Evolution and Human Behavior 37 (2016) 281–286

Contents lists available at ScienceDirect

Evolution and Human Behavior

Original Article

Eat first, share later: Hadza hunter–gatherer men consume more while
foraging than in central places
J. Colette Berbesque a,⁎, Brian M. Wood b, Alyssa N. Crittenden c, Audax Mabulla d, Frank W. Marlowe e
a
Centre for Research in Evolutionary, Social and Inter-Disciplinary Anthropology, University of Roehampton, London, UK
b
Department of Anthropology, Yale University, CT, USA
c
Department of Anthropology, University of Las Vegas, NV, USA
d
National Museums of Tanzania, Dar es Salaam, Tanzania
e
Department of Archaeology and Anthropology, Cambridge University, Cambridge, UK

a r t i c l e i n f o a b s t r a c t

Article history: The foraging and food sharing of hunter–gatherers have provided the backdrop to several different evolutionary
Initial receipt 17 March 2015 hypotheses about human life history. Men's foraging has often been characterized as primarily targeting animals,
Final revision received 24 January 2016 with high variance and high rates of failure. To the best of our knowledge, however, there are as yet no quanti-
tative studies reporting the amounts of food that men eat while foraging, before returning to their households
Keywords:
either empty-handed or with foods. Here, we document this under-reported part of forager's diets—men's eating
Central place foraging
Energetics
while out of camp on foray. Our dataset consists of 146 person/day follows (921 hours total) collected over a pe-
Foraging riod of 12 years (from 2001–2013, including 12 camps). Hadza men consumed a substantial amount of food
Hadza while out of camp foraging. Men did more than just snack while out of camp foraging, they consumed a mean
Hunter–Gatherers of 2,405 kilocalories per foray, or approximately 90% of what is estimated to be their mean daily total energy ex-
Paleodiet penditure (TEE). The characterization of men's foraging strategies as “risky”, in terms of calorie acquisition, may
Provisioning be exaggerated. Returning to camp empty-handed did not necessarily mean the forager had failed to acquire
food, only that he failed to produce enough surplus to share. Surprisingly, the vast majority of the kilocalories
eaten while out of camp came from honey (85%). These observations are relevant to evolutionary theories
concerning the role of male provisioning. Understanding primary production and consumption is critical for un-
derstanding the nature of sharing and the extent to which sharing and provisioning supports reproduction in
hunter–gatherers.
© 2016 Elsevier Inc. All rights reserved.

1. Introduction anthropology often have the impression that all foods acquired are
brought back to the residential group to be shared with others. This im-
Knowledge of food sharing and the sexual division of labor in pression is easy to understand in light of statements like those of
hunter–gatherers is mostly based on the distribution of foods at central Marshall (1998:71,77) who, despite documenting out of camp eating
places (e.g. Bahuchet, 1990; Kitanishi, 1998; Ziker, 2007), or in mixed- by the Nye Nye !Kung, goes on to write “!Kung are quite conscious of
sex or single sex groups (e.g. Kaplan, Hill, Hawkes, & Hurtado, 1984; the value of meat-sharing and they talk about it. The idea of sharing is
Ziker, 2002). In groups that split apart (fission) to forage as individuals deeply implanted and very successfully imposes its restraints…. The
or in smaller groups and then bring foods back to camp to share (e.g. idea of eating alone is shocking to the !Kung. It makes them shriek
central place provisioners) (Marlowe, 2006), it is logistically difficult with an uneasy laughter. Lions could do that, they say, not men."
for researchers to record the behavior of those in camp and those Nevertheless, many ethnographers report hunters eating spoils be-
foraging out of camp at the same time. It is probably largely due to fore returning to camp, including the Ache, Aka, Batek, G/Wi, Lengua,
these logistical problems that studies of food sharing in central Mbuti, Nukak, and !Kung (Endicott 1988; Grubb, 1911:190; Lee, 1979;
places are so much more common than studying out of camp behavior Marshall, 1976; Politis, 2009; Silberbauer, 1981; Walker & Hewlett,
(e.g. Bahuchet, 1990; Bird & Bird, 1997; Gurven, 2004; Kaplan, Hurtado, 1990). However, analyses of producer generosity and patterns of shar-
& Hill, 1990; Speth, 1990; but see Crittenden (2013) as a noteworthy ing, to date, have not systematically taken this out of camp eating into
exception). In fact, these studies are so common that students of account. The difficulties of systematically capturing these data are
sometimes lamented (e.g. Politis, 2009; Speth, 1990). In other cases,
ethnographers document total quantities of foods acquired but do not
⁎ Corresponding author. mention whether any of the foods were consumed before foragers
E-mail address: [email protected] (J.C. Berbesque). returned to camp (Endicott, 1988; Hart, 1978).

http://dx.doi.org/10.1016/j.evolhumbehav.2016.01.003
1090-5138/© 2016 Elsevier Inc. All rights reserved.
282 J.C. Berbesque et al. / Evolution and Human Behavior 37 (2016) 281–286

Patterns of eating while out of camp may lead to small or even large providing no direction whatsoever to the Hadza about where or how
corrections to estimates of the total diet of hunter–gatherers. These data to forage or behave during any of the observations.
also inform studies of food sharing practices in camp. Decisions to share Focal individuals selected using simple random sampling without
foods, or to request foods from others, are necessarily affected by how replacement, with the goal of following all males in residence in each
hungry or satiated individuals are when they return to camp. Thus, camp at least one time, regardless of whether the focal individual was
studies that are based solely on in-camp behavior can provide only alone or in a group. In contrast to some other ethnographically docu-
part of the larger picture of the diet and food sharing practices of central mented hunters (e.g. Alvard, 2002; Hill, 2002), Hadza men very often
place provisioners. For example, Hadza men have been documented forage alone (Marlowe, 2010). A total of 118 follows were conducted,
eating on average only 8% (median 0%) of the total caloric value of most of which were of men foraging alone, though in 13 cases (11% of
foods they brought back to camp (Wood & Marlowe, 2013). In the follows), more than one person was present (group foray) and data
absence of information on out of camp eating, this paints an unrealistic were also collected on non-focal individuals. Data on non-focal individ-
picture of food distribution and overall diet. Men eating while out of uals in a group follow were only analyzed when all group members
camp sheds light upon their patterns of sharing when in camp. Not con- were present and observed throughout the foray. Due to group follows,
sidering data on out of camp eating would lead to overestimations of these 118 follows constitute our focal sample of 146 person/follows. The
both the failure rate of men's foraging decisions, and the degree to mean number of men present in group focal follows (as opposed to fol-
which their energetic budgets are subsidized by others. These data call lows of a single individual was 4.6 (mode = 3, maximum = 8). Our focal
attention to the fact that characterizing patterns of diet by sex, age, mar- follow data consists of totals 146 person/day follows (921 hours total)
ital status, or other factors should involve careful considerations of how collected over a period of 12 years from 2001 to 2013, with follows in
individual diets vary across space, relative to where researchers make every region of Hadzaland and in every season (see Table 1 for a break-
their observations. Our analysis indirectly bears on previous interpreta- down of follows by region and season). The average duration of follows
tions of data on patterns of food distribution in hunter–gatherers, and was 6.3 hours per foray, with a range of 30 minutes for the shortest
has consequences for our understanding of the evolution of hunting. foray to 770 minutes (or 12.8 hours) for the longest foray. On average,
each of the 75 men followed was observed 1.95 times (median = 1
2. Materials and Methods and mode = 1), with a range of 1–9 observation days per man. However,
only 8 (11%) of the 75 men were followed on more than three person
2.1. Subject Population days, and many of these repeat follows of the same individual happened
in different years. The men followed ranged in age from 16 to 59 years
The Hadza are a group of traditional, central-place hunter–gatherers old, with a mean age of 35 years (median = 34 years, mode =
who number approximately 1,000, however only approximately 250 in- 41 years). Most forays (90%) lasted 2 hours or longer.
dividuals still acquire the majority of their diet by foraging. They live in a Amounts (kilograms) of foods eaten on focal follows were estimated
savanna–woodland habitat that encompasses about 4,000 km2 around using methods similar to those outlined by Rothman, Chapman, and
Lake Eyasi in northern Tanzania. They live in mobile camps, averaging Van Soest (2012). This entails visual estimation of units of foods con-
30 individuals per camp (Marlowe, 2006). Camp membership often sumed (e.g. three handfuls of berries) and the collection of correspond-
changes as people move in and out of camps (Blurton Jones, Hawkes, ing data that allows one to estimate the weight of such units (e.g. the
& O'Connell, 2005). Hadza camps move about every 6 weeks, on average
(Marlowe, 2010).
While foraging, Hadza men typically search for animals, honey, and Table 1
Person/Follows by Region and Season.
sometimes fruit. Hadza men rarely dig for tubers, which is a task that
women and sometimes children specialize in. They typically go on Year Region Season Camp Follows
walkabout every day, and they usually go alone. They hunt birds and 2001 Dunduyia Early dry Sungu 2
mammals using only bow and arrows. In large game kills poison arrows 2002 Tli'ika Early wet Bashana 3
are used, whereas with small game poison is not used. They always have 2002 Tli'ika Early dry Gibanola 2
2003 Siponga Early wet Sedaiko 4
their bow and arrows with them, even when they carry an ax to access
2003 Tli'ika Early dry Sangeli 3
honey (Marlowe, 2010). 2004 Siponga Early wet Sedaiko 13
The Hadza diet can be conveniently categorized into six main food 2004 Tli'ika Early dry Kisanakwipi 8
types: honey, meat, berries, baobab (Adanosia digitata), and tubers, 2004 Tli'ika Late dry Sanola 1
and in one region only, marula nuts (Sclerocarya birrea). The berries 2005 Dunduiya Early dry Mayai 13
2005 Dunduiya Late dry Wamkwimba 3
that the Hadza eat consist mostly of seed encased in a small amount of 2005 Mangola Early dry Gola 6
high-sugar pulp. Baobab fruit is common across much of Africa, and it 2005 Mangola Late dry Gola 1
is a major food in terms of kilocalories and kilograms in the Hadza 2005 Siponga Early wet Tuwa 1
diet. Many tubers are continuously available throughout the year, and 2005 Siponga Early wet Siponga 1
2005 Siponga Late wet goandeka 1
are a source of carbohydrates and an important ‘fallback food’ for the
2005 Siponga Late dry Tuwa 22
Hadza (Marlowe & Berbesque, 2009). 2005 Tli'ika Early dry Gangidape 4
2005 Tli'ika Early dry Bashana 4
2.2. Procedure 2006 Mangola Late wet Gola 4
2006 Mangola Early dry Gola 10
2006 Tli'ika Early wet Kisanakwipi 7
Men were followed on walkabout, their behaviors were continuous-
2006 Tli'ika Late wet Lelangidako 9
ly recorded from the time they departed camp to the time they returned 2006 Tli'ika Late dry Hukumako 12
to camp. Men usually begin their day of foraging early, between 6 and 2009 Han!abe Late dry Setako 4
7 am. Hadza men forage opportunistically and even if they have a partic- 2010 Tli'ika Late wet Sangeli 1
2010 Tli'ika Early dry Sangeli 4
ular goal in mind, such as looking for bee nests in a particular stand of
2011 Tli'ika Early wet Sangeli 1
trees, they are generally alert for other foraging opportunities. The re- 2012 Tli'ika Early wet Sangeli 1
searcher followed approximately 5–10 meters behind the focal individ- 2013 Tli'ika Early dry Nyalaida 1
ual(s), recording a variety of behavioral data, including every instance in 12 Years 5 Regions 4 Seasons 29 Camps 146 Follows
which they ate foods. While observing the Hadza, the researcher walked In this study, we define a camp by both the geographic location and the season of research-
as silently as possible, attempting to minimize observer effects, and er presence.
 J.C. Berbesque et al. / Evolution and Human Behavior 37 (2016) 281–286 283

number of berries in an average hand full). Honey acquisition was esti- Amounts of foods, as well as types of foods eaten out of camp varied
mated using units of volume that were intuitive for the purposes of vi- by season (see Fig. 2 and Table 2), with more honey and small game
sual estimation (e.g. teaspoons, tablespoons, and golf-ball sized bites eaten in the wet seasons and more large game eaten in the dry seasons.
of honey), following Wood and Marlowe (2014). Kilocalories of Hadza In terms of overall contribution of calories throughout the year, honey
foods were calculated using formulae including the percentage of was by far the food type most frequently eaten by men out of camp,
water in each food and the caloric values per 100 grams (dry weights) whether analyzed in number of observations or in kilocalories
of each food. (see Table 2). Of all forays where any foods were acquired, honey was
Nutritional values for honey, berries, and baobab were based on pre- eaten in 58% of those follows, and contributed 85% of the total kilocalo-
vious studies of Hadza foods (Murray, Schoeninger, Bunn, Pickering, & ries eaten out of camp.
Marlett, 2001; Schoeninger, Bunn, Murray, & Marlett, 2001). Large The number (and percentage) of person/days out of camp where a
game animals have more body fat than small game (Cordain et al., particular food type was eaten, the range of kilocalories eaten, the
2000), thus two caloric estimates of kcal/weight for animal carcasses mean amount of kilocalories eaten by food type, and the mean hourly
were used. To estimate the caloric value of small animal carcasses consumption rates by food type are presented in Table 3 (but see Ap-
(used for all animals under 32 kilograms), the kcal/kg values from stud- pendix 1 available on the journal's website at www.ehbonline.org for
ies of New Zealand White rabbits Octolagus cuniculus (Daszkiewicz, unusual eating events not listed in Table 3). More than a single food
Gugołek, Janiszewski, Kubiak, & Czoik, 2012) were used, and in the type was eaten on many forays, although in some forays no food was ac-
case of large game, N32 kilograms, estimates based on white tailed quired. Thus, the total (166 events of food acquisition) is more than the
deer carcasses Odocoileus virginianus were used (Weiner, 1973). number of person/follows (n = 146).
Neither data on duration of foray nor the sum of kilocalories eaten For each food type, we examined the ratio of what was consumed on the
per foray were normally distributed variables, so we used non- spot relative to the total amount acquired (consumed / consumed + brought
parametric tests, except when analyzing repeated measures, when we back to camp). Although honey was frequently encountered (and
used log transformations of non-normal variables in GLM models. We eaten) while out of camp, the foods with the highest percentage of kilo-
used SPSS version 21 for all analyses. calories consumed on the spot rather than taken back to camp were
berries (99%), followed by honey (84%), and finally baobab (63%). The
average percentage of meat from complete animal carcasses that was
3. Results immediately consumed by individuals while out of camp differed great-
ly by the carcass size of the animal (small game = 55% versus large
There was a great deal of variation in kilocalories eaten while on game b 1%) (see Figs. 3 and Appendix 2 available on the journal's
walkabout, with a range from 0 to 22,007 kilocalories consumed by a website at www.ehbonline.org).
single individual on a single foray. The median number of kilocalories Forays in which small game animals were acquired were most com-
consumed per foray was 910, with a mean of 2,405, and standard monly individual forays (mean = 1.9 individuals, median = 1). In con-
deviation of 3,637 kilocalories. The data were highly positively skewed trast, the majority of forays during which large game animal meat was
(see Fig. 1), with a skewness of 2.693 (S.E. = .201). The interquartile eaten were forays that were intended to retrieve a large game animal
range of kilocalories acquired per person/day was 37–3047. Men carcass killed in a previous foray (which are much more often solo).
consumed 1,000 kilocalories or more in 55% of forays observed, less All five forays in which meat from large game animal carcasses was
than 500 kilocalories in 45%, and zero kilocalories in only 20% of forays eaten were collaborative group forays (range = 2–7 people, mean =
(see Fig. 1). 4.2, median = 4), with several Hadza sharing portions of the carcass.

Fig. 1. Histogram of kilocalories consumed per foray. Fig. 2. Median kilocalories eaten on foray by season.
284 J.C. Berbesque et al. / Evolution and Human Behavior 37 (2016) 281–286

Table 2
Descriptive Statistics of Food Types Eaten by Season.

Season Food type Median Mean SD range IQR Skewness S.E. skewness N

Early wet Honey 2488 3933 4380 194–17117 194–5874 1.655 0.501 21
Small game 2342 2068 587 748–2342 2056–2342 −1.92 0.687 10
Large game NA NA NA NA NA NA NA 0
Baobab 89 89 NA NA NA NA NA 1
Berries 912 830 603 190–912 NA −0.601 1.225 3
Late wet Honey 4713 6015 6008 97–20777 1114–8969 1.4 0.564 16
Small game 165 554 820 0–1496 NA NA NA 3
Large game 588 588 NA NA NA NA NA 1
Baobab 99 99 NA 0–198 NA NA NA 2
Berries 63 197 356 2–1231 31–257 2.903 0.661 11
Early dry Honey 2197 3087 3131 194–14039 748–4098 1.792 0.388 37
Small game 0 0 NA NA NA NA NA 1
Large game 600 524 482 75–1236 75–935 0.68 0.913 5
Baobab 272 272 NA 247–297 NA NA NA 2
Berries 65 73 67 5–216 29–65 1.658 0.637 12
Late dry Honey 798 992 894 0–3178 397–1339 1.772 0.687 10
Small game 157 1073 2067 0–7590 0–1289 2.797 0.597 14
Large game 150 150 NA NA NA NA NA 1
Baobab 0 60 103 0–179 NA NA NA 3
Berries 60 80 82 10–288 15–109 1.637 0.637 12

Foods with a minimum of zero are foods that were acquired, however they were not eaten. Foods not acquired are not shown.

Very small amounts of meat from large game were shared by the ‘recov- the more absolute kilocalories of honey that were acquired, the more ki-
ery team’, which helps to carry most of the meat back to camp. When localories of honey were immediately eaten (ρ = .904, n = 84 cases of
stalking is involved, the ‘recovery team’ most commonly only involves people acquiring honey, p b .0001).
men; however, when the carcass location is known, this group can con- In a Spearman's rank correlation, longer forays were not significantly
sist of men, women and children old enough to help. associated with more kilocalories being eaten out of camp, whether or
There were a total of 24 animal acquisitions. Nineteen out of the 24 not we included forays in which no foods were eaten (ρ = .000, n of
(or 79%) of the game animals acquired in our sample were small game forays where foods were eaten = 116, p = .998;ρ = −.067, n of person
animals. In a Spearman's correlation, more kilocalories of meat acquired forays = 146, p = .421). Longer forays were also not associated with a
(larger carcass size) was associated with a smaller percentage of the higher percentage of kilocalories eaten on the spot versus taken back
animal eaten before bringing the remainder back to camp to share to camp (ρ = .062, n of forays where foods were eaten = 116,
(ρ = − .334, n = 35 people eating animal carcasses, p = .050) (note: p = .509). In univariate GLM models controlling for repeated focal indi-
this includes multiple individuals eating on group foray acquisitions). vidual observations, we found no association between the age of males
Additionally, in absolute terms, there was no clear association between and the amount of kilocalories eaten (log transformed to correct for
kilocalories of meat acquired and the amount of kilocalories immediate- heteroscedasticity) while out of camp per foray (F = 1.013, df = 85,
ly eaten per person present (ρ = .216, n = 35 instances of meat eating, p = .503) (see Fig. 3). There was also no significant association found
p = .214). However, when analyzing large and small game categories between the age of the males and the percent of kilocalories eaten im-
separately, more kilocalories were eaten in small game animals when mediately on the foray versus returned to camp by (F = 1.034, df =
these animals were larger (ρ = .473, n = 27 instances of people eating 85, p = .475), or in the duration of forays by age (F = 1.303, df = 100,
small game, p = .013). However, within the category of large game ani- p = .161).
mals (weighing over 32 kilograms) there was no association between
the number of kilocalories acquired (the mass of the animal) and the 4. Discussion
number of kilocalories immediately eaten (ρ = −.077, n = 8 instances
of people eating large game, p = .857). Most Hadza men consumed a substantial amount of kilocalories
Similarly, smaller percentages of honey were eaten when more while foraging. The mean number of kilocalories consumed by males
honey was acquired (ρ = − .572, n = 84 cases of people acquiring per foray was 2,404. The mean daily total energy expenditure (TEE)
honey, p b .001). In contrast with how large game is consumed, however, for Hadza men has been measured to be 2,649 ± 395 (range

Table 3
Acquisition and Consumption Rates on Foray by Food Type.

Food Mean/Median Mean/Median SD kcal Mean kcal eaten/hour Range† kcals IQR kcals IQR kcals N person forays Mean foray
type kcal acquired kcal eaten/foray acquired/eaten eaten/foray acquired eaten acquired/% duration
person forays (minutes)

Honey 6822/2515 3582/2398 10371/4191 614 97⁎–20776 726–6393 559–4784 84/58% 350
Meat (total) 26364/2342 1164/628 90481/1486 205 75–7590 299–5236 75–2342 35/24% 341
Small game 10211/2342 1335/942 31130/1610 206 90⁎–7590 355–2536 22–2342 28/19% 389
Large game⁎ 90974/1236† 480/588 90974/420 196 75–1236 75†–64627 75–633 7/5% 147†
Berries 181/64 171/64 323/314 21 2–1388 29–172 29–172 38/26% 483
Baobab 655/220 126/134 1271/120 15 89⁎–297 181–284 0–234 8/5% 491
⁎ For categories indicated, range does not include values of zero. In the other categories (not indicated) instances of individuals acquiring a food type without eating any of it are included
in this analysis, but we have chosen to report the minimum (non-zero) value eaten on foray.
†
The large game category includes eating of large game as gifts of meat from others, in which cases the amount acquired will be substantially less than the body size of the animal. We
analyze percent of carcasses eaten later in this manuscript.
 J.C. Berbesque et al. / Evolution and Human Behavior 37 (2016) 281–286 285

women rarely eat meat out of camp. Given the amount of both meat
and honey eaten on foray by Hadza men, it is likely they are eating a
far more energy-dense diet than are Hadza women. Substantial sex dif-
ferences in the Hadza diet have recently been documented in in-camp
eating frequencies (Berbesque et al., 2011), resulting in sex differences
that have been documented in Hadza dental wear patterns
(Berbesque et al., 2012), as well as in the gut microbiome (Schnorr
et al., 2014). The sex-difference in foraging patterns and return rates
can even be seen in Hadza childhood (Crittenden, Conklin-Brittain,
Zes, Schoeninger, & Marlowe, 2013). Sex differences in many hunter–
gatherer diets may be greater than previously appreciated, as most re-
ports on hunter–gatherer diets are from the in camp portion of diet
only—which is the portion of the diet that is taken back to camp and is
more likely to be shared or used for provisioning. It is important to
note that major cross-cultural studies of the diets of hunter–gatherers
(e.g. Cordain, Eaton, Miller, Mann, & Hill, 2002), do not report out of
camp eating, nor do they discriminate between the diets of men and
women. These are almost always based on per capita estimates of
foods arriving in camp, and assume perfect sharing among all
adults—which, in the case of the Hadza, is grossly inaccurate.
Some researchers have described Hadza men as targeting large
game animals to the exclusion of small game or other foods in order
to signal their phenotypic quality to potential mates (Bunn & Gurtov,
2014; Hawkes, O'Connell, Blurton Jones, Oftedal, & Blumenschine,
1991; Hawkes, O'Connell, & Jones, 2001). However, in our analysis,
Fig. 3. Boxplot of kilocalories acquired versus eaten. 79% of the kills made by adult men (ages 16 years and older) were
small game animals (weighing less than 32 kg). This is consistent with
data from Wood and Marlowe (2013, 2014), which show that 79%
2,008–3,363) kilocalories per day (Pontzer et al., 2012). Using the aver- of the animals that men brought back to seven different camps weighed
age values for men's consumption/foray and TEE/day, we estimate that less than 10 kg. Thus Hadza men should not exclusively be considered
men are consuming approximately 90% of their TEE on average while large game specialists.
foraging out of camp. Of course, there is significant variation in these Understanding men's provisioning and sharing with others are im-
values across forays, with only 20% of forays where no food was ac- portant components of understanding foraging decisions and family
quired. Also, bear in mind that upon returning to camp, Hadza men structures, but choices related to prey selection are also influenced by
eat foods brought by other men as well as women—so much so that re- the goal of eating—which has not been well-documented to this point.
searchers have represented Hadza men's diets using only in-camp data Hadza men's foraging is driven by the goals of getting enough calories
to this point. Patterns of out of camp consumption are rarely (if ever) re- to eat and potentially to provision their families. In the Hadza ecology,
ported in other central place provisioning groups, yet these values for hunting is not an exclusive activity that entails a high rate of foraging
Hadza men show that out of camp eating constitutes a significant per- failure, requiring men to be fed by others upon returning to camp
cent of their average TEE. empty handed. More accurately, hunting is scheduled in a way that
Hadza men's high rate of eating while out of camp complements re- also permits high levels of energetic self-sufficiency. Men take advan-
cent research that examined Hadza men's sharing of foods in camp tage of a suite of high-quality foods – especially honey – and this strat-
(Wood & Marlowe, 2013), which showed that men consumed little of egy allows them to both feed themselves and pursue riskier food types
the food they brought back to camp themselves, but instead shared that have higher chances of failure upon pursuit.
the foods with their wives, children, and co-resident kin. This sharing An adjustment of the characterization of the riskiness of men's forag-
in camp is less costly to men when they have already fed themselves ing strategies may be in order—since coming back to camp empty-
while foraging out of camp. handed does not always mean that their strategy to acquire food failed,
Hadza men's diet while foraging is strikingly different from their diet only that it failed to produce enough surplus to return to camp to share.
in camp. Honey contributes the most kilocalories to the out of camp diet
of men, followed by meat. For example, only 14% of kilocalories brought 5. Conclusions
into camp are from honey (Marlowe, Berbesque, Wood, Crittenden, &
Porter, 2014), while honey contributes 85% of the kilocalories consumed Descriptions of contemporary hunter–gatherers have played a cen-
out of camp by men. Honey has recently gained attention as a potential- tral role in models of past human forager societies, and their evolution.
ly important source of energy in human evolution (Crittenden, 2011; Our understanding of the evolution of human life history partly depends
Marlowe et al., 2014; McGrew, 2001; Wood, Pontzer, Raichlen, & Mar- on knowing possible sources of bias in data on contemporary hunter–
lowe, 2014; Wrangham, 2011), and it is the most preferred food of the gatherer populations. There are several aspects of this study that bring
Hadza (Berbesque & Marlowe, 2009). A recent study using a large to light possible biases in our understanding of men's production.
cross-cultural database showed that most hunter–gatherers in warm First, if small game is more likely to be eaten on the spot rather than
climates exploit honey (Marlowe et al., 2014). Given this ubiquity of brought back to camp to be shared (and thus documented), perhaps
honey consumption, and the possibility that honey consumption in the characterization of men as large-game hunters to the exclusion of
these populations may be under-reported due to it being consumed small game is exaggerated in the archaeological record. Second, if our
out of camp, the importance of honey as an energy source may be study population is comparable to other warm climate populations,
underestimated generally for hunter–gatherers. honey consumption is undoubtedly under-reported in ethnographies,
Hadza men have been documented eating meat in camp significant- and owing to preservation bias, is also practically invisible to the archae-
ly more frequently than are women (Berbesque, Marlowe, & Crittenden, ologist or paleontologist studying fossilized materials. Third, Hadza men
2011), and meat constitutes 11% of men's out of camp diet—whereas (and perhaps any solitary foragers) bring the spoils of their foraging
286 J.C. Berbesque et al. / Evolution and Human Behavior 37 (2016) 281–286

back to camp, whether to provision others or to signal, only after they Gurven, M. (2004). Reciprocal altruism and food sharing decisions among Hiwi and Ache
hunter–gatherers. Behavioral Ecology and Sociobiology, 56, 366–380.
have already eaten most (if not all), of what they need. As these Hart, J. A. (1978). From subsistence to market: A case study of the Mbuti net hunters.
data show, out of camp food consumption has real consequences for Human Ecology, 6, 325–353.
our understanding of hunter–gatherer diets and the dynamics of food Hawkes, K., O'connell, J. F., Blurton Jones, N. G., Oftedal, O. T., & Blumenschine, R. J. (1991).
Hunting income patterns among the Hadza: big game, common goods, foraging goals
sharing systems. and the evolution of the human diet [and discussion].". Philosophical Transactions of
the Royal Society of London. Series B, Biological Sciences, 334, 243–251.
Supplementary Materials Hawkes, K., O'Connell, J. F., & Jones, N. B. (2001). Hadza meat sharing. Evolution and
Human Behavior, 22, 113–142.
Hill, K. (2002). Altruistic cooperation during foraging by the Ache, and the evolved human
Supplementary data to this article can be found online at http://dx. predisposition to cooperate. Human Nature, 13(1), 105–128.
doi.org/10.1016/j.evolhumbehav.2016.01.003. Kaplan, H., Hill, K., Hawkes, K., & Hurtado, A. (1984). Food sharing among Ache hunter-
gatherers of eastern Paraguay. Current Anthropology, 25, 113–115.
Kaplan, H., Hurtado, A. M., & Hill, K. (1990). Risk, foraging and food sharing among the
Acknowledgements Ache. In E. Cashdan (Ed.), Risk and Uncertainty in Tribal and Peasant Economies
(pp. 107–143). Boulder: West-view Press.
Kitanishi, K. (1998). Food sharing among the Aka hunter-gatherers in northeastern
We would like to thank the Leakey Foundation, (to CB and BW), Congo. African Study Monographs, 25, 3–32.
Wenner Gren (to BW), and NSF (1062879 to BW) and (0544751, Lee, R. B. (1979). Men, women, and work. Cambridge: Cambridge University Press.
0242455, and 0650574 to FWM). We would also like to thank the John Marlowe, F. W. (2006). Central place provisioning: The Hadza as an example. In M.
Robbins, C. Boesch, & G. Hohmann (Eds.), Feeding ecology in apes and other primates
Templeton Foundation for their support for CB. The opinions expressed (pp. 359–377). Cambridge: Cambridge University Press.
in this publication are those of the authors and do not necessarily reflect Marlowe, F. W. (2010). The Hadza: Hunter-Gatherers of Tanzania. Berkeley: University of
the views of the John Templeton Foundation. We thank Costech for per- California Press.
Marlowe, F. W., & Berbesque, J. C. (2009). Tubers as fallback foods and their impact on
mission to conduct research. We would also like to thank Andrea Hadza hunter-gatherers. American Journal of Physical Anthropology, 140, 751–758.
Migliano, and an anonymous reviewer for their helpful suggestions. We http://dx.doi.org/10.1002/ajpa.21040.
would like to thank Daudi and Trude Peterson and Carla Wood Mallol Marlowe, F. W., Berbesque, J. C., Wood, B., Crittenden, A. N., & Porter, C. (2014). Honey,
hunter-gatherers, Hadza, and human evolution. Journal of Human Evolution, 71,
for all of their help and support, and Todd C. Rae and Peggy Berbesque 119–128. http://dx.doi.org/10.1016/j.jhevol.2014.03.006.
for their comments on drafts. Finally, we thank the Hadza for their endur- Marshall, L. (1976). The !Kung of Nyae Nyae. Cambridge: Harvard University Press.
ing tolerance and hospitality. Marshall, L. (1998). Sharing, talking, and giving: Relief of social tensions among the
!Kung. In J. M. Gowdy (Ed.), Limited wants, unlimited means: A reader on hunter-gatherer
economics and the environment, (pp. 65–85). Washington D. C.: Island Press.
References McGrew, W. (2001). The other faunivory. In C. B. Stanford, & H. T. Bunn (Eds.), Meat-
eating and human evolution. Oxford: Oxford University Press.
Alvard, Michael S. (2002). Carcass ownership and meat distribution by big-game cooper- Murray, S. S., Schoeninger, M. J., Bunn, H. T., Pickering, T. R., & Marlett, J. A. (2001). Nutri-
ative hunters. Research in Economic Anthropology, 21, 99–132. tional composition of some wild plant foods and honey used by Hadza foragers of
Bahuchet, S. (1990). Food sharing among the pygmies of Central Africa. African Study Tanzania. Journal of Food Composition and Analysis, 14, 3–13.
Monographs, 11, 27–53. Politis, G. (2009). Nukak: Ethnoarchaeology of an Amazonian people. Walnut Creek, CA: Left
Berbesque, J. C., & Marlowe, F. W. (2009). Sex differences in food preferences of the Hadza Coast Press.
hunter-gatherers. Evolutionary Psychology, 7, 601–616. Pontzer, H., Raichlen, D. A., Wood, B. M., Mabulla, A. Z. P., Racette, S. B., & Marlowe, F. W.
Berbesque, J. C., Marlowe, F. W., & Crittenden, A. N. (2011). Sex differences in Hadza eat- (2012). Hunter-Gatherer energetics and human obesity. PloS One, 7, e40503.
ing frequency by food type. American Journal of Human Biology, 23, 339–345. http:// Rothman, J. M., Chapman, C. A., & Van Soest, P. J. (2012). Methods in primate nutritional
dx.doi.org/10.1002/ajhb.21139. ecology: A user’s guide. International Journal of Primatology, 33, 542–566.
Berbesque, J. C., Marlowe, F. W., Pawn, I., Thompson, P., Johnson, G., & Mabulla, A. (2012). Schnorr, S. L., Candela, M., Rampelli, S., Centanni, M., Consolandi, C., Basaglia, G., ...
Sex differences in Hadza dental wear patterns. Human Nature, 23, 270–282. Crittenden, A. N. (2014). Gut microbiome of the Hadza hunter-gatherers. Nature
Bird, R. L. B., & Bird, D. W. (1997). Delayed reciprocity and tolerated theft - The behavioral Communications, 5.
ecology of food-sharing strategies. Current Anthropology, 38, 49–78. Schoeninger, M. J., Bunn, H. T., Murray, S. S., & Marlett, J. A. (2001). Composition of tubers
Blurton Jones, N. G., Hawkes, K., & O'Connell, J. F. (2005). Older Hadza men and women as used by Hadza foragers of Tanzania. Journal of Food Composition and Analysis, 14,
helpers: Residence data. In B. S. Hewlett, & M. E. Lamb (Eds.), Hunter-gatherer child- 15–25.
hoods: Evolutionary, developmental and cultural perspectives (pp. 214–236). New Silberbauer, G. (1981). Hunter/gatherers of the Central Kalahari. In R. Harding, & G. Teleki
Brunswick: Transaction. (Eds.), Omnivorous primates. Columbia University Press.
Bunn, H. T., & Gurtov, A. N. (2014). Prey mortality profiles indicate that Early Pleistocene Speth, J. D. (1990). Seasonality, resource stress, and food sharing in so-called “egalitarian”
Homo at Olduvai was an ambush predator. Quaternary International, 322, 44–53. foraging societies. Journal of Anthropological Archaeology, 9, 148–188.
Cordain, L., Eaton, S. B., Miller, J. B., Mann, N., & Hill, K. (2002). Fatty acid analysis of wild Walker, P. L., & Hewlett, B. S. (1990). Dental health, diet, and social status among Central
ruminant tissues: Evolutionary implications for reducing diet-related chronic disease. African foragers and farmers. American Anthropologist, 92, 383–398.
European Journal of Clinical Nutrition, 56, 181–191. Weiner, J. (1973). Dressing percentage, gross body composition and caloric value of the
Cordain, L., Miller, J. B., Eaton, S. B., Mann, N., Holt, S. H., & Speth, J. D. (2000). Plant-animal roe deer. Acta Theriologica, 18, 209–222.
subsistence ratios and macronutrient energy estimations in worldwide hunter- Wood, B., & Marlowe, F. (2013). Household and kin provisioning by Hadza men. Human
gatherer diets. The American Journal of Clinical Nutrition, 71, 682–692. Nature, 24, 280–317.
Crittenden, A. N. (2011). The importance of honey consumption in human evolution. Food Wood, B., & Marlowe, F. (2014). Toward a reality-based understanding of Hadza men's
Foodways, 19, 257–273. work: A response to Hawkes et al. Human Nature, 25, 620–630.
Crittenden, A. N., Conklin-Brittain, N. L., Zes, D. A., Schoeninger, M. J., & Marlowe, F. W. Wood, B. M., Pontzer, H., Raichlen, D. A., & Marlowe, F. W. (2014). Mutualism and manip-
(2013). Juvenile foraging among the Hadza: Implications for human life history. ulation in Hadza–honeyguide interactions. Evolution and Human Behavior, 35,
Evolution and Human Behavior, 34, 299–304. 540–546.
Daszkiewicz, T., Gugołek, A., Janiszewski, P., Kubiak, D., & Czoik, M. (2012). The effect of Wrangham, R. W. (2011). Honey and fire in human evolution. In D. Pilbeam, J. Sept, & C.
intensive and extensive production systems on carcass quality in New Zealand Vogel (Eds.), Casting the Net Wide: Papers in Honor of Glynn Isaac and His Approach to
white rabbits. World Rabbit Science, 20, 25–32. Human Origins Research (pp. 149–167). Oxford: Oxbow Books.
Endicott, K. (1988). Property, power and conflict among the Batek of Malaysia. Hunters Ziker, J. (2002). Raw and cooked in Arctic Siberia: Diet and consumption strategies in
and gatherers 2, 110–127. In T. Ingold, D. Riches, & J. Woodburn (Eds.), Hunter-gath- socio‐ecological perspective. Nutritional Anthropology, 25, 20–33.
erers. Property, power and ideology, vol. II, . Berg Berg. Ziker, J. P. (2007). Subsistence and food sharing in Northern Siberia: Social and nutritional
Grubb, W. B. (1911). An unknown people in an unknown land. London: Seeley. ecology of the Dolgan and the Nganasan. Ecology of Food and Nutrition, 46, 445–467.