(Warning: these notes are not a substitute for reading the book!)

Chapter 5

Variance- (risk-) reduction models of sharing

Land tenure, sharing, and exchange are all part of a permission granting behaviors. They deal with kinds of access that people have to resources.

Sharing as the sine qua non of foraging society in the minds of most anthropologists both in Man the Hunter symposium and in constructions of hominid evolution. However, their is considerable diversity in terms of whether food is shared or reciprocal access given to land.

Sharing

Man the Hunter symposium suggested that sharing was a way to reduce risk in foraging and a way to create social bonds between people.

In some cases there is demand sharing (tolerated theft?) and in other cases children are trained carefully to share. Denial of sharing leads to ill feelings and conflict. Kung may become obsessed with shares and make cracks regarding stinginess. Table 5.1 summarizes some rules regarding sharing in foraging and horticultural groups (note Yanomamo data is not correct in terms of kin bias - there is none).

Speth’s research indicates that women get less meat and fat compared to men. Among Tlingit elders get better meat portions. Also, large game seems to be shared more commonly than small game.

"A balance sheet of sorts is kept, sharing is not a product of the innate generosity of hunter-gatherers." (p. 167).

Gould’s suggestion that the environmentally stressed Ngadadjara share more than the Tolowa who live in abundance.

Variance reduction models of sharing

A. Little incentive to share (since all do equally well or poor) so storage as a mode of risk reduction occurs. Alternatively, they could migrate to better areas or exchange with other groups.

B. Great incentive to share since high daily variance for individual hunters and across all hunters.

C. Individuals get food predictably and so do all other members of the group, storage at household level is expected..

D. Predictable returns for individuals but group is not synchronized. This, by itself, suggests that specialization is occurring.

Winterhalder’s simulation shows that increasing diet breadth to reduce variance only reduces variance by 8% while reducing efficiency by 6%. However, if two hunters pool resources then variance can be reduced by 58%.

Boyd and Smith show that hoarding will develop even though sharing would be more adaptive. Standard tit-for-tat model. This occurs in the instantaneous model but sharing will win out over the long run as hoarders and not included in sharing networks. Furthermore, Winterhalder shows that sharing for variance reduction ceases when about six families are involved, game depletion increases, and free loaders may become a problem

Is Reduction in Sharing Sufficient to Explain Sharing?

Kaplan and Hill on modes of sharing

I. kin selection

II. tolerated theft

III. reciprocity

A. tit for tat or when the value of that which is given is greater than that which may be received in the future

B. trade or specialization between exchangers (carbohydrates versus game, or meat for sex)

IV. public goods or cooperative acquisition: all should have a share because all helped.

• little support for tolerated theft since no contests over resources. But large game shared more than small.
• kin selection seems to account for narrow range while in mission camps but in wild kin selection has too small a radius
• amount received by family is in proportion to the number of dependents in family (ask Kim)
• ultimate they go for the trade option: men trade meat for sex since reciprocity is not equalized after a period of three years (lousy hunters are always net recipients and good hunters always give more.

Over a sufficiently long period of time good hunters would get back what they put in. But (a) computer simulations suggest short-term memory is better than long-term; (b) people discount the value of a good in direction proportion to the length of time between one exchange and the next anticipated; and (c) empirical data over time suggest that good hunters continue to give to poor hunters over time.

Good hunters may give as a form of old age insurance. This is unlikely since who would subsidize them (children of poor hunters?).

1. oldsters may have considerable value as story tellers, keepers of important information useful to the group, and care for children.

2. If elders are not productive then one must ask what are the costs of not helping them. He suggests that their children will become angry but he admits that he has no data to prove this point.

Land Tenure

Reviews Speck’s arguments about territoriality among Algonkian groups as a mechanism for conservation and later research which suggests that it may be an adaptation to the fur trade. Mention of The Territorial Imperative followed by Devore and Lee’s claim that all foragers have means of gaining access to land or for moving from group to group.

Three functions for reciprocal access:

• adaptation to fluctuating resource levels
• security networks about information regarding people and resources to permit continual adjustments of people to land.
• sentimental ties to land which would prevent over exploitation of rich land.

The economic defensibility model

The standard Dyson-Hudson and Smith model followed by Thomas’s Kawich, Owens River, and Reese River Paiute examination which shows that territoriality could exist on a seasonal basis for particular areas winter pinyon villages.

Social Boundary Defense

Permission must be asked, therefore it is possible that it may be denied. The point is that people cannot go where they want.

Cashdan argues that social boundary defense may occur in areas of scarce resources. People who don't ask run the following risk:

• trespassers may not have sufficient knowledge to exploit the area effectively.
• they may be caught and retaliated against.

 

• they may need access next year to the visitor's territory
• cost of exclusion may be too high

According to Smith, social boundary occurs when:

• when residents are more knowledgeable about resource locations than visitors (when would this not be the case?)
• when uncoordinated search through crowding reduces efficiency
• when today's visitors are likely to be tomorrow's hosts
• when residents can impose effective sanctions

 

Case A.
(Roughly corresponds to Dyson Hudson and Smith’s case C.) In this situation groups are strongly correlated in their successes and failures so we would expect strong perimeter defense. Since their is high intragroup variance storage will buffer shortfalls (however, he should realize that sharing may also develop in the group if storage is not possible)

Case B
According to Cashdan, this is the case for most foragers. Since intergroup correlation is low, then reciprocal access through social boundary is likely. So is storage is done during periods of plenty but this storage may be social storage by allow other groups access during times of plenty and expecting reciprocity in the future.

Case C
Intrgroup variance is low (in synchrony) so no need to call on groups for assistance. Furthermore, intergroup correlation is high which means these groups are in about the same situation as you are at any time. So resources problems would lead to migration, perhaps to another area either unexploited or where people have a different resource base.

Case D
Here we have low intragroup variance (in synchrony) but low intergroup correlation which, in times of stress, would lead to social boundary defense through reciprocity since the neighboring group may have different resources that you could use.

Chapter 6 

  Group Size and Reproduction

Man the Hunter demographics, the standard generalizations:

• group size of 25
• population at about 20-30% of carrying capacity
• intentional population regulation

 

no records

difficulties in age estimation

the role of disease and contact as modifiers of population

information on infanticide, paternity, and sex difficult to collect

small numbers (e.g., San crude death can vary from 4 to 26 per thousand over a several year stretch.

 

Magic number of 25 and 500

500 from Birdsell’s work on gene flow. He had to estimate deme size and empirically thought that it was about 500 (maximal band) when it was actually between 175 and 1,000.

Wobst using simulations found that equilibrial populations ranged from 175 to 475 and that 475 was more likely as the minimum number.

The 25 band size tends to hold up moderately well from an empirical basis but some sedentary groups are much larger. Johnson argues that the number 25 has to do with decision making processes for about 6 families working together. If group is larger then additional level of hierarchy is needed. Wobst suggests that 25 is a compromise between reproductive and economic factors.

Winterhalder (1986) focus on coefficient of variance, correlation in success (synchrony) of foragers, and number of independent foragers. In general, there is little advantage in terms of reducing CV regardless of synchrony rates when there are more than about 7 foragers. Further, the larger the group the rapidly resources are depleted. Based on age structure of population, 7-8 foragers works out to a group size of about 25.

Aggregation and Dispersal

Discussion of Horns model

Horn’s Model

Resource Availability	Observed spatial organization
Evenly distributed, stable, and predictable	Small and dispersed settlements
Clumped, mobile, and predictable	Large settlement in central location
Clumped and predictable	Large, semi-permanent settlement in central location

 

Heffley found that Horn’s model does not account for storage, which, in her cases, led to aggregation.

Suggests that Horn’s model parallels Winterhalder because unpredictable resources would lead to low synchrony which in both cases increase group size.

Communal versus Individual Foraging

Pygmy Hunting and Technology (factors leading to bow and net hunting):

• Turnbull suggested that since the Congo forest was so rich, the Mbuti could either net or bow hunt, therefore the differences were cultural (a good example of possibilism).
• Harako suggested that net hunting was recently introduced and more efficient and should take over .
• Abruzzi said it was a matter of population density: Bantu horticulturalists moved in which forced a more efficient method. Bow hunting occurs because Pygmies could work for horticulturalists in fields to offset lower hunting yields.
• Milton suggests the opposite: net hunters live in a poor environment and must resort to higher yield methods.
• Bailey and Augner show that there are no correlated environmental differences between net and bow hunting. Further, when net making taken into consideration bow hunting is slightly more efficient. Bailey and Augner show that where there is labor opportunities for Mbuti women they work which leaves little time for hunting which forces men to hunt with bows since net hunting requires larger groups that must be composed of women.

Smith's work on foraging group size: prediction that the size that optimizes individual return rate will be chosen. In general, modal size was the most common (especially for individual hunts) but too frequently group size was too larger. Explanation on joiner's rule: people were allowed to join even though it reduced per capita rate in the hopes that when they wanted to join they would be allowed in. Joiners who depress per capita have strong motivation because solo hunting return rates would be inferior to per capita group hunting rates. Also suggests additional benefits such as pooling for entire settlement, training of young hunters, mutual aid, and recreation.

Aggregation could be for a number of important social reasons such as marriage, ceremonials (which may be an effect and not a cause of aggregation), and information exchange.
 

Carrying Capacity, Foraging, and Population Growth

Research by Baumhoff and others shows a strong association between measures of food density and population density among foragers.
 
Two measures of carrying capacity:

• number of people who can be supported given subsistence technology (forgets long-term equilibrium);
• density dependent growth rate. General problem with carrying capacity model is equilibrial assumptions that foragers want to limit growth.

populations fluctuate widely to moderately under different environmental productivity rates

sensitive factors include how much children worked and when they began to work, nutritional requirements, and harvestable biomass

if boom bust periods occur rapidly this suggests that people could recognize them and perhaps innovate subsistence technology (agriculture)

 

Handwerker suggests that foragers do not exercise contraception or abortion to significant effects and therefore these practices no little or no effect on forager demography. He also suggests in the same article (although not mentioned by Kelley) that foragers have low fertility because they have low coital rates because they are so tired from their foraging activities.
 

Rates of infanticide in some foraging groups

Hiwi	4% (female bias)
Anbarra	5-11%
San	1%

 

Preferential female infanticide

Rasmussen estimated that 67% of Netsilik females killed at birth. Problem in the estimates.
Our knowledge of the incidence of infanticide is problematic because:

• males may not be a good source of information since females bear without male presence
• Bugos and McCarthy found that males knew very little about female reproductive histories
• secondary sex ratios seem to vary greatly (116 for Hiwi and 117 for Ache compared to 91 for San)
• problem with small samples
• age estimation difficulties

Freeman says that it is a way to balance people with land by reducing fertility by limiting the number of females. This standard ecological model stresses human's ability to overrun the productivity of the environment. However, data analyzed by Hill on the !Kung indicate that they harvest about 0.7% of the available animal biomass per year. Where deer hunting is regulated sustainable rates of harvest are up to 20% kill of the population per year.

Balikci says that it was to reduce male competition for mates since males died so frequently in economic tasks. Irwin showed an inverse correlation between temperature and sex ratio: more females killed (i.e. fewer girls) where temperature was the coldest (cold as a proxy for dangerous hunting and male hunting deaths)

Smith found no evidence for population regulation of sex ratio balancing and tried Fisher’s sex ratio theory and costs of raising boys. Since boys were costlier (died more often and marry at a later age) there should be fewer boys. However, boys help feed the family which may make them cheaper. They concluded that "the most plausible hypothesis to account for female infanticide in the Arctic is that sons were favored because they could contribute more to the fitness of their parents and siblings than could daughters". (p. 238).

 

Blurton Jones shows that birth spacing infanticide actually maximizes a woman's reproductive success or fertility. Pennington and Harpending show that the more children a woman has the more that survive to reproductive age. However, this is not good evidence against Jones since he was concerned with birth intervals and its effects on RS.

Hewlett points out that children drastically decrease a woman's work load (cite your data and that of Ache) and that foraging populations adapt to this by having multiple caretakers for children which is uncommon in horticultural societies.

Harpending and Pennington show that women who have been married more than once have higher child mortality rate and the Ache data show that children who have lost a father have a 9% chance of dying before age 15 while those with fathers have a 1% chance of dying.
 
 

Hewlett says the more time husband and wife are together the more time husbands will spend in childcare.
 

The TFR of foragers is lower than agriculturalists. However, reanalysis by Bentley shows that if we have foragers, agriculturalists, and horticulturalists there is no difference between horticulturists and foragers and both are lower than rates of agriculturalists.

Maternal Nutrition, activity, and fertility

Frisch’s critical fat hypothesis. Forager studies show that births cluster nine months after the highest food intake month.

Precise relation may work in one of two ways according to Ellison:

1. low food intake could reduce nursing rates (because of low milk supplies) and therefore increase infertility
2. low food intake could directly impair ovarian function. And the issue may be in long term energy balance and not short-terms. Wilmsen’s data shows that a chronically marginal diet impacted by occasionally sever dietary intake may lead to very low fertility (through cholesterol and estrogen storage in adipose tissue).

Hill and Hurtado found that net caloric intake (product of gross caloric intake less work expenditure) predicted fertility.

Mortality

High infant mortality rates although no more so than in horticulturalists (his list on page 252 includes the Batak who are dying out) and infant mortality could be the result of benign neglect as a kind of infanticide.
 
Most infant death the result of parasitic and infectious diseases. Some have suggested that sedentism increases mortality rates but Harpending and Wandsnider argue that mobile San have twice the mortality rate as sedentary San. It could be that there is more milk at settlements but LuAnn and Henry opt for traveler's diarrhea.

 
Mobility and Population Growth

Origins of agriculture.

Agriculture may increase woman's workload in two ways:
 

1. increase female workload and lead to earlier cessation of breast feeding and use of weaning foods (from Draper and Cashdan it is appears that sedentary women are busier and Hitchcock found that Nata River Bushmen nursed less and were weaned earlier than mobile Kua)
2. or female absence leads to a reduction of frequency of breast feeding (I doubt that they would be more absent and it is more likely that they would be better able to carry children around)

 Possibly that poorer diet higher in carbohydrates would increase fertility.

More realistically, evidence points to the fact that younger children can do more work and thus may (though helpers at the nest) increase a mother's fertility by sparing her work and/or increasing overall food supply.
 

Chapter 7  

---

Man the Hunter proclaims the norm of

• bilateral descent
• bilocal post marital residence
• sexual division of labor
• egalitarian political organizations
• male-female equality

Kelly finds a negative correlation (r=0.605) between effective temperature and male contribution to subsistence. Male contribution can rise to 100% but only falls to about 25%.
 
Halperin notes that Ember’s research tends to underestimate important things that women do such as prepare and store food, make clothing, and repair equipment.

Halperin argues that division of labor in band societies is unspecialized because of seasonal variation in activities and male-female complimentarity in productive activities. Kelly feels this an important qualifying point but I am not convinced.

Women do hunt regularly but the focus is on small game and many of the kills are opportunistic (kills made while doing other things).

The Agta are a strong exception because:

• they hunt regularly
• hunt even when menstruating or nursing
• have a higher success rate than men (31% compared to 17%, but the men are going after bigger game and their net rate of return is higher?
• use bows and spears

but

• hunt less than men
• travel shorter distances than men
• focus more on small game than men

Hurtado’s research on the Ache and the role (a la Brown) on child care responsibilities as the major constraint on female hunting. Actually, changes the cost benefit ratio in favor of men. Also, women will gain greater fitness benefits from investing in children.

Postmarital Residence, Descent, and Marriage
 
Postmarital Descent and Residence among hunter-gatherers

Residence

About 60% are patrilocal and 70% have bilateral descent.

 
Postmarital Residence

Ember finds gathering leads to matrilocality, fishing to patrilocality, and hunting dependence does not predict much of anything and certainly not patrilocality.

Ember’s relationship between external warfare and female contribution to subsistence and matrilocality and internal warfare and patrilocality. Model extended by Perry who argues that matrilocality among sub arctic people is a result of male long distance hunting (and absenteeism?).

Characteristics of Bilocal Groups:

• Bilocal found in unstable rainfall regimes while unilocal patterns found in stable regimes.
• Bilocal found also in depopulated groups in contact with the West supporting Service’s contention.
• Bilocal with small group size and also unstable rainfall, but could not separate the effects.

errors
seasonal variation
lack of correlation between ideal statements and actual residence
 

 According to Kelley on page 277:

Residential groups reflect the collective result of individuals’ decisions to join (or allow other to join) a foraging group. These decisions should be rooted in the foraging rate-maximization process described in Chapter 6 but they are written in a kinship idiom.
 

Residence and descent are correlated but poorly so. Service and others suggest that bilateral descent is a recent phenomenon and a shift from patrlineality as a result of contact (however, research by Cashdan and Gardener show that the shift can be to patrlineality).

Discussion of complex Australian section and sub-section system. Sections and subsections are exogamous units which prescribe which other sections you must find a particular kind of cousin to marry.

Westerners use blood to classify kin relations but others do not use blood relations to classify. While this may be true, there is a confusion between emic and etic models here.

Marriage

Service suggests that cross-cousin marriages forces marital alliances between bands because father's sister or mother’s brother live in different bands under patrilocal rules.

The degree to which groups fit their own norms is variable: among Walbiri it is 92% while Anbarra it is 17% for proper marriages.

Explanations of marriage range from

• alliance (Service, Levi-Strauss, and Tylor) for offensive and defensive purposes
• access to territories in times of economic need (Yengoyan)
• Kelly suggests than men and women have different goals and choices represent different cost-benefit decisions.

Collier suggests problems with groom working for his own family or his in-laws and classifies into brideservice, equal bride wealth, and unequal bridewealth societies.

• In brideservice societies men must earn wives through their own efforts and "people credit men (and women) with foraging their own destinies".
• In equal-birdewealth societies a man must depend on his seniors and a bride on hers to get married "people perceive themselves as being dependent on the beneficent help of seniors and supernatural beings."
• In bride wealth societies we have considerable wealth inequality and demand for a high bride price "people consider hereditary rank the most important determinant of a person's fate".

Kelly: Chapter 8  

Old view: male domination but after Man the Hunter symposium it changed to sexual equality with the following characteristics (but the situation is more complex):

• peaceful
• sexually egalitarian
• equality in wealth, opportunity, and status

Table 8.1 revised from Keeley (1988) presents contrasts of 14 characteristics of simple and complex hunter-gatherer societies.

Kelly suggests that the terms simple and complex are unfortunate and his alternative is to focus more on the origin and evolution of social inequality. [In class, point out that simple and complex are value free terms when properly used and therefore are legitimate monikers for societies in a classificatory scheme].  This seems to be a replacement term but is it possible to have complexity (say in political leadership) without inequality?

Egalitarianism

Woodburn’s immediate return societies (versus "delayed return") have the following traits:

• short time between acquisition and consumption of resources
• people have equal access to resources
• mobility is the mode of dispute settlement

Says that prestige cannot be used to gain power over others because individuals are fiercely egalitarian. This may be a reaction to attempts by those of high prestige to take advantage. Lee uses example of disparagement of kills made by good hunters.

Male-Female Egalitarianism

Hayden’s study shows no association between female status and economic contributions.

Sanday’s research suggests that when men and women spend a great deal of time away from one another (and when the environment is deemed hostile), female status is lower. This seems to contradict studies that show male absenteeism leads to greater female autonomy.

Other studies:

• Lee shows that men dominate 2/3’s of band discussions
• Speth and Speilman show that women have a poorer meat diet
• Hewlett shows that Aka women have more caries
• Leacock suggests that bias of male ethnographers and contact reduce female status (add notions of autonomy and equality)
• Bell suggests for Aboriginals that males in attempts to conform to Euro standards took their frustrations out on women who conformed more readily
• Begler suggests that we use the concept of authority instead of status because it is easier to measure and it measures control and autonomy.
• Woodburn and others see that older men have higher authority, regulate marriages and control younger men. Further, men cooperate in this regulation of marriage by retrieving stolen brides but women will not organize among themselves to prevent this.

Male ritual more highly regarded in Australia but this does not translate into oppression on a daily level? [But then, why should it since we don't have a model of how religious practice relates to "oppression".]

White argues Aboriginal husbands and wives were partners although women were the junior partners because of the large difference in age between husband and wife (20 years not uncommon). As women get older they get more power. [I would argue that this may have to do with declining mate value for women which makes them less useful or beneficial to control]

In aboriginal society we have a situation where egalitarianism prevails with female domestic autonomy and structural inequalities that favor males in ritual and domestic matters.

Yaraldi women gain status from husbands. Also, double standard in sex.

Hayden shows that when resource stress is high female status is low. He suggests that this is a population regulation mechanism because women are overworked and polluted which allows males to control female reproduction and production for the good of the group. He eschews this argument.

Sanday also finds that strong belief is sexual pollution and societies with unreliable food supplies.   What this association means is unclear.

Nonegalitarian Hunter-Gatherers

 Characteristics:

• high population densities (or high population?)
• nearly sedentary or sedentary
• occupational specialization
• perimeter defense
• resource ownership
• focal exploitation of a particular resource (commonly fish)
• large residential group size
• inherited status
• ritual feasting complexes
• standardized valuables
• food storage

Cohen asks what causes a change from humble giving to arrogant giving?
 

Keeley research indicates that stratified foragers have the following characteristics:

• sedentism
• moderate to high food storage
• high population densities (here Keeley argues that it is population pressure that is major factor that leads to stratification)

Disputes increase as density increases and standard modes of settlement (movement and family and kin mechanisms) fail can lead to three results:

1. fission
2. sequential hierarchies where nuclear families merge into higher level units (lineages) but they are likely to be short-lived.
3. vertical hierarchies where groups have leaders and groups of groups have leaders whose role is the processing of information

Riches points out that people do not acquire prestige but are given it by others.

Riches' approach has two problems:

1. Do stratified foragers really live in areas of abundant resources "Is elaborate material culture a sign of affluence or a sign that the cots of the particular social relationships signified by this material culture has increases? (p. 307).
2. Abundance can be used carelessly because it ignores the costs of production.

Hayden says that competition will occur if there is a way to transform abundant resources into highly desired scarcer goods or resources. It should occur where resources are abundant and highly invulnerable to excessive human exploitation.

The idea that there are scarce goods (culturally created) that an accumulator can acquire (through followers) that are given in feasts and then returned which followers can acquire.

First he says that the scarce resources are (1) trade goods; (2) labor to construct commemorative monuments, or: (3) hold feasts. [I could look at these as tokens that measure a group's power and productivity and organizational capacity]

Second these conditions also promote population growth [Not clear how these conditions would promote population growth.] [I would add that this is a strong measure of the military power of a group and would make them feared and induce others to join]

These are important observation:

• Demographic kick: hard work by women in food processing which increases sedentary or non-aerobic work load and lessens on demand nursing which leads to population or fertility increase.
• Sedentism leading to high levels of exploitation, lack of ability to move because niches are all filled up and storage becomes critical which leads to warfare (going after a dense and predictable resource)
• According to Hayden labor was the limiting factor since resources were available in extreme abundance for short periods individuals had to be mobilized and organized to do the job (salmon, caribou, or bison) and this required leadership.
• Overproduction was necessary because one was never sure that how lean the lean season would be plus it would put the group in an advantageous position in relation to others who might have resource problems.

Spatially heterogeneous resource variability

When resources are patchily distributed then social relationship useful for reducing variance would be adaptive. When environment becomes saturated this make exchange more problematic and Kelley suggests that brokers or middle-men become more important. Such leaders develop agendas that are hidden from and perhaps at odds with local members. And it only becomes viable when groups attempt to restrict social access (Kwakiutl would be an example).

Examines the role of middle men who have contact with modern outsiders (or pastoralists or settled horticulturalists) and their role in trade.

Spatially homogenous resource variability

The costs of admitting outsiders is great because resource distributions are similar. This leads to household extension, restriction of sharing, polygyny as a way to increase labor, and high bridewealth, and decrease in female status. (Tolowa would be an example)

An Example: The Northwest Coast

The south to north gradient: an increase in the use of fish as a basic staple to the diet. The south to north pattern:

• increase in social group size (sequential hierarchies) that own resource areas
• women as producers to women as alliance builders
• increase in stratification
• an increase in formal social alliances that are more firmly constructed and tightly controlled
• resources become more defensible because they are more localized

• Tolowa and Yurok more of a big man system with high bride wealth and polygyny possible for wealthy men
• Wakashan and Salishan groups were more sedentary, all individuals and family lines were ranked, large resource owners would allow poorer to exploit their areas so long as they returned some of the food they acquired. Lineage ownership of resources along with individual ownership. Among Kwakiutl villages were ranked according to the size of the salmon run and reliability.
• Tlingit, Tsimshian, and Haida had largest and most socially complex villages. Chiefs highly regarded, had regalia, and did no menial tasks. Slaves up to 25% of group but 7-15% more common. The potlatch was a way of establishing one's power and prestige. The requirement to repay at higher rates and if could not repay followers might leave and join another or the group could be annihilated