The evolution of morality

Abstract

Complex animal societies are most successful if members minimise harms caused to one another and if collaboration occurs. In order to promote this, a moral structure inevitably develops. Hence, morality has evolved in humans and in many other species. The attitudes which people have towards other humans and individuals of other species are greatly affected by this biologically based morality. The central characteristic of religions is a structure which supports a moral code, essentially the same one in all religions. A key obligation to others is to help to promote their good welfare and to avoid causing them to have poor welfare. Human views as to which individuals should be included in the category of those to whom there are moral obligations have broadened as communication and knowledge have progressed. Many people would now include, not only all humans but sentient animals, e.g. vertebrates and cephalopods, as well. Amongst sentient animals, coping with adversity may be more difficult in those with less sophisticated brain processing.

Full text

Broom, D.M. 2006. The evolution of morality. Applied Animal Behaviour Science., 100, 20-28.
Pre-publication copy
The Evolution of Morality
Donald M. Broom
Department of Veterinary Medicine, University of Cambridge, Madingley Road,
Cambridge, CB3 0ES UK.
Abstract
Complex animal societies are most successful if members minimise harms caused to one
another and if collaboration occurs. In order to promote this, a moral structure inevitably
develops. Hence morality has evolved in humans and in many other species. The
attitudes which people have towards other humans and individuals of other species are
greatly affected by this biologically–based morality. The central characteristic of
religions is a structure which supports a moral code, essentially the same one in all
religions. A key obligation to others is to help to promote their good welfare and to avoid
causing them to have poor welfare. Human views as to which individuals should be
included in the category of those to whom there are moral obligations have broadened as
communication and knowledge have progressed. Many people would now include, not
only all humans but sentient animals, e.g. vertebrates and cephalopods, as well. Amongst
sentient animals, coping with adversity may be more difficult in those with less
sophisticated brain processing.
1. Concepts and attitudes
Morality is not an obscure topic which is difficult to comprehend. Something is moral if
it pertains to right rather than wrong (Broom 2003). Every person has ideas about what
is right and some actions are considered to be right by a very high proportion of people.
People take account of morality in their actions and most discuss moral issues with
others. Ethics is the study of moral issues. Is morality an issue related to biology and are
there links between the functioning of humans and non-humans in relation to decisions
about which actions to carry out because they are moral and which to avoid because they
are not?
There are many ideas which biologists would take as axiomatic but which other people
do not necessarily accept. The arguments presented by biologists can lead to conclusions
which are initially regarded as surprising by some members of the general public but

which may well come to be more widely accepted as time goes on. One example of such
a biologically-based statement, which seems entirely normal to all biologically trained
people but which is not part of normal thinking for many people even now, is that
humans are animals . Another is that the organ involved in decision-making in humans
and other animals is their brain and that the heart is not directly involved in decision
making. These and many other biological facts are relevant to the discussions about the
biological basis of morality which are presented here.
The topic of morality is one which some people would not accept as suitable for
discussion, from a biological or other perspective, because it is thought of as sacred or
God-given. The influential philosopher G.E. Moore (1903) went so far as to state that:
“It is illegitimate to argue from the facts of nature to human values”. Even a biologist
might regard morality as in some way outside biology. In the midst of a strong argument
about the importance of evolution by natural selection in social life, Dawkins (1976) said:
“We, alone on earth, can rebel against the tyranny of the selfish replicators”. Somewhat
similar views are stated by Alexander (1979) and Williams (1988).
The idea of the “selfish gene” proposed by Richard Dawkins was embedded in a very
illuminating and influential book which promoted the understanding of genetic and
behavioural mechanisms. However, the term itself is misleading. Selfish describes an
individual acting in a way which increases its fitness at the expense of the fitness of one
or more other individuals whilst being aware of the likely affects on itself and on the
harmed individual or individuals (Broom2003). The word selfish is thus limited to
individuals and it could not describe a gene. If there is no awareness, it is not selfishness.
As Midgley (1994) points out, a word which is widely used with one set of connotations
cannot be transferred to another set without causing the reader or hearer to misunderstand
either the breadth of its implications or the concept itself. One consequence of Dawkins’
usage of “selfish gene” is that people will argue that we are not responsible for the effects
of our genes, genes are often selfish, and hence there is nothing wrong with being selfish.
It would be better to produce another term to refer to genes that promote the fitness of the
bearer, i.e. the actions benefit the subject, at the expense of others that are harmed by the
action. The terms “harmful subject-benefit” (Broom 2003), or “subject-benefit at the
expense of others”, are more accurate if more cumbersome.
The desirability of considering the biological basis for morality has been expounded by
many authors, for example Kropotkin (1902), Kummer (1978), de Waal (1996) and

Ridley (1996). Wilson (1975) said: “the time has come for ethics to be removed
temporarily from the hands of philosophers and biologicized”. The idea that
consideration of the biological basis of morality is at odds with the concept of humans
being able to take important ethical decisions, was criticised by Midgley (1978): “The
notion that we ‘have a nature’, far from threatening the concept of freedom, is absolutely
essential to it. If we were genuinely plastic and indeterminate at birth, there is no reason
why society should not stamp us into any shape which might suit it”.
In order to explain the basis for morality we often refer to altruism. An altruistic act by
an individual is one which involves some cost to that individual in terms of reduced
fitness but increases the fitness of one or more other individuals. Trivers (1985) said:
“There can hardly be any doubt that reciprocal altruism has been an important force in
human evolution”. Reciprocal altruism occurs when an altruistic act by A directed
towards B is followed by some equivalent act by B directed towards A or by an act
directed towards A whose occurrence is made more likely by the presence or behaviour
of B.
Although altruistic actions may spring to the mind of many people when the term
morality is used, what else do people think of? The response of some people is to
assume that sexual activity is being discussed. However, a major confusion exists in the
usage of the term morality to refer especially to aspects of sexual activity. There are
some sexual and other actions, which might be criticised by many in human society, but
which are to do with customs rather than with true morality. Many sexual taboos serve a
mate-guarding function for certain males rather than being in the general interest of the
members of a social group. A straightforward example is the view that it is morally
wrong for women to derive pleasure from the act of copulation. The practice of
clitorectomy is a consequence of this view.
There are actions which are always wrong, in my view, so they cannot be justified by
cost-benefit analysis of consequences. However, I consider that sexual acts are not in
themselves wrong. The consequentialist argument is useful here in that moral judgements
about sexual activity should concern whether or not there are harms to individuals as a
consequence of the acts (Broom 2003). Hence, whilst rape would always be immoral
because it would always have harmful effects, no sexual act would necessarily be
immoral. . It is necessary to consider the context of the act, including the individual to
which it is addressed, and its consequences in order to determine whether or not there are

harms as a result of the act. Indeed, some sexual acts result in the production of much
desired offspring, help to cement bonds between partners, or calm individuals and reduce
the risk of anti-social behaviour.
Codes and rules of conduct, which include issues of great importance, are widespread in
human society. Some of these codes are specified as laws, for example those to prevent
murder, theft, rape and fraud. Other selfish acts are the subject of sanctions which,
although social rather than legal, are important nonetheless. Indeed Ridley (1996) refers
to a taboo against selfishness. Codes of conduct have been written down in many
societies, for example the ten commandments of the Jews and Christians in the Bible
(Exodus, 20, 3-17 and Deuteronomy, 5, 7-21) and the Greek rules of conduct. The
Qu’ran makes it clear that it is the morality of the individual’s actions which determines
reward and punishment (Sura XLIV, 40).
Society condemns, albeit to different degrees: those who injure another deliberately,
those who cause injury by careless contact with another such as a push which leads to a
head injury, and those who are negligent with the consequence that an injury is caused to
another, for example leaving a large hole in the ground uncovered in the dark or giving a
child a dangerous weapon. There might be circumstances in which the extent of
condemnation by a section of society is reduced, for example the view of killing during
war. However, even in this circumstance, some killing is not accepted.
There are also rules relating to the use of important resources. If plentiful quantities of
food are occasionally obtained by individuals in a social group, there is likely to be an
expectation within the group that these will be shared. Many of these rules seem to exist
in other social species.
2. Cooperation and cheating
Humans and other animals which live in social groups cooperate in many ways which
benefit the cooperating individuals more than would occur if they just competed with one
another. Those who have watched stable groups of cattle, horses or various primate
species will have observed allogrooming. The individual groomed will often, at a later
time, groom the groomer. In all species, some body areas are difficult for the individual
itself to groom efficiently and, where mutual grooming occurs, ecto-parasite infestation
tends to be less. Benham (1982,1984) describes mutual grooming relationships in

unrelated cows in suckler herds which spend much time together and seldom show
aggression to one another.
There are many ways in which individuals can gain more food by responding to or
collaborating with others. Broom (1981) lists joining others who are likely to have found
food; observing others in order to find food sources or learn how to acquire food;
collaborating in hunting for, acquiring, handling and defending food or avoiding depleted
sources; sharing food; and giving food to others. A vulture soaring over the plains of
Africa or a pied wagtail joining a post-roost gathering gains useful information about the
location of food from conspecifics (Ward and Zahavi 1973, Broom et al, 1976). Wolves
hunt more effectively in packs, pelicans move in formation and synchronise scooping for
fish, and crocodiles hold prey while another crocodile grasps and twists its body to
remove flesh. In all of these actions, the individuals gain by collaborating but there is a
potential for cheating. The sharing of food by vampire bats, ravens and chimpanzees
(Wilkinson 1984, Heinrich 1989, Savage-Rumbaugh and Lewin 1994) is clearly
reciprocal altruism. The characteristics of shared foods are summarised by de Waal
(1996) as follows:
• “ Highly valued and concentrated put prone to decay.
• Too much for a single individual to consume.
• Unpredictably available.
• Produced through skills and strengths that made certain classes of individuals
dependent on others for access.
• Most effectively procured through collaboration”.
Cooperation is also necessary in the production and defence of communal nests by
weaver birds, termites and ants. Indeed in the social insects and in naked mole rats the
feeding environment is partly produced by social collaboration. The edifices produced
by ants, termites and humans are valuable means of environmental control.
Defence against predators is more effective in many social species because of
collaboration. House sparrows will sit chirping on a fence or branch until they can move
safely en masse into a feeding area (Elgar 1986). Alarm calls when a predator is detected
often draw attention to the caller but also help to maintain group stability. Terns in a
colony may collaborate by diving onto and pecking an intruder which would not be
repelled by the actions of a single tern. Fieldfares will dive on intruders and defecate on

them with similar effect. Vigilance in flocks and herds, often with sharing of
responsibility for doing so, benefits the vigilant individual when it is reciprocated and
because group stability is promoted. Communication about danger may be more
sophisticated. Savage-Rumbaugh and Lewin (1996) reported walking in woodland with a
bonobo and dogs. A large feline potential predator was detected in a tree. On their
return, the bonobo vocalised to five other bonobos. The researcher also talked to the
bonobos. When two of these bonobos that had not previously detected the feline predator
next went to that area of woodland they showed signs of being frightened.
In addition to the more obvious kinds of cooperation, the commonest kind of altruistic
behaviour in social groups, which is often reciprocated, is to avoid injuring other
individuals (Broom 2003). Great care is usually taken by individuals to avoid collisions,
which would benefit the avoider as well as the avoided, but also not to step on others, or
injure them with horns or teeth, or push others out of trees, or over cliffs, or into places of
danger from predators. If any accidental and perhaps avoidable harm to another does
occur, this can be followed by changed behaviour on the part of the harmed individual
and on the part of the one who has harmed. Harm may be followed by some form of
retribution but either accidental or deliberate harm may also be followed by
reconciliation, at least in primates (de Waal 1996). The individuals which take part in
reconciliation may form alliances in order to achieve social and other objectives.
Once altruism occurs and is reciprocated, the possibility of cheating becomes important.
A variety of characteristics of individuals, any of which would tend to promote altruistic
and moral behaviour is listed in Table 1. Amongst these are ways of detecting and
responding to individuals who cheat, in that they fail to avoid harming others or make no
effort to reciprocate to an individual or contribute in a more general way with in a group
if benefit is received.
Table 1 Individual characteristics which may promote altruistic or moral behaviour
1. Affection for certain types of individuals, perhaps those which are close
relatives or group members, or are likely to be, which reduces the chances that
harm will be done to them.
2. Affection for those same individuals which increases the likelihood of
carrying out behaviour which is beneficial to them.
3. Ability to recognise individuals which might be beneficiaries or benefactors.

4. Ability to remember the actions of others which resulted in benefit to oneself
or to others in the group.
5. Ability to remember one’s own actions which resulted in a benefit to another
individual.
6. Ability to assess risk or benefit of own and other actions and either to compare
these or to avoid high risk and try to attain high benefit.
7. Ability to detect and evaluate cheating.
8. Ability to punish or facilitate the punishment of those who cheat.
9. Ability to support a social structure which encourages cooperation and
discourages cheating.
10. Having a desire to conform.
(based on Broom 2003)
The characteristics listed in Table 1 are discussed further in Section 4. Feelings are
important parts of the mechanisms which individuals need in order to cope with the
various problems of life.
3. Selection and Altruism
A key question in relation to morality and its evolution is whether or not genes which
promoted cooperative, altruistic behaviour would be out-competed by those which
promoted harmful subject benefit at the expense of others. Would a gene which
promoted altruistic behaviour spread in a population of a social species? Dawkins (1979)
said that gene U (Universal)– altruistic to every other individual would lose out to a gene
K (Kin)– altruistic to kin but never to others. However, altruism is not universal, it has
limits. In the following argument (Broom 2003), other possible strategies resulting from
genes are considered, firstly in the situation where there is reciprocation and secondly in
the situation where there is not.
A gene UR leads to the bearer being altruistic except where the bearer’s monitoring shows
that it is inadvisable because reciprocation is unlikely. With a gene K(UR) the bearer is
altruistic to kin and also to others if monitoring shows that reciprocation is likely. Either
of these might out-compete K.
A further aspect of this argument is that UR might be relatively more successful in a
population in which many of those benefited also carried UR. A similar argument is
relevant for K(UR). UR is likely to be more successful because it would increase the
stability of the social group and hence benefit the bearers of UR.

In the circumstance where no reciprocation occurs, another possible strategy, linked to
the gene U* would involve being altruistic with no reciprocation necessary but with
monitoring which allows recognition of X (bearers always competitive) or K when
bearers are not interacting with kin. U* could spread if its effects promoted group stability
and hence the bearers of U*. It may be concluded that genes promoting altruism in
defined circumstances are likely to be successful in social species. Also, such genes are
likely to be old and to be present in all social species in which individual recognition is
possible.
Game theory has been used by several authors in order to try to explain whether
strategies which are altruistic could be as successful as strategies which promote personal
gain whatever the effect on others. One model is the prisoner’s dilemma in which two
individuals may either co-operate, or defect and betray one another, for specified pay-
offs. Several different strategies have been tried and the strategy which was most
successful was “tit-for-tat” in which the individuals: begin by co-operating and then do
what the opponent had last done (Ridley 1996). However, as Kitcher (1993) and Axelrod
(1997) have pointed out, there are problems with using such game theory results. In
general they are too simple when the sophisticated intellects of any social vertebrate are
considered. In order to understand how altruistic behaviour might have evolved, it is
necessary to incorporate into modelling some information obtained from studies of real
life situations. Riolo et al (2001) found that tolerance and cooperation in a group-living
species could arise from a strategy involving benevolence to individuals bearing a
recognisable characteristic that is the same as one borne by the actor. This could occur
even if there was no direct reciprocation.
4. Capabilities needed to show moral behaviour
In order to behave in a moral way, i.e. in a way that results in at most avoidance of harm
to others and that may lead to benefit to others and to the stability of the social group,
animals need to have brain function which allows some degree of recognition, awareness,
decision-making and feelings (Barton and Dunbar 1997). There must be social living for
a long enough time. Many studies of social behaviour include implied evidence for
individual recognition, for example, in vervet monkeys (Cheney and Seyfarth 1990).
Experimental studies with cattle (Hagen and Broom 2003) show that they can be trained
to approach one individual rather than another in order to gain a food reward. Kendrick
et al (2001) found: that sheep could be trained to discriminate between individual

conspecifics and individual humans. In these animals, specific cells in the medial
temporal and pre-frontal cortex fire when a particular face is seen, and the same cells fire
and the same behavioural discrimination occurs 8-12 months later. Other aspects of
cognitive ability relevant to the development of moral behaviour have been demonstrated
in many animals. Most of the literature on non-humans is on primates (Harcourt 1992,
Byrne 1995, Lee 1999, Heyes and Huber 2000). In a study of sheep in alpine pastures,
Favre (1975) found that flocks led by an old ewe would graze a pasture and then avoid
revisiting it until about 30 days later, by which time it had regrown. Many non-human
species would appear to be capable of some degree of awareness, as defined and
categorised by Sommerville and Broom (1998), complex decision making and a variety
of feelings (Broom 1998). An indication of the possible awareness of own actions and
functioning comes from the studies of Hagen and Broom (2004) on young cattle. The
heifers were put in a pen whose gate could be opened by pressing a panel with the nose,
thus giving access to food 15m away. They learned to do this and at the time of learning
showed an excitement response of increased heart rate and jumping or galloping. This
“Eureka” effect was not shown by controls which just gained access to the reward or by
heifers which had learned earlier how to open the gate.
In order that an individual behaves in a moral way, there must be an appropriate
motivational system. Motivational state is the sum of the states of a set of causal factors
and the mechanisms which give rise to these will have evolved like any other biological
mechanism (Broom 1981). The evolution of morality will therefore depend substantially
on the evolution of the motivational system. The higher brain processes which are used
when using all available information to make complex decisions will be a very important
aspect of the biological processes underlying morality.
5. Morality and its evolution
The key points of the arguments presented here about morality and its evolution and
elaborated by Broom (2003) are as follows.
1. Morality is defined. True morality does not include customs, or attitudes to
sexual behaviour stemming from mate guarding etc., except indirectly by
effect.

2. Laws may indicate what is morally right but may protect the persons and
property of the powerful or perpetuate tribal or other customs. Although
more likely to do so in a democracy, laws will not always indicate what is
right.
3. There is widespread occurrence of co-operative and altruistic behaviour in
social animals.
4. Awareness, feelings and cognitive ability are clearly demonstrated in
mammals, birds and other animals to a lesser extent.
5. There is great overlap in the gene complement of humans and other animals
which suggests that any genes that promote moral actions are not likely to be
unique to humans..
.
6. The likely success of strategies that involve moral action is demonstrated by
modelling and the actual success is apparent from behavioural and other
observation.
7. Reciprocal altruism is important in the evolution of morality but is not all of
the biological basis. Some actions which do not harm, or which directly
benefit others, are not reciprocal but are directed towards individuals who
need help and who have not previously provided benefit to the actor. Such
actions may make a contribution to the stability of the social group.
6. The moral core of religion
A religion is a system of beliefs and rules which individuals revere .and respond to in
their lives and which are seen as emanating directly or indirectly from some intangible
power (Broom 2003). All religions have a moral code that is central to their functioning.
The differences among religions are in peripheral aspects, including tribal components.
Holy books are a source of information about what is moral but they also include much
history. Religions have a guide to behaviour and a system for discouraging cheats or
those who harm others. The moral code in each religion is very similar and includes a

variety of commandments used by those who adhere to the religion.
7. What does the biological basis for morality tell us about our treatment of
humans and other animals?
Moral actions are directed more towards those identified as “us” than towards those
considered to be “them”. The most limited category of “us”(a) includes only individuals
readily recognised as close relatives. A wider range of individuals is included (b) if “all
of those who know who I am” is the category. Still wider is the group (c) who “might
have access to the same information that I have” or (d) “all sentient beings who share
characteristics with me”. A sentient being is one that has some ability: to evaluate the
actions of others in relation to itself and third parties, to remember some of its own
actions and their consequences, to assess risk, to have some feelings and to have some
degree of awareness. Increased communication efficiency is revolutionising our degree
of concern for other humans and extending our area of moral concern. For many humans
who own pets, category (a) will include companion animals. Serpell and Paul (1994)
found that many pet owners stated that they regarded their pets as part of their family.
Whether or not they do so, most pet owners would include their pet in category (b). As
scientific knowledge about animal functioning advances and the media presents ever
more information about the sophisticated abilities of many animals, the number of people
who include non-humans in the category of “sentient beings who share characteristics
with me” increases. In many societies now, education levels are high and there is easy
access to good quality information about people in other countries and about animals
whose abilities are complex. Hence the likelihood will decline that people will cause, or
tolerate poor welfare in foreign people or animals perceived to be aware. Closely allied
with this is a desire to help individuals who are disadvantaged for some reason such as
being subject to a natural disaster, or a disease outbreak, or a famine. Similarly, animals
used by man are more likely to be helped, or at least harmed less. It is of particular
interest that changed attitudes to animals appear to be linked more closely with the
education level of people than to their affluence. In countries which are relatively poor,
but well educated, interest in animal welfare may be such that people are willing to incur
some degree of financial loss rather than benefit from poor welfare in animals.
The idea that humans have a relatively broad view as to which individuals, human or not,
should be helped in some circumstances may have parallels in other species. There are

many accounts of animals which seem to have formed inter-specific friendships and
which help one another. Groups of individuals who collaborate may sometimes include
two or more species. For example, this is evident in tropical bird flocks. Although there
are many examples of inter-specific competition, the nature of inter-specific interactions
is not always competitive.
If we use a living animal in a way which gives us some benefit, we have an obligation to
that animal. The nature of the obligation is affected by the level of awareness of which
the animal is capable. One obligation is to avoid causing poor welfare in the animal
except where to do so would lead to net benefit to that animal, or to other animals
including humans, or to the environment,. A utilitarian approach is not sufficient to
determine all obligations, however, and a deontological approach is also needed because
there are some degrees of poor welfare which are never justified by benefit to others. It is
my view (Broom 2003, p 130) that all human behaviour and laws should be based on the
obligations of each person to act in an acceptable way towards each other person and to
each animal that is used. It is better to base strategies for living on our obligations rather
than to involve the concept of rights. Many so-called rights can result in harm to others.
The most widely accepted obligation to animals which we use concerns avoidance of
poor welfare so learning about animal welfare and its scientific basis is very important for
all who have frequent contact with animals.
The link between level of awareness and welfare is complex. Welfare concerns how well
individuals are able to cope with good or bad environments (Broom and Johnson 1993).
Some animals, for example, Protozoa, seem not to have any awareness and would not be
called sentient. Animals which are sentient would have a wider array of ways in which
their welfare could be poor, because the complexity of their brain function is above a
threshold level, than non-sentient animals. We should be concerned about the welfare of
all animals and more concerned about those that are sentient, for example all vertebrates
and cephalopod molluscs. However, within this category of sentient animals, more
sophisticated brain processing will provide better opportunities for coping with some
problems. There seem to be means of dealing with pain which humans have but fish do
not. As a consequence, a certain degree of pain may cause worse welfare in fish than in
humans (Broom 2001). This argument would also be valid for other causes of poor
welfare. It also seems likely that more complex brains allow more possibilities for
pleasure, which contributes greatly to good welfare. The same type of human action may

sometimes be more cruel if inflicted on a simpler animal than on a human or other more
complex animal.
8. References
Alexander, R.D. 1979. Darwinism and Human Affairs. Pullman: University of
Washington Press.
Axelrod, R. 1997. The Complexity of Cooperation. Princeton: Princeton University Press.
Barton, R.A. and Dunbar, R.I.M. 1997. Evolution of the social brain. In Machiavellian
Intelligence II. ed A. Whiten and R.W. Byrne, pp. 240-263. Cambridge University
Press.
Benham, P.F.J. 1982. Social organisation and leadership in a grazing herd of suckler
cows. Appl. Anim. Ethol., 9, 95.
Benham, P.F.J. 1984. Social organisation in groups of cattle and the interrelationship
between social and grazing behaviours under different grazing management
systems. Ph.D. thesis, University of Reading.
Broom, D.M. 1981. Biology of Behaviour. Cambridge: Cambridge University Press.
Broom, D.M. 1998. Welfare, stress and the evolution of feelings. Advances in the Study
of Behavior, 27, 371-403.
Broom, D.M. 2001. Evolution of pain. In Pain: its nature and management in man and
animals, ed. Soulsby, Lord and Morton, D. Royal Society of Medicine International
Congress Symposium Series, 246, 17-25.
Broom, D.M. 2003. The Evolution of Morality and Religion. Cambridge: Cambridge
University Press
Broom, D.M., Dick, W.J.A., Johnson, C.E., Sales, D.I. and Zahavi, A.1976. Pied
wagtailroosting and feeding behaviour. Bird Study, 23, 267-79.
Broom, D.M. and Johnson, K.G. 1993,/2000. Stress and Animal Welfare. Dordrecht:
Kluwer.

Byrne, R. 1995. The Thinking Ape: Evolutionary Origins of Intelligence. Oxford
University Press.
Cheney, D.L. and Seyfarth, R.M. 1990. How Monkeys see the World. Chicago:
University of Chicago Press.
Dawkins, R. 1979. Twelve misunderstandings of kin selection. Zeitschrift für
Tierpsychologie, 51, 184-200.
Elgar, M.A. 1986. House sparrows establish foraging flocks by giving chirrup calls if the
resources are divisible. Animal Behaviour, 34, 169-174.
Favre, J.-Y. 1975. Comportement d’Ovins Gardés. Ministère de l’Agriculture Ecole
Nationale Supèrieure Agronomique de Montpelier.
Hagen, K. and Broom, D.M. 2003. Cattle discriminate between individual familiar herd
members in a learning experiment. Applied Animal Behaviour Science, 82, 13-28.
Hagen, K. and Broom, D.M. 2004 Emotional reactions to learning in cattle. Applied.
Animal Behaviour Science, 85, 203-213.
Harcourt, A. 1992. Coalitions and alliances: are primates more complex than non-
primates? In Coalitions and Alliances in Primates and Other Animals. Ed. A.H.
Harcourt and F.B.M. de Waal, pp 445-471. Oxford: Oxford University Press.
Heinrich, B. 1989. Ravens in Winter. New York: Summit.
Heyes, C. and Huber, L. (Ed.) 2000. The Evolution of Cognition, Cambridge University
Press.
Kendrick, K.M., da Costa, A.P., Leigh, A.E., Hinton, M.R. and Peirce, J.W. 2001. Sheep
don’t forget a face. Nature, London, 414, 165-166
Kitcher, P. 1993. The evolution of human altruism. The Journal of Philosophy, 90, 497-
516.
Kropotkin, R. 1902. Mutual Aid: a Factor in Evolution. London: Allen Lane.

Kummer, H. 1978. Analogs of morality among non-human primates. In Morality as a
Biological Phenomenon. ed. G.S. Stent, 31-47. University of California Press:
Berkeley and Los Angeles.
Lee, P.C. 1999. (Ed.) Comparative Primate Socioecology. Cambridge: Cambridge
University Press.
Midgley, M. 1978. Beast and Man: the Roots of Human Nature. Hassocks, Sussex:
Harvester Press.
Midgley, M. 1994. The Ethical Primate. London: Routledge.
Moore, G.E. 1903. Principia Ethica. Cambridge: Cambridge University Press.
Ridley, M. 1996. The Origins of Virtue. London: Viking.
Savage-Rumbaugh, S. and Lewin, R. 1994. Kanzi: the Ape at the Brink of the Human
Mind. London: Doubleday.
Serpell, J. and Paul, E. 1994. Pets and the development of positive attitudes to animals.
In: Animals and Human Society, ed. A. Manning and J. Serpell, 127-144. London :
Routledge.
Sommerville, B.A. and Broom, D.M. 1998. Olfactory awareness. Applied Animal
Behaviour Science, 57, 269-286.
Trivers, R. 1985. Social Evolution. Menlo Park Ca: Benjamin Cummings.
Waal, F. de 1996. Good Natured. Cambridge Mass: Harvard University Press.
Ward, P. and Zahavi, A. 1973. The importance of certain assemblages of birds as
'information centres' for food-finding. Ibis, 115, 517-34.
Wilkinson, G.S. 1984. Reciprocal food sharing in the vampire bat. Nature, London,
308,181-184.

Williams, G.C. 1988. Reply to comments on "Huxley's evolution and ethics in
sociobiological perspective". Zygon, 23, 437-438.
Wilson, E.O. 1975. Sociobiology. The New Synthesis. Cambridge, Mass.: Harvard
University Press.