Cavalli-Sforza Luigi Luca. Genes, Peoples and Languages
Подождите немного. Документ загружается.
society democratic principles have established
more
complex polit-
ical processes, although a small
number
of
people from
the
political
or
business world continue
to
control an important
number
of
daily
decisions.
The
hierarchical structure
of
society can help a transition
proceed, starting with
the
upper
echelons
of
power
and
descending
to
the
lower ones.
On
the
other
hand,
under
the
fourth transmission mechanism,
which we have
called."concerted" (where many people transmit
the
same cultural trait to a Single person, usually
of
the
next generation,
and
do so in
tum
with each
member
of
it),
an
innovation has a very
small chance
of
success. A Single person sympathetic to
the
change
would first
meet
resistance among
the
allies
he
needs. Unless
the
innovation is unusually useful,
or
the
proposer prestigious, success
is
unlikely.
Most cultural characteristics are transmitted by a variety
of
means
that
often conflict. This conflict
is
common; for instance it
occurs when
one
learns rules
of
behavior
at
school
that
differ from
those taught
at
home,
or
when
companions at school have very dif-
ferent opinions from school authorities
and/or
the
family.
The
con-
sequences
of
these conflicts vary Significantly among individuals
and particular cultural traits.
Examples
of
Cultural
Transmission
Cultural transmission comprises education received from both
family and school.
It
also includes all
the
habits and customs
that
are
not
explicitly
part
of
one's education. Certainly,
an
individual
acquires
these
through personal experience,
but
here
again, con-
scious
or
unconscious imitation
must
play
an
important part .
.
It
is not easy
to
distinguish
between
relative contributions.
The
similarity
between
two friends
or
between two individuals with a
more
intimate relationship, .like a husband
and
wife who have lived
together
for a very long
period
of
time, are partly
the
expreSSion
of
what
the
two have learned from each
other
and
what may have
187
attracted
them
in
the
first place. These forces are often vel)' strong,
and
we sought
to
examine
them
by surveying a
group
of
students
about similarities
between
husbands
and
wives, parents
and
chil-
dren,
and
between
frie~ds.
We asked
about
forty questions,
and
queried
the
students about themselves
and
their
parents,
as
well as
the
parents about
the
students
and
themselves.
On
average,
the
correlations (the similarities)
between
husbands and wives (the stu-
dents' parents)
were
the
greatest, followed
by
those between par-
ents
and
children,
and
finally
between
friends.
The
characteristics
studied addressed social activities, habits, leisure activities,
super-
stitions, beliefs,
and
so on.
The
most interesting result
of
this study is
that
the
highest cor-
relations
were
shown
by
characteristics
of
two
categories-religion
and
politics. Both show
the
major role played by parents,
that
is,
by
vertical cultural transmission.
In
the
first case, children resembled
their
mother
to a remarkable degree,
both
in
the
choice
of
religion
in mixed marriages,
and
in
the
frequency
of
prayer.
The
choice
of
religion is
not
surprising, since a child's religion is almost always
chosen by his parents,
or
at
least
by
one
of
them, at
an
age when
the
child cannot express his own preferences. Conversions do occur,
but
only rarely,
and
later in life.
That
a twenty-year-old should
continue
to
pray
to
God, however, does seem
to
imply
more
than
familial constraint. Unfortunately
our
data
do
not
indicate
whether
prayer continues to
be
an important activity throughout
the
life
of
children raised
to
pray.
If
the
mother's influence prevails in
the
choice
of
religion,
the
father appears
to
exercise influence only
on
the
regularity with which a religion is practiced, which is a social
rather
than
a spiritual
decision-and
even in this case,
the
mother's
influence is as great as
the
father's. Both parents
appear
to
con-
tribute equally
to
a child's political outlook.
It
is always possible
that
some
part
of
the
Similarity between
parents
and
children has a genetic basis.
The
distinction between
biological
and
cultural transmission
is
not
always an obvious one.
For
example, it was long believed
that
the
similarity between
the
IQ
of
parents
and
children was entirely genetic
in
origin.
The
famous British psycholOgist Sir Cyril Burt, undoubtedly carried
188
away by enthusiasm, even stooped so low as to publish false data to
"proven a genetic basis for IQ.
It
is thanks
to
the
American psychol-
ogist
Leon
Kamin
that
Burt's fraud was discovered.
At
the
beginning
of
work
on
IQ,
the
French
government asked
Alfred Binet
to
develop a
method
for identifying children with
mental handicaps in
order
to
provide them with special schooling
at
an early age. But
it
was primarily American psychologists who tried
to
change Binet's
IQ
scores into a measure
of
"pure
n
intelligence--
independent
of
the
culture
or
social milieu in which
the
tests
were
conducted. This misplaced enthusiasm led to several serious social
mistakes,
of
which
not
all have
been
corrected.
The
study
of
adopted
children was decisive in showing
that
cultural transmission
exerts a strong influence
on
the
determination
of
IQ. American
studies in
1980
and
1981 established
that
only one-third
of
the
vari-
ation in
IQ
among individuals was
due
to biological heredity.
Another third can
be
explained by cultural transmission, while
the
last
third
appears mostly
due
to
other
unspecified, mostly random
differences in personal life experience. This
is
a far cry from
the
80
or
90
percent
genetic contribution suggested by
Burt
and
his many
American colleagues. Similarly, Arthur jensen's statement
that
the
low
IQ
average
of
African Amelicans relative to Whites is genetic
was contradicted by studies
of
Black children adopted by Whites in
England
and
the
United States.
Theories about
the
role
IQ
plays in social stratification have
also
been
disproved. Some researchers have claimed, without real
evidence,
that
the
difference in
IQ
observed between high
and
low social classes was genetic, because people with a high I Q
automatically became
part
of
the
high
soCial
classes. A
French
study
on
adoptions again showed
that
the
difference was primarily
sociocultural and
not
genetic.
There
is probably still very widespread prejudice in America
concerning
the
low
IQ
of
Black Americans:
the
majority is likely
to
be
still convinced
that
it is
the
result
of
a real genetic difference
and
not
of
a strong social handicap
that
cannot
be
reversed in a short
time. Contrast
the
enthusiastic acceptance
of
the
book
The Bell
Curve
and
its racist message with
the
response to
the
information
189
that
the
average Japanese
IQ
is
greater
than
that
of
White
Ameri-
cans
by
11 pOints, almost as
much
as
the
average difference
between
White
Americans
and
Black Americans.
Then,
the
response was:
it
is clear
that
American
high
schools
are
very bad.
Adoption studies provide
the
best
guarantee against
the
confu-
sion
of
biological
and
cultural transmissions,
but
these studies are dif-
ficult
and
costly, largely because
there
are
so few subjects.
The
most
ambitious studies use
~dentical
twins who have
been
raised sepa-
rately.
But
these studies
are
hampered
by small samples
and
because
the
very early environments
of
twins, pairs,
and
their
rearing
is
not
always independent.
But
other
methods exist
that
help limit confu-
sion
between
cultural
and
biological inheritance.
For
example,
in
the
case
of
religiOUS
or
political similarities
between
parents
and
chil-
dren,
we
used
published research data comparing identical twins,
fraternal twins,
and
regular siblings. Fraternal twins should
not
resemble
each
other
more
than
regular siblings
if
biolOgical heredity
were
the
only important factor.
In
the
case
of
religiOUS
or
political
creeds,
the
similarity between fraternal pairs was almost
the
same as
for identical twins, indicating
that
genetics plays
no
(or only a vel)'
small part)
in
this trait. Family background does have a major effect.
The
purely
or
predominantly maternal transmission
of
some
religiOUS
characteristics would
be
difficult
to
explain in a strictly biological
way.
Maternal transmission exists for biolOgical characteristics deter-
mined
by
the
mitochondrial genome. However,
it
would
be
very sur-
prising
if
tl,ese cytoplasmic organelles, which supply
the
cell's energy,
had
any effect
on
individual
religiOUS
beliefs.
We
can
study
cultural transmission directly, instead
of
taking
the
indirect approach
of
twin studies.
In
this way
we
avoid confusing
biological
inheritance
with
other
mechanisms.
We
can
question
people
directly for certain characteristics,
and
the
depth
of
memory
revealed
in
subjects is
often
surprising.
In
collaboration,
the
anthropologist Barry
Hewlett
and
I asked African Pygmies from
whom
they
learned
certain basic knowledge essential
to
their
life:
information
about
hunting, gathering,
preparing
food, building
houses,
and
so
on.
They
perfectly
remembered
learning
these
things
and
often
they
even
remembered
the
time
and
place
where
190
they
were taught particular skills.
The
information collected could
be
verified by questioning also
the
teachers
of
this knowledge.
From
80
to
90
percent
of
the
time, parents
were
responsible for
the
transmission. Because some
of
the
skills
were
known
to
only
one
sex, teaching usually occurred via
the
parent
of
the
same sex. Only
for important communal activities like dancing, singing, rules for
dividing food;
and
other
characteristics
of
Pygmy society did mem-
bers
of
the
wider community participate with
the
parents in educat-
ing children.
The
contribution
of
other
African villagers with whom
Pygmies associate during parts
of
the
year is restricted mostly to
farming, which Pygmies do in a very limited
way,
having
been
exclusively hunter-gatherers until recently.
The
Pygmies have also
learned
how
to
make
and
use certain hunting weapons, like
the
crosshow, from villagers. This knowledge has diffused very quickly
among
the
Pygmies.
Our
notes show
that
several Pygmies learned
how to make crossbows directly from a villagel; although
we
also
observed
one
instance
of
a Pygmy father teaching his son. A tradi-
tional society like
that
of
the
African Pygmies
that
lacks
leaders
and
schools,
and
is organized in very small social groups, tends to
remain
independent
from nearby villagers, even when those vil-
lagers try to establish control over
the
Pygmies. Therefore, cultural
transmission tends
to
be
vertical within
both
groups, with very lim-
ited horizontal exchange between them. Vertical transmission and
the
social pressure exerted by
the
group's members
tend
to
make
Pygmy society very conservative. African farmers,
on
the
other
hand, have more contact with
the
exterior, missionaries, for
instance,
and
have radios
and
schools.
C.ritical
(Receptive/Susceptible)
Periods
and
Their
Importance
Most culturally
determined
characteristics are more easily changed
than genetic ones. Even for clear-cut genetic diseases, onset can
occur very late in life, with
much
variation from
one
individual to
another. Huntington's chorea can strike individuals ranging from
191
two to eighty years old, although: most cases. manifest themselves
around
the
age
of
forty.
But
the
pattern
of
inheritance is very strict.
Some genetic diseases disappear with age, as do certain types
of
allergy.
But
in general, genetically
detennined
characteristics
are
rather
stable
and
rarely reversible.
The
same is
not
true
of
cultural
characteristics.
We
have already observed
that
religious conver-
sions take place. Political affiliations also change with appreciable
frequency.
Nevertheless,
some
cultural traits change less readily
than
oth-
ers. Stability
of
celtairl behaviors may
be
favored by biological fac-
tors
that
render
changes
more
or
less likely
at
certain ages.
In
other
words,
there
may
be
sensitive
or
critical periods in life,
the
phe-
nomenon
sometimes called "imprinting.»
The
most obvious critical period, although inadequately stud-
ied, is undoubtedly
the
one
governing
our
ability
to
learn a first
and
a second language.
The
first language must
be
acquired in
the
first
years
of
life.
One
can learn
other
languages after
the
first,
but
rarely,
if
ever,
as
well;
it
is particularly difficult to learn
proper
pro-
nunciation
of
a foreign language after puberty.
The
time before
puberty
is·
also a sensitive
period
for acquiring
the
incest taboo.
The
psychologist
Edward
Westermarck has sug-
gested
that
the
cohabitation
of
brothers and sisters before
puberty
could diminish sexual interest
and
explain why incest
between
brothers and sisters is
as
rare for humans
as
it is for
other
mammals.
There
have
been
notable exceptions in some ancient dynasties such
as
those in Egypt
and
Persia,
where
the
marriage
of
Siblings was
encouraged,
but
this custom quickly-disappeared.
In
some
commu-
nities, especially in
the
Middle East
and
India, marriages
between
close relatives (e.g., uncle
and
niece, first cousins) are still frequent,
but
this is a different phenomenon.
Westermarck's hypothesis was tested by Arthur Wolf
(1980) in
Taiwan,
where
marriages have occurred
between
a boy
and
an
adopted
sister
of
similar age.
The
daughter would
be
adopted
by
the
boy's parents after his birth.
In
a society where spouses
are
bought, adoption
at
a very young age guaranteed a lower buying
price. This custom also afforded
the
mother
the
opportunity to
192
instruct
the
future daughter-in-law ill
the
art
of
serving he.r hus-
band. Wolf showed
that
these marriages
were
less successful than
others;
they
ended
more often
in
divorce
and
produced fewer chil-
dren
on
average. This result is consistent with
data
from Israeli kib-
butzim where children are raised together in a
SOlt
of
communal
nurselY,
and
have little contact with
their
parents. These children
essentially have a very large family
of
adoptive brothers
and
sisters,
and
very few marriages occur
between
children from
the
same kib-
butz.
It
is
harder
to fall in love with someone whom you are used to
seeing on
the
potty.
TIlere are celtainly many
other
Clitical peIiOds
in
the
fonnation
of
human societies, about which
we
don't currently lmow much. Even
those
tllat I've just cited have not been studied in sufficient detail. I
could mention just
two
other
areas deserving further inquiry. Gianna
Zei,
Paola Astolfi, and Suresh Jayakar have shown that daughters
of
an older father
tend
to
marry
men
considerably older tllan them-
selves.
TIm
may
be
palt
of
a more general phenomenon, which
deselve5 to be investigated in great detail: it
is
likely that we llave a
tendency to choose spouses who share some physical resemblance
(and
perllaps behavior) with
our
opposite-sex parents. This phenom-
enon could explain tlle pronounced physical resemblance obselVed
among individuals
of
the same social
group-especially
obvious in
the
small and isolated ones.
The
same phenomenon broadens differ-
ences between groups.
In
another investigation aided by psychologists, we studied
the
propensity
of
our
Stanford students to identify with a pruticulru·
region
or
physical habitat.
Our
preference for mountains, plains,
seashores, lakes, big cities,
or
small towns is probably set at
an
early
age. I
became
interested in this when I realized
that
I
had
no partic-
ular preference.
The
desert, countryside,
or
city
were
all
the
same
to
me, so long
as
the
humidity was not too high. I thought this might
be
due
to
my parents' frequent changes
of
residence before I was
four years old.
In
AmeIica
the
importance·
of
the
environment in
which
one
lives can
be
seen
in
the
frequency with which immi-
grants establish themselves
in
areas
that
resemble those they have
left.
Our
studies
of
Stanford students confirmed
that
those who
193
moved frequently during childhood
had
trouble identifying with a
. particular environment and adapted more easily to all environments.
Our
data did not allow us to determine
the
most sensitive age,
but
the
study did succeed h, showing that a nomadic tendency can
be
culturally inherited,
and
that a psychological imprint received while
young
is
difficult to erase later in life. Governments
or
countries
with large nomadic populations (e.g., Gypsies, Bedouins, Berbers,
Tuaregs, Pygmies) have difficulty changing their nomadic habits.
This poses serious problems for the schooling
of
their children.
Moreover,
the
freedom
of
nomadism
is
fascinating,
and
if
that is
the
way in which
one
was raised,
it
must
be
very difficult
to
settle down.
Linguistic
Evolution
as
an
Example
of
Cultural
Evolution
Amazingly, linguistic evolution has
not
been
studied much. There is
great potential for rigorous quantitative analysis, and research
is
not
very expensive. Interest in language evolution began in the second
half
of
the
nineteenth century,
by
applying
the
methods
of
evolution-
ary trees
to
the
history
of
language differentiation-especially
the
Indo-European languages. I have already mentioned that August
Schleicher constructed a tree
of
this family
tha,t
resembled one based
on a recent study. Even though the phenomenon
of
borrowing words
from
other
languages, especially from neighbors,
is
well established,
the most studied evolutionary trees give the impression that a lan-
guage changes in ways that are largely independent
of
changes taking
place in
other
languages. This
is
a prerequisite for the applicability
of
tree analysis. We know that languages are often spread over large
areas with different varieties (or
dialects) developing
locally.
We are
aware
that
a language changes slightly even within an individual life-
time,
but
knowledge
of
old languages
is
limited, making variation in
time somewhat less easy to study than variation in space. Neverthe-
less, variation in time almost automatically implies variation
in
space,
and
the basic rules are the same for both.
194
What
precisely is linguistic variation?
There
are
several aspects.
Phonological variation is easily perceived.
In
any
European
country,
and
even
in
the
United
States, significant differences
in
accent occur
between
north
and
south, east
and
west. With a little experience,
we
can
.easily guess someone's native region.
The
pronunciation
of
words changes
in
time
and space----often significantly.
Another
aspect
of
phonological variation is
the
richness
or
poverty
of
sounds
in
differ~nt
languages. Polynesian languages have
among
the
fewest sounds.
They
have only
three
vowels: a, i,
and
u.
English
is
at
the
opposite extreme with twenty
or
so vowel sounds
(including diphthongs), which
are
different from those found
in
all
other
languages, making
the
acquisition
of
English very difficult for
foreigners.
The
speed
with which vowel sounds change is particu-
larly astonishing. To paraphrase Voltaire:
if
the
consonants
are
not
very useful for etymolOgical reconstruction,
the
vowels
are
com-
pletely useless.
Semantic variation is
the
change
of
word meanings.
For
exam-
ple,
the
French
word femme has acquired a second meaning, to
include
both
"woman"
and
"wife."
In
Italian,
the
word
donna (origi-
nally
denved
from
the
Latin
word
domina meaning "household mas-
ter") means "woman,"
but
Italians use moglie (from
another
Latin
word,
mulier) for "wife." Italian also uses femmina (from
the
same
root
as femme
in
French)
to
mean only "woman"
and
not
"wife."
Although
grammar
is
the
most stable
part
of
a language,
it
too
can
change with time.
In
English, as
in
French
and
Italian,
the
nor-
mal
word
order
in
a
sentence
is subject (S), verb
(V),
and
object
(0
)--<>r SVO.
But
all eight possible
word
orders exist
in
various lan-
guages, even
if
SVO
and
SOy
languages
are
much
more
common.
The
rarest
are
OVS
and
Osv.
In
the
film
Return
of
the
Jedi, Yoda,
the
master
of
the
Jedi, uses
the
OSV style: "Your father
he
is."
For
each
of
these
three
modes
of
linguistic variation (phonologi-
cal, semantic,
and
grammatical), change in space is more obvious
and
easier to study
than
change in time. We can illustrate
on
a geographic
map
the
variation a word experiences by drawing a curve delimiting
the
areas
where
the word has a particular pronunciation. This curve,
195
which separates
one
homogeneous region from another, is called all
"isogloss." By tracing
the
isoglosses
of
many words, we notice that
most words display a unique pattern:
the
limits
of
pronunciation
eliffer for every word.
Where, then, is
the
region where a single and
eliscrete language
or
elialect is spoken?
The
representation
of
lan-
guages with a tree-like monolith, where languages are differentiated
from each
other
in a totally regular malmer without being influenced
by
other languages, is ollly
an
approximation.
Five years after Schleicher's work,
in
1872,
one
of
his students,
Johannes Schmidt, emphasized
the
importance
of
local linguis-
tic variation
and
proposed a theory
that
in some ways opposed
Schleicher's. According to Schmidt, each new fonn
of
a word spreads
over a geographic area like
the
waves spreading from a rock thrown
into a pond, influencing neighbOring speakers to various degrees.
This
metaphor
is particularly apt;
it
sets itself apart from
the
model
of
a tree, which presents completely isolated languages.
Can
these
two views
be
reconciled?
TIleories
of
biolOgical variation in space, developed in
the
mid-
dle
of
the
twentieth century by several different mathematicians,
resulted in very
sinlilar models. They have
the
generic
name
of
"iso-
lation
by distance"
and
show
that
genes vary randomly in geographiC
space, follOwing exact rules derived from statistics and probability.
TIle most
Significant regularity is
the
relation between genetic dis-
tance (calculated from averaging a
number
of
genes)
and
geographic
distance. We have seen
that
genetic distance increases regularly (but
always more slowly)
as
geographic distance increases, until it reaches
a maximum.
The
shape
of
the
theoretical and empirical
CUlves
is
determined by two measurable variables:
the
mutation rate, which
increases genetiC differences between two places, and
the
rate
of
genetiC exchange between neighbors
due
to
migration, which tends
to increase genetic similarity between
them-so
these forces are
opposed, to a certain extent,
and
balance each other.
The
same mathematical theory can
be
applied
to
linguistiC evo-
lution:
the
equivalent
of
mutation (which produces
new
forms
of
genes,
or
alleles) is innovation, which in linguistics is
the
generation
of
new sound, meaning,
or
grammar. Migration propagates
these
196