Matt Corks
mvcorks [AT] alumni [dot] uwaterloo [dot] ca
7 April 2000
The theory of political realism holds that without a world state, warfare is an intractable part of the international system. The argument is that in the absence of a benevolent and all-powerful world government to protect the interests of states, states must be concerned primarily with their own national self-interest, and should therefore not let mere moral arguments dissuade them from a course of action necessary to maintain their sovereignty. To support the underlying assumptions about human nature in this argument, realists often turn to the writings of Thomas Hobbes, and in particular his work Leviathan. However, in doing so, they ignore sophisticated contemporary models which describe conflict resolution between egoists and predict the emergence of mutual cooperation. Specifically, the model of the Iterated Prisoner's Dilemma can be used to counter Hobbes's basic claims about conflict in the state of nature. Since these claims are central to realist writings, this has profound implications for the doctrine of realism as a whole.
We will begin by outlining the essential argument of Hobbes's Leviathan, and refuting certain of his conclusions using a model from game theory. Next, a contemporary Hobbesian answer to these objections will be discussed, and itself refuted.
Thomas Hobbes penned Leviathan, the “crowning achievement of his political science” (MacPherson, in Hobbes, p. 22), in 1651. In the section on political philosophy, he begins by claiming that a basic characteristic of humans egoism, that is, “individuals are primarily concerned with their own well-being, and act accordingly” (Kavka, p. 33). He claims that all humans have a “general inclination” toward “a perpetual and restless desire of Power after power, that ceaseth only in death” (Hobbes, p. 161). From this, he concludes that “during the time men live without a common Power to keep them all in awe, they are in that condition which is called War; and such a war, as is of every man, against every man.” (Hobbes, p. 185) From this he reasons that there must be a state, in fact, one with absolute power. The alternative, he says, is “continual fear, and danger of violent death; And the life of man, solitary, poor, nasty, brutish, and short.” (Hobbes, p. 186) He also specifically refutes the belief that moral arguments can mitigate this suffering, when he writes “To this war of every man against every man, this also is consequent; that nothing can be Unjust. The notions of Right and Wrong, Justice and Injustice have there no place. Where there is no common Power, there is no Law: where no Law, no Injustice. Force, and Fraud, are in war the two Cardinal virtues.” (Hobbes, p. 188)
So, Hobbes has claimed that all people are egoists, and that if left to their own devices they will therefore be forced to spend their time and effort fighting in defense of their life and property. He explicitly rejects the idea that two egoists could maintain a non-aggression pact between themselves, since it will be in the interests of both parties to renege on their agreement. This will therefore lead to a violent anarchy ungoverned by any norms.
From this he concludes that “the only way to erect such a common power as may be able to defend from the invasion of foreigners and the injuries of one another,... is to confer all their power and strength upon one man, or upon one assembly of men,... in such manner as if every man should say to every man I authorise and give up my right of governing myself to this man, or assembly of men” (Hobbes, p. 227, emphasis in original). This all-powerful government, or Leviathan, is able to restrain the natural drive for power amongst people and allow them to live productive lives. Whatever else one may say about Hobbes, it must be conceded that his writings imply that even the most self-centred egoists will wish to bring about a peaceful world for their own benefit, and that he describes a world in which anarchic chaos is the exception rather than the norm. We will see later that those same realists who support their views with quotes from Leviathan do not reach the same conclusions as Hobbes himself.
The main objections to Hobbes's writing at the time were religious in nature (MacPherson in Hobbes, p. 737), but since then a strong atheistic response has become possible. The argument is known as the Prisoner's Dilemma, and is a part of game theory, a multidisciplinary science developed in the past fifty years. It proceeds as follows.
Consider the case of two egoists must make a decision. Both may choose between the moves C or D (cooperate or defect). The payoffs to each will be as shown in the payoff matrix below.
| C | D | |
| C | (R,R) | (S,T) |
| D | (T,S) | (P,P) |
In this matrix, each pair (v1,v2) represents the value of a particular outcome to player 1 or 2, with higher numbers indicating a preferred outcome. This is a strictly ordinal game, meaning that T ≠ R ≠ P ≠ S. In the Prisoner's Dilemma, 2R > T+S. Typical values are those used by Axelrod, 1984, as shown below.
| C | D | |
| C | (3,3) | (0,5) |
| D | (5,0) | (1,1) |
It is immediately clear that each player's strictly dominant move (that is, the one which would be best regardless of their opponent's move) is D. For example, player 1 would receive 5 if player 2 cooperated and 1 if they defected, and thus player 1 is better off in either situation than if they had played C. In a one-round prisoner's dilemma game, the rational move for both players is to pick D. The outcome {D,D} is the Nash equilibrium of this game, meaning that it is the only state from which each player could only do worse by unilaterally changing their move. Choosing this outcome leads to a paradox, however, since it is Pareto-inferior to {C,C}. That is, every player is at least as well off in {C,C} as {D,D}, and at least one is better off. In the language of game theory, Prisoner's Dilemma as a model is of interest because it is the only strictly ordinal 2×2 game that has a Nash equilibrium which is Pareto-inferior to some other state.
It is only in the iterated case when the move D is not superior. When the game is iterated over a large number of rounds, trust may be established between players over time which would lead to cooperative strategies. For several years there had been debate in the literature of game theory regarding optimal strategies for the Iterated Prisoner's Dilemma (IPD), until Axelrod held a tournament between several strategies, submitted in the form of computer programs (Axelrod 1985). Similar tournaments have since been done to test various strategies under a wide variety of conditions. To eliminate possible end-game effects in tournaments between various strategies, Axelrod's second tournament (and those of his successors) now end the game with some small probability p after each round (such as ((1)÷(1000))). The most successful strategy in both of Axelrod's tournaments was Tit-For-Tat (TFT), submitted by Anatol Rapoport. This strategy was also the simplest in the tournament. It always cooperated on the first round, and then proceeded to copy its opponents' last move. In this way, it defected on defectors, and cooperated with cooperators. Axelrod saw the chief advantages of this strategy as follows: it was ‘nice’ (since it opened by cooperating), it was provokable (since it never cooperated with an opponent which had just defected), it was forgiving (since it returned immediately to cooperation when its opponent did), and it was simple, so that its opponents could understand its behaviour easily. Other strategies have since been found which outperform TFT under various circumstances, but they have all shared these properties (Nowak and Sigmund 1992, 1993).
Many variations on IPD have since been explored using similar tournaments. Some of the most interesting are spatial games, in which various strategies are fixed on a grid and only interact with a few neighbours, games with imperfect communication, in which a strategy intending to play C may occasionally play D by mistake, and tournaments involving genetic algorithm techniques, in which strategies can gradually evolve and change over time. This last technique corrects for the main flaw in Axelrod's initial tournaments: that they took place between an arbitrary subset of the many possible strategies for IPD, determined solely by the creativity of the contestants. The goal of this recent work has been to more closely simulate the processes of biological evolution. According to Hobbes's theories, Darwinian evolution and its mechanisms for selection which lead to the survival of the fittest would lead to a war of all against all, with no room at all for cooperation. However, game theorists have reported that even in variants of IPD carefully designed to simulate of evolutionary biology, cooperation emerges in almost all of their trials given sufficient time, and is evolutionarily stable (Nowak and Sigmund 1992, 1993; Nowak, May, and Sigmund 1995). We therefore have good reason to believe that even self-interested egoists may be able to cooperate and work together for their mutual benefit, rather than expend their energy in competition with each other.
Game theory notwithstanding, Hobbes and his writings are not without their contemporary apologists. One who has attempted to specifically refute these implications of the Prisoner's Dilemma model is Gregory Kavka. However, his arguments betray a lack of familiarity with the literature of game theory in general and IPD in particular.
To follow Hobbes, instead of ‘cooperation’ and ‘defection’, Kavka employs the terms ‘lying low’ and ‘anticipation’. He defines anticipation as the willingness to “use force aggressively on occasions when it seems worth the risk”, and lying low as avoiding the use of force “except in immediate defense” (Kavka, p. 121, emphasis in original). Like Hobbes, however, he states that an anticipator will be preemptive wherever possible (Kavka, p. 96), which makes this functionally identical to an aggressive strategy. Lying low is described as a strategy that “provides no assurance of safety other than the hope that no one stronger will ever have the occasion to attack them” (Kavka, p. 122) We may therefore read Kavka correctly substituting ‘aggression’ or ‘defection’ for ‘anticipation’ and ‘cooperation’ for ‘lying low’.
Kavka attempts to answer the objection raised by the Prisoner's dilemma by stating that “anticipation leading to a worse outcome for all does not imply, as the objection assumes, that it is not the ‘most reasonable’ strategy for each to follow.” (Kavka, p. 112) He writes that “so long as individuals remain in that state [of nature] and act rationally, they will inevitably produce worse outcomes for themselves than they could obtain under other conditions,” (Kavka, p. 113) but this presumes a definition of ‘rational’ that implies an overriding interest in immediate gains. This ignores the fact that reasoning humans could see for themselves that a different approach could ultimately be more beneficial. In response to evidence of the emergence of cooperation from evolutionary biology, he states that “People [are] intelligent and have language; hence they are generally much more capable of coordinated action than are other animals. As a result, humans are especially vulnerable to others of their own kind acting in concert... even if it generally pays other animals to lie low in dealing with [others], it may well pay people to anticipate in the state of nature”. (Kavka, p. 122) This, too, is an extremely pessimistic reading of human nature, which conveniently ignores a number of anthropological studies (Nowak, May, and Sigmund, 1992). It is also not faithful to Hobbes, who saw other animals as being in the ultimate state of nature, hence his choice of words. Kavka acknowledges that there are anthropological examples which contradict his claims, and explains them by saying that the people in question are “so poor and malnourished that no one has much to steal or defend, and they cannot afford the energy expenditure of violent conflict” (Kavka, p. 123) He concludes that Hobbes's state of nature is characteristic of environments which “neither too poor nor rich”, meaning that there is enough material wealth to make stealing worthwhile but not enough to remove scarcity as a source of conflict. Again, this differs greatly from Hobbes's claim that there is a “perpetual and restless desire of power” in all humans.
These arguments all work by misapplying the IPD model. To strengthen his claim, Kavka also attempts to give several reasons why the IPD model is not rich enough to be applied usefully to human conflict. Of these, the five most important will be examined below. Although some of these are valid objections to IPD as presented above, all can be refuted with a few further enhancements drawn from the literature of the IPD.
1.) If two egoists are playing an IPD with a fixed number of rounds, both are likely to defect in the last round, since their opponent will have no way to punish them. Having concluded that their opponent is likely to defect on them in the last round, however, both players are likely to defect on the penultimate round, in an attempt to obtain the payoff T. Again, since each player knows that this is likely to happen, they will defect on the move before that. The logical extension of this process is that neither will cooperate even in the first round of the game. This can be generalized to cases in which the players do not know the number of rounds, but do know that this number has some upper bound u, in which it is in their interest not to cooperate. Each knows that there must be a (u-1)th move, in which they should also not cooperate, and will proceed by mathematical induction to defect in the first round and thereafter. (Kavka, pp. 130-131)
There are several flaws in this argument. Firstly, in most actual conflicts, it is highly unlikely that any player would be able to predict the number of interactions with their opponent. As in computer tournaments, the end of the game is likely to come as a surprise. Secondly, a rational player can see that the payoff for both themselves and their opponent will be T × u if they cooperate and P × u if they do not, and so they will prefer to play cooperatively.
2.) People occasionally act irrationally in various ways. For example, if they “devalued long-term benefits relative to short-term ones”, they would not be able to take advantage of mutual cooperation. (Kavka, p. 134)
This point has been discussed at length in the literature of game theory. In all iterated games, it is assumed that players will prefer a gain now to a long-term gain, with the factor w representing the discount per round of future benefits. This factor is known as the shadow, or discount, of the future. For example, with w = 1, the sequence {C,C}, {C,C}, {C,C} is worth 3 + 3 + 3 = 9, but with w = ¾, it is 3 + (¾)×3 + (¾)²×3 = 6.9375.
To properly explain the game theoretical response to this, it is necessary to now introduce a concept from evolutionary biology. We say that a new (or mutant) strategy (or organism) M invades a population of a native strategy N if the new strategy receives a higher payoff with a native strategy than a native strategy receives with another native strategy. If this is not true, the old population will be able to resist the new strategy, and it will die out. Symbolically, we write V(M|N) ≥ V(N|N). From this, it is straightforward mathematically to show that if 1 ≥ w > 1/2, then V(AllD|TFT) < V(TFT|TFT), where AllD is a strategy which unconditionally defects (see Appendix). In other words, a population of conditional cooperators (such as TFT) will be evolutionarily stable (that is, will be able to resist invasion) by an unconditional defector if each player considers benefits in the next round at least half as valuable as those in the current stage of the conflict.
3.) The IPD model is unrealistic in that it assumes that every possible matching of players in a population will occur. In fact, particular pairs of individuals may interact very few times, and the model cannot properly be applied to these situations. (Kavka, p. 147)
An empirical response is appropriate to this objection. Several studies have been done of spatial games, in which each player is fixed on a grid and interacts with its closest neighbours by playing an IPD. The sum of its scores in these games is its total score for the round, and strategies that do well take over less successful cells. Although the outcomes are dependent on the precise values of the parameters T,R,P,S, and w, all of the authors found that cooperation can still emerge in many circumstances, in both IPDs (Axelrod, 1985) and, interestingly, even single round PDs (Nowak, May, and Sigmund, 1992).
4.) In the state of nature, individuals may have difficulty identifying each other, and thus will not be able to establish trust with specific opponents. (Kavka, p. 147)
Again, several simulations have been done which attempt to mimic this inaccurate identification and other random errors. For example, Nowak and Sigmund used genetic algorithms to study strategies defined as a pair (p,q), which would respond to a cooperation with cooperation with probability p, and to a defection with cooperation with probability 0<p,q<1. In this case, they found that the most stable strategy was the TFT variant generous tit-for-tat (GTFT), which had p=1 and q=min{1-((T-R)÷(R-S)),((R-P)÷(T-P))}=⅓, and so would forgive a defector with probability ⅓. Their later study involved strategies defined by the vector (p1,p2,p3,p4), which correspond to the probability of cooperation after receiving the payoff R, P, respectively. To ensure at least a minimum amount of randomness, they fixed 0.001<pi<0.999. In both experiments, they found that a degree of random noise still allowed the emergence of cooperation (Nowak and Sigmund 1992, 1993).
5.) Finally, a player may take into account their opponent's previous behaviour towards third parties. If you cheat a person, others may observe this and withdraw their trust of you. (Kavka, p. 147)
This objection misunderstands a central premise of the IPD, namely, that it is a game played amongst self-centred egoists. As such, they have no reason to care how their opponent has treated third parties. For them to care would imply the existence of a universal norm in the state of nature against exploitation shared amongst all players, which is precisely what Hobbes and Kavka are attempting to refute. It is not that the IPD cannot be applied to such a situation, but that it is superfluous.
Having rejected Hobbes's claim that egoism will inevitably lead to violent chaos, we must also question his conclusion that only after ceding sovereignty to an all-powerful government can entities cease their futile war of all against all, in which nothing can be unjust. First, however, it will be established that Hobbes's claims concerning the actions of entities in the state of nature are central to contemporary realist writing. These writers agree with Hobbes's claim that all humans are driven by a lust for power, and that this can only be avoided with the creation of a Leviathan. They proceed to argue that the formation of a Leviathan is fundamentally impossible in the current international system, and that an amoral war of all against all is therefore inevitable. Two influential American realist thinkers will be discussed: George Kennan and Hans Morgenthau.
George Kennan, a former ambassador to the Soviet Union and Yugoslavia, gave a series of four lectures on American foreign policy at Princeton University in March, 1954, which were later published as Realities of American Foreign Policy. They serve as a succinct introduction to the objectives of American foreign policy. Although they do not deal with philosophy as such or Hobbes's writings in particular, the basic elements of his theories are evident in many of Kennan's unstated assumptions about the nature of humans and of the international system. They are of interest because, as a member of the American state department, Kennan not only advocated but actually employed these policies.
He begins his lectures by defining American foreign policy as a means to the “general objects of American society”, that is, American citizens (Kennan, p. 5). He sees the purpose of the American state as protecting “life, liberty, and the pursuit of happiness” and states that “progress and improvement... [are] most apt to ensue if men were left as free as possible to pursue their own self-interest and happiness” (Kennan, p. 7). A true liberal, he sees “the objects of government [as] primarily protective and subsidiary” (Kennan, p. 9). He contrasts this with the purpose of the Soviet state, which he sees as the “cultivation of a given social theory” and facilitation of its dominance (Kennan, p. 6).
Given that the reason for American foreign policy is to further the individual goals of its citizens, he sees it as having two primary objectives. The first is to “protect... out national life from any external military or political intrusion” (Kennan, p. 11). It should hardly come as a surprise that an objective of the American state to ensure its own continued existence as an institution. Kennan's second objective has interesting implications, and will be dealt with separately.
Kennan describes the types of means necessary for a state to preserve its sovereignty. He agrees with Hobbes that humans and therefore states have a tendency to lust for power. For example, he states that American expansionism and the Monroe doctrine of the nineteenth century were necessary to save America from being conquered by larger European countries which were also seeking to expand (Kennan, pp. 13-15). This type of justification is typical of the Hobbesian: they explain their own excesses of exploitation by explaining that had they not done them first, these acts would have been done to them. Kennan sees diplomatic concessions to “world peace” in general as precisely the opposite of national interest (Kennan, p. 21), indicating that the US cannot afford to let its diligent armed guard over its national interests lapse for a moment without loosing ground in the international war of all against all. As a Hobbesian, he believes that true peace could only follow from the formation of an effective world government, but states that American interests could never be guaranteed by a democratic body such as the United Nations. The problem, to him, is that the UN is numerically dominated by “smaller, newer, and less-developed countries” who can outvote the US, NATO, and the Soviet bloc combined (Kennan, p. 40). American apologists of the UN, he states, have ignored the problem of these countries, and “taken for granted... their predisposition to ourselves”, implying that impartial judgement should not be acceptable to Americans (Kennan, pp. 40-41). As long as it is not in the national interest of the US to surrender its sovereignty to any consortium of foreign states, there can be no great Leviathan which will create a world without war. In the meantime, the United States cannot make progress without a reputation for military strength, and “learn how to create a respect for our possible disfavor at least as great as the respect for our favor” (Kennan, p. 58).
Moral and ethical arguments against military interventions overseas are unfounded for Kennan, since the State department, as an agent of the American state, should be morally accountable to American citizens and not to foreigners (Kennan, p. 47). Individual citizens may allow morals to temper their actions, but when a state is formed, individual behaviour “undergoes a general transmutation, and the same moral concepts are no longer relevant to it” (Kennan, p. 48). He advocates using any means necessary, and in particular advocates strict control by the State department over “those facets of American activity from which people in other countries tend to profit” (Kennan, p. 59).
Kennan's second purpose of American foreign policy is very interesting. He states “that insofar as the activities of our citizens in pursuit of their private interests [spill] over beyond our border and into the outside world, the best possible arrangements [should be] made to promote and to protect them.” (Kennan, p. 11) Taken to its logical extreme, this second statement implies that any country in which an American citizen has invested capital should not be allowed to protect its national life from external, American military or political intrusion. Does Kennan's theory allow for such an extreme reading? The historical evidence available indicates that this is the case. Americans have consistently and insistently attempted to mould the economies of foreign nations, especially those in the developing world, into forms which would be convenient for those of its citizens with investments abroad. Particularly blatant examples of this include the Opium Wars in China and the American-sponsored 1954 coup d'état in Guatemala. Although Kennan, attempting to distance his country's policies from Soviet expansionism, states that “we [Americans] did not come along bearing in our hands a patent medicine of social reform which we were prepared to recommend to all comers as the cure for every ailment” (Kennan, p. 13), it is simply not true that his country has not had an agenda in international development. Kennan wholeheartedly endorses the Marshall plan (Kennan, p. 55-56), which pioneered the awarding of aid with stipulations attached. His government has long attempted to use aid packages and other means to make foreign economies subservient to American economic imperialism. This is clearly in the interest of American investors, and like Adam Smith, Kennan assumes that it is therefore in the interest of the American populace at large, via an invisible hand.
Kennan also justifies military and other interventions for economic purposes on the grounds that access to certain foreign raw materials has become essential to American business (Kennan, p. 51). He goes as far as to state that the Monroe doctrine should be extended to include any country that produces something which the US depends upon, and that the State department should tell foreign presidents that “they must not permit our great and in many cases delicate economy to become dependent on them... unless they are prepared to acknowledge a clear obligation to guard the durability and reliability of the respective arrangements”. In cases where this obligation is not taken seriously enough, or when the government in question is not strong enough to ensure continued access to this resource, any military means necessary should be employed to protect the American economy (Kennan, p. 53). He acknowledges that in many cases, the producing countries are dependant upon American trade for their economic well-being, but states that this alone is not a strong enough assurance for the continuation of trade (Kennan, p. 51). This statement ignores the fact that foreign governments often have no economic choice other than to sell their raw materials to Americans.
This willingness to go to war for economic gain is only ever justified by appeal to Hobbesian egoism. Kennan's book never finds time to consider the possibility that by surrendering immediate gains for the sake of international cooperation, American citizens may find themselves more secure from military, political, or economic invasion. An end of the Hobbesian war of all against all is almost by definition in the interests of all parties in the international system, and yet Kennan considers any move towards this, such as drafting the most basic international treaties, a waste of time and effort (Kennan, pp. 18-19). He openly states that American diplomats should never accept any compromise which requires them to cede away power, although both Hobbes and the model of the Iterated Prisoner's Dilemma show that by doing so, an egoist can greatly increase their own rewards. This refusal is perhaps motivated his extreme distrust of the Soviet Union, and fear that if his country hesitates for a moment it will lose its power to them. His motivations for this struggle are, for an avowed pragmatist, surprisingly ideological.
Morgenthau, at least, does not exaggerate Soviet power to the same degree as his colleague Kennan. Although also firmly grounded in what he sees as pragmatism, Hans Morgenthau's defense of realism is developed with a more elaborate basis in theory in his Politics Among Nations than Kennan was able to in his lectures. This book a transcript of a course in international relations he taught in the Political Science department of Princeton University. In it, he develops a detailed theory of national character which leads him to his judgements concerning foreign policy. These theories have their basis in Hobbes's Leviathan, which Morgenthau quotes directly to support his claims that all states have a natural drive towards power and conquest. This can only ever be mitigated by the creation of a world state, which he sees as impossible given the nature of the international system. This leads him to conclude that only a balance of power between two expansionist countries can forestall war, and that when war comes it must be wholly amoral. Only frank and secret negotiations between two career diplomats with no illusions as to the strengths of their countries, or the means available to countries seeking to maintain their power, can possibly avert war on the international scale.
While struggling to build a definition of the term ‘imperialism’ which would exonerate the United States and Britain, he quotes Hobbes's account of how the unlimited desire for power affects citizens and heads of state, and leads them to attempt ever larger gains (Morgenthau, p. 36). He eventually states that imperialism is any expansionist policy which seeks to overthrow the “existing status quo”. This means that to conquer, whether militarily, economically, culturally, or otherwise, is only imperialistic if one's intention is to change the balance of power in a region (Morgenthau, pp. 47-48). By way of example, he states that Spain's involvement in South and Central America after 1800 was not imperialistic because it was not meant as a threat to American military supremacy in the region, and that British foreign trade policies are not imperialistic, because they “aim at economic survival through favorable trade balances, not at the maintenance or acquisition of political power over foreign nations” (Morgenthau, p. 48). It follows that imperialism is any expansionist policy which is not sufficiently moderated by domestic security concerns, and that this is the reason that such imperialists as Hitler, Genghis Khan, and Alexander the Great have ultimately failed (Morgenthau, p. 36). It is this concern for national security which makes the balance of power an effective means of preventing or limiting war. Morgenthau writes that “the balance of power and policies aiming at its preservation are not only inevitable, but an essential stabilizing factor in a society of sovereign nations” (Morgenthau, p. 125).
Like Hobbes, Morgenthau sees the state as essential for maintaining domestic peace, but he adds that it is not sufficient. Without support and a popular mandate from society, the state will collapse into revolution and civil war. The formation of a state requires that internal conflicts be suppressed and overridden by internal loyalties and external threats, and that a strong and impartial enforcement mechanism is known to exist to carry out the state's wishes. Morgenthau argues that an international state is impossible because neither of these exist (Morgenthau, p. 398). States are unwilling to agree on issues or part with their independent sovereignty (Morgenthau, p. 401), will never cede power to an international democratic body because, unlike in domestic parliaments, the numerical relational between minority and majority “does not in any sense correspond to the actual distribution of power and interests among the members of the UN” (Morgenthau, pp. 434-435).
Another example of the Hobbesian nature of Morgenthau's arguments is the section of his book devoted to discussions of the role of international morality as a limitation on international power. A true Hobbesian, he begins by stating that the it is human nature to lust for power, and that the “function [of] normative systems has been to keep aspirations for power within socially acceptable levels” (Morgenthau, p. 169). This means that the fervour of the war of all against all has been offset to some degree by concerns of international morality, since those who seek power must temper their goals with the appearance of morality (Morgenthau, p. 169).
However, to Morgenthau, this alone is not sufficient to avoid war. To explain this, he gives the example of relations between states in Europe between 1600 and 1900. He states that the degree of peace and cooperation achieved was due to a common moral sense shared between a small aristocratic class of rulers and career diplomats. These rulers were accountable only to themselves, and could thus take a moral stance in state affairs without the fear of losing power (Morgenthau, p. 184). They often had more in common with each other, such as ties of kinship, than with their subjects. More importantly, decisions of state could be properly ascribed to a single person with a certain set of objectives, and so a degree of trust and understanding could develop between countries. This, he argues, cannot happen in a democracy, because heads of state change with the whims of the electorate. A similar degree of trust is impossible with the growth of nationalistic standards of action. “Where responsibility for government is widely distributed among a great number of individuals with different conceptions as to what is morally required in international affairs, or with no such conceptions at all, international morality as an effective system of restraints upon international policy become impossible” (Morgenthau, p. 189).
This argument, of course, contradicts the realist claim that introducing moral arguments into debates concerning war will lead to moralistic crusades with greater bloodshed and unlimited wars (Morgenthau, pp. 194-195). As well, even if we conveniently ignore the almost constant territorial wars fought in Europe between 1600 and 1900, it remains to be established how the populace of a nation, who invariably suffer more from war than their head of state, will call for more wars than an aristocratic ruler personally immune from suffering. Instead, Morgenthau states that democracy has lead to the growth of total war, meaning that its industrialization has made civilians both legitimate (i.e., ‘good’) and possible targets, and nationalistic universalism, meaning the belief that one's one ‘way of life’ is somehow superior and should thus be adopted worldwide (Morgenthau, pp. 181-182). Although he concedes that there are things which are universally held as immoral by the general public, and hence heads of state, these are very limited, since the nations of the world do not share a common moral sense. The only examples he gives of these are mass extermination and assassination (Morgenthau, pp. 175-177). However, Morgenthau still feels that morals can only hinder the advancement of national interest in international relations. He writes that in a democracy, “when war comes, it must come as a natural catastrophe or as the evil deed of another nation, not as a foreseen and planned culminations of one's own foreign policy. Only thus might the moral scruples, rising from the violated ethical norm which holds that there ought to be no war at all, be stilled, if they can be stilled at all.” (Morgenthau, p. 181)
Like Kennan, then, Morgenthau agrees with Hobbes that people, and thus states, are egoists, but disagreed with Hobbes's claim that the overwhelming benefits of a world government are enough to make the nations of the world give up their power and sovereignty to one. Instead, he encourages nations to use any means necessary, regardless of moral qualms, to satisfy their immediate national interest. This includes the economic subjugation of a country. He recognizes that a people conquered with the dollar are just as subjugated as a people conquered with the sword.
To accept realist doctrine requires that one accepts without question Hobbes's claims regarding the character of interactions within the state of nature, yet believes that relations between states are so fundamentally different from relations within a country that only half of what Hobbes wrote can be applied to it. That this view is convenient to those who would profit from American economic imperialism is obvious; that it is short-sighted is demonstrable. Mutual co-operation between states, and aid freely given without stipulation, can bring about an end to the war of all against all, without the existence of an omnipotent and yet benevolent Leviathan whose impartiality we must take on faith. A beginning to this is to stop using that part of Hobbes's argument which predicts a war without morals. With the tools of game theory, this claim is today provably false, and its continued use by the self-fulfilling prophets of the American state department only delays that happy day in which humankind can enjoy a more just and more peaceful international system than those profits can provide.
For the case w = 1, both V(AllD|TFT) and V(TFT|TFT) are infinite, but we can evaluate the limit of the inequality:
V(AllD|TFT) < V(TFT|TFT)
⇒ T + P + P + ... < R + R + R + ...
⇒ (T-R) + (P-R) + (P-R) + ... < 0
Given that T>R>P, it is clear that as the number of rounds approaches infinity, the series (T-R) + (P-R) + (P-R) + ... will approach negative infinity. So, for < V(TFT|TFT), as required.
Assume now that w<1. The payoff for TFT against TFT is:
V(TFT|TFT) = R + Rw + Rw² + ...
w<1, so this series converges, to R(((1)÷(1-w))). The payoff for TFT against AllD is:
V(AllD|TFT) = T + Pw + Pw² + ...
w<1, so this series also converges, to T + P(((1)÷(1-w))) - P.
Therefore, if AllD cannot invade TFT, then
V(AllD|TFT) < V(TFT|TFT)
⇒ T + P(((1)÷(1-w))) - P < R(((1)÷(1-w)))
⇒ T - P < (R-P)((1)÷(1-w))
⇒ ((T-P)÷(R-P)) < ((1)÷(1-w))
⇒ w > 1 - ((R-P)÷(T-P))
So, if 1 ≥ w > 1 - ((R-P)÷(T-P)), then AllD cannot invade TFT. For Axelrod's values of S=0, we have 1 ≥ w > ½. QED.
Axelrod, Robert. The Evolution of Cooperation. New York: Basic Books (1984).
Axelrod, Robert and William D. Hamilton. “The Evolution of Cooperation.” Science 211 (27 March 1981): 1390-1396.
Hobbes, Thomas, C.B. MacPherson, ed. Leviathan. Aylesbury: Hazell, Watson & Viney Ltd., (1651: 1972).
Kavka, Gregory. Hobbesian Moral and Political Theory. Princeton: Princeton University Press (1986).
Kennan, George. Realities of American Foreign Policy. New York: W. W. Norton & Co. Inc. (1954: 1966)
Morgenthau, Hans. Politics Among Nations. New York: Alfred A. Knopf, Inc. (1949).
Nowak, Martin and Karl Sigmund. “Tit for Tat in Heterogeneous Populations.” Nature 355 (16 January 1992): 250-253.
Nowak, Martin and Karl Sigmund. “A Strategy of Win-stay, Lose-shift that Outperforms Tit-for-tat in the Prisoner's Dilemma Game.” Nature 364 (1 July 1993): 56-58.
Nowak, Martin, Robert May, and Karl Sigmund. “The Arithmetic of Mutual Help.” Scientific American (June 1995): 76-81.