The benefits and costs of altruism
Daniel Batson, the social psychologist, has recently brought out what is probably his defining work on the topic he has studied for 30 years, Altruism In Humans. I bought it after hearing Martha Nussbaum rave about it when she spoke at the RSA in December. She says on the dust jacket that it's "simply one of the most important books in our time for anyone who wants to ponder the problems and prospects of our species". Casting my eye around Google, I think this might actually be the first review of the book. Woohoo, first!
The theory that Batson has spent his life experimentally testing is what he calls the 'empathy-altruism hypothesis' (or EA for short). The EA theory is that, firstly, humans are capable of 'empathic concern' - which Batson defines as 'noticing someone is in need, and caring about it'. And, secondly, this empathic concern leads to altruism, which he defines as a 'motivated state with the ultimate goal of increasing another person's welfare'.
For much of Batson's career, this theory has been marginal to western psychology. The Buddhist monk Mattieu Ricard (pictured below with Batson), writes: "[The EA theory] is certainly not part of mainstream western psychology, in which the dominant view is that of universal egoism. According to the latter view, any seemingly altruistic behavior must have been driven by some kind of selfish motivation: 'Scratch an altruist, and watch the hypocrite bleed.'" But Batson's work can be credited, according to Ricard, "with proving that genuine altruism does exist". Quite an achievement.
In the new book, Batson spends the first two chapters carefully defining what he means by empathy and altrusim, and distinguishing it from other people's use of the term. This is slow reading, but important, if you think about how many ways philosophers and psychologists have used those terms over the last 2,000 years. Indeed, Nussbaum says: "Unlike some work in psychology, Batson's work is very rigorous in how it defines terms." (Think, for example, of how psychologists spray the term 'happiness' around without really defining it.)
He distinguishes his definition of empathy ('noticing someone is in need, and caring about it') from other definitions of empathy, such as 'reading another's feelings', 'mirroring their responses', 'putting yourself in another's shoes' etc. Some of these, like the last definition, might help give rise to empathic concern, but they are not 'necessary or sufficient' - you can feel empathic concern for someone without necessarily putting yourself in their shoes. I could feel empathy for my wife giving birth without imagining myself doing it.
Likewise, some have defined altruism as 'helping behaviour' or, more broadly, as 'acting morally'. But one might help for egoistic reasons. And altruism might not be moral - we might help someone because we care for them, even if it's against the common good, as Batson later examines.
Having defined his terms well, Batson briefly suggests an origin for empathic concern in parental care and nurture, making reference to work in this area by the pioneering social psychologist William McDougall. But Batson doesn't spend long exploring the roots of empathy - the work of primatologist Frans De Waal on empathy in primates might supplant this absence in those interested. The two were actually brought together to speak at a conference in 2007 by Martha Nussbaum.
Batson is more interested in whether empathy really leads to altruistic behaviour in humans. He writes: "Many, especially within religious traditions, have said that we humans ought to be altruistic. I shall not engage this issue, at least not directly. As a scientist, my concern is with what is, not what ought to be. Philosophers would say my goal is descriptive, not normative. Of course, 'ought' and 'is' are not totally unrelated. We can only be expected to do what is within our capacity."
Testing out altruism experimentally
How do you experimentally test out the EA hypothesis? Batson distinguishes two different approaches, which he calls Aristotelian and Galilean. Aristotle's experimental approach is to define an organism by all the different characteristics it exhibits. You catalogue the perceived phenomena. So, if you want to see if humans are altruistic, you might catalogue all the examples where humans have behaved with apparent altruism - heroic stories of rescues, self-sacrifice, volunteering and so on. This approach is interesting to read (the only time I personally felt 'empathic concern' while reading his book is when he actually describes some of these real-life examples of altruism). But it is not always scientifically effective - particularly if the phenomenon being explored is one of internal motivation rather than external behaviour.
The Galilean approach, by contrast, begins with a theory of the underlying law or process that gives rise to phenomena. It then makes predictions of how phenomena will happen based on this theory, and sees if things do in fact happen as predicted, or not. Batson uses this approach, which he says is better-suited to studies of motivation, because you can change variables under controlled experiment conditions, to rule out different explanations for internal phenomena.
He gives the example of a situation: Frank has two tickets for a gig. Jessica, the sexy girl in the office, hears of this and is unusually friendly towards Frank. The question is, what is motivating Jessica? Frank wonders if she is motivated by (A) the concert tickets or (B) a genuine romantic interest in Frank. Then, a variable is introduced - someone else announces they also have a spare ticket to the concert. Suddenly, Jessica is less friendly towards Frank, suggesting that theory A best explains her motivation.
Batson took this approach to try and test the EA hypothesis against rival egoistic explanations for helping behaviour, such as that we help others to feel good about ourselves, or to gain social rewards, or to avoid social censure, or to reduce our own pain at seeing another suffering. You create a controlled experimental situation, and then you change variables to see how they affect a person's behaviour, seeing if the changes fit with your theory of their motivation.
He and his colleagues at Kansas University conducted several experiments over the course of three decades, which typically followed this sort of format: undergraduate volunteers would arrive at the lab, and be told they were taking part in an experiment to test their attention or task-solving abilities. They weren't told the experiment was testing their altruism, as this would immediately affect their behaviour. They were then shown a video of someone ('Katie') who they were told was another student taking part in a parallel study, which they were asked to pay attention to. The person in the other study was failing the task she was set, and was receiving small electric shocks, which she obviously found very distressing (this was actually an actress, pretending to be electrocuted). 'Katie' explained she had a particular past trauma related to electric shocks, which is why she was so upset. The volunteers were then offered the apparently spontaneous suggestion to swap places with the person being electric-shocked - 'Katie' would do the attention test and the volunteer would do the electric-shock-memory test.
The other main technique was to play volunteers a fake radio story about a girl struggling to bring up her children while completing a degree, and then suddenly give them the opportunity to give their time to help this girl.
Batson would manipulate various factors in this controlled experiment to test out different hypotheses for why the volunteers might help 'Katie'. For example, one experiment tested out two rival theories - the EA hypothesis and the negative arousal hypothesis, which suggests that we help others in order to stop feeling empathic pain at the sight of their suffering.
Batson manipulated two variables - the level of empathy participants felt, and the ease of escape from the situation. Level of empathy was manipulated by showing the volunteers a questionnaire that suggested Katie was very similar to them, which apparently led to higher feelings of empathy; or by showing them a questionnaire that showed Katie was very different, which led to lower feelings of empathy. Ease of escape was manipulated by giving the participants the opportunity to leave the study after two electric shocks were given to Katie (easy escape) or making them watch her be electrocuted for around 15 minutes (hard escape).
So Batson created a grid of four situations: (1) Easy escape-low empathy. (2) Easy escape - high empathy. (3) Difficult escape - low empathy. (4) Difficult escape - high empathy. The Empathy-Altruism hypothesis predicts that volunteers would offer to help 'Katie' when empathy is high, even when escape from the situation is easy. The Negative Arousal hypothesis predicts participants will choose to leave the lab when escape is easy, even when feeling high empathy. Leaving the lab would be the quickest and easiest way to reduce their suffering at seeing Katie being electrocuted. It turned out that participants chose to swap places with 'Katie' even when offered an easy escape from the situation.
This gives the basic format, which was then repeated around 30 times by various psychologists to test out different hypotheses: the EA hypothesis versus the social praise-altruism hypothesis, for example. Do participants help 'Katie' less if their help is anonymous? Do they help less if they are given a good justification why they shouldn't feel bad for not helping her? Do they help a person in need less if other people have offered help? And so on.
Batson reviews the results from around 30 experiments from 1978 to 2008. In experiment after experiment, participants' behaviour fit the EA hypothesis rather than any of the rival egoistic explanations for their behaviour. He writes: "The idea that empathy produced altruistic motivation may seem improbable given the dominance in Western thought by the doctrine of universal egoism. Yet, in the words of Sherlock Holmes, 'When we have eliminated the impossible, whatever remains, however improbable, must be the truth'.
The benefits and liabilities of empathy-altruism
This is all extremely worthwhile and important research, but can be rather technical reading for the layman. What I personally found more interesting was the last third of the book, on the benefits and liabilities of altruism, and on other possible prosocial motives for helping others.
Batson notes the many benefits of empathy-induced altruism as a motive of human behaviour. It has a strong emotional base, which stimulates us to really care about a person's welfare. It makes us give more help to others, and more sensitive help. When we feel empathic concern for others, we are less aggressive, we cooperate and negotiate with others better, we are less prone to intergroup conflict, and are more sympathetic to marginalized or stigmatized 'out groups'. We're also better at friendships and relationships.
All of this is a powerful argument in defence of the humanities, as Nussbaum suggested in her new book, because the arts and literature teach us to take others' perspective, including the perspective of marginalized 'out groups', which cultivates empathic concern in us. The humanities help to 'humanize' outgroups in our eyes - and to humanize us by broadening our sympathies.
Feeling empathy and helping others also apparently leads to greater positive feelings in ourselves, less depression, and better health. These findings have been seized on by Positive Psychologists as evidence for their social and civic focus: 'Focusing on happiness isn't selfish or self-absorbed', we often hear. 'In fact, the surest way to personal happiness is to help others.' But Batson gives two reasons why this is not always true. First, "too much selfless concern for others may lead to 'caregiver burn-out'".
And secondly, Batson suggests that
those who turn to altruism as an antidote for depression, meaninglessness , and tension will find it does not work. To use altruism as yet another self-help cure - providing a means to the ultimately self-serving ends of gaining more meaning and better health - involves a logical and psychological contradiction. As soon as benefit to the other becomes an instrumental means to gain those self-benefits, the motivation shifts from altruistic to egoistic. So if it is empathy-induced altruistic motivation - rather than simply helping behaviour - that produces the health benefits noted, intentional pursuits of these benefits may be doomed to failure.
Batson also notes the costs to empathy-altruist behaviour. Firstly, it can be demeaning to those receiving help. After all, Batson suggests the roots of altruism are in parental care. So when we feel sorry for someone and want to help them, we are pitying them, and perhaps suggesting that they are not capable of taking care of themselves, like a child. That is why Stoicism, for example, has been called an 'anti-pity philosophy' by Martha Nussbaum. It is the epitome of tough love, suggesting that it is through struggling with and mastering adversity that we grow into autonomous adults. But empathy-altruism tries to shield us from adversity, like a mother, so it denies us the experience of learning to stand on our own feet.
Secondly, empathy-altruism tends to be motivated by individual cases, rather than situations affecting large numbers. That's why charities, when appealing for donations, present individual stories rather than impersonal statistics. It is also apparently more stimulated when we feel the person in need is 'like us'. It has a strong emotional base, but the feeling of empathic concern diminishes over time, so "may not be able to sustain the kind of long-term helping effort often required of, for example, community-action volunteers". And it does not always lead to the common good. Batson notes that many people who do things that are against the common good might be motivated by empathic-altruistic concerns to provide for their families. The bankers who destroyed our economy may have been prompted as much by empathy as by greed.
Egoism, altruism, collectivism, principalism
Therefore, in the final chapter, Batson suggests we should explore the 'pluralism of prosocial motives'. We're motivated to prosocial behaviour by four main motives, Batson argues. (1) Egoism. This has the advantage of being powerful, emotional, and enduring. But it only does altruistic good for indirect reasons. (2) Empathy-altruism, with all the benefits and liabilities described. (3) Collectivism - we do good to protect our group, tribe or society. This is a powerful and emotional motive (think of British volunterring during WWII for example), although it can lead to marginalizing or attacking of 'out groups'. And finally (4) Principalism, which helps others out of duty to an impersonal principal, like 'Do no harm' or 'help thy neighbour'.
Principalism is probably closest to the Stoic motive for helping others. You don't help others because you feel sorry for them. You do it out of duty to the impersonal law of the Logos. The benefit of this motive is it's enduring, it's not fickle or reliant on passing emotional states. It is perhaps more rational and clear-eyed, looking at what genuinely helps others rather than what 'feels right'. And yet its lack of a strong emotional base means that, perhaps, it is not a very ubiquitous motive, and also that it lacks human warmth.
Take, for example, the figure of the judge in Hans Fallada's excellent novel about Nazi Germany, Alone In Berlin. In the novel, a judge takes in a Jewish lady, and hides her in his flat, thereby risking his own life. He says to her he does this out of obedience to 'Justice'. He is undoubtedly a noble figure. And yet the Jewish lady feels, perhaps rightly, that "his goodness is cold...For all his goodness, human beings don't mean anything to him, the only thing that has meaning for him is his justice".
And there's something a bit priggish, inflexible and self-regarding about Principalism, as I have noticed with some modern Stoics. Those who are mainly motivated by Principalism are often rather cut-off, emotionally tone-deaf, individualist. They often seem concerned more with adhering to their own code of ethics, their own idea of themselves, rather than allowing themselves to recognize and feel the pain of another. It's not always prosocial - Principalists are never happier than when haughtily declaring their independence from the tribe in the name of their lofty principals.
Still, Batson suggests that, if we want to inspire prosocial behaviour in ourselves and our society, we should look to "strategies that combine appeals to either altruism or collectivism with appeals to principal". Much to think about, then, as we try to work out how to galvanize more volunteering, more activism, and more effective public responses to challenges like climate change.
And the book is, it seems to me a gauntlet thrown to philosophers, demanding that they try harder, work harder. Stephen Stich of Rutgers University says on the back of the book that "Batson and his associates have made more progress [on the issue of altruism] in the last three decades that philosophers have made in the previous two millennia." Hyperbole, perhaps, but there's some truth in what Stich says. Batson has taken one issue - a very important one - and carefully probed it with experiments for over 30 years. That seems worth so much more to me than the rambling post-prandial pronouncements of armchair philosophers.