Emotions are the core drivers of human actions. However, they emerged as drivers of action in the animal kingdom. What I want to do in this chapter is set out how emotions became action drivers for animals, and then explain the key aspects of humans that make their operation different. Finally, I want to set out how they can become distorted in humans and thereby become the origin of self-destructive actions.
EVOLUTION BY NATURAL SELECTION
Evolution by natural selection is the only scientific explanation for the existence of life on our planet. There are philosophical theories such as that preached by Hindu mystics, that the whole of existence and life is an illusion. But this, instead of explaining anything, solves the problem by saying no explanation is necessary. Then there is the non-scientific and non-philosophical theory that some sort of pangalactic metabeing created life and the universe. This theory merely transforms the problem into new problems that we aren’t encouraged to explore: how did the metabeing itself come into existence? If the metabeing built the entire universe in a week, what exactly was he (she? it?) doing for the previous bit of eternity? Why is it that this metabeing happens to be moral? Why should it love us? Was the devil created by the metabeing, or is it a product of natural selection, and if neither, how can we be sure that it wasn’t the devil that was the creator?
God may move in a mysterious way, but fortunately evolutionary theory does not.
The idea that the metabeing theory belongs in science is a lie. Proper scientists direct their research efforts towards the bits of their theories that seem incomplete or contradictory, but the metabeing theorists ignore their problems. People who believe this theory have been attacking Charles Darwin for over 150 years. Meanwhile, they are quietly ignoring the fact that the consequence of winning the argument (which seems unlikely) will be to spend the next 150 years attacking modern astronomy for making it pretty clear that the universe wasn’t created in a week.
There is another implausible piece of the metabeing theory (at least as expressed in the book of Genesis) that tends to get overlooked, and that is that he also created language at the same time that he created the universe; so believers in this theory will then have to go on to refute linguistics. And the philosophical problems that the metabeing theory throws up are simply mind-boggling – for example, how could intelligence, morality and aesthetic judgement pre-exist all life, all matter and (possibly) space and time? How could the designer also be the creator if he existed purely in a nonmaterialistic form – this is the design/creation ambiguity that we aren’t supposed to question. And lastly, if believing the metabeing theory is supposed to give human life a purpose, what is the purpose from the perspective of the metabeing? If he created us purely so that we could worship him, then he was just mailing himself a Valentine’s card. These questions are more difficult than the one the metabeing theory sought to answer in the first place, so why bother? It would be easier to assume that life and the universe has always been here (complete with the fossil record), and so doesn’t need a start-point. At this point, we are back with the Hindu mystics – claiming that no explanation is necessary.
There is only one way for the metabeing theory to be consistent: reject the entirety of science and philosophy. Most of its proponents would happily dump them to get rid of all the above questions, which they find embarrassing. But they want to keep the bits of science that brought them the iPhone and the internal combustion engine. This makes them at best intellectually inconsistent and at worst hypocritical. I actually respect creationists who reject all of modern science. There is something deeply honest and organic about them. They are perhaps more at peace with the world than the rest of us. Usually, to find such people, you have to hack through rainforest for weeks on end, but they crop up in some strange places such as rural Ohio and Pennsylvania.
Charles Darwin19 and Alfred Russel Wallace almost simultaneously formulated the theory of natural selection, but Darwin gets disproportionate credit because he wrote his famous books. However, anybody who troubles themselves with the facts cannot deny that this theory is one of the greatest scientific theories in history. Pieces of the evolutionary jigsaw are still missing, but every year another piece or two falls into place; so the passage of time repeatedly emphasises the genius of the original analysis.
Such was the foresight of the original vision and so comprehensive was Darwin’s writing that there has only really been one major change to the theory since: Darwin and Wallace didn’t know what a gene was. Gregor Mendel discovered the gene during their lifetimes, but the importance of this wasn’t recognised until shortly after they were all dead. It was then left to a generation of evolutionary theorists in the 1960s and 70s to reorient evolutionary theory to a gene’seye view. The emphasis shifted from the survival of the individual to the survival of the individual’s genes through the reproduction of that individual gene carrier. A newly evolved gene must enhance the chances of survival and reproductive success of the individual that carries it, otherwise the gene will not propagate in the next generation.
Darwin coined the term ‘natural selection’20, but Herbert Spencer coined the expression ‘survival of the fittest’. Darwin adopted this latter expression, but we need to look closely at what it actually means. Firstly, ‘fitness’ to a biologist does not refer to the ability of an individual to run a marathon, but to the rate of propagation of that individual’s genes in the next generation.21 Secondly, ‘survival’ does not so much refer to the survival of the individual, as to the survival of its genes as passed on through reproduction. On a preliminary view, the best way for an individual to preserve its genes is to stay alive itself, but this is only true if we either make assumptions about future opportunities to reproduce or a continuing need to nurture offspring. The best illustration why these assumptions do not always apply is to look at spider species where the female promptly eats the male after being fertilised. The male has a greater chance of passing on its genes if it gives its mate a good meal than if it stays alive and goes on an improbable search for a new female to fertilise. Fitness is promoted by behaviour patterns that are individually self-destructive but enhance the probability that one’s genes will survive one’s own life.
Let us now focus on the piece of evolutionary theory that is relevant to this book. The genes that give rise to feelings are selected only if the feeling drives an action that promotes biological fitness. The feelings of hunger and thirst serve the sole and very necessary function of making us seek food and water – they drive actions that promote fitness. The strength of the feeling is somehow proportional to the threat to survival, and as the strength of the feeling increases, gradually that feeling becomes our dominant driver. It is reasonable to say that things necessary for gene survival (eating, drinking, mating) cause feelings that are pleasurable, and circumstances that threaten survival (lack of food and water, or injury) cause feelings that are unpleasant. Such feelings are sensations, and the genes that gave rise to these sensations were naturally selected because they drive an intelligent animal to pursue actions that promote the survival of its genes, and avoid circumstances that pose a threat. An emotion is something beyond a sensation, and I want to consider how emotions evolved. The genes that give rise to emotions were selected because they drive fitness-enhancing actions.
Have you ever tried to swat a mosquito and been amazed at how they always seem able to fly off just before your hand makes impact? What is it that makes the mosquito fly off? We might imagine that the mosquito feels fear, and this makes them fly off. However, as explained in the chapter How Does a Human Work? we have no basis for knowing this. We tend to judge that certain mammals feel fear because they demonstrate behavioural characteristics of fear similar to humans. Critically, this does not just include flight, but behavioural demeanour (wide eyes, tensed muscles, etc.). With an insect, this comparison is not really possible. Because an insect has an exoskeleton, it cannot make facial expressions, and it is hard to know whether its muscles are tensed.
It is of course possible that an insect does experience fear, but with the absence of the behavioural elements that we associate with fear it is impossible for us to relate to the emotions (if there are any) of an insect.22 It is therefore natural for us to presume that the mosquito flies away not because of an emotional impetus (fear), but because of a reflex. Whether this is the case with insects or not, it is almost certainly the case if we go far enough back in the evolution of the animal kingdom.
Recognisable emotion (i.e. to a human) appears to evolve somewhere in the vertebrate stage of the animal kingdom, but we can only be certain of it in fairly advanced mammals. Crucially, the evolution of the emotion of fear can be seen purely in terms of evolution by natural selection.
To illustrate the evolution of fear, it is perhaps best to compare ungulates, or hoofed mammals. Ungulates are grazing mammals, and the problem with grazing is that eating grass requires putting your head down. This means that it is easy for a predator to whack you while you eat. The reason that almost all grazing animals form herds is that they take it in turns to lift their heads to survey the horizon for predators. Taking it in turns to look out for predators is of course pointless unless the members of the herd have the ability to communicate alarm, and different families of ungulates do this in different ways.
The family bovidae includes cattle, and the best-known example of cattle in the wild is the wildebeest. What is notable about cattle is that they have not evolved any behavioural expression of fear other than to run. When cattle are afraid, they do not call out and their faces are impassive. Running is the only sign to other members of the herd that one of their members is frightened, and so the only natural response of the herd is to follow suit. Also, when they run, they do so in the same direction. This, in very simple terms, is why cattle stampede. The problem with cattle stampeding is that it is very easy for predators that hunt in packs (lions, wolves, hyenas, dogs) to get themselves a steak dinner. If the whole herd runs the same way, then you just need one member of the pack to frighten them into running towards the rest of the pack where they will be caught. This is essentially how lions hunt wildebeest: some of the pride frightens from upwind and others lie in wait downwind.
The family equidae includes all horses, and the best-known example in the wild is the zebra. Compared to cattle, horses have developed relatively advanced ways of expressing fear. When a member of the herd detects a predator hiding in a nearby bush, it snorts, neighs and runs back and forth looking at the bush. Other herd members will first look at the zebra that is making the commotion, and will then follow its gaze to grasp the problem. Horses have evolved a complete fear emotion, i.e. the feeling (we presume) and the full expression. This permits them to develop much more sophisticated measures to defend against pack-hunting predators. Unlike cattle that all run the same way, alarmed zebras tend to zigzag and run in different directions to each other. By scattering and regrouping later, it makes them much harder to catch because pack-hunting predators will tend to scatter in the pursuit. Zebras’ stripes make them hard to focus on when they are zigzagging, but stripes on wildebeest all running together wouldn’t work so well. The zigzag strategy only works if all the members of the herd know the initial location of the danger. A zebra has the ability to communicate this to other members of the herd in a way that a wildebeest does not.
The evolution of fear (including the body language associated with it) can be explained in terms of natural selection and survival. However, although horses express fear with complex behaviour and cattle do not, horses do not obviously express any other emotion. This too can be explained in terms of natural selection. Grazing animals only collaborate in looking out for predators. Therefore, they benefit in being able to communicate fear, but (provided they have bonding mechanisms to keep the herd together) they have little benefit in being able to communicate anything else.
I suspect that fear is the earliest form of emotion in evolutionary terms.23 However, it is not the only emotion that can be explained purely in terms of natural selection. Robert Trivers, William D. Hamilton and Richard Dawkins are the leading figures in the relatively new science of sociobiology. Hamilton demonstrated that reciprocal altruism24 (where two species provide a benefit to the other that has a cost to the giver) can be explained in evolutionary terms.
An example of this is a cleaner fish that eats parasites out of the mouth of a host fish.25 The host fish not only refrains from eating the cleaner fish, but has a mechanism where it warns the cleaner fish of approaching predators. This reciprocal altruism benefits both parties, but only because the cleaner fish tends to stay loyal to one host.
The concept of the ‘evolutionarily stable strategy’ (or ESS) is a concept derived by evolutionary theorists from game theory to mathematically predict evolutionary outcomes. A Nash equilibrium in game theory is a pair (or combination) of strategies in a game where no player of the game can improve his position by adopting an alternative strategy unless another player also changes his strategy. Biologists use a modified form of game theory to demonstrate that an evolutionarily stable strategy cannot be invaded by a gene mutation that causes an alternative strategy. This has been critical in describing how social behaviour evolves in animals. For the cleaner fish and his host to be an ESS the cost to the giver has to be smaller than the benefit to the receiver. If each party was receiving less benefit than the cost of giving, then the genes that gave rise to the reciprocal altruism would not be selected in either species.
Reciprocal altruism between species is interesting, but for the purposes of this book it makes more sense to look at reciprocal altruism within a species. Let me start with the concept of nepotistic altruism – that is, reciprocal altruism between family members. The sociobiologists point out that natural selection leads not to the survival of the fittest, but to the survival of the genes of the fittest, and this means the fittest should be the one who passes on his genes to the greatest number of offspring – as measured by the individual’s reproductive success. Nepotistic altruism has an explanation in natural selection because an individual ensures his genes pass to the next generation by ensuring the survival and reproductive success, not only of himself, but of his close relatives. This is because many of an individual’s genes are replicated in his close kin, and evolution selects for the survival of the gene, not the survival of the individual. “An individual that assists his brother to be an ancestor may thereby ensure the survival of the gene-pool of the genes ‘for’ brotherly assistance . . . Parental care is really only a special case of caring for close relatives with a high probability of containing the genes for caring.”26
Kinship theory27 explains that we can compute the degree of relatedness (r) of two individuals within a species. This is the probability that a specific gene in one party’s genome will also be present in another. Humans are a diploid species, where half of an offspring’s genome comes from each parent. For your identical twin, r=1; in other words, you have an identical genome to your identical twin. For each of your parents, r= ½ because you receive half your genes from each of them, and for full siblings r= ½ because there is a ¼ chance that your sibling shares a gene that you inherited from your father and a ¼ chance that they share a gene that you inherited from your mother and you have to add these together. Clearly, this means that for half-siblings r= ¼. For your grandparents, aunts and uncles r= ¼; and for your first cousins r= 1/8. Generally, the arithmetic is straightforward until we consider non-diploid species or families with incest, where it gets a little messy.
Where an action has a cost (C) and a reciprocal benefit (B), then evolution will tend to select against acts where Cr > B. Cost and benefit here should be understood to be cost and benefit in fitness terms – i.e. a cost is a relative decrease in the chances of gene survival in the next generation. Kinship theory is important because if natural selection is concerned with the selection of genes and not individuals; then an individual should have an interest in the survival of a family member proportional to their degree of relatedness. As an example: an animal should attempt to save its drowning full sibling if the chance of being drowned itself was less than ½; but an individual should only attempt to save their cousin if the chance of being drowned itself is less than 1/8.
Kinship theory accounts for some curious aspects of animal behaviour, for example where females mate with many males, siblings will tend to have r= ¼, and where females tend to be monogamous siblings have r= ½. In the latter case, sibling cooperation tends to be higher. An example of kinship theory and nepotistic altruism can be seen in monkey troupes, where the time individuals spend time grooming each other is approximately proportional to their degree of relatedness. Another surprising example is found in ground squirrels. In this species female offspring tend to remain close to their mothers, while male offspring wander off. The giving of alarm calls for predators is altruistic: the benefit (B) is that you warn your relatives (who carry many of your genes) of the approaching predator; and the cost (C) is that you also alert the predator to your own location. Female ground squirrels give warning calls because they can be confident that their close relatives will hear the warning, but this is not the case for males who rarely give warning calls.28
Dawkins gives a simple example of how we can use mathematical models of kinship theory to explain altruism:
I am an animal that has found a clump of eight mushrooms . . . I estimate that they are worth +6 units each . . . The mushrooms are so big that I could only eat three of them. Should I inform anybody else of my find by giving a ‘food call’? Who is within earshot? Brother B (his relatedness to me is ½ ), cousin C (related- ness ?), and D (no particular relation: His relatedness to me is some small number which can be treated as zero for practical purposes). The net benefit score to me if I keep quiet about my find will be +6 for each of the three mushrooms that I eat, that is +18 in all. My net benefit score if I give the food call needs a bit of figuring. The eight mushrooms will be shared equally between the four of us. The pay-off to me from the two that I eat myself will be the full +6 units each, that is +12 in all. But I shall also get some pay-off when my brother and cousin eat their two mushrooms each, because of our shared genes. The actual score comes to (1 x 12) + (1/2 x 12) + (1/8 x 12) + (0 x 12) = +19½. The corresponding net benefit for the selfish behaviour was +18: it is a close run thing, but the verdict is clear. I should give the food call; altruism on my part would in this case pay my selfish genes.29
This gives us a simple theoretical model of why we are altruistic towards family members. However, animals cannot do arithmetic. If they could do arithmetic (like a human can), they would work out that the optimal strategy would be to eat a couple of mushrooms and then give the food call. However, for non-arithmetic animals the only options (in this model) are food call or no food call – selfish act or altruistic act. The model demonstrates that the altruistic act causes the better outcome for the individual’s genes in terms of biological fitness, so, over evolutionary time (that is hundreds or thousands of generations) the genes that create the driver to give the food call will be selected. What is this driver? We can just call it ‘instinct’, or we can be more specific and say that the driver is a guilty feeling whenever we secretly eat food within earshot of our relatives. We can now see how sympathy and guilt have their evolutionary origins. Sympathy and guilt are action drivers. They make us behave altruistically, especially towards kin. Kinship theory can also account for some complex social arrangements. As Trivers explains:
Consider parents with two offspring. Imagine that the first offspring is considering an altruistic act toward the second. It is only selected to act altruistically whenever B > 2C [because for full siblings r = 0.5]. The parents are equally related to the two offspring and would therefore enjoy a gain in reproductive success whenever one offspring acted altruistically towards the other at B > C. There must exist situations in nature when C < B < 2C. In such situations parents are selected to socialize their offspring to act altruistically, while offspring are selected to resist the socialization.
A similar argument applies to selfish behaviour. The offspring is selected to act selfishly toward a second offspring whenever C < 2B. But parents are selected to discourage all selfish acts in which C > B. Since there must be situations in which B < C < 2B, we expect to see conflict between parent and offspring over the selfish tendencies of the offspring.30
Trivers is using a mathematical modelling approach to demonstrate the evolutionary origin of a parent-offspring conflict that every human parent of more than one child must surely recognise.
Reciprocal altruism extends beyond nepotistic altruism, and this too can be explained in evolutionary terms. Dawkins creates a simple model for demonstrating why non-nepotistic altruism can be an ESS.31 Assume an animal needs to groom itself for parasitic ticks, but cannot groom the top of its head, and assume that the animal eventually dies if it is not groomed. (Situations like this are not uncommon in the animal kingdom.) The species will become extinct if nobody grooms anybody else.
Now imagine that we divide the population into ‘suckers’ and ‘cheaters’. A sucker will groom the top of anybody else’s head without expecting the altruism to be reciprocated. A cheater will accept altruism but never bothers to reciprocate. Initially cheaters will succeed because they can always be groomed, and don’t have to waste energy grooming others. This should cause cheating to spread throughout the population. However, as the percentage of cheaters grows, then it makes finding a sucker to groom you difficult, and then even the cheaters won’t be groomed, and eventually nobody will be groomed and the ticks will start to kill off the population overall. In other words, cheating will only survive if the ratio of cheaters remains below a certain threshold. A population solely comprising cheaters is not an ESS.
Now imagine that a new type of individual enters the population: a ‘grudger’. This type of individual acts like a sucker until it remembers that someone it has groomed in the past acts like a cheater, and then it will act like a cheater towards such individuals. The grudger optimises the strategy because it only wastes energy grooming a cheater once. It therefore wastes less energy than a sucker, and it also has a much greater chance of being groomed itself than a cheater. We can build computer simulations of mixed populations and see how they shift over time. This enables us to simulate thousands of generations of natural selection. The pattern is roughly as follows:
Thus grudgers will eventually come to dominate the population. It appears that, given the simple assumptions made here, ‘grudger’ is the ESS. Thus we have an evolutionary basis for the emotion of gratitude.
We need to remember this is only the case because the cost of grooming someone else is less than the benefit of being groomed. If we take another set of assumptions, for example if we assume that the ticks are not a nuisance to the species (i.e. the benefit of being groomed is very low) then it is probable that cheater is the ESS. By playing around with the assumptions, we can work out mathematically that there might be other ESS’s, and it is possible that there could be a stable mix of cheaters, suckers and grudgers all coexisting together. However, we can work out certain general conditions that are necessary for same species non-nepotistic altruism to be an ESS. These include: 1) Individuals live long lives; 2) Individuals have effective memories of who has acted altruistically towards them; 3) Individuals tend to stay in relatively stable groups for long periods; 4) Offspring remain dependent upon their parents for long periods. In such circumstances non-nepotistic altruism has advantages to survival and becomes an ESS.
In fact, many animals live in groups where altruism works as a hybrid of nepotistic and non-nepotistic altruism. For example, in a pride of lions there might be several adult females but only a couple of adult males. Female offspring remain with the pride, but male offspring are evicted when they mature and spend time living alone before trying to oust the males of another pride. Lions raise their young collectively, i.e. they don’t know which lioness gave birth to which offspring. However, it is possible to build a statistical model for each pair of lions in a pride, and work out the average degree of relatedness of all the lions to one another within the pride. It is actually slightly below ¼, or the degree of relatedness of half-siblings or grandparents. Over time, natural selection would ensure a degree of altruistic behaviour within the pride that was in equilibrium for this average degree of relatedness.
This gives us a theoretical framework to account for the evolution of many emotions that act as drivers of social functioning, since these apply not only to humans but to other advanced animals that live in complex social groups. The psychologist Jonathan Haidt has compiled a natural history of emotions that exactly corresponds with what the work of Trivers would predict: sympathy or pity drives altruistic acts without immediate hope of return; gratitude drives bonding with people who show altruism to us; shame drives us to return favours; and guilt deters cheating.32 These emotions are action drivers, each of which drives a set of actions that enhances biological fitness. We therefore have an evolutionary explanation for what philosophers call the ‘moral emotions’. Importantly, we can see evidence of the operation of these emotions in chimpanzees and other advanced mammals.
As shown by Dawkins’ theoretical example, reciprocal altruism requires the evolution of mechanisms to detect cheaters, and this can account for the evolution of anger. Anger as a mechanism for punishing cheaters can be observed in chimpanzees, which throw temper tantrums in a fashion very similar to the way humans do.33 This is why the study of primatology is so important to our understanding of ourselves. Trivers further refines this analysis by making the distinction between ‘gross cheaters’ (who return no altruism) and ‘subtle cheaters’ (who return altruism only sufficiently to avoid detection as cheaters). He points out that this accounts for the fluid relationships between friends where altruistic pairings shift when one party detects subtle cheating in his soon-to-be ex-friend.34 Again, all of the behaviour that this theory would predict has been observed in chimpanzees, and clearly provides clues to the origins of such relationships in humans.35
Sociobiology is not the only newly emerging science that provides clues as to the evolution of emotions. Evolutionary psychology is the discipline that explains human behavioural characteristics purely in terms of evolutionary theory. Evolutionary psychology could be said to be an extension of sociobiology applied solely to humans. Martin Daly and Margo Wilson are evolutionary psychologists who have focussed on homicide and male violence. “Should a more aggressive type of male appear in a population of long-lived pacifists, for example, and should the new type tend to fertilise more females but die younger than the old, then that new type will supplant the old by natural selection, and male life span will decline.”36 In a society where competition exists for mating partners or resources, conflict arises. Daly and Wilson demonstrate that male violence is generally prompted by threats to fitness (i.e. the ability to pass on genes) and is directed at the threat. Individuals who gain a reputation for violent retaliation are very unlikely to be the victim of opportunistic exploitation by either females (who might deceive a male into assisting in raising another male’s offspring) or rival males (who might steal his females).
The reason that violence is a particularly male trait also has an evolutionary cause. To understand this, we first need to understand the concepts of ‘sexual selection’ and ‘parental investment’. The concept of sexual selection is the process of choosing mates: in mammals it is generally the case that females choose males and males will mate with almost any female. Darwin was the first to explain this37, but its significance was ignored for almost a century. In mammals, the female invests much more in parenting than the male, since she has to carry the foetus in her womb and then suckle the newborn infant. The female investment in parenting therefore can take months or years, but the male investment can be limited to a single teaspoonful of semen. For example: Moulay Ismail Ibn Sharif, a seventeenth-century Moroccan King, fathered more than 850 children.38 Needless to say, he couldn’t invest in the care of his offspring because he was too busy conceiving their siblings. Female mammals do the selecting because reproduction places a much greater burden on them – it matters more to them that they get the right mate because it will be a long time before they get another chance.
This logic would not apply to a species where both parents make an equal parental investment. The greater parental investment of females also makes them the bottleneck in the reproductive process, which is why males compete for females and not the other way around. A successful male can father offspring with many females, but an unsuccessful male will father no offspring, so males have a higher variability in reproductive success. This leads to violent conflict among males. The tendency for successful males to breed with many females is known as polygyny, and if we compare the degree of polygyny across different species key characteristics emerge that are exactly what the evolutionary theory would predict. As the degree of polygyny increases:
For example, red deer, which are highly polygynous, have substantially larger males, equipped with huge antlers for fighting, and male mortality in fighting is high. In contrast, monogamous antelope species that share parental care exhibit two sexes of identical size, with identical antlers, and almost identical life expectancy.
If this theory of male violence is correct, we should expect female violence to be pronounced in species where the reproductive roles are reversed. And indeed we do! A Polyandrous species is one where successful females are likely to mate with several males. For this reversal of reproductive roles to be possible, we would need to have males making the higher parental investment. This would mean that males were the bottleneck in reproduction, so females would compete for males, males would have the role of selecting mates, and females have more variability in reproductive success. In these circumstances, we find that females are more aggressive. This set of circumstances isn’t possible in a mammal because embryos must be carried by the female and the young suckled by her, but neither of these factors apply to birds. Several wading bird species are polyandrous (spotted sandpipers, Wilson’s phalaropes and jacanas).39 In these species, females are larger and more brightly coloured; they aggressively compete for male partners; and crucially, the male incubates the eggs after the female has laid them and then cares for the young. A successful (and therefore aggressive) female can raise several broods with different males, and an unsuccessful female will raise none. Polyandry, together with these resulting behavioural consequences also arises in seahorses and several insect species.
Humans are a polygynous species. Most societies legislate monogamous marriage, but even in these societies males are more prone to have further offspring from second marriages, and it is likely that successful men have more covert reproduction. Every study that is not motivated by some dubious political agenda consistently demonstrates that male humans in every society are more violent than females. This debunks the idea that males are more violent because they are “socialised” differently.40 Greater male violence can be seen in tribal societies, where violence leads to power and status and dominance of breeding rights over females. However, in advanced societies this effect is masked because the state usurps the legitimate use of violence.41
If successful males of a polygynous species (such as humans) breed with more than one female, then some males inevitably fail to breed. A male who is facing complete reproductive failure should (in evolutionary theory) be prepared to adopt desperate measures to pass on his genes. Thus violent and risky behaviour in human males is explainable in evolutionary terms even if a possible outcome is the death of the individual. This explains why human males of low social status are more prone to violence, and why they are most violent at the age at which they are seeking breeding partners – early adulthood. Reproductively successful males subordinate unsuccessful ones, and unsuccessful males need to break free from this subordination to have any chance of breeding.42 To optimise our biological fitness, men need to have sufficiently violent revenge acts to counteract humiliation. This gives us an evolutionary basis for why human males enact violent revenge as a response to humiliation. This is an important conclusion, and later I will draw upon it to explain some of the most grotesquely violent aberrations in human history.
The evolutionary theory surrounding parental investment and sexual selection also gives us a basis for explaining jealousy. Male animals have a problem of certainty of paternity. Certainty of maternity is not a problem at all for female mammals, but can be a problem for female birds where another female may dump eggs in her nest. Different species cope with this differently: In many species, the male copes by mating with as many females as possible, but makes no parental investment beyond copulation. In species where parental care is shared (most birds, most primates and many rodents) males are jealous because they are selected to resist cuckoldry: The genes of a male who makes a parental investment in another male’s child won’t pass to the next generation and so aren’t selected. The male will either watch the female during her fertile period or will actually confine her and guard her. For male birds, there is a double problem of guarding the female against fertilisation by another male, and guarding the nest against egg dumping. This explains why birds are so intensely territorial during the breeding season – it is necessary for the couple to stay away from all other birds of the same species.
Humans share parental care, so jealousy has an evolutionary basis. However, experiments of infidelity role-playing performed on dating college students produced completely different responses in men and women: men had fantasies about sexual contact between their girlfriend and their rival, whereas women were mainly concerned with their boyfriends investing time, money and attention on their rival.43 This is absolutely what evolutionary theory would predict. In evolutionary terms, a female whose partner cheats is only having male parental investment in her offspring diluted; whereas a male whose partner cheats is risking raising another male’s offspring – the ultimate evolutionary fail, which explains why men react to cheating more severely than women. Cuckoldry is the most serious problem for males, but benefits females, who can take sperm from a strong but uncaring male and deceive a weak but caring male into helping her raise them – the evolutionary slamdunk for females. Jealousy exists for evolutionary reasons, and the different form of jealousy between the sexes similarly has an evolutionary basis.
To summarise, evolutionary theory, particularly that which has emerged since the 1970s, gives us a basis for explaining the existence of many human emotions. The moral emotions drive actions that promote cooperation in animals that live in social groups. Humiliation, revenge and jealousy drive actions that enhance male fitness, maximising the chances of a male breeding and preventing him from being a victim of cuckoldry.
THE IMPACT OF LANGUAGE
A critical stage in the evolution of emotion is emergence of language.44 Language is, more than any other thing, what distinguishes humans from other animals. There is a boundary separating thought processes that require language from those that don’t. Scientists studying animal behaviour approach this boundary from one side, and philosophers and scientists studying technical linguistics approach from the other. For the most part there is not much disputed territory regarding where the boundary lies.45 There are two separate (but related) consequences of the development of language in humans; the first is that it permits us to form concepts; and the second is that it considerably increases the potential range of emotions that humans can experience.
There are two separate concepts that humans can form that I want to explore. The first is that we form concepts of abstract space and time. This completely transforms the way that we humans can conceive of ourselves in our environment. It permits us to think outside of ourselves, and imagine being in another place or time. Without this, we could not imagine our future, and without that we would have no incentive (or, possibly, ability) to plan. It has become orthodox thinking among animal behaviourists that animals cannot plan. A squirrel that hides nuts does so instinctively, but has no understanding of the approaching winter, or any knowledge that soon the trees won’t bear nuts.
Secondly, humans form concepts concerning emotion and behaviour. The combination of this and the concept of time permits a human to conceive of future emotions. This has a curious effect: it creates a situation where the causality of action is reversed. Let us consider hunger. We don’t usually think of this as an emotion, but it is a good example because nobody is confused about what it is.46 Hunger, like an emotion, is a feeling that is a product of evolution by natural selection: it drives an action that promotes biological fitness. Hunger drives an animal to seek food and to eat. Until it is hungry, it will not perform these actions. So how is this different in a human? A human can conceive of future hunger, and can therefore plan its avoidance. So we take action in advance of our hunger to strategise its avoidance. This creates a reversal of the causality of action. We could say that in an animal, the feeling pushes the action, but in a human the action pulls the feeling. The same can be said of emotions. Fear drives an animal to run away, but for humans, the concept of hypothetical future fear drives a human to strategise its avoidance.
It is common for evolutionary biologists to talk about animals being “survival machines”. If a human is solely the product of an evolutionary process, then the fact that a human can self-destruct is an apparent contradiction that this book is focussed on explaining. The apparent contradiction is resolved by looking at the ways that humans strategise emotional outcomes. So whereas we can think of an animal as a “survival machine”, we have to think of a human differently. I am going to assume throughout this book that a human is a “machine that goal-seeks emotional outcomes”. Goal-seeking is a computer science technique47 that enables us to calculate backwards, and a human brain is a computer: we seek the output (an action) of a formula that would result from a given input (an emotional goal); since I am exploring the reversal of an evolved causal mechanism for driving actions, this analogy is apt. Goal-seeking emotional outcomes involves humans forming concepts of future emotions, and implicit theories as to what causes them. A human is driven by the need to create for itself an environment that achieves favourable emotional outcomes and avoids unfavourable ones. What I am seeking to demonstrate is that all our actions are determined this way.
*
So if language fundamentally changes how emotions drive human actions, can it also create new emotions? In the previous part of this chapter, I examined emotions that could be said to have evolved by way of natural selection. In other words, they arose because they enhanced the chances of gene survival. Let us compare fear with anxiety. Fear is an emotion that is common in the animal kingdom. But we could consider fear and anxiety to be related. We could also argue that anxiety has an evolutionary origin in that it drives us to evade foreseeable threats before they present themselves to our sensory perception. But could this arise in an animal that doesn’t have language?
Generally, we consider fear to be triggered by the perception of immediate danger. I recall a photograph of a zebra that had just been grabbed by a crocodile on a riverbank. The zebra’s eyes bulged, its lips were drawn back to reveal its teeth and all the veins in its neck bulged with the tensing of its muscles. Anxiety is generally triggered by the belief in a continuous danger. Define it like this, and already the idea of anxiety in animals is running into problems because it is difficult to understand how belief can arise without language. (Your dog can see you put its food in its bowl, but can we really say that it believes it is about to be fed?) Anxiety, in comparison to fear, is more of a slow burn emotion – more constant rather than immediate. In humans, it also has a different body language – furrowed brow, rather than the muscle tensing required of fight or flight. It is difficult to imagine an equivalent to this body language in an animal. Without the body language, we have no way of knowing if they experience the emotion itself.
Does this mean that animals do not feel anxiety in the way they feel fear? We could return to the example of the grazing herd where each member of the herd looks up periodically to survey the horizon for predators. I have argued that their urge to run is caused by fear, but is their urge to look up caused by anxiety or is it just an instinctive habit?
I think this question can be clarified if we look at the difference between fear and anxiety another way. Fear could be said to be a response to a present danger, and anxiety could be said to be a response to a non-present danger. Anxiety has a speculative quality: it needs a “what if . . .?” element to frame impending doom. This transforms the question: can an animal that has no language have awareness of non-present danger? To be aware of a non-present danger, we need to be able to conceptualise sources of danger in abstract time and space. This is surely only possible with language. A child can worry that there is a monster under the bed, but surely a puppy cannot.48
Philosophers have long doubted that animals experience anxiety. Wittgenstein asked: “A dog believes his master is at the door. But can he also believe his master will come the day after tomorrow? – And what can he not do here? – How do I do it? – How am I supposed to answer this? Can only those hope who can talk? Only those who have mastered the use of language.”49 Similarly, Kierkegaard said “Time does not really exist without unrest; it does not exist for dumb animals who are entirely without anxiety.”50 And Rousseau argued that an animal cannot have a concept of its own death.
Once we have realised the way in which language expands the possible range of emotions, we can attempt to divide emotions into pre-language and post-language emotions. Clearly, fear is a pre- language emotion. I believe that anxiety is a post-language emotion because awareness of dangers that we cannot currently detect with our senses requires us to have a concept of the danger separate from the perception of the danger itself. This is language-dependent, but difficult to clearly define. Certainly there is a grey area between fear and anxiety and the boundary is not always clear. Hope is almost certainly a post-language emotion because it requires a concept of future.
There is nothing in our experience of happiness that would lead us to believe that it is a post-language emotion. However, in the next section I will argue that it certainly is. This is a surprising conclusion, and makes it clear that we cannot tell from our experience of an emotion whether or not it is pre or post-language. The reason for this is that we can no longer remember what life without a language is like. This is because, among its myriad applications, language is an instrument of memory – humans remember things through the medium of language. Our inability to know this by examining ourselves is why the study of chimpanzees and other animals is crucial to the understanding of humans.
Evolutionary psychology and sociobiology also give us clues as to certain other emotions having origins before the evolution of language. Are there other examples of pre-language emotions that I have not already mentioned? It is difficult to be certain, but one likely possibility is grief. There is evidence that elephants sometimes cover their dead with sticks and leaves, and chimpanzee mothers have been known to carry their dead offspring for days after death.
Certainly, grief is a very old emotion in humans. Translations of ancient poetry give rise to dangers because poets tend to assume an understanding by the reader of the words they use. So-called ‘translations’ of poetry are often recreations of poetry in modern language from ancient texts. A translation of a word relating to an emotion from a poem therefore runs the risk of not representing the intended feeling (if there was any) of the original poet, and we can have no means of knowing this unless the poet describes each of the cause of the emotion, the feeling and the behaviour. This comprehensive description of emotion in ancient poetry is perhaps only found in the case of grief. Certainly, many ancient poems are loaded with it. Gilgamesh, probably the oldest known work of literature, was written approximately 4,700 years ago.51 It contains extensive descriptions of its central character suffering from grief following the death of his friend. However, as I will explain soon, ancient Chinese philosophers saw grief in a way that makes it sound not completely like an emotion, which illustrates the problems of knowing for sure.
Arguably, the distinction between a pre-language and post-language emotion has little theoretical significance. However, it is important to realise that for pre-language emotions, the story does not end with natural selection because the subsequent emergence of language can further alter these emotional states. Maybe anxiety is a post-language distortion of the pre-language fear emotion; but language changes the fear emotion even when we regard it in isolation. An animal is fearful of anything that chases it, but without a concept of time and space (which are language-dependent) it cannot visualise its own death. A human can visualise its own death, so fear is altered by language: animals fear threats, and humans fear the consequences of threats. A zebra fears the lion that is chasing it, but a human chased by a lion fears the consequence of being caught. This transformation occurs simply because a speaking human can conceptualise the consequences, whereas a non-speaking animal cannot.
There is one other sense in which language alters a pre-language emotion, and this is the origin of morality in humans. As I argued in the previous section, the moral emotions are pre-language emotions. But as critics of W.D. Hamilton point out (irrelevantly in my opinion) his explanation of the evolution of altruism explains behaviour that has the effect of altruism without the intent. This is correct. But language gives humans the concept of their own behaviour, and they understood the intent after the behaviour evolved. In other words, altruism (without intent) could be driven by pre- language emotion, and humans observing their own behaviour worked out why they were doing it. This means that the behaviour (without intent) evolved first, and the intent was an ex post facto discovery that was dependent upon language. Chimpanzees have moral emotions without having a morality because without language they lack the intent.
THE IMPACT OF PHILOSOPHY
Evolution by natural selection can explain how human emotions evolved, but can it account for happiness?
Every owner of a dog assumes that tail wagging is the body language of a dog that is happy. Dogs are pack-hunting animals that develop complex behaviour and social structures. This includes behaviour demonstrating dominance and subservience. Dominance and subservience are necessary for enforcing mating rights (of the individual) and territorial rights (of the pack). However, dominance and subservience must be suspended during hunting. Hunting requires teamwork, but no member of the pack can give orders without language. Teamwork requires bonding, and in particular it requires the identification of pack members and the ability to distinguish members of rival packs. It is difficult to explain how the survival of dogs is improved if they have an emotional state of happiness. Simply put: what is the benefit to the survival of the species in being able to communicate a sense of wellbeing to other members of your herd or pack? Observation in the animal kingdom does, I think, provide almost no evidence that animals experience happiness. The laws of survival dictate that any weak member of a pack will simply be left behind to die. Even animals that demonstrate powerful maternal instincts will show little sentiment towards sick offspring. For example, elephants will ultimately abandon orphaned young that can no longer suckle. It therefore seems likely that tail wagging is not a sign of happiness, but is more a sign of greeting. This is not the same as saying that absence of tail wagging is not an indicator that something is probably wrong with your dog. It is simply saying that it is incorrect to say that tail wagging is a sign of an emotional state.
Tail wagging is not a way for a dog to say it is happy. It is a way for a dog to say “let’s suspend the dominance behaviour for a minute because I am about to sniff your genitals.” In any case, how can humans imagine that they can identify the emotional state of another species that demonstrates entirely different body language? Tail wagging in humans does not indicate happiness, unless they are dancing the samba. It is entirely a coincidence that dancing the samba often precedes genital sniffing in humans.
So if the emotion of happiness does not arise in the animal kingdom, when and how did it first appear?
A baby’s first smile is always a memorable moment. Clearly, this is something instinctive – innate. But can it be taken as a sign of an emotional state, or is this bonding behaviour? As bonding behaviour, it has an evolutionary explanation. For an infant to be able to demonstrate wellbeing is difficult to explain in evolutionary terms. To survive, a baby only needs to tell us when something is wrong which, as any new parent knows, they manage perfectly well. Charles Darwin admitted that he could not distinguish between pleasure or joy as emotions, and affection or greeting as bonding mechanisms .52
To many of us, smiling is considered to be an indicator of happiness. But could this really be a bonding signal that has become habitual? Certainly, there are places in the world where people hardly ever smile. In parts of China, people will only smile if you smile at them, and this is more of a greeting (like tail wagging) than a sign of happiness.
Searching for the origins of happiness in humans is tricky because we have no way of observing historical behaviour. However, we can find clues in ancient writings. Ancient poetry and drama seems to refer to happiness, but translation problems cloud the issue. A close reading makes it appear unlikely that the poet or dramatist considered a feeling to be involved. For example Aeschylus, writing in the fifth century BC, talks almost continuously of every conceivable grief and lamentation, but only obliquely about happiness. George Thomson’s translation of Agamemnon refers to a “happy state”,53 but Anne Carson’s translation interprets the exact same passage as a “happy change”54 which doesn’t sound like an emotional state. Thomson’s translation is closer to a literal translation, but Carson’s is more a recreation of ancient drama in contemporary language. Again, Thomson’s translation refers to a “happy life”,55 but Carson’s version makes no such emotional reference in translating the exact same passage. However, it isn’t clear that Aeschylus ever seems to consider a happy feeling. It is therefore difficult to conclude that he sees this as an emotion. Likewise, Euripides writing around the same time delivers negative emotion to the point of overload. Indeed, he took it so far that Aristophanes mocked him.
Looking for clues in poetry is not very reliable because poets simply assume that their readers understand the words they use. It is therefore possible that a word that we think we know how to translate might have had a completely different meaning at the time it was written. What I therefore want to do is to analyse ancient philosophical writings because philosophers need to be more precise, not in their choice of words, but in explaining exactly what they mean by them. Crucially, this is less likely to be lost in the translation process because translators of philosophy must retain the meaning intent of the philosopher, whereas translators of poetry are more concerned with retaining the poetic quality.
My aim here is to demonstrate that happiness is not innate in humans because there was a relatively recent time when it didn’t exist. This in turn means that, unlike the emotions discussed in the first part of this chapter, it certainly did not evolve by way of natural selection and, additionally, it is certainly a post-language emotion.
One of the earliest Western references to happiness arises in the Nicomachean Ethics56 of Aristotle (384–322BC). Aristotle wrote the Ethics in 350BC.
Aristotle is seeking to explain human behaviour: we want money because it permits us to buy things; we want health because it permits us to work and live an active life; we want honour because it enables us to interact with our fellow man and have their respect. But is there anything that men want purely for its own sake? Aristotle’s answer is that happiness is the ultimate end, since we want it for its own sake. We want other things solely as a means to happiness.
Aristotle’s view would seem to be in agreement with how most people would consider happiness today. In my shorthand terminology, a human is a machine that goal-seeks happiness. However, a careful reading makes it clear that Aristotle’s conception of happiness is quite different from how we think of it today.
The Ancient Greek word that Aristotle uses is ‘eudaimonia’. This word is generally translated as “happiness”, but a literal translation means something closer to “living well” or “having a good life”. Most scholars now agree that Aristotle was not referring to an emotional state when he used this word. Rather, eudaimonia is an “activity of the soul”.57
Aristotle sees the soul as divided into two parts – the rational and irrational. The irrational soul is further divided into the vegetative soul (which plants also possess) and the appetitive soul representing the desire to satisfy bodily needs (which is common to all animals). Aristotle distinguishes between intellectual virtue that comes, via the rational soul, from learning; and moral virtue that comes, via the irrational soul, from habit. Aristotle believes that eudaimonia comes from developing good habits. In time, one learns to develop pleasure in performing good actions.
For Aristotle, eudaimonia is not an emotional state; it is an action. He says that it cannot be measured until close to the end of your life. But eudaimonia cannot be an emotion if one cannot know it concurrently with life’s experience.
It has been suggested that early languages were too simple to fully convey emotion. That they were structured in such a way that they could only be used to describe material things and their properties,58 and that this could imply that happiness has always been a human emotion, but the language to describe it only emerged later. However, I don’t think this is true of Aristotle. This is made clear when he discusses other emotions; for example he is explicit in stating that shame is a feeling, and not a state of character.59
Prior to Aristotle, the word eudaimonia occurs several times in Ancient Greek philosophy, but none of these occurrences come with a thorough definition of what it is. Socrates (469–399BC) held a number of beliefs on eudaimonia that would also indicate that it was not an emotion: for example, that it was caused by scientific thinking60 or virtue alone.61 Again, Socrates does not go on to define this in terms of feeling, but it seems difficult to imagine it as an emotional state if it is solely the result of correct behaviour rather than being correlated to the outcome.
One other Ancient Greek reference connected to happiness is in Democritus (approximately 460 to 370BC). Democritus became known as the “laughing philosopher”. However, don’t imagine that this is laughing as you and I now know it. The laughing for which Democritus was famous is often translated as ‘scoffing’, as in “scoffing at the foolishness of men”. His contemporary nickname was the Mocker, which seems to support the view that his laughing was not mirthful. This reputation notwithstanding, Democritus wrote several works on ethics, one of which is On Cheerfulness in which he says that cheerfulness (‘euthumia’) is the goal of life: “Cheerfulness arises in people through moderation of enjoyment and due proportion in life.”62
However, like Aristotle’s eudaimonia, it is almost certain that Democritus did not think of euthumia as an emotion. The principal biographer of the Greek philosophers, Diogenes Laertius, writing some six centuries later summed it up like this: “The goal or summation is euthumia, though this is not the same as pleasure (as some have mistakenly understood it to be . . . ), but is the state according to which the soul is in a calm and well-balanced condition, disturbed by no fear, superstition or other emotion.”63 This does not sound like a positive emotional state, but an absence of negative emotional states, something that is consistent with my view that the concept of positive emotional states had not yet arisen. Democritus referred to the idea of a calm absence of emotional states as wellbeing (‘euesto’).
My view that happiness didn’t exist as an emotion in ancient times is also supported by early writings from the East. Kung-Fu Tzu means “The Master Kung-Fu”, but he is generally known in English as Confucius (551–479BC). He is China’s most famous philosopher. Mencius (Meng Tzu, or The Master Meng, 372–289BC) is considered to be one of the principal interpreters of Confucius, although he is clearly a philosopher in his own right. Both lived in Eastern China. As a comparison, the life of Confucius just preceded Socrates and Democritus, and Mencius overlapped with Aristotle. Confucius and Mencius lived during a period of Chinese history known as the Warring States Period. Confucius sought to end the constant civil war by demonstrating that moral behaviour was to the benefit of all people. He set out to demonstrate the correct behaviour of both rulers and the ruled. In effect, Confucius sought to re- establish altruism by reason, since its prior establishment by evolution had become disrupted – perhaps because humans began to suppress their moral emotions.
Confucius was concerned with teaching morality. However, his ethics are based in doctrine. There is really no evidence that Confucius considered human nature to be a distinct concept. In particular, he seems largely unaware of emotion.
The Chinese word ‘wen’ in Confucius is often translated as ‘culture’, but more properly means “what beautifies human life”.64 However, Confucius never discusses how this makes one feel. Confucius discusses mourning, but he does not seem to specify what the feeling element of it is. He describes mourning in terms of ritual and considers that breaking from this ritual would result in the mourner not “feeling at ease”.65 However, this is far from a description of the emotion of grief – in fact, the cause of the feeling he refers to is ignoring the ritual, not the death of a family member. And yet, he does have a concept of insincere mourning: “the forms of mourning observed without grief – these are things I cannot bear to see!”66 This is an important statement because it makes it clear that humans understood that they performed tactical deception with their emotional behaviour in the fifth century BC. But Confucius is not clear what separates sincere from insincere grief – a distinction that we would see in terms of the presence of the feeling. (By way of comparison, ancient European literature tends to refer to grief more than any other emotion.) In ancient China, music was not listened to for pleasure, but was considered to have magical power. Confucius saw his response to music not as an emotion, but as evidence of its magical power.
Happiness gets very little mention in Confucius, and there is nothing to enlighten us as to what he thinks it is. We think of happiness as a positive emotional state, but contentment is more an absence of negative emotional states. With ancient writings, it does not appear that they had any concept of positive emotional states, but in translation, the word ‘happiness’ is often used when it isn’t clear that an absence of negative emotional states wasn’t the original intent. For example “The Master said, He who seeks only coarse food to eat, water to drink and a bent arm for pillow, will without looking for it find happiness to boot.”67 A reference to unhappiness also appears, where it is unlikely that the original meaning is how it sounds in translation. The Three Ways of a true gentleman are “He that is really good can never be unhappy. He that is really wise can never be perplexed. He that is really brave is never afraid.”68 The Chinese word ‘jen’ is generally translated as ‘good’ but more appropriately means, “possessing the qualities of one’s tribe”.69 Arthur Waley compares this with one Englishman saying that another is a “true Englishman” as intended to imply a high form of praise. However, it occurs to me that there is another clear explanation: the sign of tribal bonding being taken as a sign of personal wellbeing. Compare the wagging of a dog’s tail as it bonds with its pack members and the perception that this means that it is happy. Humans are, like dogs, pack animals. Happiness and being within your tribe or pack are not the same thing, but could over time come to seem so.
Further evidence of the absence of a concept of emotion at the time of Confucius can be seen in the Tao Te Ching. The date of this is uncertain, although it precedes Mencius, and there is uncertainty as to whether its author Lao-Tzu (literally “The Old Master”) was a historical figure, so the Tao Te Ching might be a collection of sayings from many people. There is a story about Lao-Tzu and Confucius meeting, but it is uncertain whether this is historical fact. However, there is no evidence the writer (or writers) of the Tao Te Ching showed any awareness of the concept of happiness. In fact, there is plenty of encouragement in the Tao Te Ching to be rather dour. Desire in all its forms, according to the Tao Te Ching, is seen as a cause of incorrect behaviour. It teaches that knowledge does not lead to virtue, and that contentment comes from letting things just happen – a philosophy of non-intervention in the natural course of events.
The concept of emotion (at least positive emotion) has not emerged in China, and the concept of feeling is only just appearing. The concept of feeling appears in a more evolved form two centuries later in Mencius, and even then it is not really quite how we understand it today.
The Chinese word ‘xin’ is often translated as ‘feeling’, but is also sometimes translated as ‘heart-mind’. These alternate translations seem to have little in common, and this highlights the difficulty in translating ‘xin’, though the term ‘heart-mind’ “is most adequately understood as a function rather than an organ”.70 This likens it to Aristotle’s use of the term, where eudaimonia is an action, not an emotion. Crucially, the word ‘xin’ appears only six times in the Analects of Confucius, but 119 times in the writings of Mencius.71
Confucius considered that morality could be learned but does not appear to have considered whether men have a nature – either good or bad. Mencius considered that men were fundamentally good; so morality was innate in men. Feeling is important for Mencius because it is the origin of the moral force. It is what we should look to for moral guidance. He does not see it as a source of selfish motivation in humans, but it is starting to emerge as a concept of a human disposition. “Feeling,” says Mencius, “guides us more surely than doctrine in charting the most productive course”.72 Mencius is moving away from the Confucian concept of a morality defined entirely by ritual patterns of behaviour, towards a more humanistic morality that builds on natural human responses.
The following is perhaps the most famous passage in Mencius. It gives us an insight into his entire philosophy:
Each and every human has feelings sensitive to the suffering of others. The former kings had such feelings, and these were manifested in their compassionate governing. Putting such feelings to work in governing, they ordered the world as easily as turning it in their palms.
As for each human having feelings sensitive to the suffering of others, suppose a person suddenly sees a child about to fall into a well. Each would feel empathy for the child – not in order to gain the favour of the child’s parents, nor to win the praise of villagers and friends, nor out of concern for a potentially blemished reputation.
From this we observe the following: without such a feeling of commiseration one is not human, without a feeling of shame one is not human, without a feeling of deference one is not human, without a feeling of discrimination one is not human. A feeling of empathy is associated humanity in its germinal state. A feeling of shame is appropriateness in its germinal state. A feeling of deference is ritual propriety in its germinal state. A feeling of discrimination is wisdom in its germinal state.
People have these four sprouts just as they have four limbs. For one to possess these four sprouts yet consider oneself incapable of developing them is self-mutilation; for one to consider the ruler incapable of doing so is to mutilate the ruler.
For anyone having these four sprouts in him or herself, to realise their enlargement and bring them to ‘fullness’ is like having a fire catch or a spring break through. If these germinal beginnings are brought to fullness, one might safeguard the whole empire; if they are not, one might not even tend to one’s own parents.73
This passage is often referred to as Mencius’ philosophy of the four tender shoots of human nature. Feeling creates a genuine moral impulse that requires no deliberation before acting. This analogy is important because the Confucian concept of a moral ruler is modelled on his idea of the instinctive protection of parents for their children. Compare Mencius in this passage with the sociobiologists and evolutionary psychologists who came over two millennia later. They argue that natural selection provided humans with the emotions that are necessary for altruism and social cohesion – in essence evolution provides us with the tools required for morality. Mencius is arguing that we should trust in nature. Correct action follows from the impulse that nature provides us.
In the progression from Confucius to Mencius, we can see the beginnings of the human awareness of feeling, but there is still not a concept of positive emotion. Confucius parallels Socrates in equating happiness with virtue without describing what happiness is in any way. Mencius can loosely be compared to Aristotle in seeing feelings as a motivator without having a concept of positive emotion. It remains a great mystery why Eastern and Western thought should be so convergent at this stage of history. The mystery extends beyond Greece and China. At approximately the same time as Confucius and Socrates, Zoroaster in Persia, Isaiah in Judah, and Siddhartha Gautama in Northern India and Nepal were developing broadly similar ideas without there being any evidence that they had knowledge of one another.
Siddhartha Gautama is popularly known as the Buddha. Buddhism teaches that enlightenment or virtue comes from the avoidance of desire or craving, including the desire for continued existence. This is very similar to the Tao Te Ching. This could be interpreted for my purposes as the avoidance of negative emotional states without really defining positive ones. The origins of Buddhism are deeply entrenched in the concept of negative emotions. According to the legend (or history, depending on your belief) Siddhartha Gautama lived in luxury in a palace. The beginning of his journey in becoming the Buddha started when he left his palace for the first time and saw old age, sickness, death and finally a monk. Siddhartha Gautama chose the life of a monk and set out on his course of seeking a way to escape from the cycle of suffering – or negative emotional states. Buddhism teaches that enlightenment is the goal, but there is no suggestion that enlightenment is happiness, or even an emotional state at all.74
Approximately 100 years after these teachings, Patanjali, the author of the yoga sutras, talked about achieving “bliss” through meditation, ethical practice and the study of philosophy and metaphysics. This might be said to be an early emergence of an awareness of positive emotional states. However, it still isn’t clear that this is happiness. Bliss could be seen as an early concept of happiness, or it could also be seen as a form of inner peace – the absence of negative emotional states.
Putting all these strands of ancient teaching together, it seems clear that happiness as an emotion is not something natural or instinctive. It was absent in ancient times, and our concept of it being a distinct emotional state emerged later. This leaves open a key question: what was the process by which happiness came into being? It appears to be the bringing together of an instinctive gesture of greeting or bonding, namely smiling, with teachings of morality and virtue. We could perhaps argue that Aristotle (for example) invented it. But with philosophers from history we can never really be sure whether they were the first to have the idea, whether they copied the idea from someone else, or whether they merely recorded an idea that was already emerging in popular consciousness.
THE EMERGENCE OF BEHAVIOURAL STRATEGY
There is one final point about the evolution of emotion that I want to raise. This is the ability to tactically affect or suppress emotional behaviour. Charles Darwin demonstrated that the expression of emotions through facial expression and behaviour was universal among humans, as was our ability to recognise them.75 He hinted at an evolutionary basis, without being able to work out the mechanism. Darwin also understood that the tendency to poker face this expression is highly variable from one culture to another. However, although scientists have looked at poker-facing emotional expression, few seem to have studied the habitual affectation of such expressions of emotion. This is also variable between cultures; for example the affectation of happiness is endemic in America, but a Romanian told me that if you smile too much in Eastern Europe people will think you stupid, so Eastern Europeans tend to suppress smiling – the opposite tactical deception.
You might disagree with my view that the emotional range of a zebra does not extend much beyond fear. But could a zebra that is experiencing fear ignore the fact? For that matter, could they fake it? My sorry little philosopher-brain always assumed that the ability to fake or suppress emotion was language-dependent. However, the science demonstrates that this isn’t quite true. Experiments with the famous chimpanzee Nim Chimpsky76 lead to the conclusion that chimpanzees have a basic ability to learn vocabulary but no ability to master grammar.77 They also show no urge to gobble up language the way a human child does. But the only example of deliberate emotional behaviour suppression that I can find in the scientific literature occurs in chimpanzees. A chimpanzee bares its teeth when it is frightened, and there are recorded observations of a male chimpanzee after a fight pushing its lips shut with its fingers before he turned to face his adversary.78 This leads us to the conclusion that the ability to do this is dependent upon something other than language, possibly some more basic form of self-awareness. A simple experiment checks this: leave an animal or person in front of a mirror for some time, and then take them away and alter their appearance (e.g. by putting paint on their face), and then put them back in front of the mirror. If they stare at themselves in the mirror after their appearance has been altered this would seem to indicate an awareness of self when looking at the mirror. Humans record a positive result in this test from the age of about 16 months; but several other species provide a positive result (principally the great apes).79 Note that a chimpanzee can only suppress its facial expressions by using its fingers. Imagining that humans needed to do this would be a pretty good basis for a Ricky Gervais movie. This, I think, makes it inconceivable that a chimpanzee could suppress behaviour habitually or systematically – something that humans constantly do.80
The suppression of an emotion could either mean suppressing the behaviour, or it could mean suppressing the feeling that underlies it. However, surely it is not possible to suppress the feeling while letting the behavioural response flow normally. So suppression must start with concealing the behaviour. A person who is experiencing fear can, with an act of will, suppress the behaviour associated with that fear. But they can only do this because they have a concept of the behaviour that they can separate from its performance. This is a concept that is certainly dependent upon self-awareness, and which is greatly strengthened by language. An animal with no concept of self sees danger, and reacts to it. When it does this, it does not see its reaction as a separate entity. Similarly, social animals react to the behaviour of other herd or pack members without seeing this behaviour as a separate concept.
I cannot find an example in the scientific literature of an animal tactically affecting emotional behaviour. Affectation is a more complex tactical deception than suppression because it requires memory of the behaviour. I believe therefore that affectation is unique to humans. The faking of an emotion is similar to suppression: it starts with the faking of the behaviour – for example, pretending to be happy. This requires us to be able to see the behaviour as a separate concept to its cause and the feeling associated with it. This certainly requires a concept of self. It certainly requires a memory of the relevant behaviour, and it may well also be language-dependent.
Emotion is not irrational; it is pre-rational. The very first piece of human reasoning started with someone thinking, “Soon I will be hungry.” A belief in a future emotion was the original axiom of logic. Originally, the chain went
Belief in a future emotion => Logic => Action
But then somebody realised that the action could be to pretend an emotion: they could go
Belief in a future emotion => Logic => Pretend emotion
and tactical deception with behaviour was born. At this point humanity became horribly confused. The birth of man’s problems came at this moment, not when Adam and Eve ate the apple. But worry not! Sorting it out is just a logic problem that I want to persuade you can be solved. A human is a machine that goal-seeks emotional outcomes. Tactical deception with emotional behaviour results in us deducing the identity and cause of our own emotions from falsified data. We then perform self-destructive acts, not because we are irrational, but because we goal-seek emotional outcomes based on false beliefs about the causes of emotions.
False behaviour is the original false assumption from which self-destructive actions are derived.