473
The
British
British Journal of Psychology (2009). 100. 473-479 MiOam
<i ' tv
© 2009 77ie British Psychological Society * * * = ^ iOCiety
www.bpsjournals.co.uk
Commentary
The roots and branches of human altruism
Dale R Hay*
School of Psychology, Cardiff University, Cardiff, UK
The authors' rigorous and ingenious programme of work documents young humans'
capacity for prosocial action (Warneken & Tomasello. 2009). Their laboratory findings
are corroborated by centuries of observational data, spanning cultures, and historical
epochs. Two general questions are raised. Firstly, what are the rules of evidence needed
in using comparative data to make claims about the evolution of human social
behaviours? Secondly, what ontogenetic processes contribute to the transformation of
toddlers' helpfulness into mature, cognitively informed and rule-governed altruism?
These findings alone do not provide complete evidence for the phylogenetic roots of
altruism. I argue that selective pressures in human evolution have favoured sociability,
which could lead to aggression or altruism, depending on context and the nature of the
rearing environment. Any social behaviour shown in infancy may have multiple
functions, at phylogenetic. ontogenetic and episodic levels of analysis.
The human infant has traditionally provided a Rorschach test for biological and social
theory. Although behaviours that appear early in development are not necessarily innate
(see Lehrman, 1970), and biologically underpinned traits may appear late in life (e.g. hair
loss), theorists nonetheless look to infants as a mirror of human nature. For example, the
study of aggression is currently dominated by a neo-Hobbesian approach that
emphasizes infants' innate aggressive tendencies (see Tremblay, Hartup, & Archer,
2005). The present article presents a most welcome alternative view of human nature.
The authors' rigorous and ingenious empirical work has revitalized the study of the
development of altruism.
The ubiquity of early prosocial behaviour
The argument rests on experiments on human infants. Some might query whether the
infants tested in these laboratory experiments are indeed representative of all young
humans. However, there is much corroborative evidence. Early forms of sharing,
helping, cooperation, and sympathy have been recorded throughout history, by the
* Correspondence should be addressed to Professor Dale F. Hay, School of Psychology, Cardiff University, Park Place, Cardiff
CFIO 3AT, Wales, UK (e-mail: haydf@cardiff.ac.uk).
DOl: 10.1348/000712609X442096
474
Dale F Hay
early baby biographers (e.g. Shinn, 19OO; Tiedemann, 1787, translated by Murchison &
Langer, 1927) and pioneers of developmental psychology (e.g. Murphy, 1937; Stern,
1924; Valentine, 1942). Young children's prosocial behaviour has been observed across
cultures, as well as historical epochs (Whiting, 1983).
Among the qualitative observations reported by these early scholars are clear
instances of instrumental helping. For example, one diary-keeping mother described her
18-month-old as follows: 'She works alongside me in the garden, manages to rake or use
a trowel fairly well... In the house, she helps push the vacuum or mop. . . anticipates
her father's needs in dressing or in building a fire in the fireplace' (Church, 1966,
pp. 71-72). Valentine (1942) recorded that his toddler son was 'Very thoughtful
nowadays. When I came in this morning, he said, 'Daddy want slippers', and ran off to
get them' (p. 321). Rheingold (1982) sought to corroborate these anecdotes, and so
transformed her laboratory into a home with many undone tasks (a table to set, a bed to
make, books and cards to pick up from the floor, scraps of paper to sweep up, and
laundry to fold). The majority of 18- to 30-month-olds enthusiastically helped their
parents with the housework. Supplementary home observations revealed that all 2-yearolds studied helped their mothers (and the observer), and clearly stated their helpful
intentions, e.g. 'I help you, I hold that little light bulb', which suggests that they
understood what it means to help (Hay & Rheingold, 1983).
Given that toddlers' helpful behaviours have been noted for centuries, why do most
parents and developmental psychologists still believe that children are naturally
aggressive and need to be taught how to help or share, despite the fact that toddlers'
spontaneous sharing occurs at higher rates than their aggression (Hay, Castle, & Davies,
2000)? In Rheingold's view, the descriptive facts were not assimilated by most
psychologists because the facts ran counter to most developmental theory:
'The prosocial behaviours are. . . predictable outcomes of infants' biological
characteristics and the experiences afforded to them in human homes. Yet typically,
the behaviours have not been remarked; no current theory of social development - not
psychoanalytic, attachment theory, cognitive-developmental theory, nor even social
learning theory - would predict a prosocial infant. Hence, the data warrant a new
conception of infants and their social development' (Rheingold & Hay, 1978, p. 104).
The present authors' programme of work indeed sets forth a new conception of
the human infant and calls attention to a more compatible theoretical framework,
evolutionary biology. Whilst I am in complete agreement with the authors that
prosocial behaviour emerges early in life, and is underpinned by the biological nature
of the human infant, I do have some reservations about the very strong claim that
infants' instrumental helping is a pure form of human altruism, and a behavioural
phenotype that has been selected for in the course of human evolution. Rather, I
believe that multiple functions of infants' helping can be identified at the
ontogenetic and episodic as well as the phylogenetic levels of analysis (Sommerhoff,
1950), and that one should not assume a trait is an adaptation favoured by selection
until competing explanations are tested (Gottlieb, 1997; Williams, 1966). It is
important to understand how ontogenetic processes refine the development of the
ability to help others with their tasks, and also to identify factors in the immediate
episode of social interaction that support helping, including features of the task that
elicit the toddler's 'executive competence' (Wenar, 1972) and problem-solving skills
(Chen & Siegler, 2000).
These considerations lead me to pose two questions to the authors: Firstly, what rules
of evidence need to be applied in comparative studies in order to identify the
Roots and branches of altruism
475
'phylogenetic roots of human altruism'? Secondly, what ontogenetic processes influence
the emergence of helpful behaviour in infancy and its differentiation over time?
Comparing helpfulness ¡n different species: Analogies or homologies?
Which species and which ages should be compared?
In making their evolutionary claims, the present authors adduce evidence from two
sources: laboratory studies of human infants and analogous studies of nursery-born and
wild born chimpanzees. They argue that demonstration of instrumental helping in a
close biological relative provides evidence about the phylogenetic roots of the
behaviour. However, one may ask, if species-typical helpAilness has been selected for in
evolution, should species be more different? Should we in fact expect chimpanzees
to perform less well on tests designed for human babies? What exactly does it mean to
compare human infants with adults of another species? Why not compare infants with
adults of their own species?
The present comparison rests on the implicit assumption that non-verbal,
instrumental helping is a behavioural phenotype that may take a homologous form in
the non-verbal interactions of another species. But does this then imply that the nonverbal form of instrumental helping shown by infants and chimpanzees is not akin to
more complex forms of helping in human adulthood? Obviously the different types of
help offered by adult humans to their conspecifics might look very different from the
tasks invented by these authors, but is adults' rate of helping all that different from
infants and chimpanzees? For example, in one observational study, we found that
toddlers behave sensitively to distressed peers at about the same rate that adults help
distressed bystanders in airports (Demetriou & Hay, 2004). It would be helpful for the
authors to spell out the logic of further comparisons of species and ages that are needed
to support their hypotheses.
The interpretation of null results
The authors note that not every research group has found chimpanzees to be all that
helpful. How should one interpret null results from one research group versus evidence
for empathy or altruism from other laboratories? What is needed is a meta-analysis of
published and unpublished findings from the various samples of non-human primates
observed w^ith reference to potentially cooperative or altruistic behaviour. However, in
undertaking such a task, we need to consider whether the absence of altruistic
behaviour in any particular species, tested under specific conditions, necessarily
undermines the case for the evolutionary basis of human altruism. It may prove
important to scrutinize the qualitative observations made in the very many
observational studies of primate species to look for analogues as well as homologues
to human helpfulness, such as protection and alliance in conflict situations
(e.g. Kummer, 1978).
Ontogenetic processes: Development and differentiation
Change and differentiation over time
Once a rudimentary ability to help other people emerges in the human repertoire, what
happens next? Do children become more helpful with age? There is consistent evidence
476
Dale F Hay
for increases in the rate of prosocial behaviour as children grow older (Eisenberg, Fabes,
& Spinrad, 2006); on the other hand, in the early childhood years, prosocial behaviour
also becomes more selective, directed to friends and relations, not all conspeciñcs, and
increasingly governed by display rules and social norms (Hay & Cook, 2007). Heritable
individual differences in empathy and prosocial behaviour soon emerge (Knafo &
Plomin, 2006; Zahn-Waxler, Robinson, & Emde, 1992). Thus, the very high rates of
helpfulness observed in nearly all toddlers are likely to decline in favour of more
selective displays of helpfulness by particular children to particular people, under
particular conditions.
It's not just learning
Early prosocial behaviours are not easily explained by straightforward learning
processes. As the present authors have demonstrated, rewarding toddlers for prosocial
behaviour may actually depress its frequency. In our laboratory, we similarly found that
rewarding 18-month-olds for sharing objects did not increase the rate of sharing
(Rlieingold, 1973). Furthermore, we found that modelling sharing did not increase its
rate, as opposed to modelling a 'selfish' social action, i.e. asking for an object with
outstretched hand; asking for objects, and especially modelling both actions together,
promoted immediate sharing with the experimenter and later sharing with the mother
(Hay & Murray, 1982).
In their day-to-day lives, children's prosocial behaviour only rarely meets with
approval or reward from adults; preschool teachers are more likely to reward
helpñilness in the context of pretend play than in children's real interactions with
classmates (Caplan & Hay, 1989; Eisenberg, Cameron, Tryon, & Dodez, 1981). Thus,
I concur with one of the main conclusions of the present authors: infants' and toddlers'
near universal impulse to share, cooperate, and be helpful is not easily explained by
simple learning processes.
However, we must not conclude that children learn nothing about prosocial
behaviour. What is acquired over the first years of life is not the motivation to share, or
be helpful and sympathetic. Infants don't have to learn to care about other people, but
tbey learn how to care for tbem. One- and two-year-olds' pretend play with dolls and
teddies reveals extensive knowledge about how to look after the young (Rheingold &
Emery, 1986). By 3 years of age, children can tell an interviewer a great deal about how
to minister to their distressed classmates, although they do not think it is their personal
responsibility to do so (Caplan & Hay, 1989). Thus, early versions of the social norms
that govern prosocial behaviour in adults - responsibility, deservingness, and reciprocity
norms - already influence young children's sharing and sympathy (Hay, Castle, Davies,
Demetriou, & Stimson, 1999); might they also begin to influence instrumental helping?
What has been selected for: Altruism or sociability?
I began by comparing the authors' welcome focus on the roots of altruism with other
contem]îorary accounts that focus on the intrinsically aggressive nature of human
infants. These opposing interpretations of the nature of human nature derive from
observations of infants' non-verbal interactions with their conspecifics; helpful infants
are also capable of using force against other people. Developmental psychologists have
long noted that, in very young children, cooperation can co-occur with conflict, and
prosocial behaviour may be positively correlated with aggression (e.g. Murphy, 1937).
Roots and branches of altruism 477
In a longitudinal study of toddlers observed with familiar peers (Hay, 2006), we found
that, at the time of the first home visit, sharing was positively associated with the use of
force (especially tugging on toys), and what might be interpreted as the egoistic use of
possessive pronouns (e.g. saying 'Mine!'). However, 6 months later, sharing and the use
of force were no longer significantly correlated; aggression had disaggregated from
other forms of sociability. Furthermore, the 'egoistic' act of claiming possession over
objects predicted sharing, not the use of force at the second visit.
The fact that different forms of toddlers' behaviour thought to measure opposing
traits are initially correlated raises important questions about the operational definitions
of phenotypes in evolutionary studies. (Similar problems plague genetic studies of
children's psychological problems: the human genome may not be precisely lined up
with DSM-rV.) I suggest that the authors take a closer look at earlier-appearing social
phenotypes, the fundamental skills that support all forms of social intercourse among
humans - altruism, cooperation, enlightened self-interest, and aggression. Infants come
equipped with perceptual biases that promote attention to the human face and voice
and sensitivity to emotion that may lead to empathy and contagion of fear and anger;
they soon develop species-typical forms of non-verbal and verbal communication.
I believe that more complex, cognitively informed social activities, such as intentional
aggression, instrumental helping, and sympathetic responses to distress derive from
these basic social skills, refined and deployed through emergent processes of motor
maturation, learning and neurocognitive development. Studies of aggression in humans
and other mammals clearly show the interplay between genetic traits and rearing
environments (Gottlieb, 1997); for example, after 28 generations of artificial selection,
mice who had been selected for aggressive genotypes onlyfightat high rates after being
subjected to social isolation (Hood & Xie, 2007). In humans, genotype interacts with the
experience of maltreatment to promote antisocial behaviour (Caspi et al, 2002). Similar
interplay between the genetic heritage of our species and the characteristics of human
environments no doubt influences prosocial development. We do not need to speculate
whether human infants are naturally aggressive nor altruistic. Infants are indubitably
social, and their innate social skills and motivation underpin their proclivities for both
helpfulness and aggression.
Over the childhood years, humans acquire the ability to balance their prosocial
tendencies with their individual needs. They learn how to share, sympathize and be
helpftil to others while negotiating their own entitlements to space and resources. The
move from universal, non-verbal sociability to cognitively informed judgements about
the conditions under which it is appropriate to be helpful is one of the main
achievements of early childhood. Developmental scientists need to study how children
construct representations of self within the framework ofthe norms of their societies, if
we are to understand exactly how the distinctly human form of altruism branches off
from its phylogenetic roots.
References
Gaplan, M. Z., & Hay, D. F. (1989). Preschoolers' responses to peers' distress and beliefs about
bystander intervention./owni«/ of Chitd Psychology and Psychiatry, 30, 231-243.
Gaspi, A., McGlay, J., Mofñtt, T., Mül, J., Martin, J., Graig, I. W, et at. (2002). Role of genotype in the
cycle of violence in maltreated children. Science, 297, 851-854.
Ghen, Z., & Siegler, R. S. (2000). Overview of toddlers' problem solving. Monographs of the
Society for Research in Child Development, 65, 32-43.
478
Dale F Hay
Church, J. (Ed.), (1966). Three babies: Biographies of cognitive development. New York: Random
House.
Demetriou, H., & Hay, D. E (2004). Toddlers' reactions to the distress of familiar peers: The
importance of context. Infancy, 6, 299-319.
Eisenberg, N., Cameron, E., Tryon, K., & Dodez, R. (1981). Socialization of prosocial behaviour in
the preschool classroom. Developmental Psychology, 17, 723-729.
Eisenberg, N., Eabes, R. A., & Spinrad, T. (2006). Prosocial development. In N. Eisenberg (Ed.),
Handbook of child psychology: Social, emotional, and personality development (6th ed..
Vol. 3, pp. 646-718). Chichester, UK: Wiley.
Gottlieb, G. (1997). Developmental psychobiological theory. In R. B. Cairns, G. H. Elder, Jr., &
E. J. Costello (Eds.), Developmental science. Cambridge: Cambridge University Press.
Hay, D. E (2006). Yours and mine: Toddlers talk about possessions with familiar peers. British
foumal of Developmental Psychology, 24, 39-52.
Hay, D. E, Castle, J., & Davies, L. (2000). Toddlers' use of force against familiar peers: A precursor
of serious aggression? Child Development, 71, 457-467.
Hay, D. E, Castle, J., Davies, L., Demetriou, H., & Stimson, C. A. (1999). Prosocial action in very
early childhood./owrw«/ of Child Psychology and Psychiatry, 40, 905-916.
Hay, D. F, & Cook, K. V. (2007). The transformation of prosocial behaviour from infancy to
childhood. In C. A. Brownell & C. B. Kopp (Eds.), Sodoemotional development in the toddler
years: Transitions and transformations. New York: Guilford.
Hay, D. E, & Murray, P. (1982). Giving and requesting: Social facilitation of infants' offers to adults.
Infant Behavior and Development, 5, 301-310.
Hood, K. E., & Xie, H. (2007). A comparative-developmental account of aggressive behaviour in
Mus tnusculus and Homo sapiens. Intemational Society for the Study of Behavioural
Development Newsletter, 51, 2-5.
Knafo, A., & Plomin, R. (2006). Prosocial behaviour from early to middle childhood: Genetic and
environmental influences on stability and change. Developmental Psychology, 42, 771-786.
Kummer, H. (1978). Analogs of morality among nonhuman primates. In G. Stent (Ed.), Morality as
a biological phenomenon: The presuppositions of sociobiological research. Berkeley, CA:
University of California Press.
Lehrman, D. S. (1970). Semantic and conceptual issues in the nature-nurture problem. In
L. R. Aronson, E. Tobach, D. S. Lehrman, & J. S. Rosenblatt (Eds.), Development and evolution
of behaviour: Essays in memory ofT.C. Schneirla (pp. 17-52). San Erancisco, CA: Ereeman.
Murchison, C, & Langer, S. (1927). Tiedemann's observations on the development ofthe mental
faculties of children. Pedagogical Seminary andfoumal of Genetic Psychology, 34, 205-230.
Murphy, L. B. (1937). Social behaviour and child personality: An exploratory study ofsome roots
of sympathy. New York: Columbia University Press.
Rheingold, H. L. (1973). Independent behavior of the human infant. In A. D. Vick (Ed.'), Minnesota
symposia on child psychology (Vol. 7). Minneapolis, MN: University of Minnesota Press.
Rheingold, H. L. (1982). Little children's participation in the work of adults, a nascent prosocial
behavior Child Development, 53, 114-125.
Rheingold, H. L., & Emery, G. (1986). The nurturant acts of very young children. In D. Olweus,
J. Block, & M. Radke-Yarrow (Eds.), Development of antisocial and prosocial behavior:
Research, theories and issues (pp. 75-96). New York: Academic Press.
Rheingold, H. L., & Hay, D. E (1978). Prosocial behaviour of the very young. In G. Stent (Ed.),
Morality as a biological phenomenon: The presuppositions of sociobiological research.
Berkeley, CA: University of California Press.
Shinn, M. W. (1900). The biography of a baby. Boston, MA: Houghton Mifflin.
Sommerhoff, G. (1950). Analytic biology. Cambridge: Cambridge University Press.
Stern, W. (1924). Psychology of early childhood up to the sixth year of age (3rd ed.). New York:
Holt, (translated by A. Barwell).
Tremblay, R. E., Hartup, W. W., & Archer, J. (Eds.), (2005). Developmental origins of aggression.
New York: Guilford.
Roots and branches of altruism 479
Valentine, C. W. (1942). The psychology of early childhood. London: Metheun.
Warneken, E, & Tomasello, M. (2009). The roots of human altruism. British Joumal of
Psychology, 100, 455-471.
Wenar, C. (1972). Executive competence and spontaneous social behaviour in one-year-olds.
Child Development, 43, 256-260.
Whiting, B. B. (1983). The genesis of prosocial behaviour. In D. Bridgeman (Ed.), The nature of
prosocial development: Interdisciplinary theories and strategies (pp. 221-242). London:
Academic Press.
Williams, G. C. (1966). Adaptation and natural selection: A critique of some current
evolutionary thought. Princeton, NJ: Princeton University Press.
Zahn-Waxler, C, Robinson, J. L., & Emde, R. N. (1992). The development of empathy in twins.
Developmental Psychology, 28, 1038-1047.
Received 18 March 2009; revised version received 18 March 2009
Copyright of British Journal of Psychology is the property of British Psychological Society and its content may
not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written
permission. However, users may print, download, or email articles for individual use.