ARTICLES
Hunter-Gatherers and Human Evolution
FRANK W. MARLOWE
Holocene foragers may indeed have
ﬁlled a new niche in many places, but
by analyzing how climate, ﬂora, and
fauna inﬂuence ethnographic foragers
we can gain insights into how Pleistocene
foragers should have differed
from them. Contact with agriculturalists
poses a greater challenge. When
more powerful societies that practice
horticulture, pastoralism, or intensive
agriculture (all referred to here as agriculturalists)
come in contact with
foragers, they can rapidly alter forager
subsistence and culture. The signiﬁcance
of such contact, however, varies
considerably. Many Pygmies, such as
the Mbuti, speak only the Bantu language
of their farmer neighbors for
whom they do some labor in exchange
for food. On the other hand, the
Hadza have had contact with agropastoralists
for a long time but maintain
their language. Their interaction has
been limited mainly to the trade of
meat and honey for iron and tobacco,
and has altered the Hadza surprisingly
little over the past century.7 All
we can do is perhaps give extra weight
to those with less contact. We might
also give extra weight to those in
richer habitats, though the bias toward
marginal habitats is not as great
as usually assumed.
The greatest obstacle to using foragers
as analogs of our ancient ancestors
is that virtually all foragers in the ethnographic
record have complex technology
compared to premodern hominins.
Human niches are deﬁned to a
large extent by technology, even
among foragers. There are horsemounted,
bow-and-arrow hunters of
bison, harpoon hunters of walrus who
travel in kayaks, salmon weir-ﬁshers,
and spear, blowgun, and net hunters.
Technology has allowed human foragers
to occupy the full spectrum of terrestrial
habitats, producing a wide
range of cultural variation. Our ancestors
must have varied considerably as
well, so it seems pointless to reconstruct
a single type of society. However,
because we think the ﬁrst modern
humans arose from a particular
population living in Africa 160,000 to
200,000 years ago,8 there would have
been a limited range of variation in
that particular population at that particular
time. As modern humans
spread to diverse habitats and developed
diverse technologies, socio-cultural
variation increased. Because the
archeological record, with a few exceptions
like Tasmania,9 reveals a
trend toward increasing efﬁciency in
technological evolution, we can assume
that there was less efﬁciency (in
extracting resources, for example) as
we go further back in time. It is
mainly the effects of increased productivity
associated with technology
that we need to subtract from ethnographic
foragers when extrapolating
to much earlier times.
Ironically, while foragers may be
problematic analogs of humans in the
past, they are certainly the most useful
exemplars of humans in the present. To
know which of our traits are derived
and therefore require a separate explanation
from traits we share with Pan,
we must ﬁrst characterize extant species.
When we compare humans with
other species with respect to traits like
diet, group size, home range, mating
system, or mortality rates, we need to
measure these traits in foragers, not agricultural
populations, if we are to understand
the relevant selective forces
that shaped modern humans. Here I
provide an overview of foragers to put
humans in primate perspective and to
examine some basic relationships between
social organization, and habitat,
and technology so that their effects can
be accounted for when modeling hominin
behavioral evolution.
THE FORAGER SAMPLE
The sample I use here is, to my
knowledge, the largest forager dataset
that exists (Fig. 1). I borrow heavily
Frank W. Marlowe is Associate Professor
of Anthropology at Harvard University. His
research focuses on the behavioral ecology
of mating systems and cooperation.
For the past decade, he has worked with
Hadza hunter-gatherers of Tanzania. Email:
fmarlowe@fas.harvard.edu.
Key words: foragers; Hadza; hunter-gatherers;
human behavioral ecology
© 2005 Wiley-Liss, Inc.
DOI 10.1002/evan.20046
Published online in Wiley InterScience
(www.interscience.wiley.com).
Although few hunter-gatherers or foragers exist today, they are well documented
in the ethnographic record. Anthropologists have been eager to study them since
they assumed foragers represented a lifestyle that existed everywhere before
10,000 years ago and characterized our ancestors into some ill-deﬁned but remote
past. In the past few decades, that assumption has been challenged on several
grounds. Ethnographically described foragers may be a biased sample that only
continued to exist because they occupied marginal habitats less coveted by
agricultural people.3 In addition, many foragers have been greatly inﬂuenced by
their association with more powerful agricultural societies.4 It has even been
suggested that Holocene foragers represent a new niche that appeared only with
the climatic changes and faunal depletion at the end of the last major glaciation.5
Despite these issues, the ethnographic record of foragers provides the only direct
observations of human behavior in the absence of agriculture, and as such is
invaluable for testing hypotheses about human behavioral evolution.6
Evolutionary Anthropology 14:54–67 (2005)
from previous researchers, especially
Lewis Binford’s book Frames of Reference.1
Many other data come from tables
in The Foraging Spectrum by Robert
Kelly.10 Both Kelly and Binford
relied heavily on George Peter Murdock’s
Ethnographic Atlas,11 and I
have added more from the Atlas and
the Standard Cross-Cultural Sample
(SCCS) in their updated, electronic
form on the World Cultures CD.2 I
have also taken data from many other
original sources (see Appendix). There
are 478 societies altogether, with almost
all deriving less than 10% of
their diet from domesticated foods.
On several important variables (Box
1), I report descriptive statistics in Tables
1 and 2.
Comparative analyses face the
problem of non-independent data
(Galton’s Problem) whether the units
are species, cultures, or individuals.12
When the phylogenetic relationships
between species are known, they can
be used to avoid Type 1 or Type 2
errors. Across human societies it is
less clear whether genetic or linguistic
relatedness would be preferable for
constructing phylogenies because of
genetic mixing and language replacement.
The SCCS was assembled to
avoid Galton’s Problem,13 but consists
of only 186 societies, of which about
one-ﬁfth are foragers. The results I
present on kinship, marital residence,
and the mating system are mostly
from my previous analyses of the
SCCS. Because here I am more concerned
with the full range of variation
and central tendencies, I use the complete
forager sample even though
there is no satisfactory phylogeny yet.
Because I present some correlations, I
have broken the sample into Old
World versus New World. If relationships
found in the whole sample also
hold within each hemisphere, we can
have more conﬁdence that the relationship
is real.
In general, evolutionary theory predicts
that individual behavior will be
adaptive and track ecological conditions,
subject to constraints. Populations
should also track local conditions
since groups are seen as
collections of ﬁtness-maximizing individuals.
This assumption should usually
hold despite the fact that individuals
are so often in conﬂict with each
other. Two individuals might compete
for scarce foods but both go about
acquiring them with the same “most
efﬁcient” foraging strategy in a given
environment. Thus, to the extent that
certain aspects of social organization
follow from foraging strategies, which
in turn track ecology and technology,
we should be able to explain relationships
between habitat variation and
social organization.
To illustrate how ecology and technology
shape social organization, consider
the equestrian foragers. Several
North American Indian tribes became
specialists in hunting bison from
horseback during the 1700s, after the
Spaniards introduced the horse.14 In
Box 1. Glossary
Absolute latitude—absolute value of latitude; that is, distance
from the equator whether north (designated with positive
numbers) or south (designated with negative numbers).
Primary biomass—the amount of plant biomass (kg/m2
),
using Binford’s equation.1
Effective temperature (ET)—18W–10C/(W Ϫ C) ϩ 8, where
W ϭ mean temperature of the warmest month and C ϭ mean
temperature of the coldest month.
Hunting, gathering, ﬁshing—The percent of foods acquired
by each foraging activity (in some cases these are actual weight
or calorie measurements, but in most cases only estimates of the
ethnographer).
Ethno-linguistic group—the set of people sometimes referred
to as a tribe who share a language.
Ethno-linguistic population—the population of the ethno-linguistic
group.
Ethno-linguistic area—the total area occupied by the ethno-linguistic
group.
Population density—ethno-linguistic population/total area
of ethno-linguistic group.
Local group population—the mean population of the residential
band (the people who regularly reside in the same camp or
settlement). When maximums and minimums for a society were
cited, these were averaged.
Number of moves—the number of times per year camp location is
moved.
Male/female contribution to diet—the percent of the diet
acquired by females or males. (In some cases these are actual
weight or calorie measurements, but in others the values are
calculated from the percent contributed by each sex in hunting,
gathering, and ﬁshing.
Polygyny rate—the percent of married men or married women
that are in a polygynous marriage. This is not the same as the
percent of all men and women; for example, 100% of married men
could be polygynous even with an operational sex ratio of 1, since
many other men can be bachelors.
Local group area (home range)—by dividing the ethno-linguistic
population by the local group population we can derive the number of
local groups. We can derive the minimum area used by a local group
if we divide the total area of the ethno-linguistic group by the number
of local groups. This is a minimum estimate because it assumes
nonoverlapping areas, whereas local groups tend to move into areas
that overlap with those of other local groups. Even though females
may have shorter day ranges than males, the main factor determining
area covered in a year is moving the camp to a different area, not
forays out from camp.
Total fertility rate (TFR)—the mean number of children a
woman bears over her reproductive career.
Infant mortality rate—the percent of children who die
within the ﬁrst year of life.
Juvenile mortality rate—the percent of children who die
within the ﬁrst 15 years of life.
Interbirth interval (IBI)—the mean number of months between
births regardless of survivorship.
Weaning age—years of age when nursing completely ends.
Age at ﬁrst reproduction—mean age when females ﬁrst give
birth.
ARTICLES Hunter-Gatherers 55
the sample here, equestrian foragers
have much larger home ranges and signiﬁcantly
larger local groups than do
nonequestrian foragers. The horse increased
hunting success rates and lowered
travel costs, allowing the Plains Indians
to specialize in hunting the highyielding
bison and follow them over
vast distances in large groups.15 When
reconstructing periods before horse domestication
with data from ethnographic
foragers, we can ignore the
equestrian foragers. If we are interested
in the period before 30,000 years ago,
we might exclude the arctic foragers because
it was only during the last 30,000
years that very cold areas were occupied
by modern sapiens.16 Like the
equestrian foragers, arctic foragers
have larger home ranges than do others,
but in their case it is due to the low
density of foods, which requires covering
large areas.
When we are interested in how human
foragers respond to different environments,
the full sample is most
useful because it reveals how factors
like very cold climates or horses alter
traits like diet, group size, and home
range. When we are interested in estimating
parameters for modeling earlier
periods, on the other hand, we
need a more relevant sample. For this
reason, I created a subsample by excluding
those foragers who hunt from
horseback and those where effective
temperature (ET) is Ͻ13°C, which
eliminates the arctic foragers (Table
1). I use the warm-climate, nonequestrian
subsample for exploring traits
during earlier periods, but results refer
to the whole sample unless otherwise
stated.
HABITAT VARIATION AND DIET
Primary biomass (kg/m2
) is a measure
of the productivity of a habitat
calculated from climate data.1 The
plant matter it measures provides
food for animals and so limits their
numbers (secondary biomass). It excludes
animal biomass and includes
inedible plant parts, so it is not an
ideal measure of food abundance for
humans. For example, rainforests
contain less human food per kilogram
of primary biomass than savannas do,
but because there are so many more
kg/m2
of primary biomass in rainforests
there is usually more total food
energy present.10 Therefore, even
though primary biomass is not ideal,
it does provide a rough estimate of
habitat quality, and since it has been
calculated by Binford1 for most of the
foragers, I use his data. In the sample
here, primary biomass peaks around
latitude 15° North and approaches
zero at the poles (Fig. 2). The northern
and southern hemispheres are not
mirror images since there is more
land mass in the northern hemisphere.
Mainly as a consequence of
temperature effects on primary biomass,
as Lee,17 Kelly,10 and Binford1
have previously reported, gathering
falls off in colder temperatures at
higher absolute latitudes, where there
are fewer plants, whereas hunting and
ﬁshing increase (Fig. 3). Males, therefore,
contribute more to the diet at
higher absolute latitudes (r ϭ .480,
P Ͻ .0005, n ϭ 158). Male contribution
to diet is also higher where ﬁshing
is more important (r ϭ .510, P Ͻ
.0005, n ϭ 155).
The “marginal habitat criticism”
noted earlier assumes that areas ideal
for agriculture would have been ideal
for foraging and that areas bad for
farming would have also been bad for
foraging. This is not necessarily the
case. Some areas unsuitable for planting
can be quite good for foraging.
The map of foragers (Fig. 1) shows
that the main bias in the sample is a
geographic one due to history. The absence
of foragers in the circum-Mediterranean
and most of Eurasia is not
because that huge area has rich arable
soil and other continents do not, but
because complex state societies arose
earlier there and had incorporated or
replaced all the foragers before ethnographies
were written. In addition,
a large fraction of the foragers in marginal
(low primary biomass) habitats
are those in Central Australia, who
were not pushed there by agriculturalists
since there were no agriculturalists
on the continent prior to about
1800. The habitats with the lowest primary
biomass are those of the arctic
foragers. When the cold-climate foragers
are excluded, a comparison reveals
that forager habitats are not less
productive than those of agriculturalists
(Box 2).18
Population density increases with
primary biomass, as expected from
basic ecological theory, but it levels
off at a primary biomass of about
30 kg/m2
(Fig. 4). The high-density
foragers at primary biomass about 35
kg/m2
are mostly the foragers of the
northwest coast of North America.
These complex foragers occupy habitats
where salmon return to their
freshwater spawning areas. Their
Figure 1. Geographic location of the total forager sample (n ϭ 478).
56 Marlowe ARTICLES
rich supplies of ﬁsh, which these foragers
preserve and consume year
round, allow them to live in small territories
in sedentary, socially stratiﬁed
groups, often with frequent warfare
and even slavery.19 It is not habitat
richness per se that explains the
special qualities of the complex foragers
but the seasonally abundant
anadromous ﬁsh, which promote storage
and investment in time-consuming
technologies, such as weirs and
smoke houses, with delayed returns.20
Complex foragers may not have been
rare toward the end of the Pleistocene,
but for modeling earlier periods we
should exclude them, and most are
absent from the warm-climate sam-
ple.
GROUP SIZE, HOME RANGE,
AND MOBILITY
Among foragers, there are typically
three types of groups, the ethno-linguistic
group (tribe), which may never
assemble in one place; the residential
or local group (camp or band); and
the daily foraging party. Most foragers
have several types of ﬁssion-fusion.
First, every day the local group splits
into smaller parties to forage (while
some may remain in camp), then returns
to camp to sleep. Foraging parties
may also ﬁssion or fuse. One
camp (local group) may fuse with
other camps and, two months later,
ﬁssion off again. Individuals also
move back and forth between camps.
There is also residential mobility of a
different sort when the whole local
group moves the location of the camp
to access new resource patches. The
vast majority of foragers are quite mobile,
especially in the warm-climate
sample, where the median number of
moves per year is 7 (Table 1).
Dunbar argues that the number of
individuals one interacts with on a regular
basis is limited by the ability to
keep track of social interactions.21 On
TABLE 1. Traits for the Total Sample and the Warm-Climate, Nonequestrian Subsample, by Hemispherea
Region Statistic
Primary
Biomass
(kg/m2
)
% Diet
Gathering
% Diet
Hunting
% Diet
Fishing
% Male
Contribution
to Diet
Ethno-
linguistic
Pop.
Ethno-
linguistic
Area
(km2
)
Pop.
Density
(Persons
per
km2
)
Local
Group
Pop.
Local
Group
Area
(km2
)
Number
of
Moves
per
Year
Total Sample of Foragers
Old W. Mean 15.17 50.64* 25.1009 23.72 51.16 1393 24146 .23 32.10 592.35 8.87*
N 105 114 114 114 35 160 105 105 91 82 105
SD 16.17 23.00 13.95 25.86 18.78 2035 51889 .24 31.94 1079.78 9.15
Min .12 .00 .00 .00 25.00 4 45.50 .00 13.10 21.76 .00
Max 59.27 90.30 65.00 90.00 100.00 11800 386880 1.23 250.00 5516.32 45.00
Median 9.86 55.00 25.00 10.00 50.00 619 6880 .16 25.00 162.50 8.00
New W. Mean 11.35 29.64 36.29* 33.52* 64.81* 1991* 43966* .26 55.37* 2055.91* 7.00
N 236 287 287 287 123 236 235 235 203 177 235
SD 11.99 22.19 19.93 26.59 17.96 2594 89051 .40 51.08 3060.26 9.42
Min .02 .00 5.00 .00 20.00 23 270 .00 14.50 19.00 .00
Max 46.25 80.00 90.00 95.00 100.00 14582 660000 3.09 275.00 21079.00 58.00
Median 6.08 30.00 30.00 30.00 63.00 1163 12200 .09 35.00 708.00 4.00
Total Mean 12.52 35.61 33.11 30.74 61.79 1749 37845 .25 48.17 1592.54 7.58
N 341 401 401 401 158 396 340 340 294 259 340
SD 13.51 24.27 19.09 26.72 18.96 2399 79900 .36 47.20 2687.00 9.36
Min .02 .00 .00 .00 20.00 4 46 .00 13.10 19.00 .00
Max 59.27 90.30 90.00 95.00 100.00 14582 660000 3.09 275.00 21079.00 58.00
Median 7.08 35.00 30.00 25.00 60.00 895 9050 .11 29.50 426.28 5.00
Warm-Climate, Nonequestrian Foragers
Old W. Mean 15.59 54.78* 24.23 20.78 46.83 991 18476 .24 30.52 473.13 8.86
N 96 98 98 98 29 97 96 96 77 73 96
SD 16.54 20.12 13.23 24.69 15.73 1526 34364 .24 31.30 838.55 9.40
Min .12 .00 .00 .00 25.00 35 46 .00 13.10 21.76 .00
Max 59.27 90.30 55.00 90.00 80.00 11800 230000 1.23 250.00 4500.00 45.00
Median 10.14 55.00 25.00 7.50 45.00 528 5500 .17 24.85 152.50 8.00
New W. Mean 13.28 49.01 26.84 23.78 59.90* 1451 7185 .40* 47.53* 419.82 8.75
N 79 81 81 81 32 78 78 78 53 52 78
SD 14.04 14.19 13.31 19.97 13.98 1571 8968 .49 45.04 414.74 12.78
Min .02 10.00 5.00 .00 33.33 23 310 .01 14.50 46.72 .00
Max 46.25 76.00 62.00 70.00 88.89 6500 40500 3.09 250.00 1830.00 58.00
Median 5.65 50.00 25.00 20.00 57.50 875 3185 .23 32.00 278.76 5.50
Total Mean 14.55 52.17ϩ 25.41Ϫ 22.14Ϫ 53.68Ϫ 1196Ϫ 13414Ϫ .31ϩ 37.46Ϫ 450.96Ϫ 8.81
N 175 179 179 179 61 175 174 174 130 125 174
SD 15.46 17.87 13.29 22.66 16.12 1558 26758 .38 38.28 692.63 11.01
Min .02 .00 .00 .00 25.00 23 46 .00 13.10 21.76 .00
Max 59.27 90.30 62.00 90.00 88.89 11800 230000 3.09 250.00 4500.00 58.00
Median 9.86 55.00 25.00 15.00 52.50 565 3905 .18 25.58 174.90 7.00
a
Signiﬁcantly higher mean values in one hemisphere are designated by (*). In the row for the warm-climate sample totals, (ϩ)
indicates a signiﬁcantly higher, and (Ϫ) a signiﬁcantly lower mean value for the warm-climate sample than the cold-climate
sample.
ARTICLES Hunter-Gatherers 57
the basis of a correlation between group
size and the size of the neocortex relative
to the rest of the brain across anthropoid
primates, he argues the group
size of humans over some evolutionarily
signiﬁcant period must have been
about 150.21 This is certainly at odds
with the size of the most salient group
among foragers, the local group, which
has a median population of 30 (Table
1). Could local groups have been closer
to 150 if foragers occupied richer habitats
in the past? The data do not support
this, since there is no correlation
between local group size and primary
biomass. In richer habitats, where concentrated
foods can support larger
groups, we see a higher population density
since the area occupied by the local
group shrinks to minimize travel energy
expenditure, but there is no increase
in local group size. Across the
full range of primary biomass, local
groups show a conspicuous tendency
toward a population of 30 (Fig. 5).
Hadza camps are at their smallest
during the rainy season and their largest
at the end of the dry season. This is
because there are only so many permanent
waterholes, which limits the
number of possible camp locations.
When rain removes this constraint,
the default is to disperse into smaller
groups in which the Hadza say there
is less bickering. This suggests a possible
explanation for the remarkable
consistency in local group population
across forager habitats. Free-rider
problems probably set an upper limit
TABLE 2. Demographic Traits for the Total Forager Samplea
Region Statistic
% Infant
Mortality
% Juvenile
Mortality
Total
Fertility
Rate
Female Age at
1st
Reproduction
(Years)
Age at
Weaning
(Years)
Interbirth
Interval
(Years)
% Married
Males
Polygynous
% Married
Females with
Co-Wife
Total Sample of Foragers
Old World Mean 24.83 41.36 4.96 18.34 3.20* 3.21 17.64 20.16
N 12 12 29 6 15 9 76 26
SD 9.89 11.79 1.50 1.66 .82 .82 18.87 28.11
Minimum 10.30 20.00 .81 15.90 2.00 1.75 .00 .00
Maximum 46.00 56.40 7.00 19.96 4.50 4.35 70.00 90.00
Median 21.50 45.50 5.30 18.63 3.00 3.40 8.50 5.95
New
World
Mean 18.15 47.83 5.97 19.17 2.32 3.39 12.09 22.68
N 4 7 18 3 25 7 136 25
SD 5.22 9.62 1.84 1.53 .68 .66 10.18 19.21
Minimum 11.60 35.00 2.80 17.50 1.00 3.00 .00 2.00
Maximum 24.00 61.00 8.50 20.50 4.00 4.83 57.00 60.00
Median 18.50 45.00 6.20 19.50 2.50 3.08 10.00 18.00
Total Mean 23.16 43.74 5.35 18.61 2.65 3.29 14.08 21.40
N 16 19 47 9 40 16 212 51
SD 9.28 11.23 1.69 1.57 .84 .74 14.14 23.95
Minimum 10.30 20.00 .81 15.90 1.00 1.75 .00 .00
Maximum 46.00 61.00 8.50 20.50 4.50 4.83 70.00 90.00
Median 20.60 45.00 5.50 19.25 2.50 3.24 10.00 10.00
Warm-Climate, Nonequestrian Foragers
Old World Mean 24.35 40.48 4.87 18.15 3.13 3.08 19.07 21.56
N 11 11 23 5 12 7 68 22
SD 10.23 11.95 1.54 1.79 .71 .81 19.22 29.47
Minimum 10.30 20.00 .81 15.90 2.00 1.75 .00 .00
Maximum 46.00 56.40 7.00 19.96 4.50 4.00 70.00 90.00
Median 21.00 45.00 5.25 18.00 3.00 3.40 11.00 6.50
New
World
Mean 11.6000 48.33 6.38 19.50 2.40 3.04 10.22 29.89
N 1 3 5 1 6 2 45 9
SD . 13.01 2.29 . .81 .06 8.91 20.01
Minimum 11.60 35.00 2.80 19.50 1.25 3.00 .00 4.00
Maximum 11.60 61.00 8.50 19.50 3.50 3.08 33.00 55.00
Median 11.6000 49.00 6.80 19.50 2.29 3.04 5.00 37.00
Total Mean 23.2917 42.16 5.14 18.38 2.88 3.07 15.55 23.98
N 12 14 28 6 18 9 113 31
SD 10.42832 12.13 1.75 1.69 .80 .70 16.46 27.01
Minimum 10.30 20.00 .81 15.90 1.25 1.75 .00 .00
Maximum 46.00 61.00 8.50 19.96 4.50 4.00 70.00 90.00
Median 20.6000 45.50 5.50 18.75 3.00 3.08 8.00 10.00
a
Signiﬁcantly higher mean values in one hemisphere are designated by (*).
58 Marlowe ARTICLES
on optimal, or equilibrium group
size.15 Beyond 30, conﬂicts between
families may cause ﬁssioning, and it is
only when there are external constraints,
like the prevalent warfare of
some equestrian and complex foragers,
that there are larger local groups.
Home range is the area an individual
occupies over a whole year. Because
few home range data exist, I
calculated a minimum estimate of
home range, which I call local group
area (see Box 1). The more hunting
contributes to the diet, the larger is
the local group area and the more frequently
camps move (Fig. 6), especially
when we exclude the sedentary
foragers. Only 25% of the foragers in
the sample are sedentary, and most of
these are the complex foragers of the
Paciﬁc Northwest. There is less mobility
where ﬁshing, rather than hunting
or gathering, accounts for more of the
diet (Fig. 6). Thus, our ancestors may
have been quite mobile before ﬁshing
was important. Home ranges are
smaller in the warm-climate sample,
where hunting contributes less to the
diet, so they may have been somewhat
smaller in the distant past if hunting
was less important. Among warm-climate
foragers, local group area (home
range) decreases as primary biomass
increases, since resources are more
concentrated. However, once primary
biomass reaches about 10 kg/m2
, local
group area does not continue to decrease
(Fig. 7). This implies that most
warm-climate Late Pleistocene foragers,
even in rich habitats, probably
had very large home ranges of about
175 km2
(Table 1). This is larger than
the city limits of Washington D.C., too
large an area to defend as an exclusive
territory,22 given the population density
(median ϭ .18 km2
). However,
residents might have repulsed outsiders
upon encounter as do some foragers
and some carnivores with large
home ranges (Table 3).23–25
As primary biomass increases, the
total area of the ethno-linguistic
group shrinks but the population remains
fairly stable (median ϭ 875
cold; 565 warm). Presumably, there
are only so many people who can
remain tied together in terms of language,
as Birdsell argued with his
magic number of 500.26 Foragers
tend to have friendly, co-equal relations
with other local groups but not
with other ethno-linguistic groups,
at least when those others are agriculturalists.
But what was it like before
there were any agriculturalists?
Australians are instructive since at
the time of European contact all
were foragers. While there were
ethno-linguistic boundaries, they
were not very sharp. There was some
Figure 2. Primary biomass of forager habitats by latitude (n ϭ 341). Fit line is Lowess
smoothed.
Figure 3. Contribution to diet from gathering, hunting, and ﬁshing by latitude (n ϭ 398). The
southern hemisphere is plotted with negative latitudes. Fit lines are Lowess smoothed.
ARTICLES Hunter-Gatherers 59
intermarriage between adjacent
groups,27,28 but there was also some
warfare.29 Along with the evolution
of language came both the means to
tie together several local groups and
the basis for dividing one cluster of
local groups from others. Since the
ethno-linguistic group is largely deﬁned
by language, there is no analog
of ethnicity in other mammals, and
the continuous movement of individuals
back and forth between local
groups within the ethno-linguistic
group is probably unique to humans.
KINSHIP, MARITAL RESIDENCE,
AND DISPERSAL
In most mammals, members of at
least one sex leave their natal group at
maturity, perhaps to avoid inbreeding.30
Among human foragers, couples
often live with either the husband’s
family (virilocal) or the wife’s
family (uxorilocal). Virilocal residence
is sometimes assumed to be
analogous to male philopatry since
males stay with their kin while females
disperse.31 This analogy is more
appropriate for sedentary agriculturalists
than for foragers who frequently
move in and out of camps; still, marital
residence does capture something
about potential nepotistic aid. There
has long been a debate about the most
common pattern of marital residence
among foragers.32,33 The dominant
view over the past few decades has
been that foragers, like agricultural
societies, are mostly virilocal.31,34,35
An analysis of foragers that takes account
of residence in the early and
later years of marriage does not support
this.36 Most foragers are multilocal;
the couple resides in camps with
the wife’s kin at times, the husband’s
kin at other times, occasionally with
both, and sometimes with neither.
Bilateral descent, tracing kin
through both mother and father, is
more prevalent among foragers
(75%) than agriculturalists (25%),
who are most often patrilineal.36
One advantage of patrilineal descent
is that it is easier to assemble a large
group of reliable allies in times of
warfare or disputes, since people
know which patrilineal clan they belong
to and where their loyalties lie.
This is probably one reason that patrilineal
descent dominates among
agriculturalists.37 However, an advantage
of bilateral descent is that it
maximizes the number of kin ties
across camps, which facilitates visiting,
ﬁnding mates, and moving to
access seasonal resources.32 Because
bilateral descent, which facilitates
multilocal residence, is possible only
if one knows one’s father as well as
mother, multilocal residence likely
evolved after pair bonding.
Language makes kin terms possible,
makes it easier to keep track of all of
one’s kin, and makes it easier to invent
Figure 4. Population density (people/km2
) by primary biomass of habitat (n ϭ 340). Xs
indicate subsistence at least partly based on anadromous ﬁsh. Fit line is Lowess smoothed
(r ϭ .378, P Ͻ .005, n ϭ 340).
Box 2. Habitat Productivity
To test if forager habitats are marginal
we can compare them to the habitats
of agriculturalists using the Standard
Cross-Cultural Sample (SCCS),
which includes 186 societies chosen to
maximize geographic and linguistic independence.2
Because primary biomass
estimates are available only for
the foragers, however, we need another
measure of habitat quality. Net
primary productivity (NPP) measures
yearly plant growth and can be calculated
from satellite data (see Appendix).
Using NPP, we ﬁnd that indeed there is
a lower habitat quality among the 36
foragers than the 150 agriculturalists.
However, when we exclude those societies
in colder climates where ET is
less than 13°C (the cold-climate foragers),
there is no difference in NPP (t ϭ
.183, P ϭ .856, n ϭ 139: 17 foragers,
122 agriculturalists, equal variances).
Warm-climate forager habitats are not
less productive than those of agricultural
societies in this sample.
60 Marlowe ARTICLES
ﬁctive kin, so it facilitates movement
between groups. It is difﬁcult to say
what dispersal was like before language,
but if we exclude the predominantly
virilocal agriculturalists, it is
worth noting that we come to a different
conclusion about the most likely ancestral
state. Since both sexes disperse
some in gibbons, orangutans, and gorillas,
as well as in human foragers, the
most parsimonious scenario has strict
male philopatry among chimpanzees
evolving only after the Pan-Homo split
and before chimpanzees and bonobos
separate. The same is true for exaggerated
sexual swellings and large testes
(Table 3). Among the extant hominoids,
only the two species of Pan live in promiscuous,
multi-male, multi-female
groups with extremely male-biased phi-
loparty.
THE SEXUAL DIVISION OF
FORAGING LABOR AND THE
MATING SYSTEM
Among chimpanzees 71% to 90% of
hunting is done by males,38 so it is
possible that early hominin males did
more hunting than females as well.
The specialization in hunting or scavenging
by males in contrast to gathering
by females may have begun much
later, but perhaps as early as 2.5
mya.39 The common view that males
only hunt and females only gather is
false, however. For example, Australian
females do considerable hunting
of small animals,40 while males in
many foraging societies collect honey
and gather fruit.41 Among some net
foragers, such as the Aka, men and
women hunt together.42 Still, the sexual
division of foraging labor among
humans stands in real contrast to the
behavior of our ape cousins and other
mammals (Table 3).
We know little about the actual
mating system of foragers since there
are no DNA paternity exclusion data.
What I describe is only the observable
“social” mating system, which probably
under-represents actual polyandrous
mating. There is considerable
variation, ranging from the Andaman
Islands, where polygyny is said not to
exist at all,43 to Australia, where 70%
Figure 5. Local group population by primary biomass of the habitat (n ϭ 260). Xs indicate
equestrian foragers; ϩs indicate subsistence at least partly based on anadromous ﬁsh. Fit
line is Lowess smoothed (r ϭ .000, P ϭ .997, n ϭ 260).
Figure 6. Number of residential moves per year by the percent contribution to the diet from
hunting, gathering, and ﬁshing (n ϭ 340). Fit lines are Lowess smoothed. (Whole sample,
number of moves by ﬁshing: r ϭ Ϫ.302, P Ͻ .0005, n ϭ 340; hunting: r ϭ .259, P Ͻ .0005, n ϭ
340; gathering: r ϭ .122, P ϭ .024, n ϭ 340). Fishing is also signiﬁcant within Old and New
World; hunting is only signiﬁcant in the New World; gathering by itself is not signiﬁcant within
either hemisphere.
ARTICLES Hunter-Gatherers 61
of married men among the Tiwi can
be polygynous44 (Table 2). Polygyny in
Australia is related to gerontocracy;
old men have much higher status (and
more wives) than young men do. The
percentage of married women with a
co-wife better captures the degree of
reproductive skew, since a polygynous
man may have 2 wives or 10, which
will affect the number of bachelors. In
the total forager sample this ranges
from 0% to 90% (median ϭ 10). In the
SCCS, about 10% of foraging societies
are monogamous, 60% slightly polygynous
(Ͻ20% have co-wives), and
30% generally polygynous (Ͼ20%
have co-wives), making foragers as
polygynous as agriculturalists.45 Even
where there is considerable polygyny,
however, most marriages are monogamous45
and, except in Australia, two
wives is often the maximum. In the
SCCS, in societies where male contribution
to diet is higher, there is less
polygyny.45 Hence, human foraging
societies ﬁt a pattern seen across
mammalian species, where males are
most likely to invest in offspring in
socially monogamous species, while
the vast majority of species in which
males invest little or nothing are po-
lygynous.46
Among most foragers, men’s foods,
especially larger game, are shared
widely outside the household.10,47
This means that the best hunter’s
household may eat no better than the
households of poor hunters, which
forces us to ask what beneﬁt a woman
would gain from marrying a good
hunter. Some have suggested that a
man’s hunting success signals something
about his quality other than his
provisioning value to his own household.47,48
While that may be true, I
found that Hadza men who were not
stepfathers brought more calories
back to camp when their wives were
nursing an infant and had decreased
productivity.41 Hadza men, therefore,
appear to compensate for their nursing
wives’ lower productivity, which
suggests that pair bonds may be maintained
by the beneﬁts of a sexual division
of labor within households.
Age at weaning ranges from 1 to 4.5
years (median ϭ 2.5), which is considerably
younger than the 4.8 years for
chimpanzees and bonobos (Tables 2
and 3). Infant mortality ranges from
10% to 46% (median ϭ 21) and juvenile
mortality from 20% to 61% (median
ϭ 45), slightly lower than the rate
for chimpanzees.49 Total fertility rate
ranges from 0.81–8.5 (median ϭ 5.5).
As I have reported elsewhere,50 higher
mean male contribution to diet across
foraging societies predicts a younger
age at weaning, a higher total fertility
rate, and greater female reproductive
success, but not a lower offspring
mortality rate. Even when infant and
juvenile mortality is high, where men
contribute more to the diet a greater
absolute number of offspring survive
to the age of 15 years. This implies
that women may use food from men
mainly to speed up their rate of repro-
duction.
CENTRAL PLACES, FOOD
SHARING, AND
EGALITARIANISM
Central place foraging is typical of
most human foragers, as it is among
many birds and social carnivores
(Table 3). Because there are species
that return to central places without
food (among them hamadryas, Papio
hamadryas, and spider monkeys, Ateles
geoffroyi), a better term for those
who do take food back might be central-place
provisioners.51 One reason
for having a central place is to leave
vulnerable or burdensome young in a
safe place. Hadza mothers take their
nursing infants with them when they
go foraging but leave weanlings in
camp because they are too young to
keep up and too big to carry. Someone
needs to stay in camp to look out for
them and so food must be taken back.
Perhaps this sets a lower limit on local
group size among foragers; with less
than 25 people and most adults out
foraging, there might often be too few
people available to baby-sit.
As Glynn Issac52 long ago suggested
and Nicholas Blurton Jones53 later
modeled, food sharing among foragers
probably began with tolerated scrounging
not so different from the meat sharing
seen in chimpanzees, who sometimes
hold out their hands in a begging
gesture, often turn their backs to beggars,
and sometimes grab meat away
from others.54 Resisting the demands of
several other hungry adults could be especially
dangerous in a species capable
Figure 7. Local group area (home range) by primary biomass of the habitat for warmclimate,
nonequestrian foragers (n ϭ 125). Fit line is Lowess smoothed (r ϭ Ϫ.318, P Ͻ .0005,
n ϭ 125).
62 Marlowe ARTICLES
of forming temporary coalitions, as presumably
our ancestors were. Such coalitions
can create what Boehm55 calls a
reverse dominance hierarchy, resulting
in egalitarianism. Food sharing would
have lowered variation in daily food
consumption,56 which should have lowered
mortality rates and altered life-history
patterns.57 Central-place provisioning
increases food sharing, since taking
food back to camp creates more opportunities
for scrounging.51 Because central-place
provisioning makes food
sharing and egalitarianism more likely,
if central places can be detected archeologically,
their earliest appearance
might set an upper limit on the antiquity
of egalitarianism.
A HIGHLY VARIABLE AND
UNUSUAL PRIMATE SPECIES
Tables 1 and 2 show considerable
variation. They also reveal many signiﬁcant
differences between Old
World and New World foragers. Less
of the diet comes from gathering and
more from hunting and ﬁshing in the
New World, and male contribution to
diet is higher. Ethno-linguistic population
and area are larger, as is the
local group area (home range). There
are fewer moves per year, and the age
at weaning is younger.
There are also differences between
the cold- and warm-climate samples.
Many of the differences mirror those
between New and Old World. This is
partly due to the fact that there are so
many northern foragers in the New
World, with the mean absolute latitude
being 44° and ET 13°C, compared
to 15° absolute latitude and ET
of 20°C for the Old World. The latitude
difference between New and Old
World is present even in the warmclimate
sample (NW ϭ 32°, 16°C;
OW ϭ 17°, 19°C). When latitude is
controlled, many of the differences
between New World and Old World
disappear, though ﬁshing is still more
important and total area of the ethnolinguistic
group is still larger in the
New World.
In comparative perspective, humans
are so behaviorally diverse
that societies are more like different
species than different populations of
one species, although there are some
common features shared by most. In
some of their widely shared features,
human foragers resemble social carnivores
more than they do our closest
primate relatives. Like wild dogs,
foragers have long day ranges, large
home ranges, central places, provisioning,
food sharing, and bisexual
dispersal (Table 3). Many of these
traits are probably due to technology
that allows us to acquire large
amounts of food that can be transported
and is consequently more
TABLE 3. Warm-Climate Nonequestrian Foragers Compared with African Apes and Some Social Carnivoresa
Trait
Foragers
Homo sapiens
Chimp Pan
troglodytes
Bonobo Pan
paniscus
Gorilla
Gorilla
gorilla
Wild Dog
Lycaon pictus
Lion
Panthera
leo
Hyena Crocuta
crocuta
Dietary niche Omnivore Frugivore/
Folivore
Frugivore/
Herbivore/
Folivore
Folivore/
Herbivore/
Frugivore
Carnivore Carnivore Carnivore
Dispersal Bisexual
multilocal36
Female30 Female73 Bisexual30 Female,74
bisexual75,76
Male,
bisexual76
Male76
Group (local)
population
Min–max
26
13–250
40
15–120
community
ϭ 5077
34
30–6078
11, 6 mt, 9
lowland
2–2079
10
20/6079
1379 55
3/8079
Population
density
(Individuals
per km2
)
.18
.00–3.09
.09–578 2–378 .5–1080 Ͻ.0579 .12–.3879 1.7779
Ngorongoro
Group
composition
Multimale,
Multifemale
camps
Multimale,
multifemale
community,
parties
Multimale,
multifemale
groups,
parties
Unimale,
multifemale
groups
Multimale,
multifemale
packs
Multimale,
multifemale
prides
Multimale,
multifemale
clans
Mating system Pair bond,
Monogamy
polygyny
Promiscuity Promiscuity Pair bond,
Harem
Polygyny
Monogamy,
cooperative
breeding
Promiscuous
but estrus
pairs
Promiscuous
Sexual Swellings None Exaggerated Exaggerated Slight None None None
Division of
foraging
labor
Yes No No No Male provisions
pregnant
female75
No79 No79
Provisioning
beyond
weaning
Yes A bit54 A bit No Yes Yes A bit
Food sharing
among adults
Yes, lots A bit54 A bit81 No79 A lot79 Yes,
grudgingly79
Yes,
grudgingly79
Foraging
pattern
Central place
provisioning
Feed as you
go
Feed as you
go
Feed as
you go
From den only
when infants
From lair
only when
infants
From burrow or
den only
when infants
Grouping
pattern
Fission-fusion,
(s often
alone
Fission-fusion,
&s often
alone
Fission-fusion,
&s more
social than
chimp
Cohesive (
harem
group,
bachelors
out
Fission-fusion,
group hunt
Fission-fusion Fission-fusion
hunt alone,
pairs
Day range (km)
&/(
9.5/14.1
n ϭ 8/n ϭ 6
3/577 2.4/2.477 1.1/1.1 (.5–
2.1)82
1079 4.5/8 (1.9–
14.4)83
1079
Home range
Min–max
(km2
)
175 (median)
22–4,500
12.5
5–5079
45
30–6078
24.4
8.2–4082
1,700
1500–2,00074
200
20–40079
500
30–2,00079
a
Numbers are means or medians, followed by ranges.
ARTICLES Hunter-Gatherers 63
likely to be transferred. This does
not have to be meat; any food, such
as honey, that is taken back to a
central place in large quantities is
more likely to provoke demandsharing.56
While technology has intensiﬁed
greatly in just the past
100,000 years, if the hominin divergence
from Pan entailed a shift to
high-quality but difﬁcult-to-acquire
foods dispersed in a more open habitat
with increased threat of predation,
some increase in technology
may have been selected for from the
very beginning.9,58
TECHNOLOGICAL CHANGE
The behavioral evolution of our ancestors
must be reconstructed trait by
trait, with the time period and area
speciﬁed.34,59 For certain traits, the
ethnographic foragers are relevant for
only the past few thousand years; for
other traits they may be informative
about hominins millions of years ago.
Table 4 shows some of the important
technologies and their probable effects.
There is not space here to do
justice to the complexity and ambiguity
in the archeological record, so I
have chosen to show both the oldest
solid dates, as well as the oldest proposed
dates, regardless of how contro-
versial.
Many foragers acquire iron through
trade and use it to make arrowheads.
Iron ﬁrst appeared about 2000 to 1500
B.C. in Western Asia,60 and by 600
B.C. had spread far and wide in the
Old World, though not to Australia.
Iron’s superiority for durable, effective
tools and weapons explains why it
is so highly valued and spread so rapidly.
Since most contemporary foragers
use iron rather than stone or bone
points, they probably spend less time
on tool manufacture per tool than did
Pleistocene foragers and so have more
time to allocate to other activities.
Other fairly recent hunting technologies
include traps, snares, and nets
(Table 4). As noted, net hunting can
inﬂuence the sexual division of labor
because males and females more often
hunt together.42 The possession of
iron tools, cooking pots, factory-made
water buckets, and cloth make contemporary
foragers different from
Pleistocene foragers, but do not make
them so different as to be uninforma-
tive.
Because the quality of their arrow
poison can vary, the Hadza often hit
animals that do not die. Hunting large
game without poison would have been
more difﬁcult. Poison use has received
little attention archeologically and
will not be easy to document. Hadza
men with bows and poisoned arrows
usually hunt alone because the challenge
is getting close enough to prey
to get a shot before being detected,
and two people only increase the
chance of detection. Bows may have
caused a switch from group hunting
to more individual hunting, or at least
hunting in smaller groups. Even
though game may have been more
TABLE 4. Technologies, With Earliest Dates Proposed (ya ‫؍‬ years ago), Area Where Found,
and Probable Effects the Innovation Had
Trait
Earliest
Evidence Where Probable Effect
Iron 3500 ya60 South-West
Asia
Time saving in arrow making, axes, more trade
Poison 11,000 ya,
cited in84
Zambia Fast killing of large game by lone hunter, more
meat
Bow (see arrow points) 11,000 ya61 Germany Effective killing at a distance
Spear thrower 17,000 ya62 France Uni-male hunting, killing at a distance, more
meat
Nets 22–29,000 ya85 Czech
Republic
Bi-sexual hunting, effective ﬁshing
Microliths (arrow points), bows
implied
65–70,000
ya86–88
Tanzania
South
Africa
Effective killing at a distance, uni-male
hunting, more meat
Fishing harpoons 75,00088
80,000 ya89
Congo Greater male contribution to diet, lower
mobility
Shellﬁsh 125,000 ya90 South Africa Coastal niche
Spears 400,000 ya63 Germany Hunting of larger game, before this mostly
scavenging, more meat, and better defense
Fire 790,00091
1.6 mya92
Israel
Africa
Protection from predators, lower mortality,
warmth, cooking
Acheulean handaxe, cleaver 1.7 mya93 East Africa Butchering, more meat, or opening beehives
to get more honey
Oldowan chopper,
hammerstone, scraper
2.6 my94 East Africa Food processing, scavenging, tool
manufacture
Carrying devices ? Africa Infant carrying, leather, straw or wooden
devices to carry food to a central place
Wooden digging stick? ? Africa Digging tubers, thus adaptation to more open
habitat, larger home range
Pounding/throwing rocks,
hitting and jabbing sticks?
6 my? Africa Protection from predators, food processing,
preadaptation for more open habitat
64 Marlowe ARTICLES
plentiful in the Pleistocene, without
nets or the bow and arrow, iron points
and poison, it is difﬁcult to see how
earlier hominins could have matched
the hunting success of contemporary
foragers. The bow was such a technological
leap forward that it could have
led to an increase in meat consumption
and population growth rates,
eventually reducing game populations
in certain areas and hastening the
adoption of agriculture.
Foragers without the bow are the
most useful analogs for humans living
before 100,000 ya, and that leaves
only the Australians and Tasmanians.
However, even the Australians had the
spear thrower (or atlat1), which
greatly enhances efﬁciency. The oldest
known spear thrower is not much
older than the oldest known bow61,62
(Table 4). Wooden spears without
stone points have been found in Germany
associated with horse bones
dating to 400,000 ya.63 Without spear
throwers a lone hunter would have
great difﬁculty killing large game, especially
if spears were used to thrust
rather than throw, since that requires
getting very close to game. If largegame
hunting occurred, it probably
required cooperative groups. Before
effective projectiles, hunters may have
run down and clubbed small game or
extracted animals from burrows.64
The slow speed of bipeds would not
have prevented them from eating
large game acquired by scavenging,
however. Among the Hadza, even
women armed with only their digging
sticks occasionally scare off a leopard
and take its kill.65 Before effective
hunting, males could have focused
more on honey and plant foods, so
their daily hauls of food did not have
to be lower but must have been different.
In this sample, male contribution
to diet is signiﬁcantly lower in Australia,
where there was no bow, than in the
rest of the warm-climate sample (38%
versus 56%). However, primary biomass
is also lower, so this could be related
to habitat rather than technology.
As noted, male contribution to diet
is higher where ﬁshing is more important.
Even in the warm-climate sample,
male contribution to diet is significantly
higher where ﬁshing accounts
for Ն20% of the diet (58% versus
49%). Since several other traits, such
as lower mobility, are associated with
ﬁshing (Fig. 6), dating the origin of
ﬁshing is important (Table 4). In the
ethnographic record, ﬁshing is done
with spears, bows, weirs, nets, baskets,
and poison. Because many of
these technologies probably did not
appear prior to modern H. sapiens, we
might use foragers who do little ﬁshing
as guides to earlier periods.
All foragers in the ethnographic
record use ﬁre, but dates for its ﬁrst
use are controversial.66 Controlled use
of ﬁre would have had radical effects
in a variety of ways. Fires are deliberately
set by some foragers to burn underbrush
and encourage new growth
for game to eat.67 Fire is used to
harden digging sticks, curve bows,
and straighten arrows. Smoke from
torches is used to stun bees while collecting
honey. Fires probably keep
predators away from camps, which
must have lowered mortality rates.
Fire allowed occupation of colder
habitats, and cooking probably increased
the value of some foods, perhaps
expanding the hominin diet.68
Stone tools appear 2.6 mya, but it is
implausible that these represent the
earliest tool use. Since chimpanzees
wield sticks and occasionally throw
stones or use them as hammers,69 tool
use could have begun even before the
Homo-Pan split. While some tool use
may have existed throughout hominin
evolution, there has been a rapid increase
in technological sophistication
in our species. Given the somewhat recent
evolution of complex projectiles
and nets, male foraging long ago had to
be quite different, but female foraging
did not; the behavior of modern females
is, therefore, more instructive for earlier
periods. Hadza females forage in
groups of about 5 women, some infants
and older children, collecting baobab
fruit that falls to the ground, gathering
berries, and digging up a variety of tubers
with digging sticks, which they
also use to defend themselves. Hadza
women, and women in many other
warm-climate foraging societies with
simple technology, might be reasonable
guides to hominin foraging patterns
over a very long time. Females may
have pioneered the earliest and still important
technologies: digging sticks,
rocks for pounding nuts and seeds, and
devices for carrying infants and foods
(Table 4).
CONCLUSION
The forager data allow us to explore
relationships between habitat
and social organization and compare
humans to other species. Such
relationships, in conjunction with
the fossil and archeological record,
can help model the behavior of our
ancestors at different times in the
past. The main obstacle is not contact
with agriculturalists or climate
change or bias toward marginal habitats,
but technology; all foragers in
the ethnographic record possess
complex technology compared to all
hominins before modern sapiens.
However, even Lower Paleolithic technology
may have afforded some surplus
production, which in some ways made
hominins more like social carnivores
than other apes. Contemporary foragers
are not living fossils, but because
they are pre-agricultural they are the
most relevant analogs for at least Late
Pleistocene humans. As long as we take
into account the effects of technology,
the behavioral ecology of contemporary
foragers can provide important insights
into human evolution.
APPENDIX
There are 478 societies in the sample
used in this paper, though not all
are necessarily equally distinct ethnoThe
bow was such a
technological leap
forward that it could
have led to an increase
in meat consumption
and population growth
rates, eventually
reducing game
populations in certain
areas and hastening the
adoption of agriculture.
ARTICLES Hunter-Gatherers 65
linguistic groups. Societies are included
if less than ten percent of their
diet comes from domesticated foods,
but I err on the side of inclusion to
present the most comprehensive sample
where good quantitative data exist.
Thus, some may eat more than
10% of domesticated food through
trade or periodic horticulture, or live
at missions but go on lengthy treks
during which data are collected, as do,
for example, the Ache of Paraguay.
Many sources of data not cited in
the text are cited in the appendices of
my analyses of male dietary contribution50
and mating systems.45 Additional
data come from various other
sources such as Boﬁ,70 and 56 societies
on the island of New Guinea where
very little is known beyond estimates
of the percent of diet coming from
foraging.71 For these and a few others,
I ﬁlled in missing data on exact location,
ethno-linguistic populations,
and a few other variables using the
Ethnologue.72
Productivity Comparison
Using satellite-gathered data on
temperature, leaf cover, and solar radiation,
the MODIS algorithm calculates
an estimate of annual net primary
production (NPP) or the amount
of new plant growth in grams/meter2
/
year as a proxy for habitat quality
(equation in: Allen RG, Pereira LS,
Raes D, Smith M. 1998. FAO Irrigation
and Drainage Paper No. 56: Crop
evapotranspiration guidelines for
computing crop water requirements:
www.fao.org. For MODIS data: http://
www.ntsg.umt.edu/). Foragers vs. agriculturalists:
NPP ϭ 572 vs. 788, P Ͻ
.0024, using only those societies
where ET Ն 13°C, NPP ϭ 848 vs. 869,
P ϭ .856.
ACKNOWLEDGMENTS
I thank Eric A. Smith, John Fleagle,
Claire Porter, David Pilbeam, Ofer
Bar-Yosef, and four anonymous reviewers
for comments on earlier
drafts, and Claire Porter for the calculation
of NPP estimates of habitat
quality.
REFERENCES
1 Binford LR. 2001. Constructing frames of reference.
Berkeley: University of California Press.
2 World-Cultures. 2001. J Comp Cross-Cultural
Res 12.
3 Lee RB, DeVore I. 1968. Problems in the study
of hunters and gatherers. In: Lee RB, DeVore I,
editors. Man the hunter. Chicago: Aldine. p 5–12.
4 Wilmsen EN. 1989. Land ﬁlled with ﬂies: a
political economy of the Kalahari. Chicago: Chicago
University Press.
5 Foley R. 1988. Hominids, humans and huntergatherers:
an evolutionary perspective. In: Ingold
T, Riches D, Woodburn J, editors. Hunters and
gatherers: history, evolution and social change.
New York: St. Martin’s Press. p 207–221.
6 Tooby J, DeVore I. 1987. The reconstruction of
hominid behavioral evolution through strategic
modeling. In: Kinzey WG, editor. The evolution
of human behavior: primate models. Albany:
State University of New York Press. p 183–237.
7 Marlowe FW. 2002. Why the Hadza are still
hunter-gatherers. In: Kent S, editor. Ethnicity,
hunter-gatherers, and the “other”: Association or
Assimilation in Africa. Washington, D.C.: Smithsonian
Institution Press. p 247–275.
8 White T, Asfaw B, DeGusta D, Gilbert H, Richards
GD, Suwa G, Howell FC. 2003. Pleistocene
Homo sapiens from Middle Awash, Ethiopia. Nature
423:742–747.
9 Oswalt WH. 1973. Habitat and technology.
New York: Holt, Rinehart and Winston.
10 Kelly RL. 1995. The foraging spectrum. Washington,
D.C.: Smithsonian Institution Press.
11 Murdock GP. 1967. The ethnographic atlas: a
summary. Ethnology 6:109–236.
12 Borgerhoff Mulder M. 2001. Using phylogenetically
based comparative methods in anthropology:
more questions than answers. Evol Anthropol
10:99–111.
13 Murdock GP, White DR. 1980. Standard
cross-cultural sample. In: Barry H, Schlegel A,
editors. Cross-cultural samples and codes. Pittsburgh:
University of Pittsburgh Press. p 3–43.
14 Shimkin DD. 1983. Introduction of the horse.
In: D’Azevedo WL, editor. Great Basin, vol. 11,
Handbook of North American Indians. Washington,
D.C.: Smithsonian Institution Press. p 517–
524.
15 Boone JL. 1992. Competition, conﬂict, and
the development of social hierarchies. In: Smith
EA, Winterhalder B, editors. Evolutionary ecology
and human behavior. New York: Aldine de
Gruyter. p 301–337.
16 Vaughan R. 1994. The Arctic: a history. Dover,
NH: Sutton.
17 Lee RB. 1968. What hunters do for a living,
or, how to make out on scarce resources. In: Lee
RB, DeVore I, editors. Man the hunter. Chicago:
Aldine de Gruyter. p 30–48.
18 Porter CC, Marlowe FW. ND. Forager habitats
are no more marginal than those of agriculturalists.
Submitted. J Anthropol Archaeol.
19 Ames KM. 2003. The Northwest Coast. Evol
Anthropol 12:19–33.
20 Woodburn J. 1980. Hunters and gatherers today
and reconstruction of the past. In: Gellner E,
editor. Soviet and Western anthropology. London:
Duckworth. p 95–117.
21 Dunbar RIM. 1992. Neocortex size as a constraint
on group size in primates. J Hum Evol
20:469–493.
22 Mitani JC, Rodman PS. 1979. Territoriality:
the relation of ranging and home range size to
defendability, with an analysis of territoriality
among primate species. Behav Ecol Sociobiol 5:
241–251.
23 Cashdan E. 1983. Territoriality among human
foragers: ecological models and an application
to four Bushmen groups. Cultural Anthropol
24:47–66.
24 Dyson-Hudson R, Smith EA. 1978. Human
territoriality: an ecological reassessment. Am Anthropol
80:21–41.
25 Brown JL, Orians GH. 1970. Spacing patterns
in mobile animals. Ann Rev Ecol Syst 1:239–262.
26 Birdsell JB. 1968. Some predictions for the
Pleistocene based on equilibrium systems among
recent hunter-gatherers. In: Lee RB, DeVore I,
editors. Man the hunter. Chicago: Aldine de
Gruyter. p 229–240.
27 Peterson N. 1976. Tribes and boundaries in
Australia. Canberra: Australian Institute of Aboriginal
Studies.
28 Tindale N. 1953. Tribal and intertribal marriages
among the Australian aborigines. In:
Lasker GW, Brozek J, Glass B, Mainland D, Spuhler
JN, Straus WL, editors. Human biology: a
record of research. Baltimore: Johns Hopkins
Press. p 169–190.
29 Hiatt LR. 1996. Arguments about Aborigines.
Cambridge: Cambridge University Press.
30 Pusey AE, Packer C. 1987. Dispersal and philopatry.
In: Smuts B, Cheney DL, Seyfarth RM,
Wrangham RW, Struhsaker TT, editors. Primate
societies. Chicago: University of Chicago. p 250–
266.
31 Rodseth L, Wrangham RW, Harrigan AM,
Smuts B. 1991. The human community as a primate
society. Curr Anthropol 12:221–254.
32 Murdock GP. 1949. Social structure. New
York: Macmillan.
33 Service ER. 1967. Primitive social organization.
New York: Random House.
34 Foley R. 1995. The adaptive legacy of human
evolution: a search for the environment of evolutionary
adaptedness. Evol Anthropol 4:194–203.
35 Ember CR. 1978. Myths about hunter-gatherers.
Ethnology 17:439–448.
36 Marlowe FW. 2004. Marital residence among
foragers. Curr Anthropol 45:277–284.
37 van den Berghe P. 1979. Human family systems.
Prospect Heights: Waveland.
38 Stanford CB. 2001. A comparison of social
meat-foraging by chimpanzees and human foragers.
In: Stanford CB, Bunn HT, editors. Meateating
and human evolution. Oxford: Oxford
University Press. p 122–140.
39 de Heinzelin J, Clark JD, White T, Hart W,
Renne P, WoldeGabriel G, Beyene Y, Vrba E.
1999. Environment and behavior of 2.5 millionyear-old
Bouri hominids. Science 284:625–629.
40 Tonkinson R. 1978. The Mardudjara Aborigines:
living the dream in Australia’s desert. New
York: Holt, Rhinehart and Winston.
41 Marlowe FW. 2003. A critical period for provisioning
by Hadza men: implications for pair
bonding. Evol Hum Behav 24:217–229.
42 Hewlett BS. 1992. Intimate fathers: the nature
and context of Aka pygmy paternal infant care.
Ann Arbor: University of Michigan Press.
43 Radcliffe-Brown AR. 1964. The Andaman Islanders.
New York: Free Press.
44 Goodale JC. 1971. Tiwi wives: a study of the
women of Melville Island, North Australia. Seattle:
University of Washington Press.
45 Marlowe FW. 2003. The mating system of foragers
in the standard cross-cultural sample.
Cross-Cultural Res 37:282–306.
46 Kleiman DG, Malcolm JR. 1981. The evolution
of male parental investment in mammals.
In: Gubernick DJ, Klopfer PH, editors. Parental
care in mammals. New York: Plenum Press. p
347–387.
47 Hawkes K, Bliege Bird R. 2002. Showing off,
handicap signaling, and the evolution of men’s
work. Evol Anthropol 11:58–67.
48 Smith EA. 2005. Why do good hunters have
66 Marlowe ARTICLES
higher reproductive success? Hum Nat 10:342–
363.
49 Hill K, Boesch C, Goodall J, Pusey AE, Williams
J, Wrangham RW. 2001. Mortality rates
among wild chimpanzees. J Hum Evol 40:437–
450.
50 Marlowe FW. 2001. Male contribution to diet
and female reproductive success among foragers.
Curr Anthropol 42:755–760.
51 Marlowe FW. n.d. Central place provisioning:
The Hadza as an example. In: Hohmann G, Robbins
M, Boesch C, editors. Feeding ecology in
apes and other primates. Cambridge: Cambridge
University Press. In press.
52 Isaac G. 1978. The food-sharing behavior of
protohuman hominids. Sci Am 238:90–108.
53 Blurton Jones NG. 1987. Tolerated theft: suggestions
about the ecology and evolution of sharing,
hoarding, and scrounging. Soc Sci Information
26:31–54.
54 McGrew WC, Feistner TC. 1992. Two nonhuman
primate models for the evolution of human
food sharing: chimpanzees and callitrichids. In:
Barkow JH, Cosmides L, Tooby J, editors. The
adapted mind. Oxford: Oxford University Press. p
229–243.
55 Boehm C. 1999. Hierarchy in the forest: the
evolution of egalitarian behavior. Cambridge:
Harvard University Press.
56 Winterhalder B. 2001. Intragroup resource
transfers: comparative evidence, models, and implications
for human evolution. In: Stanford CB,
Bunn HT, editors. Meat-eating and human evolution.
Oxford: Oxford University Press. p 279–
301.
57 Kaplan H, Hill K, Lancaster J, Hurtado A.
2000. A theory of human life history evolution:
diet, intelligence, and longevity. Evol Anthropol
9:156–185.
58 Bartholomew GA, Birdsell JB. 1953. Ecology
and the protohominids. Am Anthropol 55:481–
498.
59 Irons W. 1998. Adaptively relevant environments
versus the environment of evolutionary
adaptedness. Evol Anthropol 6:194–204.
60 Waldbaum JC. 1978. From bronze to iron: the
transition from the Bronze Age to the Iron Age in
Eastern Mediterranean. Goteborg: Paul Astroms
Forlag.
61 Cattelain P. 1997. Hunting during the upper
paleolithic: bow, spearthrower, or both? In:
Knecht H, editor. Projectile technology. New
York: Plenum Press. p 213–240.
62 Knecht H. 1997. The history and development
of projectile technology research. In: Knecht H,
editor. Projectile technology. New York: Plenum
Press. p 3–35.
63 Thieme H. 1997. Lower paleolithic hunting
spears from Germany. Nature 385:807–810.
64 Hill K, Hawkes K. 1983. Neotropical hunting
among the Ache of eastern Paraguay. In: Hames
RB, Vickers WT, editors. Adaptive responses of
native Amazonians. New York: Academic Press.
p 139–188.
65 O’Connell JF, Hawkes K, Blurton Jones NG.
1988. Hadza scavenging: implications for Plio/
Pleistocene hominid subsistence. Curr Anthropol
29:356–363.
66 James SR. 1989. Hominid use of ﬁre in the
Lower and Middle Pleistocene: a review of the
evidence. Curr Anthropol 30:1–26.
67 Lewis HT. 1982. Fire technology and resource
management in aboriginal North America and
Australia. In: Williams NM, Hunn ES, editors.
Resource managers: North American and Australian
hunter-gatherers. Colorado: Westview Press.
p 45–67.
68 Wrangham RW, Conklin-Brittain N. 2003.
Cooking as a biological trait. Comp Biochem
Physiol 136:35–46.
69 McGrew WC. 1992. Chimpanzee material culture:
implications for human evolution. Cambridge:
Cambridge University Press.
70 Fouts HN. 2002. The social and emotional
contexts of weaning among the Boﬁ farmers and
foragers of central Africa. Ph.D. thesis, Washington
State University.
71 Roscoe P. 2002. The hunters and gatherers of
New Guinea. Curr Anthropol 43:153–162.
72 Grimes BF, editor. 2000. Ethnologue. Dallas:
SIL International.
73 Furuichi T, Idani G, Ihobe H, Kuroda S, Kitamura
K, Mori A, Enomoto T, Okayasu N, Hashimoto
C, Kano T. 1998. Population dynamics of
wild bonobos (Pan paniscus) at Wamba. Int J
Primatol 19:1029–1043.
74 Kingdon J. 1997. African mammals. San Diego:
Academic Press.
75 Moehlman PD, Hofer H. 1997. Cooperative
breeding, reproductive suppression, and body
mass in canids. In: Solomon NG, French JA, editors.
Cooperative breeding in mammals. Cambridge:
Cambridge University Press. p 76–128.
76 Waser PM. 1996. Patterns and consequences
of dispersal in gregarious carnivores. In: Gittleman
JL, editor. Carnivore behavior, ecology, and
evolution. Ithica: Cornell University Press. p 267–
295.
77 Wrangham RW, Wilson ML. nd. Intrasexual
bonding and the economics of territorial defense
in ﬁssion-fusion communities of primates. Submitted.
Behav Ec. Soc.
78 Cheney DL. 1987. Interactions and relationships
between groups. In: Smuts B, Cheney DL,
Seyfarth RM, Wrangham RW, Struhsaker TT,
editors. Primate societies. Chicago: University of
Chicago Press. p 267–281.
79 Estes RD. 1991. The behavior guide to African
mammals. Los Angeles: University of California
Press.
80 Whitﬁeld J. 2003. Ape populations decimated
by hunting and ebola virus. Nature 422:551.
81 Nishida T, Hiraiwa-Hasegawa M. 1987. Chimpanzees
and bonobos: cooperative relationships
among males. In: Smuts B, Cheney DL, Seyfarth
RM, Wrangham RW, Struhsaker TT, editors. Primate
societies. Chicago: University of Chicago
Press. p 165–177.
82 Doran DM, McNeilage A. 1998. Gorilla ecology
and behavior. Evol Anthropol 6:120–131.
83 Schaller GB. 1983. Golden shadows, ﬂying
hooves. Chicago: University of Chicago Press.
84 Clark JD. 1970. The prehistory of Africa. New
York: Praeger.
85 Soffer O, Adovasio JM, Illingworth JS,
Amirkhanov HA, Praslov ND, Street M. 2000.
Paleolithic perishables made permanent (ﬁber
artifacts). Antiquity 74:812–821.
86 Elston RG, Kuhn SL. 2002. Thinking small:
global perspectives on microlithization. In: Johnson
JK, editor: Arlington, VA: American Anthropological
Association.
87 Barham L. 2002. Backed tools in Middle Pleistocene
central Africa and their evolutionary signiﬁcance.
J Hum Evol 43:585–603.
88 McBrearty S, Brooks AS. 2000. The revolution
that wasn’t: a new interpretation of the origin of
modern human behavior. J Hum Evol 39:453–
563.
89 Henshilwood CS, Sealy JC, Yates R, CruzUribe
K, Goldberg P, Grine FE, Klein RG,
Poggenpoel C, Van Niekerk KIW. 2001. Blombos
Cave, Southern Cape, South Africa: Preliminary
report on the 1992–1999 excavations of the Middle
Stone Age levels. J Archeol Sci 28:421–448.
90 Deacon HJ, Shuurman R. 1992. The origins of
modern people: the evidence from Klasies River.
In: Brauer G, Smith FH, editors. Continuity or
replacement: controversies in Homo sapiens evolution.
Rotterdam: A. A. Balkema. p 121–129.
91 Goren-Inbar N, Alperson N, Kislev ME, Simchoni
O, Melamed Y, Ben-Nun A, Werker E.
2004. Evidence of hominin control of ﬁre at
Gesher Benot Ya’aqov, Israel. Science 304:725–
727.
92 Bellomo RV. 1994. Methods of determining
early hominid behavioral activities associated
with the controlled use of ﬁre at FxJj 20 Main,
Koobi Fora, Kenya. J Hum Evol 27:173–195.
93 Asfaw B, Beyene Y, Semaw S, Suwa G, White
T. 1992. The earliest Acheulean from Konso-Gardula.
Nature 360:732–735.
94 Semaw S, Renne P, Harris JWK, Feibel CS,
Bernor RL. 1997. 2.5 million-year-old stone tools
from Gona, Ethiopia. Nature 385:333–336.
ARTICLES Hunter-Gatherers 67