Coping with CROWDING by Frans B

POPULATION GROWTH has been thought, since the time of Thomas Malthus, to produce dire consequences such as disease, scarcity and social deviancy. This dark view seemed conﬁrmed by rodent studies. Yet little evidence suggests that people are similarly affected: we seem to handle large crowds quite well for the most part.

Coping with CROWDING by Frans B. M. de Waal, Filippo Aureli and Peter G. Judge

76 Scientific American May 2000 Coping with Crowding Copyright 2000 Scientific American, Inc. n 1962 this magazine published a seminal But, one could argue, perhaps such a re- paper by experimental psychologist John lation is obscured by variation in national IB. Calhoun entitled “Population Density income level, political organization or some and Social Pathology.” The article opened other variable. Apparently not, at least for dramatically with an observation by the late- income. We divided the nations into three 18th-century English demographer Thomas categories—free-market, former East Block Malthus that human population growth is and Third World—and did the analysis automatically followed by increased vice and again. This time we did find one significant misery. Calhoun went on to note that al- correlation, but it was in the other direc- though we know overpopulation causes dis- tion: it showed more violent crime in the ease and food shortage, we understand virtu- least crowded countries of the former East ally nothing about its behavioral impact. Block. A similar trend existed for free-mar- This reflection had inspired Calhoun to ket nations, among which the U.S. had by conduct a nightmarish experiment. He placed far the highest homicide rate despite its low an expanding rat population in a crammed overall population density. The Nether- room and observed that the rats soon set lands had a population density 13 times as about killing, sexually assaulting and, even- high, but its homicide rate was eight times tually, cannibalizing one another. Much of lower. this activity happened among the occupants Knowing that crime is generally more of a central feeding section. Despite the common in urban areas than it is in the presence of food elsewhere in the room, the countryside, we factored in the proportion rats were irresistibly drawn to the social of each nation’s population that lives in stimulation—even though many of them large cities and controlled for it. But this could not reach the central food dispensers. correction did nothing to bring about a This pathological togetherness, as Calhoun positive correlation between population described it, as well as the attendant chaos density and homicide. Perhaps because of and behavioral deviancy, led him to coin the the overriding effects of history and culture, phrase “behavioral sink.” the link between available space and hu- In no time, popularizers were comparing man aggression—if it exists at all—is decid- politically motivated street riots to rat packs, edly not clear-cut. inner cities to behavioral sinks and urban Even if we look at small-scale human ex- areas to zoos. Warning that society was periments, we find no supporting evidence. heading for either anarchy or dictatorship, Crowding of children and college students, Robert Ardrey, an American science jour- for instance, sometimes produced irritation nalist, remarked in 1970 on the voluntary and mild aggression, but in general, people nature of human crowding: “Just as Cal- seemed adept at avoiding conflict. Andrew S. houn’s rats freely chose to eat in the middle Baum and his co-workers in the psychiatry pens, we freely enter the city.” Calhoun’s department at the Uniformed Services Uni- views soon became a central tenet of the vo- versity found that dormitory residents who luminous literature on aggression. shared facilities with many people spent less In extrapolating from rodents to people, time socializing and kept the doors to their however, these thinkers and writers were rooms closed more often than did students making a gigantic leap of faith. A look at who had more space. Baum concluded that human populations suggests why such a the effects of crowding are not nearly as simple extrapolation is so problematic. overwhelming as originally presumed. Pub- Compare, for instance, per capita murder lished in the 1980s, these and other findings rates with the number of people per square began to undermine, at least in the scientific —as we did, ck kilometer in different nations community, the idea that people and rats re- o using data from the United Nations’s 1996 act in the same ways to being packed to- estst W Demographic Yearbook. If things were gether. In modern society, people commonly straightforward, the two ought to vary in assemble in large masses—during their daily ETERSON tandem. Instead there is no statistically commute to work or during holiday-season

BRIAN P meaningful relation. shopping expeditions—and most of the time

A persistent and popular view holds that high population density inevitably leads to violence. This myth, which is based on rat research, applies neither to us nor to other primates

Grooming

0 Low Medium High Density

3 FEMALES Incidents per Hour

2 AAL W

DE Grooming M. . 1 ANS B Aggression FR RHESUS MONKEYS from three different settings show different rates of grooming— that is, of calming one another. The monkeys seem to adapt to crowded conditions by 0 grooming more frequently. Among the males, grooming of each other and of females was Low Medium High

more common when they lived in crowded conditions than when they lived in more spa- AN CHRISTIE Y

cious quarters. Among female nonkin, aggression was common and increased further with Density BR crowding but was accompanied by increased grooming, which served to reduce conflicts. SOURCE: Peter G. Judge and Frans B. M. de Waal they control their behavior extraordi- population density seemed to increase wide range of species—from macaques narily well. aggression occasionally—but the oppo- to chimpanzees—as well as strict and Calhoun’s model, we must conclude, site was also true. One report, for in- well-defined hierarchies that remained does not generally apply to human be- stance, described intense fighting and stable for decades. This view of an of- havior. Is this because our culture and killing when a group of macaques were ten anxiety-filled existence was con- intelligence make us unique, or is the released into a corral 73 times larger firmed when researchers found high management of crowding part of an than their previous quarters had been. levels of the stress hormone cortisol in older heritage? To answer this question, Then, after two and a half years in the the blood of wild monkeys [see “Stress we turn to the primates. corral, a similar increase in aggression in the Wild,” by Robert M. Sapolsky; occurred when the monkeys were Scientific American, January 1990]. Primates Are Not Rats crowded back into a small pen. As the view of primates became more Whereas the macaque study manipu- complex, and as the rat scenario was rimate research initially appeared to lated population density through envi- weakened by counterexamples, re- Psupport the harrowing scenario that ronmental change, other early research searchers began to wonder whether pri- had been presented for rats. In the did so by adding new monkeys to exist- mates had developed a means to reduce 1960s scientists reported that city- ing groups. Given the xenophobic nature conflict in crowded situations. We saw dwelling monkeys in India were more of monkeys, these tests mainly measured the first hint of this possibility in a study aggressive than were those living in their hostile attitude toward strangers, of the world’s largest zoo colony of chim- forests. Others claimed that monkeys in which is quite different from the effect panzees in Arnhem, the Netherlands. zoos were excessively violent. Those of density. The better controlled the stud- The apes lived on a spacious, forested monkeys were apparently ruled by terri- ies became, the less clear-cut the picture island in the summer but were packed fying bullies who dominated a social hi- turned out to be. Increased population together in a heated building during the erarchy that was considered an artifact density led to increased aggression in winter. Despite a 20-fold reduction in of captivity—in other words, in the only 11 of the 17 best-designed studies space, aggression increased only slightly. wild, peace and egalitarianism pre- of the past few decades. In fact, the effect of crowding was not vailed. Borrowing from the hyperbole In the meantime, the view of wild pri- entirely negative: friendly grooming and of popularizers, one study of crowding mates was changing. They were no greetings, such as kissing and submis- in small captive groups of baboons even longer the purely peaceful, egalitarian sive bowing, increased as well. went so far as to report a “ghetto riot.” creatures people had presumed them to We wondered if this conciliatory be- As research progressed, however, con- be. In the 1970s field-workers began re- havior mitigated tension and proposed flicting evidence accumulated. Higher porting sporadic but lethal violence in a a way to test this possibility. Without ig-

78 Scientific American May 2000 Coping with Crowding Copyright 2000 Scientific American, Inc. noring the fact that crowding increases behavior varied in discernible ways. We one matriline also fiercely support one the potential for conflict, we predicted collected detailed data on 122 individ- another in fights against other matri- that primates employ counterstrategies— ual rhesus monkeys at three different lines. Because of their strict hierarchy including avoiding potential aggressors sites in the U.S.: in relatively cramped and pugnacious temperament, rhesus and offering appeasement or reassuring outdoor pens at the Wisconsin primate seemed to be ideal subjects. We figured body contact. Because some of the skills center in Madison, in large open corrals that if this aggressive primate showed involved are probably acquired, the at the Yerkes primate center in Atlanta coping responses, our hypothesis would most effective coping responses would and on Morgan Island off the coast of have withstood its most rigorous test. be expected in animals who have expe- South Carolina. These last monkeys Our first finding was, surprisingly, that rienced high density for a long time. had approximately 2,000 times more density did not affect male aggressive- Perhaps they develop a different “social space per individual than the highest- ness. Adult males increasingly engaged in culture” in the same way that people in density groups. All three groups had friendly contact under crowded condi- different places have varying standards of lived together for many years, often for tions. They groomed females more, and privacy and interpersonal comfort zones. generations, and included individuals likewise the females groomed the males For example, studies show that white of both sexes. All the groups had also more frequently. (Grooming is a calming North Americans and the British keep been in human care, receiving food and behavior. In another study, we demon- greater distances from others during veterinary treatment, making them strated that a monkey’s heart rate slows conversations than Latin Americans and comparable in that regard as well. down when it is being groomed.) Fe- Arabs do. Rhesus society typically consists of a males also bared their teeth more often to number of subgroups, known as matri- the males—the rhesus way of communi- Coping Culture lines, of related females and their off- cating low status and appeasing poten- spring. Females remain together for life, tially aggressive dominant monkeys. e set about finding several popu- whereas males leave their natal group Females showed a different response Wlations of monkeys that were of at puberty. Rhesus monkeys make a with other females, however. Within the same species but that had been liv- sharp distinction between kin and non- their own matrilines they fought more ing in different conditions to see if their kin: by far the most friendly contact, but did not change the already high level such as grooming, of friendly interaction. In their dealings takes place within the with other matrilines, they also showed matrilines. Females of more aggression—but here it was cou-

CHIMPANZEES IN THE WILD have hostile territorial relations with other groups, and in captivity they are bothered by the presence of noisy neighboring chimps. By examining apes under three conditions—those living in a crowded space and able to hear their neighbors, those living in a crowded space without such worrisome sounds, and those living in isolated large compounds (photograph below)—we were able to measure the association between aggres-

AAL sion, space and stress. Aggression (photograph at left) remained W

DE the same, but stress varied with neighbors’ noise. Chimpanzees in M. . small spaces exposed to vocalizations from other groups showed

ANS B the highest levels of the stress hormone cortisol. FR

0.15 AGGRESSION r

u 0.12 o H

r s

e 0.09 r

p o e

h s b t g s i t r h d n a w g 0.06 n i L e

d d e u d d i e e o N e c

d d l t p n a a I w 0.03 w l m c o o o o r r o s C I C V C 0 Living Condition ) s 2 e ANXIETY c e f

f o 1.5 m

a l r s o r g s

i o r e h t b 1 t e g r i s r h p o d

a w g

C s i n L

d d e u m d e e o N a e

d d l

r 0.5 t p AAL a a w w g W l m c o o o o r r o o DE s n I C V C C a M. .

n 0 (

Living Condition AN CHRISTIE Y ANS B BR FR SOURCE: Filippo Aureli and Frans B. M. de Waal

Coping with Crowding Scientific American May 2000 79 Copyright 2000 Scientific American, Inc. pled with more grooming and submis- sive grinning. These findings make sense in light of the differences between kin and non-kin relationships. Related females—such as sisters and mothers and daughters—are so strongly bonded that their relationships are unlikely to be disrupted by antagonism. Rhesus monkeys are used to managing intrafamilial conflict, cycling through fights and reconciliations, followed by comforting contact. Crowding does little to change this, except that they may have to repair frayed family ties more often. Between matrilines, on the other hand, crowding poses a serious challenge. Normally, friendly contact between matrilines is rare and antagonism common. But reduced escape opportuni- ties make the risk of escalated conflict greater in a confined space. And our data et k

indicated that female rhesus monkeys ar

make a concerted effort at improving ck M o t

these potentially volatile relationships. he S T GE A V

Emotions in Check SA

n a second project, we turned our at- CHUCK Itention to chimpanzees. As our closest animal relatives, chimpanzees resemble 3.5 Crowded 0.15 us in appearance, psychology and cogni- Control tion. Their social organization is also hu- 3 manlike, with well-developed male 0.12

— — I bonding which is rare in nature re- 2.5 n r c u i o ciprocal exchange and a long dependen- d e H

0.09 n cy of offspring on the mother. In the r 2 t e s p

Control p wild, male chimpanzees are extremely s e t r n

H territorial, sometimes invading neigh- e 1.5 d 0.06 o i u boring territories and killing enemy c r n males. In captivity such encounters are, I 1 Crowded of course, prevented. 0.03 We collected data on more than 100 0.5 chimpanzees in various groups at the

Yerkes primate center. Although some 0 0 AN CHRISTIE Y groups had only a tenth the space of oth- SELF-SCRATCHING AGGRESSION BR ers, cramped quarters had no measurable SOURCE: Filippo Aureli and Frans B. M. de Waal impact on aggression. In contrast to the monkeys, chimpanzees maintained their and charging displays, which in wild The inhibition of natural responses is grooming and appeasement behavior— chimpanzees serve to ward off territori- not without cost. We know that continu- no matter the situation. If crowding did al intrusion. ous stress has the potential to suppress induce social tensions, our chimpanzees In a confined space, however, excited the immune system and therefore has seemed to control them directly. reactions trigger turmoil within the important implications for health and We usually do not think of animals as group. We found that chimpanzees in the longevity. We developed two noninva- holding in their emotions, but chim- most crowded situations had a three sive techniques to measure stress in our panzees may be different. These apes times lower tendency to react to neigh- chimpanzees. One was to record the rate are known for deceptive behavior—for bors’ vocalizations than chimpanzees of self-scratching. Just as with college instance, they will hide hostile inten- with more space did. Chimpanzees may students who scratch their heads when tions behind a friendly face until an ad- be smart enough to suppress responses to faced with a tough exam question, self- versary has come within reach. In our external stimuli if those tend to get them scratching indicates anxiety in other pri- study, emotional control was reflected into trouble. Indeed, field-workers report mates. Our second technique was to col- in the way chimpanzees responded to that chimpanzee males on territorial pa- lect fecal samples and analyze them for the vocalizations of neighboring groups. trol suppress all noise if being detected by cortisol. Both measures showed that Such noises commonly provoke hooting their neighbors is to their disadvantage. groups of chimpanzees who had little

80 Scientific American May 2000 Coping with Crowding Copyright 2000 Scientific American, Inc. we also consider short periods of acute crowding. This is a daily experience in human society, whether we find our- selves on a city bus or in a movie theater. During acute crowding, rhesus monkeys show a rise in mild aggression, such as threats, but not violence. Threats serve to keep others at a distance, forestalling unwanted contact. The monkeys also avoid one another and limit active social engagement, as if they are trying to stay out of trouble by lying low. Chimpanzees take this withdrawal tac- tic one step further: they are actually less aggressive when briefly crowded. Again, this reflects greater emotional restraint. Their reaction is reminiscent of people on an elevator, who reduce frictions by min- imizing large body movements, eye contact and loud verbalizations. We speak of the elevator effect, therefore, as a way in which both people and other primates handle the risks of temporary closeness.

AAL Our research leads us to conclude that W

DE we come from a long lineage of social M. . animals capable of ﬂexibly adjusting to

ANS B all kinds of conditions, including unnat- FR ural ones such as crowded pens and city ELEVATOR EFFECT helps to explain how chimpanzees, and people, deal with crowd- streets. The adjustment may not be with- ing. During brief periods of crowding, people often limit social interaction—a way of out cost, but it is certainly preferable to avoiding any conflict (photograph at left). Chimpanzees do the same, reducing their aggressive interactions (photograph above and chart at left). This doesn’t mean that the frightening alternative predicted on crowded situations do not induce anxiety. Chimpanzees packed together tend to scratch the basis of rodent studies. themselves more often—a sign of stress. We should add, though, that even the behavioral sink of Calhoun’s rats may not have been entirely the product of space and heard neighbors’ vocalizations Given a choice, they would prefer more crowding. Food competition seemed to experienced more stress. Space by itself room. Every spring, when the chim- play a role as well. This possibility con- was not a negative factor, because in the panzees at the Arnhem zoo hear the tains a serious warning for our own absence of noisy neighbors, chimpanzees door to their outdoor island being species in an ever more populous world: in small spaces showed the same stress opened for the first time, they fill the the doomsayers who predict that crowd- level as those with a good deal of space. building with a chorus of ecstasy. They ing will inevitably rip the social fabric So even though chimpanzees fail to then rush outside to engage in a pande- may have the wrong variable in mind. show a rise in aggression when crowd- monium in which all of the apes, young We have a natural, underappreciated tal- ed, this does not necessarily mean that and old, embrace and kiss and thump ent to deal with crowding, but crowding they are happy and relaxed. They may one another excitedly on the back. combined with scarcity of resources is be working hard to maintain the peace. The picture is even more complex if something else. SA

The Authors Further Information

FRANS B. M. DE WAAL, FILIPPO AURELI and PETER G. The Hidden Dimension. E. T. Hall. Doubleday, 1966. JUDGE share a research interest in the social relationships and Crowding. A. Baum in Handbook of Environmental Psychology, Vol. behavioral strategies of nonhuman primates. Their work on as- 1. Edited by D. Stokols and I. Altman. Wiley, 1987. pects of this topic will appear in Natural Conﬂict Resolution, to The Myth of a Simple Relation between Space and Aggression in be published by the University of California Press. De Waal, au- Captive Primates. F.B.M. de Waal in Zoo Biology supplement, Vol. 1, thor of Chimpanzee Politics and Good Natured, worked for pages 141–148; 1989. many years at the Arnhem zoo in the Netherlands before coming Inhibition of Social Behavior in Chimpanzees under High-Den- to the U.S., where he is now director of the Living Links Center at sity Conditions. F. Aureli and F.B.M. de Waal in American Journal the Yerkes Regional Primate Research Center in Atlanta and pro- of Primatology, Vol. 41, No. 3, pages 213–228; March 1997. fessor of psychology at Emory University. Aureli is a senior lectur- Rhesus Monkey Behaviour under Diverse Population Densi- er in biological and earth sciences at Liverpool John Moores Uni- ties: Coping with Long-Term Crowding. P. G. Judge and F.B.M. versity in England. Judge is an assistant professor at Bloomsburg de Waal in Animal Behaviour, Vol. 54, No. 3, pages 643–662; Sep- University in Pennsylvania and a research associate at Yerkes. tember 1997.