwell: 35 – 55
Taylor, M. (1987) The Possibility of Cooperation. Cambridge University Press.
Warrender, H. (1957) The Political Philosophy of Hobbes: His Theory of Obligation. Oxford University Press.
Watkins, J.W.N. (1965) Hobbes’s System of Ideas. Hutchinson
Ullmann-Margalit, E. (1977) The Emergence of Norms. Oxford University Press.