The Better Angels of Our Nature: what implications for peacebuilding?

Has the incidence of violence declined over history, and if so, why? These are the two questions Stephen Pinker sets out to answer in his 2011 book, The Better Angels of Our Nature: The decline of violence in history and its causes.

As with other long multi-disciplinary books of scholarly secondary research, one sometimes wonders whether it wouldn’t have been easier just to read the much shorter essay which the author must surely have published too – at 700 pages it just takes so long to read! But, as with other such books – Guns, Germs and Steel by Jared Diamond, War in Human Civilisation by Azar Gat, and The Wealth and Poverty of Nations by David S. Landes come to mind – reading the longer version is worth the time spent. In The Better Angels of Our Nature (the phrase was used by President Lincoln in his 1861 Inaugural Address), Pinker takes a scientific approach, being careful to explore his hypotheses and assumptions through the analysis of a great deal of data. The book is replete with graphs and tables, and for a non-academic reader, provides an opportunity to acquaint or re-acquaint oneself with data and ideas drawn from a variety of disciplines including history, philosophy, the social sciences, game theory, and the arts.

Pinker repeatedly refers to Hobbes’s analysis of the incentives to violence in terms of Gain, Fear and Deterrence (or as Hobbes called them, Competition, Diffidence and Glory). He reminds us of or introduces us to Kant’s framework of Perpetual Peace, underpinned by democracy, commerce, universal citizenship and international law. He takes us on a long and winding ride through modern studies of behaviour and incentives, such as those of Zambardo and Milgram, which illustrated how easily people could be persuaded to use or connive in the use of violence if they felt it was “useful” or “called for”, or simply if others appeared to be doing so without being sanctioned.

He recognises the need to look beneath the surface. For example he cites experimental data which show that people find it far easier to hurt (or fail to protect) those who are “other”, or “not like me”. This is fairly predictable stuff. But then he goes deeper, demonstrating that “people like me” is a far more complex lens than often thought: in fact it is based on people’s need to be in coalitions of mutual support, which quite easily transcend notions of race and shared identity, once mutual trust and confidence is built.

Pinker takes us on a historical excursion to the Thirty Years War, one of the bloodiest periods in European history; and to the 8^th Century An Lushan Revolt and Civil War in China which I had not heard of, and which he claims killed 36 million people, two-thirds of China’s and one-sixth of the world’s population at the time. He describes mediaeval torture and punishment regimes, the practices of Atlantic slavery, the prevalence of rape and other once-widely condoned acts of violence in excruciating detail. He explores the role of the arts in society, identifying trends and tropes.

And of course he makes repeated use of and reference to the famous Prisoner’s Dilemma, which neatly explains why people so often opt for violence against one another, when their mutual interest is genuinely better served by cooperation, if only they trusted and had more information about what the another might do.

In the classic Prisoner’s Dilemma, two prisoners are arrested for a crime for which the police have insufficient evidence to convict them. Interrogated separately, each is offered the same deal. If he testifies against his confederate, while the latter refuses to do the same to him, he goes free and his confederate gets ten year in jail. If both prisoners betray each other, they both receive a two-year sentence. If both refuse to cooperate with the authorities – i.e. they effectively cooperate with one another – they both get a three month sentence. The scenario is one in which mutual cooperation between the two prisoners is the optimal strategy when seen by an objective observer with both thier interests in mind, and yet the rational decision for each, absent any knowledge about how the other will respond to the police’s offer, is to betray his confederate. The basic game is shown below.

Pinker is well-read across a very broad spectrum of disciplines, and the book presents a great opportunity for the non-academic reader like myself to catch a glimpse of some of the fascinating data and reasoning which is available to those with more time, knowledge and intelligence at their disposal. 

Pinker’s approach does have some important limitations. It covers a huge amount of ground, and thus risks skating too thinly over it. Also, however hard he tries to be scientific and empirical, one can’t help feeling that there is a liberal bias to the book. In one section he asks whether the reduction in violence could be explained by rapid genetic evolution over the past few millennia. Hovering behind this enquiry is the tantalising concern that if so, the evolution may have been focused on only some human populations, and not others. Thus it might have happened differentially in difference races, which is a conclusion no liberal would be eager to arrive at. When he gets to the point where he can demonstrate both theoretically and empirically that this is not so, one senses a genuine feeling of relief on Pinker’s part: phew, the data are aligned with a liberal view!

Another limitation is that so many of the data he uses are from the West: Europe and the New World, especially the USA. This is understandable, given his milieu – he is a professor at Harvard – and the relative lack of historical series and experimental data from elsewhere. But it does leave a gap in his narrative and therefore in the robustness and generalizability of his conclusions.

Finally, his determination to rely on empirical data at all costs is exasperating at times. Among the most commonly repeated phrases in the book must be “holding other factors constant”, and “other things being equal”. For every hypothesis explored, he seeks to explain the phenomena in terms of the influence of exogenous factors, and of course this is right if one wishes to draw robust conclusions. And yet at times as a reader I just wanted to scream, for example “but it’s common sense that people who trade with each other are less likely to fight each other, you don’t need three more graphs to persuade me of that!”

But these are relatively trivial limitations, alongside the grand task he set himself. So what did he find?

The decline in violence

Examining the data, Pinker finds that over history, humans have on average committed less and less violence against each other. We are more peaceful in family relationships than we used to be. We use torture and other “unusual punishments” far less than we used to, indeed they are outlawed in a growing number of countries. We are more civil in our relationships with neighbours and others we encounter on a regular basis, and more tolerant of those who are different from us. We rape less, we stab less, we mug less, we murder less, we enslave less and we execute less. We go to civil and inter-state war less often, and when we do, the casualty rates are far lower than they used to be.

Pinker seems to expect his readers to be surprised by this, though I wasn’t. Nevertheless it is very useful to see the trends summarised and explained using data. What was particularly interesting to me was the rapidity of the change, and the steepness of some of the declining curves. Many are so steep that he has to use a logarithmic scale to compare the modern day incidence of violence with that of the past.

Pinker is at pains to note that plenty of people still live in consistently violent circumstances: his data on violence within societies are primarily taken from stable Western democracies. He also makes it clear that he has no evidence to support the idea that reduced levels of violence will be sustained in the future. Indeed, while societal violence and warfare trends point consistently downward, they can and have been reversed at times, and we only have to look around us to see there is no room for complacency. In fact, while the overall shape of the curves for the incidence and magnitude of warfare slopes distinctly downward, it is also saw-toothed, with sudden and jagged increases often lasting many years, before the curve resumes is downward trajectory. The up-tick in war and atrocities during the first two thirds of the Twentieth Century is an example.

This is quite difficult. To see 70 million people killed in two World Wars, and a further 60 million at the hands of Mao and Stalin, to say nothing of Cambodia, Rwanda, the Iran-Iraq war and other major conflicts and atrocities as an up-tick in data trends is hard to swallow. But on closer examination, it is not obvious that the twentieth century was the worst. Pinker identifies five wars and four atrocities in prior ages each of which killed more people than the 15 million of the First World War. And in a ranking of major wars and atrocities in which the number of deaths is taken as a proportion of the world’s population at the time, Stalin’s 20 million ranks only fifteenth, Mao’s 40 million deaths ranks eleventh, and the Second World War with 55 million is “only” number nine. First, second and third places in the list go to the Al Lushan Revolt, the Mongol conquests, and the Middle Eastern Slave Trade.

Pinker calls the sustained spike of 20^th Century violence a “hemoclysm”, but does not accept that it should legitimately be seen as a single phenomenon, and does not allow it to derail his main thesis that overall, levels of violence have declined. Instead he suggests that the horrors of the Twentieth Century can largely be explained by random factors and by the presence of sociopaths like Stalin, Mao and Hitler in positions of power, and by accidents of history such as the shot fired in Sarajevo which was famously “heard around the world”. Hmm.

The explanation

What I find particularly interesting about Better Angels is Pinker’s attempt to explain the sustained reduction in violence. While he does examine neurological and physiological aspects of this, his thesis is that humans have become less violent essentially due to five mutually interacting factors which are psychological, social, economic, cultural and political, rather than neuro-physiological. Like most of the best findings, they mostly seem like common sense, though some are perhaps more obvious and familiar than others. In the end, it’s the way we live together and how we organise ourselves to do so, which best explains the level of violence within and between human societies.

1. Governance: locally, nationally, regionally and globally, people have found ways to organise themselves so they can increasingly manage and resolve conflicts and differences without recourse to violence. The apparatus of the state is essentially a set of systems, underpinned by values and a social contract, which serve this purpose. These are reflected at the sub-state level too, as well as in the intergovernmental organisations to which states belong. Pinker demonstrates a strong correlation between the evolution of the state and the reduction in violence; and especially between the evolution of democratic governance and the reduction in violence. Effective governance increases the chance that people or states will opt to collaborate with other people or states rather than commit violence against them – i.e. they will not “defect”, in terms of the Prisoner’s Dilemma. The rule of law means that either party will be sanctioned for committing violence, thus the incentive to collaborate becomes higher; each “prisoner” can have more confidence that the other will not defect, and it thus becomes more rational in some circumstances to assume he won’t. Meanwhile, the evolution of states has engendered an increasing respect for human rights.
2. Commerce: when people trade and do business together, they get to know each other, develop trust and confidence in each other, share an interest in stability and – all else being equal – are incentivised to cooperate rather than fight.
3.  Feminization: societies which have become more peaceful have come to organise themselves less around the need to be able to fight. The typically masculine, martial culture has become less important. Women play an increasingly important role as leaders and in other walks of life, and society has – in Pinker’s words – become feminized. This is particularly true, so far, of Western society. This has also allowed men’s more feminine side to come to the fore. Experiments find that women more often opt for a positive-sum approach, while men opt for zero-sum. Feminized societies are more likely to seek and achieve positive-sum solutions to problems and issues, and thus to find ways to resolve them without violence. 
4. Empathy: one of Pinker’s most interesting findings is that the more we come to know about others, the less incentive we seem to have, to do them harm. Knowing other people or “peoples” helps us see how alike us they are, and thus reduces their “otherness”. Studies have shown that we are more likely to take care of, and less likely to be willing to harm people, if we know them or know enough about them to perceive them as like us. Pinker talks about an Expanding Circle, in which increasing numbers of people are part of the same meta-coalition or set of coalitions, and therefore the smaller identity-based coalitions (based on location, nation, race, ethnicity, etc.) become less important. The growth of this circle, and the depth of the connections within it, can partly be traced to culture: the advent of reading and writing made it possible to share information from elsewhere. The development of plays and the novel made it possible to consider what others are going through or thinking – to put oneself into others’ shoes; to develop a capacity for empathy. Pinker identifies the evolution of humanitarianism as a historical trend linked to the increasing prevalence of empathy.
5. Reason: finally, enlightenment has played a critical role. The ability to reason, to calculate our best option when faced with a problem for which one of the options involves violence, allows us to weigh up costs and benefits rationally. Reason has edged (or is edging) out superstition in communities around the world, arming people with the ability to develop less violent solutions. Pinker quotes enlightenment thinkers from Renaissance Europe who started to realise and argue – and then demonstrate – that torture is an irrational way to find out the truth, as people will say anything, true or false, under torture. (In one horrifying story he quotes a renaissance nobleman who tortured his own manservant horribly, to see if he would confess to an imaginary crime: he did, thus proving the hypothesis, but also surely demonstrating to the modern reader that humanitarianism was still nascent.) This was then followed by the progressive elimination of torture as a routine practice in country after country. For Pinker, the ability to reason and argue, to test and prove or disprove hypotheses, enabled this change.

The implications

I work for the peacebuilding organisation International Alert , so I read The Better Angels partly to see what lessons there might be for us and other peacebuilders. The first, I’m happy to say, is that Pinker’s relentless search for empirical evidence has done us a big favour, as his thesis is very much in line with our own, as expressed in our programming framework . Like Pinker, we see peace largely as a result of good governance, systems and culture which allow people to manage their conflicts without recourse to violence, while making equitable progress. What Pinker shows is that the progress itself reinforces the tendency towards peace. But like Pinker we would reject the whiggish notion of an inexorable reduction in violence that happens as an inevitable outcome of history. Peace is something to be built and constantly strengthened, rather than something which comes about of its own accord. And one has to avoid prescribing how peace will evolve in one set of circumstances, simply because it has happened like that elsewhere. One has to be particularly careful in taking lessons from the way the mature western democracies have evolved, and trying to apply them elsewhere.

Nevertheless, thinking generically about how peace can be enabled or built does make sense, provided strategies for specific circumstances are devised based on good context analysis. Pinker’s book seems to support, among others:

• The idea, very popular just now in the international development and peacebuilding communities, of working to establish and strengthen the social contract between responsible citizens and a responsive and accountable state. For example by improving the services provided by the state, and giving citizens a greater voice in how they are delivered and monitored; and by bringing more people into the tax system, so they have quite literally a greater stake.
• The integration within peacebuilding, of programmes designed to widen the Expanding Circle of empathy. For example using cultural events and artefacts – plays, music, films, novels – to increase the likelihood that people will habitually imagine and envisage the lives of others, including others outside their immediate ken. Another approach to this might involve bringing people from different ethnic and national groups together in the education system, thus getting to know people who might normally be seen as “other”.
• The promotion of non-identity-based civil society groupings, as coalitions of mutual support which rely on a non-communal concept of community and thus reduce the opportunity for inter-communal violence. These groups could be formed to provide services, to argue and advocate for common interests shared across traditional divides, or to manage commonly used natural resources such as water, farming and pasture land, and municipal resources in the cities. They could also be political parties – or embryonic political parties.
• The promotion of commercial links and joint enterprises across formal and virtual borders. For example Alert does this across formal borders (DRC: Rwanda); unrecognised de facto borders in the Caucasus; and across the virtual, psychosocial borders which separate people locally in Rwandese communities. In some ways this was one of the founding ideas of the European Economic Communities back in the 1950s.
• Supporting the virtuous circle in which incremental peace empowers men to become less martial, and women to play a greater and more explicit role in governance and business; and thus engender further incremental progress towards peace; and so on…. There may be some surprising elements to this. For example, polygamy frequently excludes young men from taking their full adult role in society, as it often excludes them from marriage until later in life (while older, wealthier men monopolise the marriage market). The evidence is clear that married young men are far less likely than bachelors to commit violent crimes or to join militias. Thus peacebuilders from polygamous societies might wish to consider trying to change this practice, as a step towards feminization and peace.

The good thing is, many of these kinds of ideas are already being practised, along with countless others that fit with Pinker’s thesis. But we can do more. Reading his book, I see arguments for peacebuilding strategies – such as the one just mentioned – which don’t often get aired. I therefore recommend those who think they have time for Pinker’s 700 pages to give it a go.