Nettle, D., and
Dunbar, R.I.M. (1997)
Social markers and the
evolution of reciprocal exchange
Current Anthropology,
38, 1, 93 - 99
p 93 ' Most anthropologists take the human propensity to form
groups based on cooperative exchange as a theoretical primitive. Groups of this kind do seem to be an
integral and fundamental part of human social structure'. The question is why. Natural selection
depends on genes and therefore the individual. Cooperative activity that
involves the transmission of knowledge does have survival value - but what
about reciprocation , how can it
be considered to be 'an evolutionary stable strategy' when you need to give
things away. However cooperation is only viable ' when there is relatedness or
guaranteed reciprocity between individuals'.
Kinship can secure cooperation, for unrelated individuals there has to be a high likelihood of 'meeting in the future '
,Axelrod (1884) , therefore
continuity is essential for securing cooperative social relations.
Suggestion from anthropological studies ' language serves an important index of
social allegiances, and this indexing could well be important for the
maintenance off group cohesion'.
p94 'why should so much of the surface form of language be
acquired from the environment, and why should that environment have come to be
so different the world over' Pinker and Bloom (1990) have addressed this
question. First, they explain, to represent a complete language, including all
the words, genetically might consume excessive genotypic space. Secondly, as
the language faculty must be expected to change by genetic drift, an individual
with an innate language might well fall out of step with his peers. It would
thus be advantageous to have a code with developmental flexibility. To home in
on that spoken in the group. Thirdly, as Hinton and Nowlan (1987) find, once
most of a trait is determined genetically, selective pressure to represent the
rest of the genotype declines, because learning can be relied upon to fill it. Divergence,
it is argued, arises as an accidental consequence of the genetic under
specification of language.
p 94 'Individuals do not just learn any language, they " construct their system of verbal
behaviour to resemble that common to the group or groups with which (they) wish
from time to time to be identified "LePage 1968' p 192
p 94/95 'Gaertner and Bickman (1971), Giles, Baker,
and Fielding (1975), Feldman (1968) and Harris and Bardin ( 1972) have all
shown in various contexts.... that use of a highly valued speech variety greatly
increases success in obtaining cooperation from strangers' .... ' that access
to cooperation depends of having the use of the right linguistic markers. This
socio-indexical role may be a function of language of some evolutionary
importance.......As Chambers (1995, 208,250) puts it " The fact that
linguistic variability is universal and ubiquitous suggests strongly that is
fulfilling some essential human need..... The underlying cause of
sociolinguistics differences.....is the human instinct to establish and
maintain social identity"
Simulation study
An organism's initial
position in an environment
is random and the probability of meeting another organism is random but the further apart they
are the less likely that they will meet.
Each organism has a memory span and each simulation a set time.
Task - to accumulate wealth. Giving costs 1 unit but the
receiver's wealth will increase by 2 units. ' the asymmetry precisely mirrors the fitness consequences
of exchange'
Four types of organisms. Each type has its own exchange
statelegy
• COOPs
always gives when it meets another, unless it can remember giving and not
receiving in an encounter with that particular individual. ' It this follows a
tit for tat strategy of the kind which is highly effective in organisms that
can reliably recognise each other'
• CHEATS
- free rider, never gives to anyone
• POLYGOTS
- dialects come into play,. POLYGOTS gives gifts only if the recipient has a
nearly identical dialect. When it receives a gift it changes it's dialect to
that of its benefactor. In addition it may change one of the numbers that
define its dialect probability of this occurring is it's CHANGERATE
• MIMICS.
Also a free rider but changes it's dialect to be like that of a benefactor when
it receives a gift.
Results
COOPS and CHEATS
A population of all
COPS does very well
Introducing just 5
CHEATS is disastrous irrespective of how the other parameters are set - could
only counteract by setting memory span at an unrealistic level
POLYGOTS and MIMICS
When all the organisms are POLYGOTS distinctive dialects emerge.
Organisms in the same dialect group exchange and therefore keep standardising
their dialects, whilst those in differ groups cease to exchange. As long as
minimal levels for memory span and duration are met, CHEATS cannot invade
although they dominate initially.
MIMICS a minimum of 5 (for this model) needed to invade and can
displace the POLYGOTS. However if CHANGERATE of POLYGOTS is increased then
MIMICS fail.
Discussion
The simulation is a simple system. Nevertheless ' it shows that
cooperation can evolve more easily in a simple system where social marking is
present than in one where it is not'
p 98 ' producing distinctive codes may be a way that reciprocal
exchange in large groups can be made more stable'
In f-f other systems e.g. Clothing, and other artefacts could be
adopted.
' our great skill in using and assessing language as a social
marker is an adaptive psychological mechanism tied up with the very development
of human exchange and communication'