ch4

The Causes of War

The previous chapters’ sketch of the history of war and of views about war-making left largely implicit the question of why those wars occurred, but entirely ignored the more general question – if it is a question at all – as to what is the cause of war: that is, why is war a pervasive and recurrent phenomenon in history?

To provide insight into why any individual war occurred, one would need a detailed historical examination of the circumstances in which it arose. In the outline given in previous chapters we only see kinds of reasons. They include: competition leading to conflict over resources such as territory, water, women and slaves; raiding by nomads on settled people to expropriate supplies; wars of expansion, involving conquest of territories and peoples to gain wealth, power and aggrandisement. And of course there is the other side of this coin: wars of defence against aggressors and invaders; desire for revenge or restitution, including irredentism; and religious and ideological wars. The list of reasons is various and long. As explanations for why any given war occurred – though they are not automatically, if indeed very often, justifications for that war happening – these reasons are familiar enough, and do the work of explanation adequately. Some writers add to this list, as a further and even less appetising explanation, the notion of ‘love of war for its own sake’, as noted in Chapter 1 in connection with the Huns, Mongols and others for whom war was a way of life and an end in itself as a central aspect of their society’s economic activity. In these cases the idea of manhood might be implicated in the imperative to make war, a way of achieving status by killing an enemy.

But is there something deeper that lies behind this variety of motivations, explaining in its turn why other means are not chosen for resolving or adjusting resource questions, or achieving the ends implied in some of the above list? One obvious suggestion is that humans are, or at least the male human is, innately violent; that war-making is built into human nature. Another and very different suggestion is that the way human societies are organised is the reason why war occurs. The first suggestion is a pessimistic one, in that if war is an expression of something natural or genetic in human beings, and not, as the second more optimistic suggestion has it, an accident of the way humans do things socially and politically, then we seem to be doomed to war – and eventually to be doomed by war. The second suggestion offers a way out, for it suggests that culture or politics is the cause, and could be within the power of intelligence to change.

There is much that has been said about both suggestions and others in between. I survey these ideas as follows.

In all discussions of war there is a ready and natural association between the concept of war and the concept of violence as a human propensity. Some distinctions and questions are required here. War is of course violent: blowing up people, vehicles, buildings, whole towns and cities, and killing and injuring people – both deliberately as when opposing combatants are targeted, and unintentionally as in ‘collateral damage’ – are violent events. But a group of generals sitting round a planning table inspecting maps, submariners watching plots on an ASDIC screen, politicians considering military options in response to a crisis, are not behaving violently. Neither are troops loading lorries, air force personnel refuelling aeroplanes, sailors scrubbing decks. Violence occurs at the sharp end of conflict, sporadically and intensely, and a commonplace of soldiers’ reminiscences is that war consists of a great deal of boredom punctuated by episodes of terror and chaos.

The trope ‘war is violent’ easily slides into ‘war is the expression of violent propensities in human nature’. As these remarks show, much of war is calm and deliberate, boring and routine. Something like the violence of a saloon bar brawl occurs when enemies come face to face, but even when one sees footage of fighters firing on enemies from rooftops or ruined buildings, say, the violence in train is not a matter of those fighters’ being in a crimson rage, of ‘feeling violent’ or behaving as a saloon brawler does: they are aiming deliberately, protecting themselves, working in concert with others, and whatever may be their sentiments about the enemy whom they are seeking to kill, their hatred, contempt, fear or indifference does not make them act as a saloon brawler does. Indeed it would almost certainly be fatal to them if they did so. Unquestionably, shooting to kill another human being is an act of violence. But the connection between violent feelings, rage, the need to attack, hit, gouge, bite and kick – the emotions of the saloon brawl – are far from being the apparently overriding emotions of warfare, except in hand-to-hand combat.

There is furthermore the question of trauma that the violent events of war cause in those who experience them. I deal with this in the next chapter, but it is relevant to anticipate that discussion by asking: if human beings are naturally violent, why does violence upset them, mark them, even derange them psychologically? Perhaps most people are capable of anger and aggression under provocation that is severe enough, but at the same time most people are not naturally murderous. It has been suggested that in conscript armies, such as the Kitchener armies of the First World War, only about 20 per cent of the drafted soldiers were effective – in the sense that they actually pointed their rifles at the enemy and fired to kill – and that effective soldiers had a similar psychological profile to recidivists in prison, in having scant remorse for past actions, and being little affected by the memory of occurrences which would upset most people.

These thoughts at least suggest that to locate the cause of war in the human emotional capacity for violent feelings and actions is not a straightforward matter. Ethological studies of chimpanzees show that bands of males fight each other much after the manner of a saloon bar brawl, chiefly for territorial and resource reasons.1 As mentioned in Chapter 1, anthropological studies of stone-age tribes show that conflict for the same kinds of reasons can result in fighting, though confrontations are often gestural and symbolic rather than actual, and sometimes fighting stops as soon as blood is drawn. Both kinds of conflict are organised, both do or can involve actual violent behaviour just as when opposing soldiers fight hand-to-hand; but they do not count as war, for which something non-violent seems to be required, namely, an accumulation of social, material and conceptual structures that contribute to the organised and orchestrated series of events that is war as such.

The fact that this is so – that war has a major non-violent component – of course makes it worse. Cold calculation aimed at destruction and death seems even less excusable than the outburst of rage that makes someone throw a punch in a saloon bar. A significant point accordingly offers itself. This is that ‘the origin of war’ only partly lies in the human capacity for moving from emotions prompted by disagreement or rivalry to threats and actual fighting, for what is required in addition is a sufficient degree of relevant social organisation.

And evidently the level of organisation in question has a reciprocal relationship to war-making. War both invites and requires organisation. Societies might get organised because of the experience of conflict or danger, as a response to it or the threat of it; a society needs to get organised in advance of going to war, as preparation for it. Social organisation does not of course by itself lead to war; inducements to war in competition and rivalry, threat or interference, duties of alliance and more, are the proximate causes of individual wars. But without a pre-existing level of organisation of the relevant kind, it is not war that a state or group engages in. It might come close; it might be an insurgency or rebellion, or an act of cross-border hostility perpetrated by a non-state agency. But the level of organisation of a state or state-like agency (such as, and typically, the contending parties in a civil war) is in those cases lacking. Manufacturing weapons, training soldiers, stockpiling resources: this goes on apace in peacetime.

If this in turn were right, it would shift the emphasis, when we ponder the origins of war, from matters of human psychology and its frailties to other matters, and arguably to two principal ones: to questions of competition over topics of importance to the contending parties, and to differences in culture and belief where the differences are perceived as unacceptable or threatening.

I return to these points below. Whether or not they have a place in explaining the origins of war, once war is a fact of history we see that its organisational aspects are part of the fabric of states and societies. Training of personnel; construction and maintenance of barracks, airfields and naval dockyards; the manufacture of uniforms, weapons, ammunition and other equipment; scientific and technological research for their improvement; logistical arrangements; the financing of a continual state of military preparedness and advancement of capability – all this is institutionalised, and the sheer expenditure of resource and effort thus represented carries within it the seeds of a danger, recognised by many – for example, Richard Cobden in the nineteenth century and President Dwight D. Eisenhower in the twentieth century – that such a degree of investment in the machinery of war itself makes war more likely. It is accordingly cited as one of the precipitating causes of war – and sometimes perhaps the chief cause, as suggested in the case of 1914.

Once war is in the offing, in the approach to it and in the course of it, the question of emotion enters in a different way, and focally involves management of psychological attitudes to war on the part of soldiers and civilians. This ranges from encouragement of patriotic sentiments, the idea of dulce et decorum est pro patria mori (‘it is sweet and right to die for one’s country’), the ceremonial honouring of the dead of past wars, the theatre of uniforms and parades, the importance of regimental histories and achievements in battle, to the encouragement of feelings of hostility and revulsion towards the enemy. War would not be possible without arousing and directing hostility against others to the point of willingness to kill them, not just on the part of the military personnel who carry out the killing, but on the part of enough of the population to support and applaud the military in its endeavours, while themselves being prepared to accept various kinds of sacrifices also – such as being bombed, giving up various civil liberties for the duration of the war, accepting privations such as rationing, and dealing with the deaths of family members in the armed forces.

In this latter case, the typical manoeuvre is to collectivise the enemy as ‘Other’ and to invoke not just a sometimes real and sometimes imagined threat to one’s own security and values, but to exaggerate the enemy’s atrocities, crimes and enormities. For example, in encouraging anti-German sentiment in the First World War much was made of the execution of Edith Cavell, an English nurse in Belgium accused by the Germans of spying. But this was the least of it: claims that German soldiers skewered babies on their bayonets during episodes in which civilians were rounded up and shot in Belgium and northern France in the early months of the war, received widespread currency after publication in 1915 of False Witness, an account of the German invasion of those countries by Danish writer Johannes Jørgensen.2

The aid given by the press to the British war effort in the First World War is indeed a paradigm of this. In addition to massacres and cruelty to children (it was alleged that German soldiers mutilated children before their parents’ faces by amputating their hands and slicing their ears off), great outrage was caused by the claim that a Canadian soldier had been crucified on a barn door with German bayonets through his hands and feet. Prime Minister Asquith set up a commission of enquiry under Lord Bryce, the ‘Committee of Alleged German Outrages’, and it duly reported in 1915 that German troops had committed civilian massacres, rape, looting, and the aforementioned barbarisms against children. Along with enraged responses to Zeppelin air raids, the sinking of the Lusitania, and the use of poison gas in the trenches, these demonisations prompted outrage in the United States and other neutral states at the moral vileness of Kaiser Wilhelm’s Germany.

Bad things were undoubtedly done, but the propaganda technique of encouraging pro-war sentiment in one’s own country by claiming that the enemy commits atrocities is both too tempting to resist and too commonplace to ignore. Its effect is to make going to war against a supposed such enemy more acceptable, providing a moral justification and the requisite preparedness to make and accept sacrifices.

What of the suggestion that war is the outcome of a natural propensity in humans, or at least male humans? Humans are primates, indeed ‘great apes’, and as background it is of interest to note the nature of organised violence among fellow great apes. Note the expression ‘organised violence’. Contrary to Mark Twain’s assertion that Homo sapiens is the only species that kills its own kind, there are many animal species that do so. For example, males of various species might kill each other in rutting combat, though among ruminants it is more usual for one of the combatants to accept defeat, suffering a resultant drop in testosterone, while testosterone surges in the victor, who has a herd of females to impregnate. Male lions invariably kill the cubs in a pride when they assume control of it after ousting its previous leader; this is to bring the females into oestrus. Ant colonies fight great battles, and appear to control their own population by culling some of their own number. More robust chicks in the nest can kill other chicks to monopolise the food provided by parent birds. Rodents eat their own young when food is scarce.

Although in the case of ants there is concerted activity in fighting off invading ants, or in culling their own numbers, conspecific killing is in general individual-to-individual or parent-to-offspring. Reports of male chimpanzees forming groups to attack and kill other chimpanzees introduce a different set of questions, not least about the implications for understanding human aggression.

Anthropologists Cláudia Sousa and Catarina Casanova argue that the evidence of aggression and conflict among contemporary great apes, plus the evidence from fossil remains and ethnography, does not support the idea that early hominins were ‘killer ape-like creatures’.3 The implication is that humans are not genetically programmed to be killers either. Observation of primate communities both in captivity and the wild has led Douglas Fry to the conclusion that co-operation, empathy, tolerance and impulses to reconciliation after conflict are fundamental characteristics of our closest genetic relatives, even if, as in the case of humans too, these behaviours show an in-group bias.4 This conclusion emphasises features different from those emphasised by a number of other palaeoanthropologists and primatologists. Richard Wrangham and Dale Peterson, the latter a colleague of Jane Goodall whose work they quote, show in their studies of chimpanzees (Pan troglodytes) and bonobos (Pan paniscus) that male chimpanzees murder, rape and form gangs to commit acts of out-group aggression. Nevertheless they temper their account of the ‘demonic male’ by showing also that the social arrangements of the other Pan species, bonobos, have eradicated almost all such behaviour.⁵ What is at stake in such contrasting views is the question of whether biological theories, rooted in the concept of evolution, provide a clue to the deepest underlying level of explanation as to why war occurs. In the essays collected by Fry for War, Peace and Human Nature, the idea that violence and war have roots in human nature as such are questioned both by the ethological observations cited and by the paucity of palaeoanthropological and archaeological evidence for war in pre-agricultural societies.⁶

The personal inclinations or wishes, the optimism or the cynicism, of those examining the question undoubtedly have a role in which features they select for emphasis. For most of the history of thinking about war, from the Greeks to contemporary theorists, the dominant theme has been that human nature is responsible for war. The misanthropic Hobbesian view has prevailed because the impact of war in history makes war salient; like mountain peaks in a landscape wars are more visible than the years and decades of peace, in which far less of note stands out. Religious conceptions of human nature, most notably those premised on doctrines of original sin and the fallen character of human nature, abet this view. Accordingly, one has to be alert to the fact that the culturally dominant view of human nature, in long having consisted in the pessimistic view that it contains all the darkness required to amplify greed, selfishness, cruelty and aggression into war as such, lies in the background of much of the debate.7

There is of course an important assumption buried in the idea that human nature is responsible for war, which is that there is such a thing as ‘human nature’. Of course in one sense the characteristic features of human beings and their societies are readily enough contrasted with (say) zebras and their social arrangements, to give content to the idea that something answers to the concept of humanness. But the question is whether the putatively essential constituents of humanness are such that we can see among them the factors, if such there are, that are war-inclining ones. The problem is that theories of human nature are cultural ones, philosophical ones, and the sciences – for naturally one would look to anthropology and genetics for objective answers if they exist – do not settle the matter. The evidence pushes both ways.⁸ Which way it pushes – to stress the point – might be heavily influenced by some very human prejudices and beliefs.⁹

It is, however, with the scientific debate that one does well to begin. Even those who are sceptical about direct inferences from primate studies and evolutionary theory agree that biology has something to offer the study of war and conflict in general. Two areas of debate, sociobiology and biopolitics, have been particularly active, made all the more urgent now that war, in the extreme – to which it could alas rapidly escalate in various plausible scenarios – threatens the human species itself. Sociobiology owes itself to E. O. Wilson in his influential 1975 book of that name; the idea of biopolitics has an even earlier origin in Thomas Landon Thorson’s 1970 book, also titled for its concept.10 But both strands of thought in turn have their roots in the application of Darwinian ideas to social and psychological explanation, put forward almost immediately after Darwin’s own seminal publications in the second half of the nineteenth century, and thereafter developed and, notably, parlayed into unattractive and not infrequently dangerous theories about race and eugenics.

The first application – or misapplication, as it is widely recognised – of Darwin’s views was to social and economic questions. The ‘Social Darwinists’ of the nineteenth century used the phrase coined by Herbert Spencer, ‘survival of the fittest’, to justify their claim that the capitalist class system of their own day was the best one because it was the adaptive outcome of a struggle that eliminated less fit social and economic orders.¹¹ It was inevitable that evolutionary ideas would very soon be used to justify ideas about the superiority of ‘Caucasians’ over the ‘Mongoloid’ and ‘Negroid’ races, and further still – and therefore – to justify ideas and even actual programmes (as in Nazi Germany) of eugenics to promote ‘superior’ human stock while eliminating ‘inferior’ stock.

A principal influence on theories of eugenics was the man who coined the word ‘eugenics’ itself, Francis Galton. He was Darwin’s cousin, and was profoundly influenced by him, especially by his remarks early in The Origin of Species about ‘variation under domestication’, that is, selective breeding of dogs, horses and cattle. Galton instantly saw the application to human beings. He was a polymath, explorer, and innovator in mathematics as well as social anthropology. His studies of heredity and statistics together led him to formulate the concept of genetic ‘regression to the mean’. He argued that high-born people should be incentivised to have many children early in adulthood, for they tended to marry late and have few children, which he regarded as dysgenic. Under his influence a Eugenics Education Society was founded to encourage eugenic breeding and to discourage ‘inferior’ people from having children. From 1909 onwards it published a journal, the Eugenics Review.12

The excesses of Nazi eugenics programmes, which involved sterilising or simply killing the ‘unfit’, so regarded either on racial or disability grounds, were the logical extension of such practices as the US restriction on immigration after the First World War of Slavs and Jews on the grounds of their supposed racial inferiority.¹³

Advances in genetics inspired a spate of popular books in the 1960s arguing that those aspects of human behaviour that so often led to war – competitiveness, aggression and territoriality – were biologically programmed. The most widely read of them were by Konrad Lorenz, Desmond Morris, Robert Ardrey and Lionel Tiger.¹⁴ Collectively they asserted the view that humans, or at least human males, are innately programmed to use aggression and violence – and by extension therefore, war – in furtherance of the ends of territoriality and competition. The idea appears to be a completely natural inference from the fact, for fact it is, that in the animal world generally many, if not all, aspects of behaviour are genetically determined: mating displays and competition, female nurturing of newborns, guarding of territory, feeding patterns, and the rest. Sociobiology had begun as an extension of ethology – the study of animal behaviour – to incorporate insights from biology in the explanation of that behaviour; this was simply a matter of bringing humans into the same explanatory framework. The tendentious point is easy to identify. E. O. Wilson explicitly subordinated cultural factors to genetic ones; culture, he said, is ‘held on a leash’ by genes, and is in fact ‘a circuitous technique by which human genetic material has been and will be kept intact’.15 The similarity to Richard Dawkins’ Selfish Gene thesis – though in other respects these two evolutionary biologists disagree about much – is closer than Wilson elsewhere allows.¹⁶

The first problem for views of this kind is how they can explain altruism. It is a datum that people sacrifice their own interests and even themselves for others; this is a commonplace of human experience. Darwin was conscious of the difficulty, and as a result of pondering the question why bees have evolved to die when they deliver a sting in defence of the hive, concluded that what mattered was the hive as a whole, which benefited from the individual’s sacrifice. This later came to be understood in terms of the gene pool, and different terms were coined to capture the idea; ‘kin selection’ by John Maynard Smith and ‘inclusive fitness’ by W. D. Hamilton.¹⁷ Wilson himself thought that there were several strata of levels at which selective pressures have their effect, from the gene up to populations and even ecosystems. He argued that altruism is mainly determined by cultural factors, which themselves have evolved as a result of kin selection, and that kin selection is the ultimate explanation for war: ‘The force behind most warlike policies is ethnocentrism,’ he wrote:

warfare evolved by selective retention of traits that increase the inclusive genetic fitness of human beings . . . [who are] strongly predisposed to respond with unreasoning hatred to external threats and to escalate their hostility . . . We tend to fear deeply the actions of strangers and to solve conflict by aggression’.

Wilson also held that we have a genetic predisposition to learn communal forms of aggressive behaviour, which in its fullest form is war.18

Leaving aside the strenuous disagreement between biologists over the concept of ‘group selection’ now favoured by Wilson, is anything like the view just described right? On the face of it, it appears plausible to see humans as social animals wired to be co-operative with fellow in-group members and aggressive towards out-groups.¹⁹ But for all the surface plausibility, the idea faces objections from a range of sources, in science, social science, ethics and politics. The most obvious is a rebuttal drawn from the very point just invoked, about humans as social animals. This implies that the genetic determination of human behaviour is towards co-operation, not aggression. In all societies and for the greatest part of the time, most people – including most males – do not engage in violent behaviour, still less in warfare; and even most of those who take part in warfare are more likely to be followers rather than fighters, ‘sheep not wolves’.²⁰ Visible all around us are the evidences of co-operation and trust-keeping, in the cities and systems, cultures and institutions built by the social impulses of human beings. In comparison to the normal forms of interaction between people every minute of every day everywhere in the world, war and even lesser forms of violence constitute a minority occupation.²¹

Nevertheless, wars happen often enough, so if a propensity to war is not genetically determined, why does it happen? The answer offered by the non-determinists is: culture. In just the same way as cultural development has resulted in many co-operative behaviours of a positive kind – science, art, literature, music, education, noble aspirations and acts – so circumstances can prompt culture to negative outcomes – disagreement, hostility and conflict prompted by tribal, national, cultural and ideological differences, leading to the organisation of means to give effect to that hostility in war. No aspect of social behaviour has been reliably identified with any specific set of genes, so the inheritance of social structures and memes must occur through cultural transmission.22 Campbell talks of ‘sociocultural evolution’, in which altruism and its extreme in self-sacrifice, ‘including especially the willingness to risk death in warfare, are, in man, a product of social indoctrination counter to rather than supported by genetically transmitted behavioural dispositions’.²³ The genetically transmitted dispositions would predispose to self-preservation, a powerful instinct with an obvious adaptive utility in evolutionary terms, and might accordingly support the idea that because genetics would select for self-interested behaviour, the opposite – altruistic behaviour – must be culturally acquired.

The idea that culture can override genetic predispositions can be supported biologically, so its ability to do this – even indeed its existence – does not have to seek support from anything non-biological as its source. Large brains, high intelligence, long childhoods and language more than adequately explain both the fact of culture, the diversity within it, and the changes that culture and cultures historically undergo. No other species compares. And this includes maladaptive as well as positive aspects. Indeed the idea that humans can both value truth and tell lies suggests a single underlying basis for both: intelligence.

Further, it has been argued that there is no basis for treating the idea of kin selection as the explanation for in-group loyalty and out-group hostility. This would only be plausible if humans were fragmented into genetic subgroups within which there had evolved kin-preferring behaviours specific to that group; but there are no human genetic subgroups. ‘Race’ is a fiction, all human beings are closely related genetically, and superficial differences in skin tone, hair type and eye shape are of recent origin in the human story.24 Ethnic differences exist, of course, given obvious cultural variations in language, religion, customs and food preferences. But ethnicity is not race.²⁵ On this view it is a function of ethnic identification to prefer and privilege one’s own tribe or nation over others, not a matter of genetic propinquity.

Arguments for and against the idea that war is a product of kinship in fact predate genetic theory. The sociologist Ludwig Gumplowicz argued in his briefly influential 1885 work Der Rassenkampf (The Struggle of the Races) that there will always be contest and conflict between groups or races until one subjugates the other. A ‘folk-state’ or nation, once it has achieved internal homogeneity, will ‘naturally’ seek to conquer neighbouring nations, he claimed: ‘Conquest and the satisfaction of needs through the labor of the conquered, essentially the same though differing in form, is the great theme of human history from prehistoric times . . . It cannot be otherwise, since man’s material need is the prime motive of his conduct.’ A follower of his, Gustav Ratzenhofer, summed up the most extreme form of the theory in two sentences: ‘The contact of two hordes produces rage and terror. They throw themselves upon one another in a fight to exterminate, or else they avoid contact.’²⁶

Racialist theory of this kind is polygenist, and it is the polygenism that is taken to explain the natural antipathy that occurs when different ‘hordes’ (races, thus putative separate genetic groups) meet. In their day these theories were novel and appeared to have the force of science behind them. But they were criticised by contemporaries nevertheless. I. A. Novikow wrote:

It was [formerly] believed that men fought their fellows in order to obtain food, women, wealth, the profits derived from the possession of the government, or in order to impose a religion or a type of culture. In all these circumstances war is a means to an end. The new theorists proclaim that this is all wrong. Men must of necessity massacre one another because of polygeny. Savage carnage is a law of nature, operating through fatality.27

Here ‘fatality’ means (natural, genetic) determinism.

Novikow offered two counters to such a view. First, until his own recent times there had been no race wars because people did not have the concept of race and accordingly did not perceive themselves as racially different from others or as fighting against a different race. Second, people of the same race fight each other: Swedes, Danes and Germans are all Teutons but have battled each other fiercely at times, while some very different peoples, for example the Welsh and the Tehuelche of Patagonia, get along very well.²⁸ These observations refute the Gumplowicz theory at a blow.

Paradoxically, the idea of kinship actually contains the germ of a powerful argument against the idea that conflict and war have a genetic basis, namely, that the development of society has occurred because people have transcended ethnic boundaries to include others in the sphere of moral interest and concern. Moreover, most states that have gone to war with each other in the past two millennia have been highly genetically mixed; it would seem that loyalty to the community rather than actual genetic kinship is the chief factor in play. Once again, that makes genetic factors subordinate to cultural ones.²⁹

A final thought is that if fighting and war are genetically determined, they would be adaptive evolutionarily. Another way of putting this is to say that they would be observably rational and productive from the point of view of the species and its members’ reproductive chances. It is not clear that they are either of these things always or even most of the time.

Some critics of sociobiology see it as the political inheritor of the Social Darwinist mantle in at least the sense that it justifies the status quo in society, supports stereotypes of (for example) male and female behaviour, and provides grounds for differential value judgements about (for example) white and black people. One of the sharpest critics of genetic behavioural determinism is the evolutionary biologist Richard Lewontin, who sees the tendency of sociobiological theory as politically conservative.30 Genetically underpinned conservativism is in essence pessimistic, implying that what happens is not within the control of human thought and choice, and that therefore it has to happen because we are programmed for it.

But manifestly, matters are otherwise. Take the case of ethical and political views: the treatment of animals, the social status of women, attitudes to people with sexualities that fall outside the narrow conventional heterosexual model, have all changed significantly as a result of ethical reflection; and the growth of more democratic forms of government in many polities, and with them respect for human rights and civil liberties, has largely displaced monarchy, aristocracy, feudal systems and forms of tyranny. If genetic determinism ruled human behaviour it is hard to see how cultural evolution could take place, but it obviously does. This would seem to imply that war is a cultural, not a genetic, phenomenon, and is or at least should be under the control of human choice. And this in turn would mean that there is a genuine possibility of removing war from human interaction.

If war is a cultural artefact, why and in what way is it? On the principle that ‘if your tool is a hammer everything looks like a nail’ we have just seen that psychologists and biologists seek the causes of war in psychology and biology; now we can expect that economists will see war’s causes as economic, while political theorists will see them as political. And so indeed they do. But the ‘hammer’ observation does not entail that any of these approaches is wrong. The right immediate inference to draw is that the explanation is unlikely to lie wholly within just one of them. After all, the drivers of economic activity are psychological; psychology, in turn, is biologically based, and economic activity involves political debate and decision, so the picture is sure to be complex. Nevertheless, examining the question from the point of view of sociology and anthropology – from the point of view of culture – is a more inclusive vantage point, not least because it bears on the question of human nature more informatively.

One prompt for the idea that war is a product of culture is that the evidence suggests it entered human history only about 10,000 years ago, along with settlement and agriculture. If war were a genetically encoded behaviour, so the argument goes, there would be palaeoanthropological and archaeological evidence for it from long beforehand.31 Moreover changes in social and political conditions have seen changes in the frequency and kinds of war, with smaller percentages of populations taking part in them and dying in them over time, which is another reason for thinking that culture determines whether and how war happens.

A point of contention arises here, between those who hold that war entered history with culture and those, such as Steven Pinker, who say that the numbers of those who have died violent premature deaths in war have declined over time as percentages of populations.32 This is because the former resist the implications of the view that war was more frequent in earlier periods. In fact evidence of the historical decline in war deaths is more than consistent with the cultural hypothesis; cultural factors are here being seen to act as much against war as in other periods or circumstances it acts for it. Where the disagreement between R. Brian Ferguson and Pinker lies is in the matter of whether war as such predated the rise of culture.³³ On the evidence of the case assembled by Ferguson, war would indeed seem to be a feature of humanity’s latest 10,000 years, and practically absent beforehand.

There is an added and even more compelling point in favour of this view in the argument that, for 99 per cent of the 2 million years that a ‘recognisable human animal’ has existed, that animal was a hunter-gatherer, and therefore anthropological data from hunter-gatherer societies is centrally relevant to the question.³⁴ And indeed the data are convincing: nearly a dozen forager societies studied variously in Malaysia, Tanzania, Australia and southern India are peaceable, anti-aggressive, compassionate, co-operative and generous, with very low rates of conflict and homicide. When problems occur, typically caused by sexual jealousy or theft, they are mediated and resolved by discussion.³⁵ These data relate to intragroup relations; much the same appears to be the case with intergroup relations among foraging peoples.³⁶

Another consideration to bear in mind in preferring a cultural approach, when remembering the ‘everything is a nail to a hammer’ point, is that if we were to take just one field of enquiry as constituting the main explanatory resource, we must be careful not to miss anything of importance that the radar of that theory is not calibrated to register. For example, until recently, political theorists tended to focus on interstate conflict when debating the causes of war, but it has been argued that civil war is more common than interstate war in history, and that civil wars, as is also the case for tribal, ethnic and colonial wars, have causes different from interstate wars, even if there are some commonalities too.

Moreover, some civil wars have become ‘internationalised’ by the intervention of other, and sometimes major, parties; and regional wars have sometimes been proxy wars for larger powers; and both these phenomena are characteristic of most conflicts in south and west Asia and the Middle East since the Second World War. So it is no longer possible even to see civil and interstate wars separately in unmixed fashion. The international and the internecine scenes are both messy, and too often merge into just one even messier scene, as in the case of Syria following the ‘Arab Spring’ of 2010–11. The danger is that a local, perhaps internal, conflict might become internationalised in its locality, only to then explode into a major international conflict well beyond that locality, even at worst into another world war. One thinks of the Balkans in the period 1912–14. Given the state of the world’s arsenals and what they contain, such developments threaten disaster on an unprecedented and perhaps terminal scale.

The moral when thinking about what causes war is that one has to be alert to the types of conflict under discussion, and their interactions, and that the likelihood is that this will change the nature of the question from ‘What is the cause of war?’ to ‘What are the causes of war?’, where this latter still treats ‘war’ as a sufficiently unitary phenomenon in human affairs to require a general explanation, but which recognises that the causes of any given war are a permutation of a number of factors out of a larger number of war-causing factors. This also accordingly means that the question is still not the same as ‘What are the causes of this and that particular war?’, for that question has to be asked about each war on its own merits, whether or not there is an answer to the general question ‘What are the causes of war?’

If this general question has no general answer of the form ‘The causes of war are . . .’ then there are two possibilities. Either each war will be sui generis, with its own unique causes that are different from all other wars. Or it might turn out that one can only venture answers to the question why types of war happen, where the causes are specific to the types. And now one can see why the general question still presses: in either case men (almost always men) will equip themselves with instruments capable of causing death, and will go out to slay men on the other side, and to impose themselves on the survivors in one or another way. The question really is, why is war the resource fallen back upon when other means of adjusting differences or conflicts have failed? And maybe the real question then is: why do those other means fail? It is easy to see, from this point in the process of getting the question right, why even on a cultural approach considerations from psychological and biological theories might come back into focus.

In the Introduction to this book, war was defined as ‘a state of armed conflict between states or nations, or between identified and organised groups of significant size and character’. The second half of the definition is designed to take account of the fact that non-state actors can be parties to war, and that war predates the existence of states as such – indeed, by a long way. It might be better to expand the definition to substitute ‘political entity’ for ‘state’, if ‘political’ is understood in the broadest sense to embrace almost any organised grouping of people. The parties to a war are always groups, not individuals, and moreover groups with cohesion, a common purpose, leadership, division of responsibilities, and a sharing of hardships and, if they come, benefits. One might not think of the Huns as a state, but they were a political entity in the foregoing sense, and the sustained periods of fighting between them and those they attacked, not least the Romans, cannot be described other than as war.

Clausewitz took the paradigm of war to be interstate war because that is where the political nature of what is at stake is clearest. In such wars there is a measure of symmetry, in that contending states have similar forces with similar weapons, and the state itself monopolises in its own interests the sanctioned use of force that can be applied internally and externally. The wars in the two centuries preceding Clausewitz’s writing of On War (1832) fitted this model. The point of the (over-quoted) definition of war that he premised on this model, that war is ‘an extension of politics by other means’, is well illustrated by an exchange between an American colonel and a Vietcong colonel after the Vietnam War, in which the United States was defeated. The former said to the latter, ‘Your side never beat us on the battlefield,’ to which the latter replied, ‘True but irrelevant.’ It was the political result alone that counted.37

In the last chapter it was noted that major wars have become less frequent over the last five centuries, though their lethality has increased dramatically. In the last five centuries almost all the major wars were fought in Europe, then the dominant quarter of the world whose states were expanding their reach both locally and globally, and therefore contesting one another. Since 1945 the former Yugoslavia, Georgia and the eastern Ukraine have seen hostilities, but Europe has been free of war until this time of writing, in major part because of the European Union, whereas elsewhere in the world civil wars have proliferated in the aftermath of decolonisation, and a combination of great power intervention and sectarian strife has precipitated the Middle East and west Asia (Afghanistan and parts of Pakistan) into what seems like perpetual conflict.38

Moreover, the form of war has changed: insurgency, civil war, asymmetric war, trans-state war, have mainly replaced the clash of states since 1945, though that trend could reverse quickly enough – the irredentism of Russia in Eastern Europe, or of China in the South and East China Seas, or further heightening of tensions on the India–Pakistan border, are all plausible scenarios for major interstate wars to break out. Some of the practices involved in asymmetric warfare, where insurgents use terror and cruelty to coerce local populations (Daesh in Syria, Iraq and elsewhere provides an example), and where powerful missile and drone armaments fail to control ‘collateral damage’ in efforts to deal with fighters who are often indistinguishable from, and anyway supported by, local populations, raise human rights concerns as serious as conventional war, and sometimes more so. But in any case, methods of coercive terrorisation are as old as war itself – Waffen SS troops massacring villagers in retaliation for resistance activity, Soviet troops raping their way towards Berlin, Daesh publicly beheading ‘infidels’, are no different from the Mongol strategy of unnerving enemies by acts of vast cruelty as happened at Samarkand in 1220 CE and elsewhere; and the practice was an ancient one even then.

Since the seminal work of Kenneth Waltz in the late 1950s, students of war and international relations have used a conceptual approach known as the ‘levels of analysis framework’ to think about the causes of war. In the basic form of this framework three levels are identified: the individual level, the state level, and the international level.39

The individual level is the level of political leaders and the outlook and interests of individual members of society, as well as of appeals to facts or supposed facts about human nature, as discussed above.

The second level is the level of the state. The nature of political and economic institutions and practices are the focus here. Political systems and ideologies, decision-making processes, the relation of leaderships to populaces, the degree of economic equality or otherwise in the system, the role of bureaucracies and lobbies, all require inspection. A government might make external war to distract the populace from internal failure. Arms manufacturers might be a powerful lobby, urging state leaderships to increase and perhaps even use their stocks of weapons. Popular sentiment might motivate a government to take aggressive foreign policy stances even to the point of war, over perceived or claimed ideological, territorial or historical grievances. On the other hand, the nature of social, economic and political organisation in a state might dispose it more to peace than belligerence.

The third level is the international order. As the failures of the League of Nations and the United Nations demonstrate, the international system is practically anarchic. In a situation where each of the players has very little external constraint on its behaviour apart from war being made upon it by another player whom it has sufficiently provoked, the factors that invite consideration include how many significant parties there are in the system, the degree of capacity they each have to make war, and the networks of relationships among them: alliances among some, and the existence and persistence of grievances and disagreements among others. And then there are the opportunities and the threats that relations between states pose to individual states and groups of states, which would naturally count among the triggers of armed conflict when other factors accumulate to the appropriate level.

What the basic model says of the third level is varied by different international relations theorists in various ways, one obvious variant being to account for subsystems at the international level, given that the dynamics, history and pressures of relations within a set of, say, South American countries differ from those within a set of, say, Far Eastern ones. But the assumption is the same both regionally and globally: that an account of the causes of war must address questions about the system of interstate relations, for it may be the system itself that is to blame.

Let us look at these levels of explanation.

Much has already been said in this chapter about the first level of explanation, the individual level, in considering psychological and biological ideas. Waltz was sceptical about the value of any explanation that locates the causes of war in human nature. As we see above, he is far from alone in this. His main reason was that human nature is a constant, whereas the causes and nature of war vary. It might be replied that individuals vary enough for those who are aggressive and violently disposed to be instrumental in making actual conflict arise out of competition and disagreement, the circumstances of the latter making for the differences in the wars thus started. But the rejoinder to this, in turn, is that peace is a fact of history also – and a more common fact than war – and this too would be explicable on the basis of facts about human nature. So if human nature is invoked to explain both peace and war, it cannot be (so this argument goes) the principal explanation for war.

These thoughts certainly suggest that human nature cannot be the whole story, even if psychological and biological explanations are part of it. To say this is not to deny that individual personalities – individual psychologies – are significant. Consider the impact in history of the ambition and character of such leaders as, say, Genghis Khan and Adolf Hitler. It is clear that leaders can at the very least be catalysts in harnessing and directing war-making energies among their followers. In the case of Ghengis and Hitler themselves it underplays their role to see them merely as catalysts, however; who and what they were as individuals must definitely be counted as part of the cause of the war-making associated with their names. Churchill might be an example of a catalyst, an inspiring war leader who nevertheless was unlikely to have encouraged a people to undertake a war in the first place, in the circumstances in which he found himself. And we can play with counterfactual hypotheses about how different history might be if, say, Al Gore had become US President in 2001 instead of George W. Bush, or if Churchill had not become Britain’s war prime minister in 1940 but the appeasement wing of government had signed a treaty with Hitler’s Germany in 1939, leaving the latter free to subjugate continental Europe and defeat the Soviet Union. The mere possibility that history might have been different is enough to show that the individual level of explanation matters.

There is also the consideration that a people – remember the Huns and Mongols – might have developed a culture of war-making as their way of life, and actively enjoy it and seek it; that the construction of masculinity in that society might largely be predicated on the activities of war-making; that its war leaders are ipso facto its cultural and political leaders too. These are first-level considerations of a different kind. A more diffuse form of them relates to the pride that a nation might take in that section of itself dedicated to the profession of arms. Contemporary US society accords great respect and consideration to its armed forces, whose size, power and technological sophistication figure positively in the nation’s self-image.

It is even more obvious that the second and third levels of explanation provide resources for investigating the causes of war. They lie on the cultural plane of analysis. As one would expect, the twentieth-century explosion of academia has produced a babel of theories about the causes of war at these levels, at just the time that war reached the point at which, in the extreme, it could extinguish humanity itself. This latter fact in part explains the babel, though it is a well-motivated one: the quest to understand why war happens is at the same time a quest for mechanisms of promoting peace. Of course the latter would benefit from there being fewer explanations, and more that were obviously correct. Nevertheless there is much of interest in many of the theories, as the following indicates.40

The dominant theory since 1945 about the causes of war is ‘realism’.⁴¹ At base realism sees individual states as interacting with each other in an anarchic international arena – ‘anarchic’ because it is without a controlling and adjudicating authority over it – in ways designed to protect or enhance their self-interest. Not only is the international arena anarchic, but no state has full information about the desires and intentions of other states; together these two facts produce insecurity. Planning to protect or enhance one’s own state’s interests in a situation of uncertainty requires taking worst-case scenarios into account, and being prepared for them. That, among other things, means arming, and being prepared to use one’s arms. Being armed offers the chance of imposing one’s will, or deterring someone else from trying to do so, in situations where other means of sustaining one’s interests are unavailable.

Robert Keohane summarises realism as holding that: (1) ‘States are the most important actors in world politics’; (2) states are ‘unitary rational actors, carefully calculating costs of alternative courses of action and seeking to maximise their expected utility, although doing so under conditions of uncertainty’; and (3) ‘States seek power . . . and they calculate their interests in terms of power.’42

Realism is a pessimistic view, which is never fully persuaded that peacemaking processes will succeed. It does not think that in situations of unequal distributions of power internationally, the stronger will invariably show self-restraint towards the weaker – remember Thucydides’ account of what the Athenians said to the Melians. It recognises that wars can start for accidental as well as deliberate reasons, and that when two or more parties fall out with each other forces can get to work that exacerbate matters, creating an uncontrollable spiral leading to war.

Classical realists are those who emphasise (3) in Keohane’s definition, that is, they are those who think that states, and the individuals who lead them, have as their primary interest the acquisition and maintenance of power. This might be for its own sake, but more likely because it provides security and the means to maintain it. In his version of realism, Waltz argued that the desire for power is instrumental in this way, and not an end in itself; for him the chief end of a state’s activities is security, given the context of insecurity in which states relate to each other; and power is the means to that end.⁴³ This is plausible given that an overriding interest of states is preservation of their territorial integrity and their freedom from outside imposition or coercion. Relative peace in the international order is achieved when states have a measure of security in relation to one another – when, in short, there is a balance of power between them. In Waltz’s view, balances of power have been achieved between most states throughout most of history.

The idea of a balance of power was the driving idea behind much of the international diplomacy of the great European powers of the eighteenth and nineteenth centuries, and it was expressly aimed at maintaining peace. It explains the shifting alliances forged to counteract other alliances or to constrain aggressors, as with the coalitions formed variously by Britain, Russia, Prussia and Austria against Napoleon’s France, and the opposing alliances in 1914 of the ‘Triple Entente’ of France, Russia and Britain on the one side, and the ‘Central Powers’ of Germany and Austria (originally with Italy also) on the other.44 But in theorising about the causes of war the phrase has a wider meaning, to denote the imbalances of power that could and too often did have destabilising effects on international relations. In the years leading to 1914, Germany perceived the Triple Entente as encirclement, and on one view of Germany’s role in the outbreak of hostilities in August of that year, it can be explained in part as a preventive or even defensive measure against the tightening of a noose.

As this suggests, a threatened disturbance to the balance of power can be an impulse to war. Long before Polybius asserted that ‘we should never consent to one state attaining power so great that none dare oppose it’, the growth towards dominance by a state had been seen as a casus belli by other states. That was what led to the Peloponnesian War, for example, in which Sparta and its allies refused to countenance the burgeoning power of Athens’s empire after the Greeks’ defeat of Persia. Churchill was iterating an ancient view when he said in 1948, with reference to the Soviet Union, ‘For four hundred years the foreign policy of England has been to oppose the strongest, the most aggressive, the most dominating Power on the Continent.’45

Implicit in the balance-of-power model is the assumption that the system which is being balanced or unbalanced by shifting alliances and changing power relations is a polyadic one, that is, with a number of states in play. But the same problems can arise when there is rivalry between just two states, usually geographical neighbours, in competition or disagreement with one another. War is unlikely if the disparity between the two is very great – in the event of force being used to resolve a problem, it is unlikely to result in a prolonged conflict in that case – so a dyadic balance-of-power relationship requires more rather than less equal parties to it. History is full of examples; the long-standing rivalry and many wars between England and France provide an example.

In more recent times the dyadism has obtained not between two states but between two blocs – NATO and the Warsaw Pact is the principal example, but even this confrontation had a geographical border along the ‘Iron Curtain’ in Europe. Any actual shooting was done by proxies in what were then called ‘Third World’ countries, in wars between states or groups armed and financed by the opposing blocs. Depending upon how they are counted, there were close to a hundred such proxy wars during the Cold War between 1945 and 1989.⁴⁶

If the assumption is made that a decision to go to war is a rational one, then given the fact that war is a very costly and generally unpredictable enterprise, the motivations have to be powerful ones, so powerful that they outweigh the costs and risks. Non-rational (purely emotional or even downright irrational) decisions to go to war are presumably those where the costs and the dangers, however great, do not weigh. It is hard to bring non-rational motives for war under a general theory; the madness of a ruler, the degree of ambition which is prepared to risk everything, a red mist of anger or a thirst for revenge, a profound sense of honour or preparedness to die even for a hopeless cause rather than live under someone else’s rule – all and any of these can prompt to war. In these cases it is at level one alone – the individual level – where explanations are to be sought.

But when rational calculation is in the picture it invites a different perspective. The question can be asked, ‘under what conditions is peace promoted and maintained between states who see the destructiveness of war as against their longer term interests?’ A compelling answer is offered by liberal arguments to the effect that economic relationships, principally trade, are prompts to peace. A litany of names subscribes this view, among them Adam Smith, Montesquieu and Thomas Paine. The latter wrote, ‘If commerce were permitted to act to the extent it is capable, it would extirpate the system of war.’47

The background to the liberal view was provided by the mercantilist system of closed trading, in which war was used as a means of securing markets while simultaneously excluding others from them. That policy was predicated on the belief that trade is a zero-sum game, the advantage of one state entailing the disadvantage of another. The liberal thinkers disagreed; free trade would be to everyone’s advantage, and increase would feed on increase. By the middle of the nineteenth century, Richard Cobden and like-minded colleagues were vehement in their opposition to war as a destroyer of trade and wealth. Cobden further argued that it was not just war but the standing preparation for war which militated against economic benefit.⁴⁸

Views of this kind had become the orthodoxy among many by 1914. The Labour MP and Nobel laureate for peace, Sir Ralph Norman Angell, argued in his 1910 book The Great Illusion that the great illusion in Europe was that war, conquest, or even just armed defiance, was a means to enhance the power and wealth of a nation, whereas in fact the mutual economic entanglement of states through trade and finance meant that war could only be harmful to them.49

Some took Angell to have argued that economic relationships had already made war impossible, and that therefore his thesis and that of his liberal forerunners was disproved by the 1914–18 war. In fact he had not said this; his point was the different one, which was indeed proved true by the war, that war in those circumstances is futile and harmful. The proof provided by the events of 1914–18 accordingly were only a temporary setback to the liberal theory, and the belief that economic interdependence secures peace revived in the aftermath of the Cold War, only to be met with the objection that considerations of trade pale into insignificance when diplomatic and military crises occur. The question is pointedly asked, ‘Were the Western liberal democracies seriously concerned about the short-term loss of trade when they made decisions to go to war against the hegemonic threats posed by Germany in 1914 and again in 1939?’⁵⁰

And perhaps interdependence – so another objection goes – if unequal and if it prompts suspicion or resentment, might even be or at least become a cause of war; that was a view held in the eighteenth century by Rousseau. Defenders of the liberal view can answer that interdependencies offer a plurality of channels through which differences and disagreements can be addressed and resolved. They also offer ways of applying pressure far short of war, such as withdrawal from agreements or the imposition of sanctions. The latter has become an increasingly common instrument in the international sphere since 1945.

Critics of the liberal view have history on their side in pointing out that war has often been a more efficient instrument, when successful, for gaining markets and increasing a state’s wealth. The liberal rejoinder is to say that this was only true when territory and maritime trade were the bases of wealth, but in the contemporary international economy, where knowledge and services are key, and where war is potentially far more destructive, such considerations no longer apply.

The period between the Second World War and the end of the Cold War saw no wars between capitalist states. Some put the point differently, and say that there have been no wars between democracies. The classes of capitalist and democratic states are not coterminous, though there is a large overlap; which gives rise to the question whether it is capitalism or democracy which is the peace-sustaining condition. There is something of an orthodoxy to the effect that it is the latter, so much so that the absence of direct conflict among advanced liberal polities since 1945 is described as the ‘democratic peace’. But as Gartzke argues – and a number of independent studies appear to support his view – it is not democracy but capitalism which is doing the trick. Gartzke writes:

Economic development, capital market integration, and the compatibility of foreign policy preferences supplant the effect of democracy in standard statistical tests of the democratic peace. In fact, after controlling for regional heterogeneity, any one of these three variables is sufficient to account for effects previously attributed to regime type in standard samples of wars, militarized interstate disputes (MIDs), and fatal disputes.51

This is at least consistent with the facts upon which Cobden and earlier theorists relied in urging the peace-productive benefits of economic relationships, for they did not live in an era of democracy, though they were able to claim that there was a greater benefit to the general population from peace-promoting trade, than the benefit of war to that restricted portion of the population constituting the aristocracy, as had been the case in history.⁵²

A major argument in favour of the European Union has been the promotion and preservation of peace in what had been one of the most war-torn quarters of the world for many centuries beforehand. Many internal critics of the EU complain about the ‘democratic deficit’ it appears to suffer from – although the rejoinder is that this is more the fault of electorates’ lack of engagement than the structures of the EU itself, together with the fact that the EU parliament is constrained by the prerogatives of the Council of Ministers (though the latter’s members are themselves democratically elected in their home countries). If the argument of Gartzke and others is right, the economic integration of the EU is the chief factor in the peace that obtains between its members. In the light of history, peace among the EU states is a magnificent success story, and it is hard not to see economic integration as a prime factor.

But have claims on behalf of democracy as a peacemaker among democracies been rebutted? Both Kant and Thomas Paine believed that popular sentiment would almost always be for peace because it is the populace that bears most of the burdens of war, whether in fighting it or suffering its privations, or both. Theorists looking at the relations between states, however, see matters through the lens of foreign policy imperatives, where popular sentiment is an irrelevance at best and an obstacle at worst. They share the view of Alexis de Tocqueville that ‘Foreign policies demand scarcely any of those qualities which are peculiar to a democracy. They require, on the contrary, the perfect use of almost all those in which it is deficient.’53 This is because ‘a democracy can only with great difficulty regulate the details of an important undertaking, persevere in a fixed design, and work out its execution in spite of serious obstacles. It cannot combine its measures with secrecy or await their consequences with patience . . . [they] obey impulse rather than prudence [and] abandon a mature design for the gratification of a momentary passion.’⁵⁴

This line of thinking suggests to some theorists that democracy is more rather than less likely to prompt war, therefore, because popular sentiment might encourage ideological crusades, for example to impose democracy where it is lacking. Woodrow Wilson justified US entry into the First World War by saying it was to ‘make the world safe for democracy’, and the extension of democracy to the Middle East was George W. Bush’s professed aim in invading Iraq in 2003.

Opposing this scepticism about democracy as a peace-ensurer, however, is the empirical claim that, as a matter of historical and current fact, democracies do not go to war with each other.⁵⁵ Both ‘democracy’ and ‘war’ require precise enough definition for a count of democracies and wars to substantiate this claim. This matters, given that the American Civil War and the First World War could be regarded as counters to it, as could the tensions and occasional actual warfare between Pakistan and India. A more circumscribed claim would be sustained by the idea that there have been no wars between states that are unambiguously democracies.⁵⁶

The reasons given as to why democracies do not fight wars with each other, although they fight imperialistic wars and wars against autocracies, are various. One is that the interests of democratic states tend to be aligned with each other in economic and foreign policy matters. A second is that grounds of dispute between democracies have already been removed, so there is little if any friction left between them. Another is that a constraint on war-making exercised by the populace of one democracy is greater because there is natural sympathy for the populace of another democracy, a sympathy lacking when a putative opponent is not a democratic state. This might be because democracies are more transparent to one another, and communicate with one another more, resulting in greater mutual understanding.

However, most of the arguments reported by Levy and Thomson in their survey do not appear to explain why democracies do not go to war with each other, but – less persuasively – why democracies are less inclined to war, a claim far harder to substantiate given the empirical evidence to the contrary. Such putative reasons are variants of the ones offered by Kant and Paine, but the numbers of wars fought by democracies do not support even their hypothesis. Those who remain sceptical about the claim that it is something in the nature of democracy itself which preserves interdemocratic peace point out that democracies are a recent phenomenon, emerging fitfully and in many cases incompletely in the nineteenth century, and the data on interdemocracy relations is only good for the post-1945 world. The lack of conflict between democracies could therefore be an artefact of the circumstances of the Cold War, or even just a coincidence.

For anyone keen on extirpating war from human affairs, the concept of democratic peace implies that states should be encouraged and helped to become democratic; for cynics, the only assurance of peace will always be the maintenance of greater strength than possible enemies. Because the latter is costly and dangerous, hopes are better attached to the former. Or perhaps, if it is not the politics of democracy but the economics of capitalism that promotes peace, even those who find capitalism unpalatable might be persuaded to seek a form of it which, while promoting international peace, is less abrasive and inequitable for populaces at home. This last thought matters, because economic inequality within a society is a major source of unrest and, in the extreme, civil war, and capitalism has a propensity to create inequality if left unchecked. What might preserve interstate peace could therefore cause intrastate conflict. As in so many things else, balance would be the key, and the inward- and outward-facing tasks of politics would meet at the striking of that balance.

The second level of analysis, the societal level, attracted the attention of Enlightenment thinkers and of Karl Marx. The former blamed the personal ambitions of ruling elites for the wars that wracked Europe’s history, and as noted they argued that if political power were diffused among the many they would be less likely to desire war because they were the ones most likely to suffer its effects. Marx focused on the economic rather than the political conditions in a society, arguing to the contrary of the theory just sketched that the exigencies and inequalities of capitalism were prompts to war. At this level too one finds the various suggestions that cultural, institutional and ideological factors play their part in disposing a people to war.

The Enlightenment view was able to invoke history in support. Almost all the wars of preceding centuries had been top-down wars, wars made by and for monarchs and ruling classes in which populations were subject to drag-along necessities. It provided Kant with his motivation for the peace-securing principles set out in his essay ‘Perpetual Peace’ (1795), and the necessary conditions for the application of those principles, namely that states should have ‘republican constitutions’ (that is, have responsible and representative government), that there should be a law of nations based on ‘a federation of free states’, and that people should be ‘world citizens’ in conditions of ‘universal hospitality’ (among other things implying free movement of people and mutual recognition of rights). Bentham’s proposal for perpetual peace focused on the idea of international law (he coined the word ‘international’ in putting forward this idea). His view was that disarmament, the abolition of colonies, and above all an effective system of international arbitration, would result in the abolition of war.57

Marx was not a pacifist, and he was greatly influenced by Clausewitz. His collaborator Engels wrote more on war than on any other subject, though to find a ‘Marxist theory of war’ as twentieth-century Marxism understood it one has to turn mainly to the writings of Lenin.⁵⁸ Engels was a close student of Clausewitz, and made a thorough examination of the American Civil War, writing about it in weekly columns in the press (over Marx’s name) as it happened. But the thinking of Marx and Engels on the subject of war took place in a period of prolonged European peace, whereas Lenin’s development of their ideas took place against the background of war.

The Marx–Engels view was that wars occur for reasons specific to the cultural and economic circumstances of their time, sometimes with positive and sometimes with negative effects. For this reason they were not opposed to war as something inherently bad, because it might sometimes result in the liberation of classes or the demolition of an oppressive order. Nor did they value peace merely for its own sake; Marx was critical of what he saw as the stagnation of China because of its long centuries of peace. In sum, they saw war as acceptable if it was a catalyst for progress, which for them meant movement towards the eventual passing away of class and state, when they believed peace would prevail because conflict-engendering inequality and exploitative relations of production would no longer exist.59

As developed by Lenin, the view is that capitalist regimes require the stimulus of aggressive foreign policies because without it they become stagnant and feeble. Over-production and lack of home consumption (because of the poverty of the proletariat) requires expansion into new markets. Surplus capital that the home economy cannot absorb likewise requires new arenas for investment. Both exert pressure for expansionist foreign policy; and as competing capitalist states vie for these domains, so conflicts arise. A variant is to say that capitalist production requires ever-new sources of raw materials and cheaper labour, again creating outward pressure.⁶⁰ Other Marxist theorists and some Keynesian theorists have added new sources of that pressure, for a chief example high levels of defence spending designed to stimulate the economy internally, but therefore spurring arms races with competitors, and the creation of conditions in which continued or increased such expenditure might actually need war to help sustain it.⁶¹

The Marxist–Leninist characterisation of imperialism and its wars as a necessary adjunct of capitalism is opposed by the claim, already noted, that war is harmful to capitalism, and that the economic relations between capitalist states ensure peace, not conflict. To accept that imperialism and warmongering are in the interests of capitalism requires accepting either that capitalists control the military, or that the military controls foreign policy. Although Dwight D. Eisenhower warned against the lobby power of the ‘military-industrial’ complex, plausibly suggesting a cartel arrangement between the military and the arms industry, it is less plausible to think that such a lobby could capture the state apparatus and apply it to its special ends. That would mean the Pentagon and its friends in the arms industry dictating the making of war and peace to president and Congress. Matters might rather be, as Waltz suggested, the other way round: economic flourishing gives rise to power, and it is the possession of power that prompts imperialist activity.62

The large and growing literature on the subject of the causes of war is such almost as to defy summary, and makes drawing conclusions difficult.⁶³ Sometimes the subject is discussed in ways that confuse the question of proximate and generic causes of war, even if making useful contributions to our understanding of each. Stephen van Evera in his Causes of War, for example, gives an interesting account of when wars tend to occur – at moments when offence–defence balances between hostile parties shift, and perceived opportunities are ripe, principally – which is of significance in evaluating the risk of war breaking out. But as one can see, this is a debate different from the general one about what causes war.⁶⁴ Culture and personality figure as key determinants in the generic causes of war according to Jeremy Black; here is an example of a choice of the level at which the impulses to war chiefly arise.⁶⁵

If anything is clear – and most of the literature acknowledges this – it is that the recurrent human phenomenon of war is the product of a logical sum of the family of typical causes of individual wars, and that the these causes operate at one or all levels of motivating forces – the individual, organisational and (for interstate wars) international levels. To say this is not, however, to capture the essence of the cause of war. That still evades analyses such as these.

What might a more promising route to an answer be? Perhaps an indirect approach would be better: to discern the sources of war by asking after the causes of peace. What might we learn from imperial paces – the pax Romana and pax Britannica for example, so far as the internal pax of each empire was mainly concerned? Cynics, not far from the mark at least in the latter case, might say: suppression of opposition, and they would note that the borders of these empires were anything but peaceful.

A yet more optimistic answer might be inferred from the great European peace of the post-1945 world, made possible by the growing interdependence and interconnection of the European nations. Operative factors in this case, and indeed in all these cases of peace, seem to be: interdependence and mutuality. The other side of this coin is, familiarly, the potentially inflammatory presence, and often mixture, of nationalism, ideological differences, economic and territorial competition, and suspicion. By an adapted Mill’s Methods one might begin to specify these latter as ‘the causes of war’ and not be surprised to find them so.66

But the really interesting outcome of this reflection, if correct – and equally unsurprisingly – would be that deliberate creation of interdependencies in economic respects and interconnections in political respects is a strong specific against war. This in turn suggests that the causes of war lie in structures of political relationships – a social matter, arising from the organisation of societies and polities, and the creation of interests self-identified with a given group and challenged by the interests similarly self-identified by another group. One can see how the emergence of states would make conflicts of interests a generator of war when discussion and diplomacy fail and the difficulties become intractable. The solution to war as a human problem is therefore integration, mutual linkages of a practical and beneficial kind, and the elimination of boundaries between interests.

In my view this would be facilitated by the end of the nation state and associated nationalism as a sentiment. At the time these words were written I had just returned from a visit to Zagreb, there to be told by my host, a lawyer, that Croatians had resisted their inclusion in the Austro-Hungarian empire, then deeply regretted no longer being part of it when it collapsed in 1919. There is a familiar ring to this sentiment, and all that it suggests in historical terms.

From the foregoing discussion the answer I infer to the question, ‘What are the causes of war?’ is this: the causes of war are the divisions and differences between self-identified groups with interests opposed to, or by, other such groups. War is therefore a product of a sufficient degree of organisation and structure to make such divisions and differences material. This implies that war is an artefact of the political, economic and cultural arrangements that evolved when settled societies emerged into history about 10,000 years ago.

Of course people fought each other before then, singly and in groups: skulls with holes in them suggest as much. But fighting is not war. War is organisation of and preparation for fighting in ways that lie above the military horizon, and that horizon is set in turn by the cultural horizon.

On this view, all the aspects and practices of civilisation that produced its great advances – involving as they do bureaucracy, engineering, literacy, education and training, social structure, law, hierarchy, centralisation, technology and more – also produced war. Among the more important lessons to be drawn is that civilisation has the capability of getting rid of war as it has got rid of poliomyelitis, if it would apply the will and means to do so. It would take co-operation, it would be difficult, it has been tried and it has at least twice in the last century failed badly (in the case of the League of Nations and so far of the United Nations), but it is necessary. ‘Necessary’ hardly captures the urgency. In a world armed with the means to destroy the very civilisation that brought war into being, it is absolutely and existentially necessary.67