Open Book Publishers logo Open Access logo
  • button
  • button
  • button
GO TO...
book cover

6. Communication between Parents and Offspring1

© 2017 Patrick Bateson, CC BY-NC-ND 4.0

How offspring respond to the behaviour of their parents raises important issues about their development. In studies of animal communication it has sometimes been argued that all the activities directed by one individual towards another are manipulative.2 Sometimes this view is clearly correct. Its most obvious form is in the interactions between species when one species manages to control the behaviour of another as if it were a puppet. A striking example is the European cuckoo. The mother cuckoo lays each egg in the nest of another species such as the reed warbler. The egg very closely resembles the egg of the host. The young cuckoo hatches before the reed warblers’ chicks and ejects the competition from the nest. Then the young cuckoo successfully persuades the unfortunate warbler foster parents to feed it, even when it is twice their size. By looking like a super-offspring, the cuckoo successfully exploits the normal pattern of interaction that exists between parent and young.

The essentially competitive process of Darwinian evolution does not necessarily imply a competitive outcome. On the contrary, a great deal of communication involves signals that carry real information and cooperation in which all the participants benefit by working with each other. One explanation for cooperation is that, at least in the past, the aided individuals were relatives; cooperation is like parental care and has evolved for similar reasons. Another is that co-operating individuals jointly benefited even though they were not related; the co-operative behaviour has evolved because those who engage in it were more likely to survive as individuals and reproduce than those that did not. Once again the force of this particular argument can be seen most clearly in communication between different species.

Some small fish, which are conspicuously marked but highly suitable as a morsel of food, clean the teeth of big predator fish. Before they do their job, the cleaner fish perform a characteristic waggling swim in front of the monster. This inhibits the normal feeding response and the great predator opens its mouth, allowing the little fish in. When the big fish needs to eat other little fish, it signals it is switching back into normal hunting mode by jerking its jaw in a particular way. The little cleaner fish scuttle for cover and the mutually beneficial symbiotic arrangement is preserved.

Small cleaner fish servicing a big predator fish. Photo by Richard Ling (2005), Wikimedia,, CC BY-SA 2.0.

In co-operating animals of the same species, the mutual benefits of working together can be greatly enhanced if information about the state of the external world is transmitted from one individual to another. One of the most extraordinary and well-analysed examples of such transmission is the so-called dance language of honey bees.3 The characteristics of the waggle dance performed in the hive provide crucial information about where the returning bee successfully foraged. The duration of the dance circuit is strongly correlated with the distance from the hive to the food. In a darkened hive the angle of orientation of the central segment of the dance with respect to the vertical is related to the angle between the food source and the sun’s position.

The most interesting test-bed of evolutionary thinking about communication is provided by those situations in which individuals compete with each other over resources. Advocates of the manipulation view urge that a trustworthy mode of communication is always open to cheating in which the cheater exploits information provided by its opponent and gives nothing away about itself. As a consequence, it is much more likely to win the resource and likely to reproduce faster. Before long, it is argued, cheating will have evolved to become the dominant mode of behaviour.

This argument sounds convincing, but most people who know a particular species well quickly develop a good intuitive sense of whether an animal is likely to attack or escape by observing its body and facial postures. After the initial location of the opponent, several levels of escalation may precede an actual fight, involving easily recognised and increasingly energetic displays. Neither party will benefit from getting hurt and, in the majority of cases, the disputes are settled without serious damage on either side. The encounter can break off at any stage during the process of escalation. The contests are usually won by the larger individual, but if the opponents are of equal size, they are usually won by the holder of the resource. The behaviour of each individual at each stage of escalation indicates how serious it is about continuing.

From a large number of quantitative studies, it is clear that escape is well predicted by certain patterns of behaviour. If one animal suddenly turns tail, it is liable to be attacked and might get injured. The advantage to the loser of not being misunderstood and expressing the animal equivalent of a white flag is obvious. The benefit to the winner from responding appropriately to such a signal is that it does not risk injury by escalating the conflict into a real fight unnecessarily. The argument is, therefore, that a form of behaviour which effectively negotiates the end of a conflict can be evolutionarily stable.

While most biologists placed great emphasis on dissimulation, one took a very different line. The Israeli biologist Amotz Zahavi argued that signalling that carried a handicap to the signaller was honest.4

Zahavi was treated as an eccentric since the initial attempts to model his ideas about handicaps failed formally to substantiate them. More recently, theoreticians have been able to satisfy nearly everybody else that if a signal carries a cost, it may also be reliable in the sense of carrying accurate information.5 Honest signalling is now widely accepted, although arguments continue about whether honest signals must carry some cost.

A crucial question is where the balance is struck between signalling real information about internal state and signalling misinformation. Consider analogies with the human game of poker. On the one hand, if players can get away with bluff, they will make money. On the other, if players bluff against opponents who have really good hands, they may end up very much worse off than if they had decided to throw in their bad hand before they had raised the bet too far. So something equivalent to negotiation might be expected. For each individual the optimal outcome of such negotiation should represent a balance between the costs of escalating the conflict to likely damage to the pocket (or in case of fights, the body) and the benefits of winning the resource easily. An individual that escalates without assessment is in danger of finding itself in a fight with a much stronger opponent.

Red deer stag roaring at competitors in rutting season. Photo by Bill Ebbesen (2009), Wikimedia,, CC BY 3.0.

Evolutionary pressure is present for some exaggeration of fighting ability and many species puff themselves up in various ways, making themselves look more fearsome than they really are. Even so, honest advertisement of strength, providing cues that cannot be faked, may count most in the long run. One example may be the male red deer, which competes vigorously with other males for opportunities to mate. Fights occur, of course, but conflicts are most often settled by bouts of roaring at each other.

The roaring is extremely exhausting for the animals and the one that can keep it up for longest is also likely to be the stronger. In the roaring contests, both individuals increase the rate of roaring until one seems to recognise it is outclassed and retreats. Recordings of red deer roars were played to a real stag.6 Unlike the stag, the tape-recorder did not get tired and when the tape-recorder roared at a high rate, the stag roared less. It seems as though the stag has been forced into accepting that it was dealing with a much more powerful opponent. While fake characteristics may gain short-term successes, those individuals that ignore such cues and focus on reliable sources of evidence about their opponents may eventually do better in the course of evolution.

Parents and offspring

Evolutionary theories about parent-offspring relationships have undergone a similar transformation to those about communication. The prediction used to be that young will demand food and care from their parents at weaning whereas the parents’ interests are best served by reserving their efforts for future offspring. The American biologist Robert Trivers revolutionised the study of parent-offspring relationships when he pointed out that the long-term interests of the parent are not identical with those of its offspring.7 A parent may increase its reproductive success by weaning its young earlier than is best for them, because it saves itself from becoming exhausted and is thus able to have a larger number of offspring than would otherwise have been the case. As a consequence, aggression between parents and offspring is to be expected, particularly at the time of weaning, and the communication between them was not thought to be reliable.

Trivers’s stimulating contribution has been responsible for a large theoretical literature. The insights derived from evolutionary theory have been questioned partly because of changes in theoretical stance about the benefits of reliable signals. More seriously, perhaps, the evolutionary argument had led to an unjustified expectation that conflict of evolutionary interest necessarily implies behavioural conflict. Behavioural aggression between parent and offspring should be referred to as ‘squabbling’, thus avoiding the ambiguous use of ‘conflict’. Even so, the behavioural evidence does not fit the theory especially well.

Two-year-old tantrums in humans have been interpreted in precisely the opposite sense to the theory of conflicts. When they occur, tantrums seem to be concerned with the difficult process of establishing some autonomy from the parent. Well-attached children with sensitive mothers do not seem to have tantrums, using their parents as a secure base from which to explore the world. Rhesus monkey mothers may actively reject their offspring around the time of weaning, while their offspring play an increasingly prominent role in initiating contact with their mother as they grow older.8 Squabbles did not arise in this species as a result of mothers’ reduction in suckling when they conceived their next offspring. Nor did mothers that failed to conceive have more demanding infants, as would have been predicted from the theory of parent-offspring conflict.

Occasionally, weaning squabbles between mother and young do occur, particularly if the mother is already pregnant again. In field studies of baboons, conflicts of interest did not invariably lead to such squabbles. Evidence from a variety of mammals suggests that maternal aggression often does not occur at all and, if it does, it is seen at quite different stages in development from that in which the process of weaning occurs.9

Perhaps the most convincing evidence against the early manipulative predictions derived from parent-offspring conflict theory was that mothers both monitor and respond to the progress of their current offspring. Mothers may actively compensate for the slow development of their young or delay conception of the subsequent offspring if the present one is progressing slowly. Rat mothers continue to lactate for longer than would otherwise have been the case when they are given pups younger than the ones they had been suckling, again suggesting that the mother is not necessarily driving the weaning process without regard for her offspring’s condition.

From the offspring’s standpoint, young laboratory rats seem to wean themselves spontaneously; a stage is reached in development when they actively choose solid food in preference to milk. When the offspring’s metabolic costs go up as they get older and larger, they may be forced to wean themselves because they simply do not get enough energy from their mother’s milk. In domestic cats, the rate of growth starts to slow down around 20 days after birth, particularly in kittens of nutritionally stressed mothers, and then suddenly speeds up again around 30 days as they start to process solid food efficiently. The discontinuity occurs at different ages in different animals and is less apparent, as a consequence in averaged data. The slowing down of growth before the discontinuity is much more pronounced in larger litters that impose greater energetic demands on their mothers. The eventual inadequacy of the mother to provide for all the nutritional needs of the young is particularly evident in fur seals in which the mother disappears for several days while she stokes up for the next bout of lactation. Eventually what she can provide is less than what her pup needs.10

The empirical study of behaviour has forced a reappraisal of the optimal route to maximum reproductive success in mothers and the optimal route to highest probability of survival in their offspring. The evolutionary arguments needed to be looked at again, not simply from the standpoint of honest signalling but also bearing in mind how an individual’s state might affect the optimal weaning time from both the standpoint of the mother and that of her offspring.11 In species that can breed more than once, the mother has a reasonable chance of successfully raising more than one offspring. The current offspring will only survive to independence after a certain minimum amount of care has been given to it. Even when the mother is well fed, the longer she cares for the current offspring, the worse her state may become, given the substantial energy costs of lactation. As a consequence, the costs in terms of the diminished probability of rearing more young may soon outweigh the marginal benefits of giving more milk to the current young.

The probability of having another offspring ought to influence the time and energy devoted to the current one. For example, the mother may have to face a winter in which she cannot breed and which she might not survive. Furthermore, matters may be made worse if finding a mate takes time. With a relatively low probability of breeding again, her lifetime reproductive success may be maximised if she gives higher priority to the current offspring than to those that are as yet unborn. Conversely, if she has mated immediately after the birth of the current young and is pregnant while she lactates, as is possible in species such as the cat and the rat, she may give lower priority to the offspring she is currently looking after than if she were not pregnant. Under most environmental conditions, the optimum weaning time should be earlier than for non-pregnant mothers.

The reactions of an offspring to its mother’s conditional responses to the environment should also be conditional. In many species, the relation between the effect of its behaviour on its probability of surviving before weaning and its probability of surviving after weaning must both be taken into account. This is because offspring survival does not simply depend on getting as much milk as possible from the mother while she is able to provide it. Other benefits, such as the protection the offspring might have derived from her in the post-weaning period, would be lost if it were so demanding in the pre-weaning period that it seriously damaged her health. In the early stages of lactation the offspring’s survival depends utterly on receiving immediate maternal attention. Moreover, since meeting its needs is relatively easy for the mother because of the offspring’s low body mass compared to that of the mother, the longer-term consequences for post-weaning care of being demanding are not heavy. As the offspring grows, these delayed effects of being demanding start to rise and, at the same time, the chances of the young surviving without maternal milk begin to improve. By the time the offspring has reached the late pre-weaning stage, its peak demands have dropped to low values.

The optimal weaning ages for offspring depend greatly on how much care is given after weaning. If little is characteristically given by the species, optimal demand by the offspring remains high until the demands of switching to primarily solid food become paramount. The greater the amount of post-weaning care the mother gives, the earlier the age relative to weaning when her offspring’s demanding behaviour should start to decline. The general point is that an offspring benefits from reducing its demanding behaviour as it gets older and also benefits from monitoring maternal availability which is a measure of her state.

Each mother is sensitive to the condition of her offspring so that, if it is still in need of care but reasonably well developed, she may forego a breeding opportunity in order to nurture it through to independence. Young are sensitive to the condition of their mothers and adjust their pattern of development accordingly, since their mothers respond not only to the state of the young but also to their own condition.

The offspring may also need to prepare for the probable world in which they have to grow up. Members of the same species, the same sex and the same age sometimes differ dramatically from one another. ‘Alternative tactics’ within a species commonly arise because an individual has the capacity to respond in more than one way according to environmental conditions or its own body state. Such conditional responses during development are well known in the social insects, in which one sister might be adapted for producing thousand of eggs, another has massive jaws used in defence of the nest and another is equipped with foraging skills never expressed by the other two. Similar environmentally-induced differences occur frequently in mammals. Young mammals may pick up crucial information from their mothers about when to wean themselves and how to develop later on the basis of cues that they pick up from their nursing mothers.

The process of weaning in mammals is not readily explained in terms of conflicts of interest that inevitably lead to squabbling. From a behavioural standpoint, the interplay between parent and offspring is dynamic. The mother has to balance the maintenance of her body condition, easily debilitated by energetically costly lactation, against responding sensitively to the needs of current offspring. If the current offspring lag behind in development, the mother may increase her reproductive success by extending her parental care.

For its part, each offspring balances demanding maximum resources and attention from its mother up to the last moment she is prepared to feed it against being sensitive to its mother’s state. The young have to pay attention to the condition of the mother because of the need to take into account both the immediate effects of maternal care on survival but also the post-weaning contributions of the mother. Moreover, to adopt characteristics that will be appropriate to the environment into which they will have fend for themselves, the young may need to respond to information provided by the mother. The mother and her offspring are unlikely to have perfect knowledge of each other either initially or during development. Therefore each needs to monitor its own and the other’s often rapidly changing state.


Communication is not simply about manipulating another individual. The emerging picture is more nuanced than that. Sensitivity to the condition of the receiver of the signal is often adaptive to the sender. So the sender benefits in Darwinian terms and the competitive evolutionary process produces behaviour that is apparently more cooperative than has sometimes been supposed. In short, the mother-offspring relationship is a system that is apparently well-designed for both parties.

1 Part of this chapter is taken, with permission, from Bateson, P. (1990), Animal communication. In: Ways of Communicating, ed. by D.H. Mellor. Cambridge: Cambridge University Press, pp. 35–55.

2 Dawkins, R. & Krebs, J.R. (1979), Arms races between and within species. Proc. R. Soc. Lond B 205.1161, 489–511,

3 Munz, T. (2016), The Dancing Bees: Karl von Frish and the Discovery of Honeybee Language. Chicago: University of Chicago Press, A video of the honey bees’ waggle dance is available at

4 Zahavi, A. & Zahavi, A. (1997), The Handicap Principle: A Missing Piece of Darwin’s Puzzle. Oxford: Oxford University Press.

5 Grafen, A. (1990), Biological signals as handicaps. J. theor. Biol. 144.4, 517–546,

6 Clutton-Brock, T.H. & Albon, S.D. (1979), The roaring of red deer and the evolution of honest advertisement. Behaviour 69.3, 135–144, A video of a male deer’s roaring is available at

7 Trivers, R.L. (1985), Social Evolution. Menlo Park, CA: The Benjamin/Cummings Publishing Company.

8 Gomendio, M. (1991), Parent/offspring conflict and maternal investment in rhesus macaques. Anim. Behav. 42.6, 993–1005,

9 Bateson, P. (1994), The dynamics of parent-offspring relationships in mammals. Trends Ecol. Evol. 9.10, 399–403,

10 Bateson, P. (1994), The dynamics of parent-offspring relationships in mammals. Trends Ecol. Evol. 9.10, 399–403,

11 Ibid.