Difference between revisions of "The biological instability of social equilibria"

Revision as of 01:49, 14 November 2008

__NOTOCNUM__ __NONUMBEREDHEADINGS__

P.P. van der Molen

In: Evolutionary perspectives on competition, cooperation, violence and warfare (Chapter 4)

Edited by J. van der Dennen and V. Falger

Published in Sociobiology and Conflict, 1st ed. 1990, ISBN 0 412 33770 3 (HB)

Acknowledgements

The writing of this chapter has been supported by a grant of the ANO foundation. Comments and criticism from Michael Kirton, Vernon Reynolds and Robin Dunbar were of great help to improve this text, which is not to imply that they are responsible for any flaws in the basic line of argumentation defended here. The help of Ben Hoffschulte in refining and presenting this text is also gratefully acknowledged. --- PPvdM

The Sociobiology of Conflict was the topic of the ninth meeting of the European Sociobiology Society, held on January 10 and 11, 1987. It was Michael Hopp's initiative to organize this conference in Jerusalem, Israel, a symbolic place in many respects. Thanks to the scientific and personal quantities of Professor Amotz Zahavi, from Tel-Aviv University, many non-Israeli participants were able to experience the naturalistic, geographical and political history of the country in an impressive guided tour which influenced clearly the presentations and discussion in the conference.

Without the hospitality and financial support of the Van Leer Jerusalem Institute the meeting would not have been possible. The Institute's director, Professor Yehuda Elkana, and Mrs Rivka Ra'am, member of the Executive Board of the Van Leer Jerusalem Institute, in close cooperation with local organizer Michael Hopp, contributed very much to the success of the meeting itself. The Board of the European Sociobiological Society expresses its gratitude for this vital support.

In the conference itself Vincent Falger, Lea Gavish, Johan Goudsblom, Anne Rasa, Avi Shmida, Jan Wind and Amotz Zahavi presented papers next to those elaborated and collected in this volume. ESS Board members Jan Wind, Hans van der Dennen and Vincent Falger organized those aspects of the conference not immediately connected with the meeting in Jerusalem.

Outline

This chapter deals with a behavioural mechanism which thwarts any systematic attempts to prevent and put a permanent end to conflicts between social groups and organizations. Essential in this mechanism is a certain kind of social-role blindness, a peculiar unawareness of what we are doing on the level of social-role interactions, whereby attraction or repulsion are effectuated. As in Tiger's contribution (Chapter 5), special provisions in our behavioural system are discussed which prevent us from utilizing our intellectual and cognitive faculties for investigation of our innermost social tendencies. We shall return to these 'no entrance' signs built into our cognitive system below.

Other elements of this mechanism are involuntary incrowd–outcast selection reflexes and a 'trait dimension', which may be described as a `readiness to comply with a submissive role'. This dimension is correlated with a great amount of social behaviour and a small amount of thing-oriented, individualistic and explorative behaviour. It is, by definition, of great importance for the distribution of social roles and for the social structure in a group; it determines, for example, the likelihood of drifting into an outcast position versus the likelihood of assuming or maintaining a compliant and socially accepted subordinate position. Knowledge of this personality trait dimension and of its effects in social groups and structures may increase our understanding of a wide range of intriguing and sometimes disquieting phenomena. These phenomena range from educational and organizational strategies to the often catastrophe-like collapses and turn-over phenomena in companies and other social structures, and from the way social roles and positions tend to be distributed to the resulting evolutionary consequences.

First I will explain why, from a purely biological point of view, differences between individuals are to be expected in any socially living mammalian species in the following situations: readiness to comply with a submissive role; sociability versus thing-orientedness; compliance versus self-will. It will be argued that the underlying biological organization must, from an evolutionary standpoint, be very old and elementary. We will investigate then the consequences of these behavioural differences on the level of social interaction. A life span theory of social structures and organizations will be introduced as one of the implications.

The first sections of this chapter comprise a concise outline of these mechanisms, omitting at this point experimental data and illustrations. The basic assumptions made will, however, be stated explicitly. In the following sections we will check these assumptions against experimental and empirical data from biological and psychological research. Finally, it will be pointed out why understanding of the way these interpersonal differences are behaviourally organized (and the way our awareness tends to be blocked in these respects) have such far reaching consequences; an increase in our understanding of the life cycles of social structures might be by far their most important result. Such understanding would enable us to map the processes underlying periodic catastrophe-like turn-over phenomena and to learn how to control their decreasing efficiency and violent backlashing on any level of organization.

Some consequences of living socially

Among socially living mammals, each individual is by necessity saddled with a conspicuous bi-polarity in behavioural urges. First, being a socially living animal, drives for social contact and interaction are an important part of its behavioural–genetic endowment; but secondly, it has a set of perhaps even more basic drives to ensure the fulfilment of a range of non-social personal needs, e.g., water, food, cover, warmth, sex, territory, etc. As far as these latter needs are concerned, the amount of resources is often limited, thus causing competition and social conflict. For that reason a very basic functional conflict does exist in every social individual: between the urges to fulfil a great variety of personal basic (physical) non-social needs and the urge to maintain social contact and interaction. A socially living mammal inescapably has to shift between these two sets of urges much of the time.

Whenever some of the needed resources are scarce, the ensuing competition will put a strain on social relations. Under such conditions an individual frequently has to choose between continuation of peaceful social relations and receiving an appropriate share of the resources, eventually at the cost of social peace and harmony. Most of the time this dilemma boils down to the question of whether or not to submit to the initiative of other individuals at the cost of fulfilling personal urges and desires. In any socially living species this conflict of needs is inescapably present in each individual day after day, the average outcome determining how the individual will deal by and large with the social situation at hand. It is most desirable to have one's own way most of the time and still maintain close social contact and interaction, but that is more or less identical to what is generally understood by a dominant social role, and such roles are rather scarce.

Fig. 4.1. Two dimensions of social-role behaviour.

It is, therefore, of theoretical interest to know what happens to the majority of individuals, the various types of subordinates (see Fig. 4.1), who are under regular pressure to comply and postpone or even abandon part of their individual desires and initiatives. For such non-dominant individuals, the balance between the strength of the desire for social contact and interaction, and the strength of the desires to fulfil other biological needs, determine the outcome of this continuous process of balancing one need against the other. Given a certain pressure to comply, it largely depends on this equilibrium of basic sensitivities within the subordinate individual as to what the behavioural outcome will be, either drifting gradually into an outcast position or assuming a compliant and socially accepted subordinate position. Such differences between subordinates have indeed frequently been observed in mammals (refer to section 4.7).

What is important for us to note here is that for any socially living mammalian species the competing sets of needs under discussion are very general and basic. We must therefore assume that the variance in the balance between those sets of basic needs has strong genetic roots. (After all, for many species, the behavioural organization is so simple that learning processes can only play a minor role in establishing behavioural variation. The equilibrium discussed above is therefore also an equilibrium between functionally competing parts of the genetic programme. As such we may consider this equilibrium, varying over individuals, as a trait in the classical sense. We could express this set of behavioural polarities as a set of (inter alia genetically based) trait differences which do have a clearly defined impact on the distribution of social roles.

Up to this point, three basic assumptions have been made about the behaviour of socially living mammals in general:

1. There is a strong functional link, on the level of behavioural orientation, between the frequency of social behaviour versus thing-oriented individualistic behaviour, and, on the level of distribution of social roles, between conformity and compliance with authority versus a self-willed attitude. These two polarities cannot be separated; they have the same behavioural basis. Therefore a range of personality characteristics have to be strongly intercorrelated, e.g., self-will, thing-orientedness, individualism and innovative creativity on the one pole, and compliance, person-orientedness, sociability, conformity, and adaptiveness to rules and traditions on the other pole.

2. Individuals differ from one another as far as the balance between these polarities is concerned.

3. This variation between individuals must have genetic components.

In the next part of this chapter we will check these assumptions against experimental data, but first we will investigate their logical consequences.

At this point one might justly retort: 'why so much ado about nothing? It seems self-evident that these polarities in behaviour are interconnected, and since for most broad behavioural characteristics it is likely that differences in behaviour are partly caused by genetic differences, in particular if they are of very old phylogenetic origin, which these behaviours apparently are, it is rather tautological to state that they have genetic roots.' The point is, first, that this notion of a biological basis of certain behaviours may be self-evident to behaviour biologists, but it is certainly not for large groups of sociologists and psychologists. Secondly, these three assumptions do have peculiar and important consequences if applied to the sociology of group structures — the incrowd–outcast dynamism and the concomitant behavioural reflexes in particular.

In order to discuss these consequences we have to add one more assumption, which is rather a definition, namely:

4. In what follows, social groups will mean groups over which individuals are distributed discretely. In other words, individuals can be recognized by one another as either belonging to the social group in question or not – and are treated accordingly.

Life cycles of social groups and structures

If social groups are defined as above, the previous four assumptions imply that within such social groups there is exercised a more or less continuous selection pressure in favour of compliance and sociability. It is such because the most compliant – and thus most socially-oriented and rule-adaptive –individuals are most likely to establish long lasting accommodation within the group. Self-willed individualists on the other hand (also being innovative and thing-oriented according to assumption 1), are most likely to run into trouble and disagreement with the dominant individuals and/or habits and rules in the group. They are least prepared to pay the price of postponing or giving up personal urges and initiatives in order to keep up peace and social harmony. As a consequence, such individuals are the ones who are most likely to either fight hard for attaining a dominant position, or, if failing, to drift into marginal omega-like social positions and eventually become outcasts and leave the social structure. For any eventual influx of individuals into the social group or structure, the opposite holds. Individuals will be most readily accepted if they do not pose a threat to the individuals and/or habits ruling group life, which of course favours rule-adaptive compliants.

The effect of such a continuous selection pressure is that the average behavioural make-up of a group will shift gradually towards compliance and sociable rule-adaptiveness. Due to assumption 1, this also implies a shift towards less and less independent creativity and thing-oriented innovativeness; because of assumption 3, this shifting of group characteristics is (genetically) consolidated. What automatically happens then with every social group (structure) is a gradual loss of innovativeness and behavioural flexibility. In the end such a gradual ossification reduces the effectiveness of the group (structure), whether its function be the preservation of a territorial area with sufficient resources to keep a deme of mice alive, or, in man, the enhancement of some sport, the maintenance of political ideals, the aim to get a better share of the market, or the preservation of a political state. Such ossification especially matters whenever novel challenges turn up in the form of environmental changes or the emergence of competing groups. The disadvantages of a lowered flexibility and innovative creativity weigh most when, because of changing circumstances, innovations and a change of habits are urgently required. In such circumstances the advantages of the old social system in terms of experience, solidly established routines, compliance, malleability of all members, and sheer size, may easily be outdone by the innovativeness, flexibility and efficiency of a younger, and often much smaller, social group (structure) on which these selection pressures have not, as yet, been working for such a long period of time. At such a moment the old structure will yield to the younger structure in a relatively sudden way.

Therefore, provided the above mentioned assumptions are valid, social groups and structures only have a limited life span, and, as I shall try to show below, these assumptions indeed seem to fit most socially living mammal species with discrete group structures, including man. The life cycle of a social institution in human society then, can be indicated roughly as:

Fig. 4.2. Change in time of the average characteristics of the prevalent social group structures and their incrowd members.

foundation → consolidation → internal selection pressure → increasing ossification and a reduction of flexibility of the social structure → eventual attempts to compensate these effects by means of more striving for growth and power → further increase of rigidity and ossification → catastrophic collapse by sudden environmental changes or competition (Fig. 4.2).

Fig. 4.3. Turn-over cycles in terms of personality characteristics and institutional functioning.

Our model implies a departure from notions of mere gradual changes in societal structures. The probability of sudden catastrophic turn-over events increasing in time with cumulating selection effects can be graphically represented and mathematically described with help of the bi-stable models from the mathematical branch of catastrophe theory (Thom and Zeeman, 1974; Zeeman, 1976; Woodcock and Davis, 1978). Figure 4.3 shows a cusp catastrophe, visualizing the relation between the continuous and the discontinuous part of the cycle. After foundation of a social structure, the level of overt challenges tends to decrease and the stability of the structure tends to increase until the inefficiency begins to take its toll, after which the stability of the structure decreases again. During this process the average level of self-will decreases. An increase in the level of experienced challenges may then sooner or later lead to a catastrophic turn-over event. In the new structure the percentage of innovators (average level of self-will) starts again at a high level.

The selection rate determines the speed of ossification; the life expectancy of a social structure is, therefore, roughly inversely proportional to the internal selection pressure. Such sudden turn-overs of social structures are therefore bound to happen at any level at which discrete social group structures are operating, as long as individuals can be recognized by one another as either belonging or not belonging to that group, and as long as there is some outflow or neutralization (and eventually a selected influx) of individuals. Depending on the level of organization, such a turn-over goes by the labels conquest, close-down, discontinuance, bankruptcy, revolution, subjugation or extermination.

Once the old, ossified social structure has been replaced by one or more younger competitor-structures, the individuals from the population as a whole have been reshuffled in favour of resourceful self-willed innovators who now occupy the 'incrowd' positions. The rule-adaptive compliants who formed the bulk of the establishment of the former social structure in power, have drifted into marginal positions and now run the worst risks. Thus the previous internal shift in genetic make-up has been undone, and a new selection cycle is started in these new structures.

The selective advantages for individuals are therefore different within and outside of social groups and structures, and are also different depending on the stage of the life cycle an institution is in. A compliant, adaptive and sociable temperament gives a selective advantage within a large, and especially older, social system, whereas a thing-oriented, innovative and self-willed temperament is selectively advantageous outside of the protective maze of established structures, or within small, young systems.

Evolutionary advantages

Notwithstanding the above mentioned unpleasant aspects of the turn-over catastrophes themselves (in the case of man labelled bankruptcy, revolution, etc.), such a scheme of automatic and unavoidable cyclical changes in social-behavioural structures does also have evolutionary advantages. It is, for instance, clear that this mechanism keeps everything moving: structures, individuals and finally genes. After every turn-over event (or catastrophe) there is a thorough re-shuffling of individuals and when in the ensuing chaos new combinations of individuals reassemble in the newly emerging social group structures, novel combinations of gene sets are eventually formed also.

Apart from this advantage at the level of interpersonal social reorganization and consequently of ensuing recombination of gene sets, there is also an advantage at the level of migration, exploration and colonization of the environment (e.g., Lancaster, 1986). Most mammals are reluctant to go beyond the limits of familiar territory – their home range – and generally must be forced one way or the other to do so (Christian, 1970). Every time an old structure breaks down, a large number of individuals is forced to move and is therefore added to the extra-group surplus population. This will produce a sudden increase in interindividual competition outside of the group (structures) and will produce, therefore, a sudden increase in the pressure on other established (group) structures. This will catalyse the impending catastrophic collapse of more systems, thus locally adding to the already existing chaos. This spatial synchronization causes migratory and related pressures to occur spasmodically and strongly instead of continuously and rather weakly. This may be an advantage where, for example, geographical barriers need to be overcome in order to enable further migratory moves of the population or species as a whole. Many authors have commented on the importance of surplus individuals in producing strong pressures for dispersal (Darlington, 1957) and from the model discussed above it may be clear that social hierarchies constitute by themselves a major force for dispersion. This is also stressed by Christian (1970) in a review of population dynamics research in mammals. He concludes that it is in general primarily the low-ranking individuals that are forced to emigrate from their birthplace (often maturing young animals), and whereas they have an extremely high rate of mortality, it also follows that by their expulsion increasingly more marginal and submarginal habitats should become occupied as density and migration pressures increase. Moreover, this process facilitates speciation:

... once in a great while a dispersing individual may, one would suppose, harbour a mutation or genetic change that increases its ability to adapt to the new surroundings and improves its chance of survival. It is such individuals that should be the basis for evolutionary changes. A sub-optimum area could be invaded repeatedly by countless numbers of individuals before a genetic change permitting adaptation occurred. Thus, the dispersal of large numbers of socially subordinate individuals into new environments may provide the wherewithal for natural selection, in contrast to the relative conservatism of dominant individuals in an optimum habitat. (Christian, 1970, p. 86)

The implication is that the Darwinian 'struggle for life' is in fact a process with much irony and relativity, since those individuals with, apparently, maximum reproductive success (the dominants) create by the very violence of their success the outcasts that carry on the process which we call evolution (Hoffschulte, 1986). Likewise, the ethologist and social psychologist Calhoun (1974), p. 302-3) comments on our own origins:

The strong remain where conditions are most salubrious to preserving the old life-style. The weak must emigrate – bodily, behaviourally or intellectually. Our more distant ancestors swung from trees. Slightly less distant ones lost that race and won another. Population pressure forced them out of forest islands to wander across the African plains in search of another patch of forest where they could renew the old ways. Successive losses and successive demands for adjusting culminated in upright walking creatures much like ourselves. So it has been through all of evolution; the weak [eventually] survive, changed, to open new routes into the future. The meek do inherit the earth.

The evolutionary advantages described here are of course most important in species inhabiting niches of a temporary character, which require the regular invasion and colonization of new environments. Christian (1970) reports that the above mentioned strong fluctuations in population density and migration pressure are indeed most conspicuous in species living in habitats which are ecologically transitory and thus of a strongly temporary and changing character. A species dependent on that type of habitat depends more on regular migratory moves for survival than species living in extensive and stable habitats. In this light it is noteworthy that the human species, by colonizing the most extreme sorts of habitats, has, in its recent evolutionary history, managed to inhabit virtually all of the earth's surface. The mechanism of social selection pressure and expulsion of outcasts should, therefore, have been of great importance in man's evolution.

This is also implied by Coser's (e.g., 1956, 1978) and Girard's (e.g., 1982) comprehensive works on scapegoating in man. Girard describes how throughout human history the distribution of social role positions has been brought about by means of violent acts of social repression. Not only is the dramatic shifting of non-average, deviant subordinate persons into outcast positions just as common as in lower mammals, but, according to Girard, the very development of our culture even depended on it. Only through acts of violence and the collective commemoration of the victim-outcast or scapegoat do human groups find the social-cognitive norms and unanimity from which culture can develop. Culture in our species is not to be considered, therefore, as an immaculate attainment with which we have overcome primitive forms of violence. On the contrary, it is precisely through the violent social collisions themselves that human culture emerged from the animal background. The threatening circle around victims who are found guilty of social disorders is, so to say, the daily bread of social cultural order (Hoffschulte, 1986, on Girard).

In summary, the mechanism of population- and group-cycles, as postulated above, would facilitate speciation through genetic adaptation to marginal habitats, would help to overcome migratory bottlenecks and would, in the case of man, serve to motor the evolution of culture. The actual turn-over catastrophes themselves may not be pleasant for the participants at all, but that is irrelevant from an evolutionary perspective. On this grand scale it is not the feeling and suffering of the individual involved that counts, but the long-term behavioural and behavioural—genetic output that does.

Having outlined these intriguing and also somewhat disquieting consequences of the four assumptions made, I will now present some data from ethological and psychological research that may help us assess the validity of those assumptions.

Experiments with behavioural differences in house mice

Some 33 groups, each containing 4 male and 2 female housemice of the same age, having grown up together from the same litter, were each placed in large observation cages in order to investigate interindividual differences in behaviour and the way these differences come about (van der Molen, 1981, 1988). The study investigated:

1. how social role differences within such groups could be manipulated;

1. which part of the behavioural differences had to be ascribed to those role differences;

1. which part of the behavioural differences was due to innate trait-factors.

Dominance appeared to determine the behaviour of an individual to a great extent, thus being an indispensable tool for ethological descriptions of interindividual differences. It could also be shown experimentally that becoming dominant or subordinate was mainly dependent on coincidence and contingencies, and only to a limited extent on individual characteristics such as body-weight, social- and fighting-experience, self-will, ferocity, etc.

Within the categories of dominants and subordinates there appeared large differences in tolerance for other individuals. Some dominant mice behaved far more aggressively towards their subordinates than did others and these differences determined to a large extent the number of subordinates eventually holding out with such a dominant.

Another role-difference which could rather easily be manipulated experimentally was the Incrowd/Outcast difference, or rather, the difference between beta and omega-subordinates (the usual terms in mouse research).

Detailed ethological data on the behavioural characteristics of 36 individuals were factor-analysed, using factor rotation with the experimentally found social-role differences as anchoring points. The remaining factors of (within-role) differences in behaviour were interpreted as active versus non-active and as self-willed versus compliant As far as the latter dimension is concerned, self-willed conflict-proneness was found to be strongly correlated with a high frequency of exploratory and thing-oriented behaviour, whereas compliance was found to be strongly correlated with a high frequency of social and partner- oriented behaviour.

Every time a group of four males and two females was placed in a large observation cage for the first time, there were at first no clear alpha-, beta-, or omega-roles. In the course of the following days (or weeks) an alpha male would emerge and the differences in behaviour between the subordinate males would still be rather vague. Subsequently, differences would gradually evolve between the behaviour patterns of the subordinates. These differences occurred in the amount of resistance to the initiatives and the manipulations of the alpha, the number of fights they had with the alpha, and the amount of patrolling and exploration by themselves through the territory. Some subordinates sat still and allowed the alpha to groom them whenever he chose to do so and in return groomed the alpha if he offered himself by 'crawling under', which is the mouse way of saying something like 'please scratch my back'. Other subordinates tended to walk away more often when the alpha started to groom or to crawl under. The latter type of subordinate eventually appeared to be attacked by the alpha more regularly and subsequently showed more 'fleeing'. Such individuals then remained for increasingly longer periods of time in their hiding places, especially when the alpha was walking around, and eventually they ended up as inhabitants of an uncomfortable and, for the alpha, rather inaccessible hiding place.

The subordinate mice who adapted to the initiatives of the alpha behaved submissively more regularly and underwent the maipulations of the alpha more often. They were however less often disturbed by aggressive attacks from the alpha, and did not much care whether the alpha was awake or asleep. The subordinates who put up more resistance towards the alpha showed on the other hand a conversely adjusted type of activity pattern; they kept silent for as long as they sensed that the alpha was active, and walked around when he was asleep. These gradually developing behavioural differences between subordinates can be described as differences in staying (beta types) and fleeing (omega types), since the latter type showed a tendency to flee the territory if possible. It should be noted here that the emergence of extreme omega behaviour patterns was an artifact of the experimental setting, owing to the fact that the mice were unable to escape. In natural settings they would probably have disappeared from the territory before showing such clear omega type reaction patterns. Indeed, in experimental situations in which opportunities for fleeing are provided, a large proportion of the (young) subordinate males do indeed flee the territory (Van Zegeren, personal communication). This is similar in many other rodent species (for example refer to Healey, 1967; Ewer, 1971; Wilson, 1975a, p. 278; Barash, 1977).<ref>In human societies there are also many occasions when a fleeing pattern is as difficult to achieve as it is with the artificially restricted mice in these experiments. Ghettos must consist of groups unwilling or unable to integrate fully and unable or unwilling to disappear. And whereas enforced ghettos are an extreme case, it exemplifies the general thresholds existing in any social structure, were it alone for overcoming the psychological bonds of habituation and attachment to the old situation and the extra risks and feelings of insecurity concomitant with breaking out.</ref>

In the process of a subordinate gradually becoming an omega, the behaviour of the alpha gradually changes towards treating the omega ever more as a stranger. What is important to note here however, is that the behavioural differences between betas and omegas seemed to develop before the alpha'began to treat the subordinates in a different way. This suggests that these beta/omega differences are caused by differences between the individual subordinates themselves. It could, in principle, also be explained by assuming that an alpha male initiates these differences by having a dislike for one of the subordinates, and that this subordinate thereupon avoids the alpha more than the other subordinates do. These differences in treatment by the alpha might initially be of such a subtle nature that even though the subordinate in question reacts promptly with increased avoidance behaviour, these differences have escaped our attention. It could however be shown in a cross-breeding experiment that the differences between omegas and betas originate primarily from the subordinates themselves (see van der Molen, 1987; 1989). (Of course these two hypotheses are not mutually exclusive; they may both be valid, supporting each other's effect.)

Experiments with beta- and omega-roles

In 30 populations (or groups of 4 males and 2 females in a large observation cage, as described above) observations were undertaken to determine whether or not the subordinate males did indeed develop into 2 distinct classes of betas and omegas. We used groups from the inbred C-57-black strain and the inbred CPB's-bagg albino strain and also from their F-1 and F-2 hybrids. These two inbred strains were chosen because of the conspicuous differences in the patterns of their aggressive behaviour and in their level of inter-individual tolerance. Two similar populations of wild mice were also incorporated in the experiment. Ethologically verifiable and clearly recognizable differences between betas and omegas developed in:

• 1 out of 9 CPB's populations
• 1 out of 8 C-57 populations
• 4 out of 8 F-1 hybrid populations
• 3 out of 3 F-2 hybrid populations
• 2 out of 2 wild populations

In the CPB's populations subordinate males tended to take up an (outcast-like) omega position, whereas in the C-57 populations the subordinates tended to take up a (compliant) beta position. The development of distinct classes of subordinates occurred quite clearly in half of the F-1 populations, and seemed to be normal in the F-2 as in the wild mice.

The hypothesis that subordinates from F-1 populations showed less individual differences in this respect than subordinates from F-2 and wild populations was tested by means of Fisher's exact test for independence. The statistic in question, having a discrete, hypergeometric distribution when the zero-hypothesis is true, rendered a significant value for stat.alpha = 0.10. This is in fact the most significant result that can be obtained with these numbers of populations.

These results suggest<ref>See, for example, East and Nilsson-Ehle in Srb, Owen and Edgar, 1965, pp. 450-74; or any other handbook on the basics of population genetics.</ref> a segregation and recombination in the F-2 generation of the genetic factors that determine the likelihood of subordinates becoming omega versus the likelihood of becoming a beta. An explanation of these effects in terms of differences in behaviour of the alpha mice does not make sense because in these data, subordinates were distinguished in behaviour only in relation to the alpha. Apart from this, an increase in behavioural variance of alpha males in the F-2 generation would imply more populations of the F-2 lacking either omegas or betas. This is contrary to what was found; thus the differences between omegas and betas stem, at least for a greater part, from genetic differences between the subordinate individuals. We label these differences accordingly as self-will, intolerance, tendency to have one's own way, or for that matter, tendency to dominate.

In these experiments it was found furthermore, that tolerant, compliant males, apt to take up a beta role instead of an omega role when in a subordinate position, were tolerant of the subordinates when performing an alpha role, contrary to males with a high level of self-will or tendency to dominate.

Other ethological research data

In many species differences between individuals have been found which resemble the above mentioned differences in male mice. From ethological field research it appears to be a general characteristic of social mammals that some individuals exert a lot of aggressive dominance, bullying their subordinates much of the time, whereas other dominants act as a sort of controller, governing the social relations in the group by social skill, sustained by the appreciation from companions rather than by aggressive intimidation. These differences are, for instance, reported from ethological research on mountain gorillas by Fossey (1972), on chimpanzees by Reynolds and Luscombe (1969), on a number of species including man by Chance and Jolly (1970) and Wilson (1977a, pp. 311-13) and exclusively on man by, for example, Lippit and White (1958), Krech, Crutchfield and Ballachey (1962, Chapter 12), Gibb (1969), Strayer and Strayer (1976), Hold (1976), and Sluckin and Smith (1977). Wilson (1975a) comments on these differences (p. 294):

It is not wholly imprecise to speak of much of the residual variance in dominance behaviour as being due to personality. The dominance system of e.g., the Nilgiri langur (Presbytis johnii) is weakly developed and highly variable from troop to troop. Alliances are present or absent, there is a single adult male or else several animals coexist uneasily, and the patterns of interaction differ from one troop to another. Much of this variation depends on idiosyncratic behavioural traits of individuals, especially of the dominant males (Poirier, 1970).

Itani et aL (1963) and Yamada (1966) describe the behaviour of extreme beta-type males in Japanese monkeys (Macaca Mcata) and indicate that a compliant temperament seems to be conditional for assuming such a role. Yamada further points out that, when such individuals eventually achieve a dominant position, a tendency for independence sometimes seems to exclude a tolerant attitude towards subordinates.

Differences of this sort between dominant males have also been described in stumptail macaques (Macaca speciosa) by Bertrand (1969), who describes both 'bullies' and 'fair alpha males' and stresses that aggressiveness is not always a necessary factor for dominance. She states that stumptail macaques differ considerably in the amount of intolerance and aggression displayed, and that in certain cases the sustained aggressiveness of some individuals, who were followed up for several years, seemed a personality trait that appeared early in childhood. Furthermore she concluded that the amount of investigative behaviour shown by an individual also depended upon the predisposition of each monkey, apart from social rank, age and conditions of captivity. Some individuals were far more adventurous than others. This personality dependence of investigative behaviour overruled age and rank dependent behaviour in particular when the stimuli were frightening or ambivalent.

In animal and social psychological research alike, variation in tolerance and acceptedness is reported between individual subordinate role styles. In general, it appears that individuals who do not manage to attain a dominant role (a-position in Fig. 4.1) may either stay in a subordinate position while (incrowd-)subordinates are often observed to gradually grow into a semi(incrowd-) subordinates are often observed to gradually grow into a semi-outcast or outcast position. Such outcast-like subordinates are then the potential migratory, running all the risks implied, whereas the better accepted incrowd-type subordinates, who show a better adaptation to existing hierarchical pressures, may eventually succeed the present dominant(s) if the latter should become incapacitated or even die. Such differences in social-role types have been observed frequently in relation to dispersal mechanisms operating through young individuals in particular (Wilson, 1977a; Barash, 1977).<ref>Similar descriptions have been given for e.g., deermice (Healey, 1967), free-living populations of black rats (Ewer, 1971, pp. 135-137), free-living lions (Bertram, 1975), rhesus monkeys (Vandenbergh, 1966), free-living Japanese monkeys (Irani et aL, 1963; Yamada, 1966) and by Eisenberg et al. (1972) for a number of primate species.</ref> Bertrand (1969) reports the occurrence of scapegoats in stumptail macaques and de Waal (1975) in Java monkeys. The latter reports that high-ranking individuals often formed alliances against the lowest ranking adults or adolescents, notwithstanding the fact that each of the high-ranking monkeys clearly dominated the scapegoat in question also without any help of others.

Human behaviour

The significance of the self-will versus compliance or individualistic, thing-oriented versus social dimension in the domain of temperament traits is not only corroborated by a substantial amount of ethological research data on animals, but also by ethological as well as personality–psychological literature on human behaviour.

Gibb (1969), Strayer and Strayer (1976), Hold (1976) and Sluckin and Smith (1977) report differences in dominance-styles of children, and of adolescents (Savin-Williams, 1977a,b, 1979, 1980); they are similar to those described above for mammalian behaviour. Gibb (1969) calls the two antagonist styles leadership and domination. With leadership, authority is spontaneously accorded by fellow group members whereas with domination there is little or no shared feeling or joint action and authority derives from some extra-group power.

Turning from dominance styles to more general differences in behavioural style, Abrams and Neubauer (1976) report that human infants differ considerably in the way they divide their attention between persons and objects. This trait dimension, which they called thing- versus human orientedness, was manifest as early as in the second month of life. They found that the more thing-oriented child shows a greater freedom in exploration. Therefore we might label this dimension of thing- versus human orientedness (or sociability) also as explorative versus social, parallel to the vocabulary in Bertrand's (1969) longitudinal research on macaques. Abrams and Neubauer (1976) furthermore report that learning processes are shaped in a way which is different for each type of child:

Training issues are characterized essentially as 'tasks' for the more thing-oriented child; for the human-disposed infant, they are characterized as acts in the spectrum of approval or disapproval ... If earlier impressions were that the more thing-oriented children are more outer-directed, by the third year of life they appeared more inclined to be motivated by inner determinants and resources, a distinction which seems to persist thereafter ... The dispositions of infants are re-inforced in the milieu, as implements in evolving strategies are cycled back into the psychological system and thus inevitably emerge as traits of character.

From her long range ethological research, Hold (1976) reports that children who rank high in the attention structure tend to set initiatives instead of complying to the initiatives of other children and that they

. . . prefer to play alone when the leading role was already taken by another high-ranking child. It seems that these children do not like to be commanded by other children.

This runs essentially parallel to what has been said in the introduction in that self-willed, thing-oriented individuals are more prone to become either dominants or loners than to become beta-type compliant subordinates. A similar trait contrast is employed by Edwin McClain (1978, 1979) in his detailed longitudinal study on the behaviour of adult women. He distinguishes between women who are dominated by a need for independence and women who are dominated by a need for affiliation. McClain, like Ausubel (1952), points out that two basically opposing patterns of maturation already occur in the parent—child relationship during a youngster's early years. He terms the resulting personality types as satellizers who tend to adapt to existing rules, versus nonsatellizers who tend to behave more individualistically.

The satellizing child establishes her life orbit about her parents, whom she perceives as the benign source of all that is good in her life. In contrast, the nonsatellizing child rejects this kind of dependency because she believes that her welfare lies in her freedom to choose her own course. (McClain, 1978, p. 436)

The material of McClain's study was derived from the behaviour of women. Kirton (1976, 1978b, 1987a) investigated a somewhat related dimension, namely the balance between adaptiveness and innovativeness in adults in general. The K.A.I. (Kirton Adaption–Innovation Inventory) was developed as a psychometric instrument for these investigations. Kirton based his instrument on the notion that a person confronted with a problem has a choice: he/she can do things 'better' or 'more' to solve the problem (adapt; the social-oriented approach) or he/she can do things 'differently' (innovate; the thing-oriented approach). Doing things better implies the acceptance of the old framework, while doing things differently means breaking accepted patterns. As Kirton says:

The Adaptor is right at home in bureaucracies, which tend to become more adaptor-oriented as time goes on ... whereas . . . the natural position of high Innovators seems to be out on a limb.

Kirton's work is of special significance for the performance of leaders (Kirton, 1961, 1977, 1987a; Thomson, 1980; De Ciantis, 1987). He shows that innovators tend to become initiating and directing task-leaders whereas adaptors tend to become consideration- oriented maintenance-specialists of social relations. This is in line with the differences between leader types as described by, for example, Bales (1953), Halpin and Winer (1957), Thibaut and Kelly (1959), Krech et al, (1962) and Reddin (1970, 1987). From a conceptual point of view, innovativeness may be considered, furthermore, as a positively appreciated creative variant of non-conformism and disobedience.

Conformity as defined by Krech et al. (1962) in their research on the dimensions of social interactive behaviour, is also related to the trait dimension thing-oriented and self-willed versus social and compliant. They found that some people are more resistant to group pressures and demands (the hard-core independents and the deviants) than are others (the easy conformists). Their research offers strong support for the proposition that conformity tendencies are significantly related to enduring personality factors in the individual. The relevance for our model becomes especially clear where they define conformity as a 'trait of the person' as opposed to conformity as a `trait of the situation' (or social role dimension in our words).

. . . conformity might be thought of as a 'trait of the situation'. [but] There are also marked individual differences in general readiness to conform, over a wide variety of situations. These differences . . ., reflect conformity as a 'trait of the person'. This distinction between conformity as reflecting the conformity-inducing properties of a situation and as reflecting the conforming propensity of a person should be kept well in mind. Much of the controversy and misunderstanding about the facts and theories of conformity stems from a confusion of these two aspects of conformity. (Krech et at, 1962)

Fig. 4.4. Feij's dimensions 'extraversion' and 'sensation seeking', the balance between them – our dimension adaptive – and their relation with 'general activity level'.

Of particular interest is the existence of a similar dimension in Factoranalytic Personality Trait Research. Feij (1978) compares the trait models of Heymann (1932), Eysenck (1953), Zuckerman (1974), Strelau (1974a, b), Buss, Plomin and Willerman (1973), and Buss and Plomin (1975), amongst others. Although these authors often use different classes of subjects and prefer different final rotations of their resulting factorial models, some of their dimensions appear closely related to our dimension self-willed and individualistic and thing-oriented and explorative versus compliant and social (Fig. 4.4).

For instance, a high score on Zuckerman's (1974) and Feij's (1978; Feij et at, 1979, 1981) trait dimension of sensation seeking indicates a strong need for change, exploration and new experiences, a tendency towards independence of other people and an anti-authoritarian attitude, while low sensation seeking implies a tendency to comply with conventional values and rules. Feij (1978) stresses that extreme sensation seekers may on the one hand be antisocial, drop-out delinquents, but may on the other hand be unconventional but fully accepted creative innovators. This is in agreement with what was postulated above, namely that highly self-willed individuals tend to become either drop-outs (omega-role) or accepted innovators in the focus of attention (alpha-role), and that individuals with a low self-will tend to assume beta-roles compliantly.

Buss and Plomin's (1975) trait dimension sociability, indicates a strong need to be together with others, a high responsiveness toward others and a predilection for social interaction above non-social reinforcers (Feij, 1978).

In Cattell's sixteen-personality-factor set, the dimension labelled as `liberalism' (Qj) is supposed to measure an underlying tendency toward nonconformity and independence versus a need for affiliation (Cattell, Eber and Tatsuoko, 1970; Karson and O'Dell, 1976; McClain, 1978). At least three other dimensions from his 16PF battery also relate to concepts discussed here, namely Cattell's higher order factor IV, indicating subduedness versus independence, the factor assertiveness (E), indicating cautious humbleness versus abrasive assertiveness, and the factor superego (G), indicating conscientiousness versus expedience (Kirton and de Ciantis, 1986; Kirton, 1987b).

In most other factoranalytic classification systems one or more dimensions may be discerned which are related to our concept of self-willed and thing-oriented versus compliant and social. Moreover, the empirical work of inter alia Goldsmith (1984, 1986, 1989; see also Kirton, 1987b, 1989) shows that the concepts emerging in all these factoranalytic dimensions from the various authors on personality are indeed statistically correlated, while forming a coherent web of conceptually intertwined behavioural characteristics.

Genetics

The above mentioned data from factoranalytic personality research are the more relevant because various writers point out that a genetic basis of these dimensions has repeatedly been firmly established (Eysenck, 1967; Vandenberg, 1967; Buss et al., 1973; Buss and Plomin, 1975; Feij, 1978; Claridge, Canter and Hume, 1973; Eaves and Eysenck, 1975; Wilson, 1975a; Plomin and Rowe, 1977, 1979).

The empirical findings of Kirton (1976, 1978a, 1987c) and Ettlie and O'Keefe (1982) are also in line with the notion of a biological basis. They report that differences in innovativeness versus adaptiveness are not significantly related to IQ, to level of education, or to previous experiences, but are apparently of a more basic (personality-trait) nature (Kirton, 1978a, 1987b, 1989; Kirton and De Ciantis, 1986). In this respect innovativeness, indicating the type of creativity differs from instruments which measure the level of creativity (Kirton, 1978a; Torrance and Horng, 1980).

In section 4.2 it was pointed out that in socially living mammals at least two sets of basic urges have to be postulated, which, independently from one another, vary over individuals, thus producing inter alia the adaptor/ innovator differences. The first set contains drives for social contact and interaction, leading to gregarious types of behaviour; the second set contains the drives for thing-oriented behaviour.

From recent neuroanatomical and endocrinological research it appears that there is probably a strong link between these two distinct sets of drives on the one hand, and specific neuro-endocrine systems on the other. Cloninger (1986, 1987) presented a biosocial theory of personality, based on a synthesis of information from family studies, studies of personality structure, as well as neuropharmacologic and neuroanatomical studies of behavioural conditioning and learning in man and other animals. He describes three dimensions of personality that are genetically independent, two of which, the novelty seeking dimension and the reward dependence dimension, relate to the two distinct sets of basic drives mentioned above.

One of his dimensions of personality trait differences is principally ruled by the monoamine neuromodulator dopamine. This system determines the heritable tendency towards intense exhilaration and excitement, leading to frequent exploratory activity (novelty seeking) and avoidance of monotony. Individuals high on this dimension are generally also characterized as impulsive, quick-tempered and disorderly. They tend to neglect details and are quickly distracted or bored. They are also easily provoked to prepare for fight or flight. The other dimension is principally ruled by the monoamine neuromodulator norepinephrine. This system determines the heritable tendency to respond intensely to signals of social reward and approval, sentiment and succour. Individuals high on this dimension are generally characterized as eager to help and please others, persistent, industrious, warmly sympathetic, sentimental, and sensitive to social cues, praise and personal succour, but also able to delay gratification with the expectation of eventually being – socially –rewarded.

According to Cloninger, a person high on novelty seeking (the dopamine system) and low on reward dependence (the norepinephrine system) is characterized as: seeking thrilling adventures and exploration; disorderly and unpredictable; intolerant of structure and monotony, regardless of the consequences; frequently trying to break rules and to introduce change; quick tempered and strongly engaged with new ideas and activities; socially detached; independent nonconformist; content to be alone; minimal ambition and motivation to please others; insensitive to social cues and pressures. Conversely, a person low on novelty seeking (dopamine) and high on reward dependence (norepinephrine) is characterized as: dependent on emotional support and intimacy with others; sensitive to social cues and responsive to social pressure; sentimental; crying easily; rigid; orderly and well organized; trying to impose stable structure and consistent routine; rarely becoming angry or excited; an analytic decision maker who always requires detailed analysis of complete information; slow to form and change interests and social attachments. The striking similarity of this polarity with descriptions of Kirton's innovator vs. adaptor dimension is obvious.

In summary, the available data, including data from neuro-endocrinological research, support the view that a biologically based trait dimension thing-oriented, explorative versus social or, in different terms, self-willed versus compliant is indeed conspicuously present, and does have genetic roots.

Selection within human social structures

The first three assumptions made at the beginning of this chapter apparently find ample support in ethological and psychological literature. Therefore, in any class of social (group) systems in which there are clear differences between members and non-members (prerequisite 4), cyclic changes should occur in the sense that each separate social group or structure only has a limited life-span, which is inversely proportional to the effectiveness of the selection pressure within the (group) structure in favour of compliance. The life cycles are then separated by turn-over catastrophes which go by various names, depending on the level of organization: territorial conquest; closedown; discontinuance; bankruptcy; revolution; subjugation; extermination; etc.

In the literature on animal ecology and population dynamics, the research data on population explosions and emigration waves, at more or less regular time intervals, are renowned (for example, Christian, 1970, on various species of lemmings, mice and voles). Whereas Christian points to the importance for evolution of these periodic changes in density and migration activity, the proximal causation of these conspicuous phenomena has up to this moment not been explained satisfactorily. It shall be clear that the present model constitutes, among other things, an attempt to fill this gap.

That selection forces do indeed operate within social groups against noncompliant, non-adapted individuals and other deviants, has experimentally been shown in various social mammals and birds (Neumann, 1981), in nonhuman primates (Kling and Steklis, 1976) and also in man (Schachter, 1951; Scherer, Abeles and Fischer, 1975; see van der Dennen, 1987, pp. 28 ff. and Flohr, 1987, pp. 200-2, for a discussion).

In the psychological literature we can also find many comments referring to the relevance of the discussed selection processes for the way our human society is run,<ref>(See, apart from the authors quoted here, also e.g., Snow, 1961; Etzioni, 1964; Weick, 1969; and Tiger, Chapter 5).</ref> including data on the personality dimensions these selection processes operate upon.

White and Lippitt (1960) and Scheflen and Scheflen (1972) give detailed behavioural descriptions of the process of creating chronic scapegoats as a fundamental process in the functioning of human social groups. They describe the physical as well as the cognitive and communicative aspects of the processes that lead either to getting stuck in a superdependent immobilized scapegoat-role or to becoming outcast (co-type). In their opinion there is a conspicuous contrast between, on the one hand, chronic superdependent immobilized persons who tend to neuroticism by accepting guilt and assuming the scapegoat role and thus getting stuck in cumulating 'double-binds' (Laing, 1967, 1970; Watzlawick and Fish, 1973), and on the other hand anti-social types who tend to deny guilt, generally refuse to be immobilized in a scapegoat role and tend to stay socially mobile, although in peripheral social roles. This is indeed what would be expected from our theory.

Parallel to what de Waal (1975) suggests in the case of Java monkeys, Scheflen and Scheflen (1972) explain how in their opinion every human social group or society generates automatically its own neurotic scapegoats, deviates and outcasts as a necessary by-product of continuous consolidation and reaffirmation of internal (cognitive) values and social order. Such marginal social roles serve for the society in question as a necessary external frame against which the internal social values and role criteria may be projected and by which the `shoulds' and 'should nots' for all its members are continuously exemplified (Erikson, 1966).

Milgram's (1974) famous experiments in which he asked subjects to administer heavy and supposedly life endangering electric shocks to stooges `for the sake of scientific progress', are also enlightening in this respect. According to Milgram, this general readiness to obey and even to torture fellow men, if urged and backed up by the authority of common opinion....

... is the psychological mechanism that links individual action to political purpose. It is the dispositional cement that binds men to systems of authority. Facts of recent history and observation in daily life suggest that for many people obedience may be a deeply ingrained behaviour tendency, indeed, a prepotent impulse overriding training in ethics, sympathy, and moral conduct.

This dependence of strongly repressive systems on a strong and dependable compliance of its employees and agents, explains what is often considered a paradox in the literature on holocausts. What is, for example, surprising is that the people who in 'das Dritte Reich' were in charge of the extermination machinery, quite generally appeared to be extremely docile, middle-class, adapted, morally rigid and reliable house-fathers and exemplary husbands, with an aversion to adventure and violence, and who more often than not were friendly and kind to children and pets in their daily social interactions, with an overtone of sentimentality. As shall be clear from the present theory, this is indeed the only type of person – the highly compliant, non-innovative, non-self-willed adaptor – that can be relied upon to carry through orders ('Befehl ist Befehl!') in situations where obedience strongly conflicts with morals and ethics. Under such extreme circumstances the selection pressure on personality characteristics, therefore, is extreme also, the not-socompliant individuals trying to avoid such ghastly agentic responsibilities. As Koestler (1967, in van der Dennen, 1987) eloquently stated:

It is not the murderers, the criminals, the delinquents and the wildly nonconformists who have embarked on the really significant rampages of killing, torture and mayhem. Rather it is the conformist, virtuous citizens, acting in the name of righteous causes and intensely held beliefs who throughout history have perpetrated the fiery holocausts of war, the religious persecutions, the sacks of cities, the wholesale rape of women, the dismemberment of the old and the young and the other unspeakable horrors ... The crimes of violence committed for selfish, personal motives are historically insignificant compared to those committed ad majorem glonam Dei, out of a self-sacrificing devotion to flag, a leader, a religious faith, or a political conviction.

Milgram (1974) labels this compliant, subordinate style of functioning the agentic mode, which expresses that somebody in that mode functions as the agent of some (personal or impersonal) authority. He points out that individuals tend to function in any one situation in either this mode or in its opposite, the autonomous mode. Milgram explains that the readiness to shift from the agentic mode into the autonomous mode in certain conflict situations differs considerably between adults, that people differ in the amount of time they spend in either mode, and that there is a complex personality basis to obedience and disobedience.

These differences between individuals in their tendencies either to comply with social standards most of the time, or to act autonomously and independently most of the time, are also of crucial importance for the way in which bureaucratic structures and other social institutions are run (Kirton, 1978b, 1987a):

... the 'adaptor' personality ... who can be relied upon to carry out a thorough, disciplined search for ways to eliminate problems by 'doing things better' with a minimum of risk and a maximum of continuity and stability ... [whereas] ... innovative change ... leads to increased risk and less conformity to rules and accepted work patterns (Bright, 1964), and for this reason it rarely occurs in institutions on a large scale ... (Kirton, 1978b, p. 611)

It is said that organisations in general (Whyte, 1957; Bakke, 1965; Weber, 1948 (published in 1970); Mulkay, 1972) and especially organisations which are large in size and budget (Veblen, 1928; Swatez, 1970) have a tendency to encourage bureaucracy and adaptation in order to minimise risk. Weber (1948), Merton (1957) and Parsons (1951) wrote that the aims of a bureaucratic structure are precision, reliability, and efficiency. The bureaucratic structure in its nature exerts constant pressure on officials to be methodical, prudent, and disciplined, resulting in an unusual degree of individual conformity in that situation. (Kirton, 1987a)

Therefore institutions tend to become more adaptor-oriented as time goes on because of selection, training and promoting policies which are in line with those aims (Drucker, 1969; Schumacher, 1975, p. 243). A negative selection pressure is continuously exerted against innovators.<ref>A similar process of selective isolation was seen by Rogers (1959), and reported in his account of the 'creative loner'.</ref> Even when an innovator finds badly needed novel solutions for pressing problems, it will often fail to render him social approval, inter alia because of inherent (sometimes insurmountable) communication problems with his more adaption-oriented colleagues. Instead of winning social approval when coming up with such solutions, the innovator finds that tolerance for his innovative style of approach is at its lowest ebb when his adaptor-type colleagues feel under pressure from the need for quick and radical change (Kirton, 1987b). Even when the novel solutions in question are accepted, it does not generally lead to a suspension of the above discussed selective forces. In an empirical study to investigate the ways by which ideas leading to radical changes in some companies were developed and implemented, Kirton (1987a) found that:

There was a marked tendency for the majority of ideas which encountered opposition and delays to have been put forward by managers who were themselves on the fringe, or were even unacceptable to the 'establishment' group. This negativism occurred not only before, but after the ideas had not only become accepted, but had even been rated as highly successful. At the same time other managers putting forward the more palatable (i.e., conventional) ideas were themselves not only initially acceptable, but remained so even if their ideas were later rejected or failed.

It can thus be seen how, much unlike the fate of innovators, failure of ideas is less damaging to the adaptor, since any erroneous assumptions upon which the ideas were based were also shared with colleagues and other influential people (Kirton, 1984).

As a consequence of these differences in selective pressure, ageing institutions suffer from the disadvantages of not having innovator type creative output available in times of change when policy and methods are required to change as well. Such necessary changes, therefore, are often brought about only when a precipitating event, or a crisis, occurs when at last the adaptor needs, and so collaborates with, the innovator (Kirton, 1961).

Ossification

Science is an outstanding example of a branch where these considerations about systematic intra-group selection pressures seem relevant. The very goal of scientific research is to find even better conceptual and instrumental frameworks, but, as Kuhn (1970) points out, changing the paradigms which are hitherto accepted without question by an entire scientific community requires a breakdown of previously accepted rules. Such breakdowns are the very process of scientific revolution and this revolutionary process is fundamental to scientific advance; thus, as a social institution, science stands out as an extraordinary oddity (Tiger, 1985). The consolidation and preservation of group cohesion and established rules are not its primary goal, but the creative expansion of conceptual boundaries. In scientific institutions, the innovator type input is not only needed at the rare times of inevitable change, but very regularly, since precipitating events or conceptual crises are the very thing that scientific efforts are supposed to be aimed at.

When ageing institutions become too adaptor- and compliance-oriented by the resulting unconscious bias in selection and in promotion policies, it may not be initially very disastrous in the case of factories or bureaucratic units. For as long as no drastic external challenges turn up, they can just go on producing their output as they formerly used to do with excellent results. But in research units it is eventually disastrous if the cognitive climate becomes more and more adaptor-oriented. The innovator type creative output, consisting of (often disquieting) conceptual challenges and explorations of the unknown and unthought, will in that case gradually be replaced by compliant adaptor type output consisting of puzzle-solving and quasi-discoveries without any conceptual threats. This means that scientific units finally reach an efficiency close to zero when becoming more adaptor-oriented with increasing age. After an initially fruitful phase of consolidation, the prevailing paradigms will become rigid and dogmatic. It is clear moreover, that in a government-protected scientific community competition does not operate freely. This may postpone organization—structural turn-over catastrophes considerably, and thus the timely rejuvenation science continually needs. As a result, . . . 'The Church of Reason [science] like all [ageing] institutions, is based not on individual strength, but on individual weakness. What's really demanded is inability. Then you are considered teachable: a truly able person is always a threat' (Pirsig, 1974).

This description of the process of ossification also fits perfectly for most of the established and institutional religions, as has been recognized by many philosophers and other scientists. Kierkegaard for instance, heavily criticized the organized churches, pointing out that in our society authentic existential religiosity nowadays has two great enemies: philosophers indulging in mere abstract speculations of a strict and limited rationality on the one hand, and church-going fundamentalists and uncritical believers on the other. Kierkegaard believed that the Church has degenerated into a bunch of unthinking fanatics, or even worse, a flock of passionless and anemic herd-mentalities, who dutifully walk into church for no other reason than that was the direction most others were walking. He resented that the church does not have the decency to recognize that whatever its teaching of watered-down, polite moral humanism has become, it isn't Christianity any more (Wilber, 1983).

Christianity may have been founded for the enlightenment of mankind, as an attempt to raise people to a higher level of autonomy and socio-psychic health, and for overcoming the frequent social tendency towards hateful and revengeful cultivation of deviants, scapegoats and other presumed enemies by institutionalized practices of denunciation and mobbing (Coser, 1956, 1978; Girard, 1982; Hoffschulte, 1986). But, like any other social institution, the Church has gradually deteriorated into a system, preoccupied with its own propagation as a system, and thus – contrary to its original goals – with power and the binding-in-dependency of its members in uncritical docility (Toynbee, 1972). The Church does not invite, any more, to mysticism or to experiencing the 'void', instead it imposes 'belief in God and promotes conformity and respect for 'respectability' (Laing, 1967). Similar considerations hold for other traditional religions and belief systems.

Social-role blindness

Apart from these specific ossification phenomena, many more areas in human society can be found where such effects of the selection mechanisms, as discussed in section 4.10, are manifest. Since these selection mechanisms are operative in lower social mammals as well as in man, they must be anchored quite solidly in the behavioural system. This is not surprising because this mechanism does have considerable evolutionary advantages, not only in animals, but, at least up to recent times, also in the case of man. As mentioned in section 4.4, it facilitates speciation through genetic adaptation to marginal habitats, helps in many species to overcome migratory bottlenecks, and even serves to motor the evolution of culture in the case of man.

It seems plausible therefore, that if in the case of man, our superior capacity for learning plays a modifying role in these matters, then the organization of our intellectual capacities will have evolved in such a way, as to enhance the occurrence of selection cycles, rather than to thwart them. The mechanism of selection cycles and periodic turn-over catastrophes is basically powered by the involuntary forces of attraction and repulsion between individuals within social groups and structures. It must have been evolutionarily advantageous, therefore, for behavioural and cognitive `masterprogrammes' to develop, serving to prevent the newly evolved intellectual capacities from interfering with the (phylogenetically very old) involuntary biases in social interactions.

This is indeed what can be found. Human beings appear to be peculiarly unable to assess objectively the quality of their own social-role behaviour and the behaviour of other people they are dealing with in the social group. There is a sort of 'social-role blindness', of specific blind spots in our cognitive capacities, safeguarding primitive, elementary tendencies of being either attracted or repulsed by other people, depending on the own and on the other's social role and position. As in the experiments with mice described in sections 4.5 and 4.6, the omega-like subordinates, the peripheral nonconforming types, are also in man most likely to be disliked by the established leaders as well as by the conforming and compliant subordinates. This is, in fact, a tautological statement, since drifting into a marginal or an outcast position ((o-type in Fig. 4.1) is just another way of saying that one is less acceptable to the in-group.

As a tool for this mechanism, a considerable part of human communication consists not of transferring pure information, but of more or less involuntary emotional expressions of praise, admiration, criticism, ridicule and insults, as is shown for instance in the ethological work of Weisfeld (1980) on social-role behaviour in adolescent boys, or in the sociological investigations by Segerstrdle (1986; Chapter 14) into the Wilson–Lewontin scientific debate as part of the sociobiological controversy. To a large extent the use of language serves to support or to camouflage non-verbal actions, actions for manipulating other people and for staking out and sustaining social roles (Scheflen and Scheflen, 1972; Mehrabian, 1972; Argyle, 1976a,b).

In factoranalytic research on the social interactions between people, the first and by far the most conspicuous principal component is the so called evaluation of positive–negative (Good–Bad) dimension, describing to what extent one appreciates or disappreciates the rated other. In questionnaire research where elucidation of the actual social behavioural attitudes and social-role distributions is the primary goal, the raw data are, in general, firstly corrected for the positive–negative evaluation or social desirability dimension by partialling out its influence (Benjamin, 1974, p. 419). Those who rate other persons or questionnaires in general colour their judgements with appreciation or disapproval to indicate, explain and consolidate the social relations between themselves and the rated person.<ref>In the case of dominating individuals this cognitive distortion of the own and the other person's qualities is labelled by Kipnis 'the Metamorphic Effects of Power' (1976, ch. 9).</ref>

The importance of this negative or positive bias in the way we think about our companions is also expressed by the fact that most behavioural attitudes and personality characteristics can be expressed in positive as well as in negative terms. We virtually have, therefore, a double set of conceptual labels for other people's actions and behavioural attitudes; a positive set and a negative set. This cumbersome and at first sight inefficient cognitive organization, in which the pure assessment of other people's behavioural qualities is blurred to a great extent by the strong involuntary bias of appreciation or disappreciation, can only be evolutionarily advantageous if it serves an essential purpose. From the above, it may be concluded that this purpose may be found in protecting the involuntary attraction and repulsion reflexes, which direct our social behaviour, against our intelligent faculties. Indeed, human individuals are hardly aware that the way they assess the other person's qualities is to a large extent coloured by their positive or negative feelings towards that other person, resulting from the involuntary forces of social attraction and repulsion in operation. People do not realize that their, say negative, labelling of (the behaviour of) an important other can easily be changed into its positive counterpart by simply regarding the same behaviour from the point of view of a supporter, and vice versa. They tend instead to attach a sense of permanence and absoluteness to their (categorically negative or positive) judgement, and in particular they are not aware of the relativity of the judgement in terms of its dependence on the mutual social positions of the rater and the ratee.

In summary, the postulated blind spots and no entry signs in our intellectual faculties apparently do indeed exist. Despite our vaunted intellects and our protestations of rational and scientific know-how, we humans show a disturbing tendency to reserve our intellectual powers strictly for certain specific tasks. In other specific areas of functioning, like the mechanisms of social attraction and repulsion mentioned above, we tend to rely on involuntary biases while allowing the intellectual faculties to be effectively blocked. Therefore, ironically and paradoxically, this specific stupidness, this social-role blindness in us humans, should probably be regarded as a special adaptation to our great cleverness.

In daily life, the result of these cognitive biases is that, in many instances, we cannot help but foster, involuntarily, a lower esteem for other persons if they happen to be less 'in-crowd' than ourselves. In more extreme cases, we cannot help tending to join others in mobbing or in scapegoating. We tend to justify the actions taken through our (biased) evaluations of the outcast's or scapegoat's qualities, attitudes and behaviour (unless we incidentally happen to be one of the outcast's supporters). Being in the agentic or systemic (Milgram, 1974) or compliant (Apter and Smith, 1976, 1985) motivational mode while dealing with a victim, we involuntarily tend to see the person in question to a larger or lesser extent as inferior, or even repulsive, detestable or evil. This cognitive distortion can in fact be considered the behavioural basis of torture. Without this psychological effect, the role of torturer would be impossible (Amnesty International, 1973), various built-in inhibitions on aggression would then in most cases take precedence. What happened in the extermination camps of 'das Dritte Reich' is an extreme – though unfortunately not very rare – type of event, exemplifying what this human faculty for selective blindness may facilitate.

The Nazis called their victims 'Untermenschen', but likewise, we ourselves have in turn a strong tendency to label the people who were in charge of the Nazi extermination projects as incorporations of evil, as devils incarnate. As was stressed in section 4.10 however, they were, if they can be characterized at all, rather the opposite, or they would not have been fit for a task where ethics and personal norms would most likely conflict with obedience. Like the spectator-subjects in Milgram's experiments, we cannot imagine ourselves doing the same in similar circumstances, but the facts are that most of those Nazis were not beasts, but very quiet middle-class, social-oriented adaptors, and we are no saints, but ordinary people, who in similar circumstances are rather likely to do similar sorts of things. The difficulty we have with acknowledging that those Nazi employees were not so very different from ourselves, and the other way around, exemplifies the all overruling strength of this type of social-role blindness within ourselves.

All in all, it seems that we cannot help but hate our (self-created) enemies and that we cannot help but love primarily just those individuals which the described selection-cycle mechanism urges us to appreciate. The effects of this same blinding mechanism can also be recognized in less extreme contexts, like for example, the social interactions between scientists or between managers. The contribution by SegerstrAle gives us a very illustrative and piquant example of how this mechanism works out in the case of scientific colleagues with strikingly different cognitive styles. In her account of Wilson's and Lewontin's respective contributions to the sociobiology debate, Lewontin plays the part of the thorough and careful adaptor whereas Wilson plays the part of the creative, speculative and daring innovator, and the subsequent mutual denunciation of each others scientific qualities is prototypical.

Thus, the turn-over catastrophes keep happening unhampered, in the case of man just disguised in cognitive ornaments which we, erringly, take for true. Whenever a turn-over catastrophe is at hand, there is still another type of social-role blindness in operation which ties in with the cognitive biases discussed above. It is discussed in more detail in the contribution by Tiger (see also e. g., Janis' book Group-think, 1982; Tiger, 1985). According to Tiger the evolution of the human intellect must of necessity have been accompanied by a simultaneous development of a set of awareness blocks, safeguarding groupthink tendencies and safeguarding the unhampered compliance with social habits and prejudices. He argues that the human intellect has primarily evolved as a tool for enhancing coordinated social action, not for independence and for critically observing the social processes one is involved in. Reason was designed to improve consent with the overriding purposes of kinship, not to challenge them. The effect of these particular blocks is that eventual intelligent attempts by any non-conforming individuals, trying to stop the precipitating catastrophe from happening, are likely to be futile. In most persons involved, this particular limitation to the use of intellectual faculties will overrule any capacity to rationally assess the personal risks and general consequences of the turn-over catastrophe at hand, and that will impel them to join compliantly in concerted mobbing or warring actions, not unlike lower social mammals.

It is clear from the aforesaid that the former type of social-role blindness ties in here seamlessly. The concerted actions of animosity towards scapegoats, outcasts or enemies are of course greatly facilitated by the involuntary and uncritical denouncement of the supposed opponent's qualities. And at higher levels of organization, for example political states, the degrees of blindness of the collective are even more disquieting, not only with respect to the systematic and collective denunciation of supposed enemies, but also with respect to what are desirable and effective political courses of action (Janis, 1982), in particular as soon as the ideals and goals chosen become fixed and rigid (Talmon, 1980). As Popper says: 'The attempt to make Heaven on earth invariably produces hell'. The most extensive, quixotic and disgusting violence is justified with the invocation of an Utopian ideology, a paradisaic myth, a superiority doctrine, an eschatological or millenarian ideal state, or other highly abstract political/ethical categories, metaphysical values, and quasi-metaphysical mental monstrosities: national security; raison d'Etat; freedom; democracy; God; Volk und Heimat; Blut und Boden; peace; progress; empire; historical imperative; sacred order; natural necessity; divine will; and so on and so forth (van der Dennen, 1987). In view of their tasks, the stream of information governments take in is even more biased and unbalanced, and their tools are even less effective, than they are in the case of individuals (Deutsch and Senghaas, 1971). This unbalance has become particularly precarious under the present late-20th century conditions.

Nature and nurture

It can be argued that, at least in the case of man, the same social structure cycles with their turn-over catastrophes might occur merely because of mechanisms on the cognitive psychological level. In that case one would not need to postulate a genetic background for these gradual shifts in social structures to occur.

Indeed, as we saw above, social-role blindness and related cognitive biases are a very powerful influence in man. Moreover, we can also find a wealth of empirical and experimental data on the various constraints on learning in man, on habit forming, traditions and the transfer of cultural information, on perceptual biases like the cognitive dissonance theory, etc., all showing that our behaviour is organized in such a way that a great inertia of ideas, concepts and habits is safeguarded in spite of our capacity to keep learning. These data would suggest that enough mechanisms, at a purely cognitive and cultural level, can be traced as to make social-structural cycles likely to occur. Admittedly, the basic requirement for the postulated selection cycles to occur is not so much that there is a genetic basis to it, but rather that individuals, once their phenotypes in these areas of functioning have established themselves, can not be reshaped into their opposites. As we saw above, this inflexibility aspect, irrespective of its causes, has been firmly established by psychological research. However, the evidence for genetic influence cannot be neglected. These mechanisms therefore are most probably implemented on the genetic as well as on the learning level. It goes without saying that in man, the learning animal par excellence, the influence of learning will be relatively important in that case.

Another, related, critique is the argument that where a multi-gene basis of these differences should be expected, a strong enough selection pressure and a quick selection response are difficult to imagine. However, no high mortality, low fecundity or whatever on the part of the declining morph needs to be assumed at all. Basic to the model rather is the existence of differences between incrowd- and outcast-individuals. No physical elimination whatsoever needs to be assumed to let the cycles run. The only thing which needs to be postulated is that the in-group/out-group and the incrowd/outcast distribution of social roles and positions is subject to reshuffling; it depends on the level of organization we are talking about whether the postulate of a genetic effect needs to be included in a description of the cycles or not. In man this will, in my view, probably only be indispensable in the case of very longterm cycles on a very large scale. On most levels of human social structures the individuals selected against just need to be shifted into outgroup or outcast positions, relative to the unit(s) of organization in question, for the selection process to proceed. The very presence of the removed individuals in the organizational periphery then increases the likelihood of a turn-over catastrophe.

Perspective

As was pointed out above, the duration of social-structural cycles is predicted to be roughly inversely proportional to speed and intensity of selection for the trait under discussion. In an industrial company the intensity of selection and the take-on/dismissal percentages are much higher than for example, the selection intensity and the immigration/emigration percentages in the much larger units of political states. Therefore the average cycle periods are likely to vary from a few decades in companies (for illustrative material refer to e.g., Schumpeter, 1939; Kirton, 1961, 1976) or in political parties (e.g., Ostrogorski, 1982), to a few centuries in political states (e.g., Olson, 1982), or even to one or two millennia in whole civilizations (refer to e.g., Spengler, 1918; Toynbee, 1972;<ref>Toynbee disagrees with what he calls Spengler's 'determinism'. Though he (Toynbee) gives abundant material to illustrate the point made here, he emphasizes that one cannot convincingly speak of some or other predetermined and fixed life span of societies. The present theory would support Spengler's view. But it would also give room for something that Toynbee stresses, namely that, as far as their life span and their spin-off in terms of disseminative effect towards other societies is concerned, societies differ greatly from one another. According to the present theory, it very much depends on incidental environmental factors how effective selective emigration can be, and how strong the differential propagation within the structures themselves. And it depends on the actual presence of competing structures how quickly the effects of the internal selection processes will precipitate an eventual turn-over catastrophe.</ref> Darlington, 1969; Davis, 1974). The small-scale turn-over cycles with a relatively stronger and quicker selection effect are superimposed, therefore, on the larger-scale turn-over cycles with a longer life span. Thus, individuals may be outcasts in terms of some small-scale social structure while at the same time being totally accepted incrowd members in terms of some larger-scale social structure. The small-scale cycles may be seen as the ripples on the surface of the long range waves of the large-scale cycles, What happens with a person at the social-role level of a sports club is not necessarily parallel to what happens to him at home or at the level of the village community, and what happens to a person at the level of a company does not at all need to be parallel to what happens to him on the level of the political state. In fact, being an incrowd group member on some small-scale level of organization may be vital for a person to keep functioning properly in case of struggling with an outcast position on a larger-scale level of organization.

If it were possible to manipulate these – hitherto involuntary – selection mechanisms, it would be possible to stop or to speed up population cycles at will. This might for instance be relevant for personnel management in industrial companies or for measures at the level of political nations. The latter might be of particular significance in our nuclear age, since population cycles at this broad level tend to be worked out and consolidated by means of war and other economic strangling techniques. Mankind as a whole, up until now, has been able to afford this luxury of genocidal praxis, but war and economic asphyxia, nowadays, threatens to come close to total nuclear destruction. It might be worthwhile, therefore, to take the pressure off the dynamic population cycles kettle and to search for a way to replace or short-circuit nature's hitherto applied selection tricks with which it powered our evolution and our spatial spreading. It seems about time to substitute alternative and less dangerous mechanisms for it.

It will be clear however from section 4.12 that such is easier said than done. The mechanisms in question apparently are anchored in our behavioural system quite solidly. Many authors are however of the opinion that this is no reason to sit down in utter despair. Girard (1982) for instance, points out that in some respects social-role blindness is gradually losing its grip on our behaviour. He calls attention to the exemplary function of Christianity. On the one hand, Christianity has its roots in ancient Jewish traditions, suffused with admonitions towards and justifications of revenge and genocide (see e.g., Deuteronomy, 20, 17, 7, 12; Joshua, 1-3, 6, 8, 10; Kings, 3, 22, 23; Isaiah, 61). The Old Testament can in fact serve as a school example of militant- ethnocentric delusions of racial superiority. On the other hand, a novel phenomenon has emerged from the Jewish tradition, and even more so in Christianity, which is the attempt to replace organized spite, hate and revenge by love and compassion. This scheme may not have been completely successful as yet – were it only for the systematically organized violent blindness in and through the Christian religious organizations themselves –but it certainly has had some effect in overcoming the all-out violent tendencies towards deviants and scapegoats. Christ's example and admonitions like 'love your enemies', and the attempts to break the old tradition of revenge and the resulting vicious spirals of violence and counter-violence, counteract the ordinary selective forces within social groups we have discussed here. They put the primordial tradition upside down by denying the guilt of the victims and scapegoats and by putting the blame on the persecuting society. They de-sanctionize social violence; but what is most important, this tradition, though not reversing our behaviour instantaneously, has opened our awareness to what is actually going on at the social-cognitive level. It has opened our awareness for the fact that we do not like to give up our scapegoats, that we are attached to them and find it utterly difficult to refrain from denouncing and persecuting them (Girard, 1982). It constitutes therefore a massive attack on certain blind spots, on aspects of social-role blindness that, since aeons, have been the cornerstone of the cyclic selection processes themselves.

That is surely not the only glimmer of hope that may be discerned. The involuntary selection forces discussed are under attack from more sides. On the level of personnel management Kirton's work – as described in the previous paragraphs – may also be interpreted as an attempt to extend our awareness beyond its age-old confines into the realm of the dynamics of social attraction and repulsion, and what is more, his approach provides practical scientific tools to undercut the involuntary selection effects, tools that are likely to be utilized more and more because of their profitable effects on the output of the social structures (industrial companies) involved.

It is my hope that this chapter may also add to our understanding of the mechanisms underlying periodic turn-over catastrophes. Admittedly, the present theory, in part, is still tentative, but its relevance for our very existence might urge us to search for further experimental evidence against or in favour of its basic assumptions.

Notes

<references />