This is a broad outline of the theory I’m developing for my PhD thesis. Consider it a work in progress. All comments and criticism are welcomed.
If humans have an evolved moral sense – and there’s mounting evidence that we have (Wright, 1994; Fehr et al., 2002; Boyd et al., 2003; Gintis et al., 2003; Lieberman et al., 2003; Fowler, 2005; Joyce, 2006; Haidt, 2008; Haidt at al., 2009) – then we should not expect it to function in an identical way between individuals.
Instead, we should expect a diversity in the function of the moral sense between individuals, and a corresponding diversity of moral intuitions and moral judgements. This is because there is no one solution to the problems that our moral sense evolved to solve that yields the best outcome in every environment or circumstance.
As such, our moral sense has evolved to produce a variety of strategies that increases the likelihood that we’ll be able to successfully respond to a wide range of situations and environments – with the ‘environment’ also including the strategies employed by other individuals.
The analogy for this view is one of an evolved ‘moral ecology’ – a diverse range of strategies that each perform well in their niches, while they last, but with no one strategy that dominates all others. In this essay I will employ the moral ecology metaphor to explain why our moral sense evolved the way it did, and how it produces the wide diversity of moral attitudes and beliefs we see today, not only between cultures, but within cultures, and discuss some of the implications for moral philosophy.
What is Morality?
The picture that emerges is one that sees morality not as a truth-seeking endeavour, as moral realists and objectivists claim (Smith, 1994; Bloomfield 2001) but as a device that evolved to regulate social interaction, encourage pro-social behaviour and punish anti-social behaviour (Westermarck, 1932; Mackie, 1977; Greene, 2002; Haidt et al., 2009).
It evolved because those individuals that were able to benefit from the rewards of cooperation – while mitigating the risks of free-riders – had a selective advantage over those who engaged in less cooperation or who succumbed to exploitation by free-riders (Hamilton, 1964; Axelrod & Hamilton, 1981; Krebs, 1998; Gintis et al., 2003; Joyce, 2006; Haidt, 2008). As such, morality evolved because it aided reproductive fitness in our evolutionary past (although that doesn’t mean that morality needs to be predicated on advancing reproductive fitness today).
The problems that morality evolved to solve are essentially the problems inherent in fostering prosocial and cooperative behaviour, which benefits all agents who interact, but opens up opportunities for free-riders to take advantage of the cooperative venture for their own gain at the other cooperators’ expense.
This phenomenon can be modelled in a number of ways, with one of the most effective being the Prisoner’s Dilemma, which will be used in this essay to illustrate different ‘strategies’ for fostering cooperation and protecting against free-riders (Binmore, 1990; Colman, 2003). In these models, particularly the Iterated Prisoner’s Dilemma, it emerges that there is no one strategy that performs optimally in every environment, given the strategies employed by other individuals within a population (Trivers, 1971; Dawkins, 1975; Axelrod, 2001). Yet there are situations in which a plurality of strategies can co-exist in a dynamic equilibrium, called an evolutionary stable strategy (ESS), whereby this equilibrium cannot be ‘invaded’ by any new strategies (Maynard Smith & Price, 1973).
I draw an analogy between this phenomenon and the existence of a diversity of moral attitudes held by individuals within a population. I invoke two theses concerning this diversity: ‘lesser moral diversity’, which suggests that there will always exist a mix of pro-social strategies within a population; and ‘greater moral diversity’, which suggests there will also always exist a proportion of self-interested/free-riding strategies in the mix of pro-social strategies within a stable population.
Evolution shaped our moral sense not by providing us with hardwired moral principles to follow, but by equipping us with a range of faculties that, when primed by the environment, tend to encourage pro-social behaviour and punish anti-social behaviour (Haidt, 2009).
Many of these faculties operate like heuristics – fast and frugal decision making processes that are capable of directing behaviour quickly and with minimal effort yet are also prone to error (Gigerenzer, 2008) – such as the moral emotions (Haidt, 2003), but we also have a highly plastic reflective reasoning faculty that can produce a wide range of behaviours in response to an equally wide range of environmental circumstances (Wilson & Tumminelli & Sesma, 2009).
There are three main components to our moral sense: perception (Murdoch, 1970; Blum, 1991; Lakoff, 1987 & 1996; Hauser, 2006); emotion (Hume, 1739/1985; Westermarck, 1906 & 1932; Haidt, 2001) and reason (Kant, 1787/1993; Sidgwick, 1907) that contribute to the production of moral attitudes and moral behaviour. When confronted by a scene, or after being exposed to an idea, the first faculty to be engaged is perception, which provides structure and meaning to the stimulus based primarily on prior experience as well as other innate mechanisms (Blum, 1991; Lakoff, 1987).
Almost immediately after scene is perceived, the emotions are engaged depending on the meaning and significance in the scene (Hauser, 2006). If the scene contains morally-charged elements, the moral emotions are engaged, such as empathy, guilt and disgust (Haidt, 2001). While the existence of these emotions are universal in all humans (except for the interesting counter-example of psychopaths), the stimuli that trigger them are largely primed by culture.
As such, one individual might find the idea of eating a dog disgusting while the idea won’t elicit a disgust response in an individual from a different culture. The emotional response to situations such as this plays a large role in determining whether an individual finds a particular behaviour or idea morally permissible or impermissible, which manifests as a moral intuition – an immediate sense of approval or disapproval giving an impression of the rightness or wrongness of the scene or its content (Haidt & Joseph, 2004).
It’s only at this point that conscious reflection is engaged, and along with it the application of rational principles and moral beliefs, to yield a conscious, and utterable, moral judgement. Often there might be only one moral intuition that is elicited from a scene, and this intuition will conform with an individual’s moral beliefs. However, sometimes multiple intuitions might emerge, and they might be in conflict with each other, such as in moral dilemmas.
At this point reason can play a role in mediating between the opposing intuitions to yield a conscious moral judgement about the permissibility or impermissibility of the content in the scene. Sometimes the moral intuition or intuitions will also conflict with the consciously held moral principles, with reason working to find a reconciliation between the two. Finally, the agent is led to performing some action (or withholding from performing some action), motivated by emotion, steered by reason.
Evolved Moral Sense
Evolution has influenced many steps in this process. First of all, each of these individual faculties is the product of evolution, from perception to emotion to reason. Many of the emotions also evolved in response to adaptive problems from our evolutionary past, such as fear promoting an aversion to the triggering stimulus, with many dangers from our evolutionary past – such as spiders, snakes, heights, the dark etc – eliciting a fear response disproportionate to the danger they represent in the contemporary world (Pinker, 1997). The very existence of certain emotions – such as embarrassment or guilt – also suggest we have evolved specific mechanisms to regulate social and moral behaviour (Haidt, 2003).
The moral diversity theses suggest that individuals will vary in their moral responses to certain situations due to environmental, developmental and experiential differences, as well as an innate variability in emotional responses or cognitive styles.
For example, two individuals might have different anger responses, with one being naturally quicker to anger and experiencing a more intense anger sensation, while the other might be naturally slower to anger and will experience a less intense anger sensation. These two individuals might respond differently to the same scene as a result of the variation in their emotional responses, even if their developmental and experiential background are identical. Over time, their variation in responses will lead to different experiences, which may further compound the variation in their future reactions. In time, these two individuals might come to hold widely disparate moral beliefs and might employ very different moral behaviours even in very similar situations.
Evolved Moral Diversity
If this account is plausible, then there needs to be a mechanism that can enable such a stable polymorphism – i.e. a persistent diversity of alleles in the gene pool that can lead to such a diversity of emotional and cognitive responses – to evolve. In fact there do exist other aspects of our genome that exhibit similar stable polymorphisms, one being the major histocompatibility complex (MHC).
The MHC is instrumental in our adaptive immune system, and functions to identify self cells from non-self cells, thus allowing our immune system to target foreign pathogens rather than attacking the body’s own cells (autoimmune disorders are what occurs when this system fails in some way). However, should a pathogen evolve such that it is recognised by the body as ‘self’, then it can elude our immune system. This means that had the MHC evolved into a single stable state, then it would have been vulnerable to a single mutant pathogen bypassing immunity – a result that could prove fatal for the entire species.
As a result, the MHC is in a constant ‘arms race’ with pathogens, and the MHC has evolved to be highly polymorphic – varying between individuals by as much as 80 per cent – effectively becoming a ‘moving target’, such that it’s less likely for a single pathogen to bypass the immune system in a large number of individuals at one time. One of the mechanisms that enabled the evolution of this stable polymorphism is frequency-dependent selection (Takahata & Nei, 1990).
This kind of selection is significant when the fitness benefit of a particular gene or allele depends on the presence of other genes or alleles in the population or broader environment. For example, the greater number of pathogens with MHC gene x, the less adaptive gene x is in the human MHC, while gene y becomes more adaptive, at least until pathogens mutate to express gene y, in which case gene x may once again become more adaptive, and so on. This situation can lead to the kind of stable polymorphism that we see in the MHC, and that I’m suggesting we find in various elements of our evolved moral sense.
Another example of a stable polymorphism is seen in other aspects of our psychology, such as personality (Bouchard & McGue, 2003; Bateson, 2004; Nettle, 2007; Buss, 2009). It appears as though the diversity in some aspects of our psychology, such as to what extent an individual is introverted or extroverted, is accountable for by genes.
One explanation for this is a similar frequency-dependent selection process that produced the highly polymorphic MHC; if one sees each personality type as promoting a particular strategy of behaviour, then the same frequency-dependent selection mechanism can establish a stable diversity of these strategies within a population.
In addition to innate variation, humans also have a tremendously plastic mind that is able to adapt to changing circumstances in the world, although often without the individual being consciously aware of their changing behavioural dispositions.
There is evidence that our responsiveness to environmental cues also influence moral judgements, such as individuals being subtly exposed to stimuli that evoke negative emotions, such as disgust, being more critical when it comes for forming moral judgements (Schnall, 2008), or individuals being more or less trusting of others, and engaging in correspondingly more or less cooperative or altruistic behaviour, based on their response to environmental cues, such as perception of the quality, or disrepair, of the neighbourhood they’re in (Wilson et al., 2008).
One case study of this phenomenon of moral ecology can be found in the Liberal-Conservative political spectrum as it’s seen in the United States and many other liberal democracies. There is evidence from political psychology that an individual’s personality, emotional responses and cognitive style influence their political attitudes, even when accounting for environmental effects (Jost, 2003; 2007).
In this context, I liken the Conservative responses to a ‘suspicious’ strategy in the Iterated Prisoner’s Dilemma. This strategy is highly risk averse and more concerned with avoiding free-riders even at the expense of missing out on some potential cooperative ventures. The Conservative strategy is also more concerned with tight social cohesion amongst a group of individuals who can be trusted and is less open to interacting with individuals outside the group. This kind of strategy performs well in an environment with many free-riders, although it sacrifices cooperative opportunities in environments with many cooperators.
The Liberal strategy is less risk-averse and is more concerned with fostering cooperation even at the expense of suffering some free-riding. It fosters social cohesion not by strict conformity amongst a core group, but by being tolerant of diversity and of outsiders, and being more ready to attribute trust. The Liberal strategy performs well in an environment with many cooperators but performs less well than the Conservative strategy in an environment with many free-riders. It is also less tightly cohesive, and is more prone to fragmenting.
While environmental factors clearly play a strong role in determining which broad strategy an individual employs, so too do emotional responses and cognitive style, and the latter are influenced by our genes, suggesting an evolutionary influence on an individual’s chosen political ideology (Suedfeld et al., 1994). The result is a kind of political ecology, where a variety of strategies exist, working in tension with one another, but no one strategy is able to consistently dominate.
While this thesis is descriptive, it does have some implications for moral philosophy. It sees morality, at least as it has evolved, as a device that regulates social interaction and encourages cooperation while punishing self-interested and free-riding behaviour.
As such, our moral faculty – including the moral intuitions and moral emotions that so influence moral philosophy – has not evolved to seek out truth per se, but is contingent on human biological interests and sociality. Even if we have a tendency to see moral judgements as being objective features of the world (perhaps also as a product of evolution), they are, in fact, subjective projections, in the spirit of Hume’s or Simon Blackburn’s projectivism.
However, they are not arbitrary; there is something about the way we are composed that influences the projections that an individual will make. This view is compatible with the error theory proposed by John Mackie (1977) and more recently reinforced by Richard Joyce (2001) and Joshua Greene (2002).
Instead of seeing morality as an objective feature of the world to be discovered, it might be more appropriate to understand it as a construct to be created by humans for our benefit. As there are no moral truths in the world – or moral1 facts, as Greene puts it – this doesn’t eliminate morality of the moral2 sort, which deals with issues of social interaction and concern for the interests of others.
This view is also compatible with social contract theories such as those by David Gauthier (1986) and Ken Binmore (1994; 1998). The underlying commonality is that morality exists to advance human interests through promoting pro-social and cooperative behaviour, and each individual agrees to sacrifice some of their freedoms – namely their freedom to free-ride – in order to benefit from the cooperation of others.
However, there is no one strategy as to how best to promote that cooperation, and there is no practical way to eliminate free-riders. As such, a moral system benefits from a pluralism at the level of strategies that work in tension to produce a diversity of responses to any particular situation, making it more likely that a better strategy will be available rather than if the society employed a moral system that adhered to only one strategy – say, Liberal or Conservative – which would likely ultimately prove unstable.
This means this picture of morality – at least as it has evolved, and given the problems it was trying to solve – leans towards a kind of pragmatic pluralist social contract theory that allows the moral ecology to flourish, and thus is more able to remain in equilibrium rather than become unstable.