International Journal of Osteoarchaeology
Int. J. Osteoarchaeol. (2014)
Published online in Wiley Online Library
(wileyonlinelibrary.com) DOI: 10.1002/oa.2422
Neonatal Mortality, Young Calf Slaughter
and Milk Production during the Early
Neolithic of North Western Mediterranean
R. GILLIS,a* I. CARRÈRE,b M. SAÑA SEGUÍ,c G. RADId AND J-D. VIGNEa
a
CNR Muséum national d’Histoire naturelle, UMR 7209, Archéozoologie, Archéobotanique: sociétés,
pratiques et environnements, Dép. Ecologie et Gestion de la Biodiversité, Paris, France
b
CNRS-UMR 5608, ‘Travaux et Recherches Archéologiques sur les Cultures, les Espaces et les Sociétés’
(TRACES), Ecole des Hautes Etudes en Sciences Sociales, Université Le Mirail, Toulouse, France
c
Departament de Prehistòria Edifici B Facultat de Filosofia i Lletres, Bellaterra, Barcelona, Spain
d
Department of Civilisations and Forms of Knowledge, Pisa, Italy
ABSTRACT
The North-Western Mediterranean witnessed a rapid expansion of farmers and their livestock during the Early Neolithic period. Depending on the region, cattle played a more or less important role in these communities; however
how these animals were exploited for their milk is not clear. Here we investigate calf mortality to determine indirectly
whether cattle dairying was practised by Early Neolithic stock herders. Age-at-death (AtD) frequencies for calves
from two sites: Trasano (Italy, Impressa culture: 7–6th millennium BC) and La Draga (Spain, Cardial culture: 6th
millennium BC) were estimated from dental eruption and development stages, and measurements of un-fused
post-cranial material. Adult age classes are well represented in the dental AtD frequencies and were interpreted
as the result of the slaughter of prime beef and retired lactating females. For calves aged less than 12 months,
there was no statistical difference in the AtD frequencies based on dental and post-cranial material indicating that
the data is a good representation of the mortality patterns of calves, either natural or deliberate. At both sites
there was a strong mortality peak at 3–6 months in all AtD profiles. At La Draga, this peak was clearly
differentiated from a peak at 0–1 month, which can be interpreted neonatal mortality possible a consequence of the birthing season coinciding with the end of winter during more humid climatic conditions
that at present. The deliberate slaughter peak around 3–6 months is discussed, and we propose that
stock herders controlled the mortality of infant classes, possibly in response to variable external environment pressures while maintaining animal productivity. Copyright © 2014 John Wiley & Sons, Ltd.
Key words: cattle; dairy husbandry; early neolithic; mortality frequencies; neonatal mortality; North-Western
Mediterranean; young calf slaughtering
Introduction
The status of milk in prehistoric societies has long been
debated (Chapman, 1982; Sherratt, 1983, 1997). However in the last 20 years, ceramic residue studies have
demonstrated without a doubt that humans have processing milk since the beginning of the Neolithic
(Evershed et al., 2008). The publication of Payne
(1973) subsistence models and further reference
ethnographic scenarios (Vigne & Helmer, 2007) have
provided means for archaeozoologists to study dairy
husbandry. These models, however, were based on
* Correspondence to: Rosalind Gillis, CNRS—Muséum national d’Histoire
naturelle, UMR 7209, Archéozoologie, archéobotanique: sociétés, pratiques
et environnements, Dép. Ecologie et Gestion de la Biodiversité, CP 56, 55
rue Buffon, F-75005 Paris, France.
e-mail: [email protected]
Copyright © 2014 John Wiley & Sons, Ltd.
small stock; all dairy species have different physiological reactions to the removal of the young (Balasse,
2003). The presence of the calf has been argued to
be necessary for Neolithic dairy husbandry (CluttonBrock, 1981; Peske, 1994; Balasse et al., 1997; Balasse
& Tresset, 2002; Balasse, 2003). This is based on the
fact that non-specialised cows can become inhibited
to let-down milk without the presence of the calf
(Labussière, 1999; Balasse, 2003); consequently if milk
is exploited, it is shared between calves and humans.
However, at Grimes Graves (Middle Bronze Age,
UK), calves were culled between 2 and 3 months,
which was interpreted as the result of dairy husbandry
(Legge, 1981, 1992). This is similar to the present industrial systems where calves are removed from their
mothers at birth or within the first weeks of birth.
Then, depending on their sex, either culled or kept
Received 28 April 2014
Revised 21 August 2014
Accepted 19 September 2014
R. Gillis et al.
for breeding stock or beef production. If cows are
inhibited by the removal/death of the calf, there
are methods to stimulate milk production such as
creating a dummy using the dead calf skin, blowing
into the uterus (Balasse, 2003; Le Quellec, 2011) or
providing a live calf substitute (Vigne & Helmer,
2007). There is long history of these techniques
being used in Europe from the medieval period onwards (Lucas, 1989). Consequently, it is possible
prior to the development of specialised dairy breeds
that calves could have been slaughtered during the
lactation period by stock herders to increase milk
supply for human consumption.
The role of cattle in the Impressa, and the FrancoIberian Cardial cultures of North-Western Mediterranean are only partially known (Rowley-Conwy et al.,
2013). Specialised cattle breeding is believed to have
taken place in South-Eastern Italy from the 7th millennium BC onwards (Vigne & Helmer, 1999; Vigne,
2003) while further west, sheep and goat dominate assemblages (Helmer & Vigne, 2004; Tresset & Vigne,
2007; Vigne & Helmer, 2007). For the Iberian
Peninsula, Early Neolithic assemblages predominately
come from cave sites with questionable representation
of the whole herding system (Saña, 2013). However,
recent discoveries of well-preserved open air sites (La
Draga, Banyoles; Sant Pau del Camp, Barcelona) indicate that cattle were important in Spain as in SouthEastern Italy. The initial review carried out by Vigne
& Helmer (2007) identified high frequencies of young
calf mortality from Early Neolithic contexts in Southern Italy, which were tentatively interpreted as the result of natural mortality due to birth accidents/postnatal infections or deliberate slaughter to optimise milk
production. Distinguishing neonatal mortality from deliberate slaughter was not possible until recently as
there were no means of predicting the age-at-death of
calves with narrow standard errors. Here we take advantage of the recent publication of predicting age-atdeath (AtD) from un-fused post-cranial bone measurements (Gillis et al., 2013a) together with dental AtD
frequencies to investigate neonatal mortality and
young calves culls (<6 months) at two sites in the
North-Western Mediterranean region: Trasano and
La Draga.
is a Cardial open air site situated on the Eastern submerged Neolithic shore line of Lake Banoyles and dates
from the second half of the 6th millennium BC (Bosch
et al., 2000). The faunal remains have been processed
by J-DV & IC, (unpub.; Trasano) and M S-S (Saña
Segui, 2001; La Draga). At both sites, the cattle dental
and un-fused bone material was studied by RG.
All contexts were sieved at both sites; however,
taphonomic conditions differed at both sites. Using
the ratio of humerus and tibia MNEprox/MNEdist of
caprines as a proxy for an index of preservation (see
Binford, 1981) indicates that at Trasano the ratios
were lower than that of La Draga (Trasano: humerus:
1/42; tibia: 6/17; La Draga: humerus: 20/28; tibia:
17/18). The low density un-fused bones are more subject to fragmentation and gnawing by dogs (Munson,
2000). The percentage of carnivore gnawing on
bones at La Draga was 0.1%, which suggests that
there were low rates of attrition (Sãna Segui, 2001).
This was not recorded at Trasano, so we can make
no comparison.
Archaeozoological mortality profiles are commonly
constructed using AtD frequencies estimated from dental eruption, development and wear. Dental remains
provide more precise and reliable estimates compared
to epiphyseal fusion timings of post-cranial elements
(Halstead, 1998; Helmer, 2000; Greenfield & Fowler,
2005; Vigne & Helmer, 2007). There are some discrepancies for eruption timings for M1 between different
authors (Silver, 1963: 5–6 months; Legge, 1992:
2–3 months). Gillis et al. (2013a) have demonstrated
that the absolute age references given by Legge
(Legge, 1992) are more accurate than those based on
traditional veterinarian observations of tooth eruption
through the gum (Silver, 1963; Higham, 1967). In addition, to Legge’s (Legge, 1992) scheme, crown height
ratios were used to identify adult mortality following
Material and methods
Trasano (Matera, Italy; Figure 1) is a dry open air site of
the Impressa tradition dating from the mid-7th to the
early 6th millennium BC (Guilaine & Cremonesi,
1987; Radi, 2003). La Draga (Banyoles, Girona, Spain)
Copyright © 2014 John Wiley & Sons, Ltd.
Figure 1. The location of Trasano and La Draga.
Int. J. Osteoarchaeol. (2014)
Calf Mortality During the Early Neolithic of NW Mediterranean
Table 1. A summary of the age at death distribution frequency of the superior and inferior teeth from Trasano and La Draga (N is the
unit of quantification)
Age classes
0–1 months 1–4 months 4–6 months 6–12 months 1–2 years 2–4 years 4–6.5 years 6.5–9 years 9+ years
Trasano
Phase I
Phase II
Phase III
Upper
Lower
Upper
Lower
Upper
Lower
Total
0
0
1
0
0
0
1.0
Upper
Lower
Total
2
3
5
2
0
0
0.3
0
0.8
3.1
0
1
0
4.6
0.3
0.8
6.7
0
0
0.5
2.1
0.8
0.8
4.2
0.5
1.0
1.0
2.1
0.8
0.5
5.9
3.5
2.8
1.5
3.5
3.5
5.5
20.3
0
1.8
0
3.5
4.5
1.5
11.3
0
0.8
0
1
5.5
0.5
7.8
0
0.8
0.2
0
0.7
1.5
2.8
9.3
11.5
20.8
0.3
0.5
0.8
2.3
0.5
2.8
5.2
7.4
12.6
12.2
11.9
24.1
15.2
6.4
21.6
2.7
1.9
4.6
0.7
1.9
2.6
La Draga
Ducos (1968). Both lower and upper teeth were used in
this study (Table 1). The mortality data is presented as
a histogram following Brochier (2013), without using
the a priori correction initially proposed by Vigne &
Helmer (2007). The a priori correction is related to
the un-equal probability of recruitment between age
classes; we have not used it here as further investigation
is required in understanding how affects the final
resulting morality profiles.
Estimates of AtD from bone epiphyseal fusion events
begin at 7 months, which excludes examination of the
perinatal mortality. Gillis et al. (2013a) have developed
linear models for the prediction of AtD from measurements following von den Dreisch (von den Driesch,
1976; GL, Bp, Dp, SD, Bd and Dd) of un-fused diaphyses and epiphyses of post-cranial long bones as well as
astragalus and calcaneus. The post-cranial elements
studied for each site are detailed in Table 2. The AtD
estimates for post-cranial bones for Trasano and La
Draga were from 45 and 46 elements, respectively.
Table 2. Number of post-cranial elements measured to predict
AtD for the two studied sites
Element
Trasano
Frequency
Scapular
Humerus
Radius
Ulna
Metacarpal
Femur
Tibia
Metatarsal
Astragalus
Calcaneus
Total elements (measurements)
0
1
4
1
6
2
10
11
4
6
45 (98)
Copyright © 2014 John Wiley & Sons, Ltd.
La Draga
Frequency
4
2
9
3
6
0
5
3
14
1
47 (210)
The excellent preservation conditions are evident at
La Draga and of the 120 possible measurements 89%
were measurable (Table S1b), whereas at Trasano of
the 236 measurements, only 42% were (Table S1a).
The number of bins chosen for the AtD frequencies
from bone measurements histograms was based on the
square root of the total sample number. To identify
whether there was one or more culls within the AtD
frequencies predicted from un-fused bones, mixture
analyses were performed. The Akaike information criterion (AIC; Akaike, 1974) is the most reliable parameter for determining the best mixture model with
reference to the actual distribution of the frequencies.
The preferred model is the one that has the lowest
AIC and that minimises information loss. The relative
probability of the model (Eq.1) can be interpreted as
the ith model which minimises the information loss
(Burnham & Anderson, 2004).
Relative probability ¼ ½exp ððAICmin AICi Þ=2Þ
(1)
The raw AtD frequencies predicted from dental and
un-fused bone remains were compared using a Chi2 test.
For processing this test, the ages predicted from the unfused bones were grouped into the following age classes: 0 to 1 months, 1 to 4 months, 4 to 6 months and
6 to 12 months following the dental age predictions
(Table S2). It is difficult to refine these age classes further due to the large intervals associated with eruption
and development stages.
The age predictions from the linear models were calculated using the free-platform R package (version
2.13.1) and the PAST program (version 3.0; Hammer,
2013) was used for the mixture analysis modelling.
Microsoft Excel software was used for all other statistical calculations.
Int. J. Osteoarchaeol. (2014)
R. Gillis et al.
Results
The mortality profiles based on dental material
The Trasano (phases I–III) dental mortality profile
(Table 1; Figure 2a) indicates a low frequency of mortality in age class 0 to 1 months (N = 1), which then increase in age class 4 to 6 months (N = 7). For adult age
classes, 2–4 years constitute the strongest peak
(N = 20), which is classically interpreted as the result of
slaughter for prime beef (Helmer, 1992). Age classes
>4 years are also well-represented (N = 22), possibly a
reflection of retired lactating females slaughter. Overall,
the dental mortality profile at Trasano suggests that the
early culling of infants coupled with adult age classes is
the result of mixed husbandry practices towards beef
and dairy production.
At La Draga, there is a strong peak in age classes 0 to
1 months (Table 1; Figure 2b). The high frequency of
neonatal mortality (N = 5) is consistent with the previous findings of Saña (Saña Segui, 2001). The peak in
the following age class (1 to 4 months; N = 21) could
be the result of deliberate slaughter rather than natural
mortality. The interpretation of the mortality pattern
for adults is similar to Trasano with the slaughter for
beef (2–4 years; N = 24) and of retired lactating females
(adults >4 years old; N = 29).
Mortality profiles based on un-fused bones
measurements
For Trasano, the histogram and the distribution of the
Kernel densities show one peak between 4 and
5 months (Figure 3a.). For La Draga, the histogram
shows two distinct peaks, between 0–1 months and
3–4 months (Figure 3b). The models for one to three
cull groups predicted from the mixture analyses of the
AtD frequencies distributions are presented in Table 3.
For Trasano, the difference of the AIC for a model
for one, two or three groups is small. Moreover, when
we compare the relative probability for information loss
between the model for three groups (lowest AIC) and
Figure 2. Cattle mortality profiles based on dental eruption, development and wear stages for a) Trasano (Phases I–III) (N = 63) and b) La
Draga (N = 95).
Copyright © 2014 John Wiley & Sons, Ltd.
Int. J. Osteoarchaeol. (2014)
Calf Mortality During the Early Neolithic of NW Mediterranean
Figure 3. Histogram and mixture analysis of predicted ages from unfused bones from a) Trasano (phases IIII) and b) La Draga. S-Table 1a
and b details the complete AtD predicted from the post-cranial unfused bone material.
one or two groups, it is very low (<0.005 for both
groups). Due to the sample size and the number of outliers, it is wise not to conclude more than one group.
The mean age of mortality for one group is 5 ± 2 months
(p > 0.05). This result is congruent with the mortality
profile (Figure 2a) based on dental material, where the
largest peak for calves is in age class 4–6 months. The
0–1 month mortality peak is very small, as previously
seen for AtD frequencies based on dental remains
(Figure 2a).
For La Draga, the relative probability for information
loss between the model for two groups and that for
three groups is 0.09. This small difference between
the two models is also evident in the predicted mean
ages (Table 3). The first centres on the neonatal age
class 0–1 month (Model 2 groups: mean = 0.6 ± 0.5,
p > 0.05; Model 3 groups: mean = 0.7 ± 0.6, p > 0.05)
and the second between 3 and 4 months (for both
Models: mean = 3.5 ± 1, p > 0.05). The final mean age
predicted by the model for three groups is 7.1,
p = 0.02. These mean ages predicted are congruent with
the dental mortality profiles, with two main peaks:
0–1 month and 3–4 months (Figure 2b) and a smaller
one between 6 and 12 months.
Teeth and bones are affected differently by taphonomic conditions (Brain, 1981; Lyman, 1984). Unfused
bones are more open to destruction from carnivore attack and abiotic processes due to their fragility. To a
lesser extent, these factors are also a threat to infant
and juvenile dental remains. The Chi2 test between
the raw AtD results (Table S2) predicted indicate no
significant differences between age distributions from
un-fused bone and dental material (Trasano phases I
to III (X2 = 3, df = 3, p = 0.4); La Draga (X2 = 0.8, df = 3,
p = 0.05). The close congruencies presents a strong case
that the age structure is the same for both dental and
post-cranial remains. Consequently, we believe that
the dental and bone AtD predicted frequencies at both
sites provide an accurate representation of prehistoric
mortality.
Interpretation and discussion
Halstead (1998) proposed that equifinality problems related to partial recovery, sampling and preservation
conditions for cranial and post-cranial elements could
lead to interpretation problems. The presence of calves
Table 3. Akaike information criterion (AIC) and mean ages for the models for 1, 2 and 3 mortality peaks generated from mixture analysis
of AtD predicted from un-fused cattle bones
Mixture analysis (no. of bins = 7)
No. of groups = 1
Trasano
La Draga
No. of groups = 2
AIC
Mean age (months)
AIC
118.5
104.3
5
2.6
119.3
97.82
Copyright © 2014 John Wiley & Sons, Ltd.
Mean age (months)
4.8, 10.8
0.6, 3.5
No. of groups = 3
AIC
Mean age (months)
57.1
93.15
4.3, 5.3, 11.5
0.7, 3.6, 7.1
Int. J. Osteoarchaeol. (2014)
R. Gillis et al.
younger than 6 months could be interpreted as any of
the following: over-representation of neonatal disposal
areas, natural mortality, specialised slaughtering practice and symbolic disposal. Over-sampling of calf disposal areas could be a concern at La Draga as the site
has not been completely excavated (Tarrùs, 2008).
There are no apparent differences in mortality frequencies between the excavation sectors, which suggest the
sample is not biased. Calf remains at both sites were
not found in ‘specialised’ contexts, which may have
suggested a ritual/symbolic disposal. Natural mortality
can be differentiated from deliberate slaughter assuming natural mortality will be highest in the first month.
In the following sections we will focus on evaluating
and determining the possible animal husbandry practices, which may have caused the mortality frequencies
from Trasano and La Draga.
Accidental death and natural mortality
At present, neonatal mortality can be as high as 25%
during the first month of life due to poor husbandry
and non-sterile birthing stalls (Mellor & Stafford,
2004). With reference to this rate, Trasano had a low
frequency of neonatal mortality; this may be due to
poor preservation of low density neonatal bones. The
low frequency of age class (0–2 months) in the caprine
mortality profile (S-Figure 1a) from Trasano supports
this conclusion. For La Draga caprines (S-Figure 1b),
where the preservation was better, there is also a low
frequency in this age class. The use of caves/rock shelters as birthing stations evident from truncated mortality profiles is a common feature in caprine husbandry
during the Neolithic (Helmer & Vigne, 2004). Absence
or presence of very young calves at La Draga and
Trasano appears to be an accurate representation of
neonatal mortality in the two sites.
High levels of neonatal mortality could be a reflection of poor husbandry practices or climatic conditions
(Halstead, 1998). The birthing season for calves usually coincides with March to May, based on regional
ethnographic references of traditional and feral cattle
(Gomez et al., Gómez et al., 1997; Soysal & Kök,
2006), The modern Mediterranean climate has been
characterised by annual levels of precipitation between
450 and 750 mm with 80% precipitation occurring
from December to March (Harding et al., 2009).
The stability of the summer temperatures and indeed
the climate in general in the western Mediterranean
is influenced by the Atlantic. During the periods of
both sites’ occupation, the climate was significantly
wetter than today (up to 800–1000 mm per year),
where winters/springs were considerably more humid
Copyright © 2014 John Wiley & Sons, Ltd.
particularly at La Draga (Aguilera et al., 2011). The
high neonatal mortality at latter could be a reflection
of very humid climate, where the wet season would
have coincided with the beginning of the birthing
season (Figure 4). This may have forced the Neolithic people to stall their animals, potentially increasing infection amongst calves (Halstead, 1998:
14). Winter periods were drier at Trasano, which
would explain the low rate of neonatal mortality with
reference to La Draga.
This difference can also result from different management practices. Stalling animals may have led to a
more direct control of milk production. If lactating
cows were affected by the loss of the calf, methods
such as substitution and artificial stimulation could have
been employed. If so, the different rates of neonatal
mortality for the two sites may be due to more controlled cattle husbandry at La Draga than at Trasano.
However, it is difficult to determine whether this difference in neonatal mortality is a direct result of poor
Figure 4. The modern average rainfall (mm) for the Matera (Italy) and
Banoyles (Spain) regions. The case studies dental AtD frequency densities have
been superimposed to aid the interpretation of the timing of mortality patterns.
This figure is available in colour online at wileyonlinelibrary.com/journal/oa.
Int. J. Osteoarchaeol. (2014)
Calf Mortality During the Early Neolithic of NW Mediterranean
climatic conditions or a deliberate husbandry practice.
Further AtD studies from un-fused bone and dental material at the scale of the North-Western Mediterranean
area will push ahead research into the effects of climate
and husbandry practices on neonatal survival.
Deliberate slaughter of calves
At Trasano, the dental and bone AtD frequencies
reflected deliberate mortality taking place between 4
and 6 months old, whereas at La Draga, it was focused
between 1 and 4 months. Evidence from lipid residues
analysis indicates that milk processing was taking place
at La Draga and in the region of Trasano (Salque et al.,
2012). As we have ruled out a ritual motive, we will
examine in more detail the possibility that the slaughter timings were the result of seasonal culls associated
with dairy production. The role of the calf is believed
to be integral to prehistoric cattle dairy management
due to cows’ sensitivity to being milked without them
(Peske, 1994). Identification of calves slaughtered at
6–12 months, during the period when they are being
weaned from their mothers, has been suggested to be
indicative of dairy management. However, the timing
of the annual culls will also reflect actions resulting
from external climatic pressures. For example, at the
Northern French site of Bercy (5th millennium BC;
Chasséen) calf mortality between 6 and 12 months
was interpreted as evidence for post-lactation slaughter
(Tresset, 1996). Further isotopic examination, confirmed that these animals were within weaning process
or had been recently weaned, therefore indicating that
the lactation period had finished or that animals had
been removed from the mother (Balasse et al., 1997).
The timing of this slaughter as well as being a direct
result of dairy husbandry practices was proposed to
have coincided with the onset of winter (Balasse &
Tresset, 2002). In Northern France the winter months
would have been a difficult time for farmers, with the
necessity of constant procurement of fodder and
shelter for 4 to 6 months. Therefore, the removal of
unwanted animals, particularly males, would have taken
place prior to this seasonal change. Consequently, calf
slaughter timings are a result of interplay between the
pressures of the external environment and of the stock
keeping practices to maintain health and productivity
of the herd.
In the Mediterranean lowlands, it was not necessary
to provide shelter and fodder for cattle during winter,
since it was rarely too cold and there would be sufficient naturally available fodder. However, previous
studies have indicated that during the height of summer (July to September) in Italy and Greece, there is
Copyright © 2014 John Wiley & Sons, Ltd.
a shortage of water and fodder (Molenat & Casabianca,
1979; Halstead & Jones, 1989; Rosati, 2000). At
Trasano, assuming the animals were born around April,
the slaughter peak of animal in age class 4 to 6 months
coincides with the height of summer (August/September; Figure 5a). During the middle Holocene in this region temperatures were similar as to today (Magny
et al., 2011, 2012). For La Draga, it is possible that herd
numbers were reduced prior to the onset of the dry season to ensure sufficient fodder sources for the remaining animals. The differences in climate between the
Atlantic-temperate and Mediterranean regions will lead
to different birthing seasons (Balasse & Tresset, 2007).
Consequently, the timing of the seasonal cull and the
age of the infants would vary between North (September to November) and South (June to August).
Central to the post-lactation model for cattle dairy
management is that the lactation length for Neolithic
cattle would be consistent between regions. In modern
animals, it can vary depending on the breed history and
Figure 5. The modern average temperatures (°C) for the Matera (Italy)
and Banoyles (Spain) regions. The study sites dental AtD frequency
densities have been superimposed to aid the interpretation of the timing
of mortality patterns.
Int. J. Osteoarchaeol. (2014)
R. Gillis et al.
the external environment. For example, lactation
lengths can vary from 152 days in Mongolian cattle to
340 days in Icelandic cattle (Table 4). The Icelandic
cattle are from improved stock whereas those from
Mongolian are not. Therefore a cull timed between 3
and 6 months could be a signature of shorter lactation
length in the North-Western Mediterranean, whereas
the slaughter of animals between 6 and 12 months at
Bercy may be a reflection of a longer lactation length.
Cattle were generally more widespread in temperate
Europe than in the Mediterranean area during the Early
Neolithic (Tresset & Vigne, 2007). The longer lactation lengths would mean that more milk was produced
and consumed in the Northern region, which may have
led to differences in the lactase persistence frequency
evident between these two regions (Gerbault et al.,
2011). Further investigation in terms of stable isotopic
analysis of dental enamel could determine the rhythm
and season of births (Balasse et al., 2003; Blaise &
Table 4. The duration of lactation for cattle European and Asian
cattle breeds
Country
Croatiaf
Czech
republicf
Estoniaf
Francea
Germanyf
Icelandf
Italyb,c,f
Mongoliae
Sloveniaf
Spainf
Turkeyb,f
UKe
Breed
type
Duration
(days)
Busa
Ceska Cervinka
Multi
Dual
180
300
Eesti Punane
Bretonne pie-noire
Salers
Aubrac
Normande
Abondance
Brune des Alpes
(Schwyz)
Holstein-Friesian
Dopelnutzung
Rotbung
Icelandic
Modicana
Agerolese
Regginia
Mongolian
Cika
Arileña-Negra Ibérica
Pirenaica
Southern Anatolian
Yellow
Anatolian Black
Grey Steppe
Kilis
East Anatolian Red
Southern Anatolian Red
Gloucester
Milk
Dual
Dual
Dual
Milk
Dual
Milk
305
259
249
248
275
285
286
Milk
Milk
308
305
Dual
Multi
Dual
Dual
Multi
Multi
Multi
Multi
Dual
340
250
200–300
305
152
300
180
290
188
Multi
Multi
Dual
Multi
Dual
Multi
257
220
238
270
275
300
Breed
a
Quittet & Denis, 1979.
Rosati, 2000.
c
Soysal & Kök, 2006.
d
Gandini et al., 2007.
e
www.embryoplus.com.
f
www.dad.fao.org. NA—not available.
b
Copyright © 2014 John Wiley & Sons, Ltd.
Balasse, 2011), and of bone collagen to investigate
the suckling and weaning phases of calves (Balasse &
Tresset, 2002; Gillis et al., 2013b). This would elucidate
in what season the La Draga and Trasano calves were
born and died, and whether they were weaned prior
to slaughter. These may provide the finer resolution
to the already complex story of calf mortality during
the Early Neolithic as described by the analysis of AtD.
Conclusions
The primary aim of this study was to investigate calf
mortality at Early Neolithic two sites in the NorthWestern Mediterranean. This was achieved by using
two methods of age determination based on dental
eruption, development and wear and on the measurements of un-fused post-cranial elements. These techniques allowed us to distinguish natural neonatal
mortality from deliberate cull of young calves. The regional differences in climate appear to have played an
important role in the variation of neonatal mortality.
The high rate of neonatal mortality at La Draga, where
the climate was wetter during the Early Neolithic than
today, could either be a reflection of death from exposure or increased stalling of animals leading to high infection rates. Whereas at Trasano the low frequency of
neonatal age classes is probably a combination of both
poor preservation and a reflection of a more favourable
climatic conditions. Further analysis of neonatal mortality at sites in the same region with different taphonomic and climatic conditions is needed for
comparison.
The secondary aim was to determine whether the
calf mortality was a reflection of dairy husbandry.
The strong presence of adults possibly retired lactating
females (>4 years), and the presence of dairy lipids in
ceramics, supports the idea of the existence of a lactating herd. Both the dental and un-fused bones’ AtD indicated that there was deliberate slaughter around
3–6 months at both La Draga and Trasano. This would
have coincided with the dry season at both sites, and
therefore the cull would have removed un-wanted
calves to ensure sufficient fodder and water supply.
The removal of animals this young would also increased the milk production as the milk would no longer be shared between the suckling calf and human
competitors. On the other hand, if the calves were an
integral part of the husbandry strategy, a cull of
3–6 months could suggest that the length of lactation
periods was shorter for Early Neolithic cattle at these
North-Western Mediterranean sites. The methods
presented here can correctly determine the timing of
Int. J. Osteoarchaeol. (2014)
Calf Mortality During the Early Neolithic of NW Mediterranean
calf slaughter. However, whether it can determine
whether dairy husbandry was practised requires other lines
of evidence to confidently confirm our initial hypothesis.
The investigation into calf culls in different
European regions is important for understanding the
evolution of cattle husbandry during the Neolithic
and assessing the effect of physical and cultural environments on slaughter practices. Overall, our work
has demonstrated that dental and un-fused bone AtD
predictions can be combined to investigate neonatal
mortality and young calf slaughter. Future studies of
this type will elucidate further how Early Neolithic
farmers in the NW Mediterranean adapted husbandry
practices to maintain herd productivity.
Acknowledgements
This study was carried out during the PhD of RG
within the framework of the LeCHE project (MarieCurie International Training Network, grant number:
FP7-215362-2) under the supervision of JDV at Museum national d’Histoire naturelle, Paris. We thank
Jean Guilaine and Giovanni Cremonesi for access to
the Trasano material and Àngel Bosch, Josep Tarrús
and Júlia Chinchilla for access to the la Draga material.
We also thank the anonymous reviewers for their
stimulating remarks to an early draft of this paper.
References
Aguilera M, Ferrio JP, Araus JL, Tarrús J, Voltas J. 2011. Climate at the onset of western Mediterranean agriculture expansion: Evidence from stable isotopes of sub-fossil oak
tree rings in Spain. Palaeogeography, Palaeoclimatology, Palaeoecology 299: 541–551
Akaike H. 1974. A new look at the statistical model identification. IEEE Transactions on Automatic Control 19: 716–723.
Balasse M. 2003. Keeping the young alive to stimulate milk
production? Differences between cattle and small stock.
Anthropozoologica 37: 3–37.
Balasse M, Tresset A. 2002. Early Weaning of Neolithic Domestic Cattle (Bercy, France) Revealed by Intra-tooth Variation in Nitrogen Isotope Ratios. Journal of Archaeological
Science 29: 853–859
Balasse M, Tresset A. 2007. Environmental constraints on the
reproductive activity of domestic sheep and cattle: what
latitude for the herder? Anthropozoologica 42(2): 71–88.
Balasse M, Bocherens H, Tresset A, Mariotti A, Vigne J-D.
1997. Émergence de la production laitière au Néolithique?
Contribution de l’analyse isotopique d’ossements de
bovins archéologiques. C.R. Académie de Science Paris, Sciences
de la terre et des planètes 325: 1005–1010.
Copyright © 2014 John Wiley & Sons, Ltd.
Balasse M, Smith AB, Ambrose SH, Leigh SR. 2003. Determining sheep birth seasonality by analysis of tooth enamel
oxygen isotope ratios: the Late Stone Age site of
Kasteelberg (South Africa). Journal of Archaeological Science
30: 205–215.
Binford LS. 1981. Bones: ancient men and modern myths.
Academic Press: London.
Blaise E, Balasse M. 2011. Seasonality and season of birth of
modern and late Neolithic sheep from South-Eastern
France using tooth enamel δ18O analysis. Journal of Archaeological Science 38: 3085–3093.
Bosch A, Chinchilla J, Tarrús J. 2000. El poblat lacustre
del Neolític de La Draga: Excavacions de 1990-1998.
Museo d’arqueologia de catalunya-centre d’Arqueologia
subaquàtica de Catalunya VII: Girona.
Brain CK. 1981. The hunters or hunted? An introduction to
African cave taphonomy. University of Chicago Press:
Chicago.
Brochier JE. 2013. The use and abuse of culling profiles in recent zooarchaeological studies: some methodological
comments on “frequency correction” and its consequences. Journal of Archaeological Sciences 40: 1416–1420.
Burnham KP, Anderson DR. 2004. Multimodel Inference:
Understanding AIC and BIC in Model Selection. Sociological Methods Research 33: 261–304.
Chapman JC. 1982. The ‘secondary products revolution’ and
the limitations of the Neolithic. Bulletin of the University of
London Institute of Archaeology 11: 119–141.
Clutton-Brock J. 1981. Contributions to discussion. In Farming practice in British prehistory, RJ Mercer (ed). University Press: Edinburgh; 218–6220.
Domestic animal diversity information systems. nd. www.
dad.fao.org
Driesch von den A. 1976. A guide to measurements of animals bones from archaeological sites. Harvard University,
Peabody Museum of Archaeology and Ethnology Bulletin
1: Cambridge, MA.
Ducos P. 1968. L’origine des animaux domestiques en Palestine.
Publications de l’Institut de Préhistoire de l’Université de
Bordeaux, Mémoire no.6, Ed. Delmas: Bordeaux.
Embryo plus. nd. www.embryoplus.com
Evershed RP, Payne S, Sherratt AG, Copley M, Coolidge J,
Urem-Kotsu D, Kotsakis K, Özdogan M, Nieuwenhuyse
O, Akkermans PMMG, Bailey D, Andeescu RR, Campbell
S, Farid S, Hodder I, Nurcan Y, Özbasaran M, Bicakci E,
Garfinkel Y, Levy T, Burton M. 2008. Earliest date for
milk use in the Near East and South-eastern Europe linked
to cattle herding. Nature 20: 1–4.
Gandini G, Maltecca C, Pizzi F, Bagnato A, Rizzi R. 2007.
Comparing local and commercial breeds on functional
traits and profitability: The case of Reggiana dairy cattle.
Journal of Dairy Science 90: 2004–2011
Gerbault P, Liebert A, Itan Y, Powell A, Currat M, Burger J,
Swallows D, Thomas MG. 2011. Evolution of lactase persistence: An example of human niche construction. Philosophical Transactions of the Royal Society of London. Series B:
Biological Sciences 366: 863–877.
Int. J. Osteoarchaeol. (2014)
R. Gillis et al.
Gillis R, Arbogast R-M, Piningre J-F, Debue K, Vigne J-D.
2013a. Prediction models for age at death estimates for
calves, using un-fused epiphysis and diaphyses elements.
International Journal of Osteoarchaeology DOI: 10.1002/
oa.2377
Gillis R, Bréhard S, Bălăsescu A, Ughetto-Monfrin J,
Popovici D, Vigne J-D, Balasse M. 2013b. Sophisticated
cattle dairy husbandry at Borduşani-Popină (Romania,
fifth millennium BC): the evidence from complementary
analysis of mortality profiles and stable isotopes. World Archaeology 45(3): 447–472.
Gómez M, Plazaola JM, Seiliez JP. 1997. The Betizu Cattle
of the Basque country. Spanish Journal of agricultural science
22: 1–5.
Greenfield HJ, Fowler KD. 2005. The secondary products
revolution in Macedonia: The zooarchaeological remains
from Megalo Nisi Galanis, a Late Neolithic-Early Bronze
Age site in Greek Macedonia. BAR 1414. Archaeopress:
Oxford.
Guilaine J, Cremonesi G. 1987. L’habitat néolithique de
Trasano (Matera, Basilicate): Premiers résultats. Proceedings
of the Atti della XXVI riunione scientifica dell’istituto italiano di
preistoria e protostoria. Firenze: 707–719
Halstead P. 1998. Mortality models and milking: problems of
uniformitarism, optimality and equifinality reconsidered.
Anthropozoologica 27: 3–20.
Halstead P, Jones G. 1989. Agrarian Ecology in the Greek
Islands: Time Stress, Scale and Risk. The Journal of Hellenic
Studies 109: 41–55.
Hammer Ø. 2013. PAST: PAleontological STudies Version
3.0: Reference Manuel. Natural History Museum: Olso.
http://folk.uio.no/ohammer/past/past3manual.pdf.
Harding A, Palutikof J, Holt T. 2009. The climate system. In
The Physical Geography of the Mediterranean, Woodward J (ed). Oxford University Press: Oxford; 69–88.
Helmer D. 1992. La domestication des animaux par les
hommes préhistoriques. Masson: Paris.
Helmer D. 2000. Discrimination des genres Ovis et Capra à
l’aide des prémolaires inférieures 3 et 4 et interprétation
des âges d’abattage; l’exemple de Dikili Tash (Grèce).
Anthropozoologica 31/ Ibex, Journal of Mountain Ecology 5: 29–38.
Helmer D, Vigne J-D. 2004. La gestion des cheptels de caprines au Néolithique dans le midi de la France. In
Approches fonctionnelles en Préhistoire. Actes du XXVe
congrès Préhistorique de France (Nanterre, 2000), P Bodu,
C Constantin (eds). Société Préhistorique Française: Paris;
397–407.
Higham CFW. 1967. Stock rearing as a cultural factor in
Prehistoric Europe. Proceedings of the Prehistoric Society 6:
84–106.
Labussière J. 1999. The physiology of milk ejection: consequences on milking techniques in Biology of Lactation,
J Martinet, L-M Houdebine, HH Head (eds). INRA: Paris;
307–343.
Legge AJ. 1981. The agricultural economy. In Grimes
Graves Norfolk, excavations 1971–72: Vol. 1, R Mercer
(ed.). Her majesty’s stationary office: London, 79–100.
Copyright © 2014 John Wiley & Sons, Ltd.
Legge AJ. 1992. Excavations at Grimes Graves Norfolk
1972–1976 Fascicule 4: Animals, Environment and the
Bronze Age Economy. British Museum Press: London.
Lucas AT. 1989. Cattle in Ancient Ireland. Boethius press:
Kilkenny.
Lyman RL. 1984. Bone density and differential survivorship
of fossil classes. Journal of Anthropological Method and Theory
3: 259–337.
Magny M, Vannière B, Calo C, Millet L, Leroux A, Peyron
O, Zanchetta G, laMantia T, Tinner W. 2011. Holocene
hydrological changes in south-western Mediterranean as
recorded by lake-level fluctuations at Lago Preola, a
coastal lake in southern Sicily, Italy. Quaternary Science Reviews 30: 2459–2475.
Magny M, Peyron O, Sadori L, Ortu E, Zanchetta G,
Vannière B, Tinner W. 2012. Contrasting patterns of precipitation seasonality during the Holocene in the southand north-central Mediterranean. Journal of Quaternary Science 27(3): 290–296.
Mellor DJ, Stafford KJ. 2004. Animal welfare implications of
neonatal mortality and morbidity in farm animals. The Veterinary Journal 168: 118–133.
Molenat M, Casabianca F. 1979. Contribution à la maîtrise
de l’élevage porcin extensif en Corse Institut National de
la Recherche Agronomique: Jouy-en-Josas. Bulletin Technique du Département de Génétique Animale 32.
Munson PJ. 2000. Age-correlated differential destruction of
bones and its effect on archaeological mortality profiles
of domestic sheep and goats. Journal of Archaeological Science
27: 391–407.
Payne S. 1973. Kill-patterns in sheep and goats: the mandibles from Asvan Kalé. Anatolian Studies 23: 281–303
Peske L. 1994. Contribution to the beginning of milking in
Prehistory. Archeologické Rozhledy 46: 97–104.
le Quellec J-L. 2011. Provoking lactation by the insufflations
technique as documented by the rock images of the Sahara. Anthropozoologica 46(1): 65–126.
Quittet E, Denis B. 1979. Races Bovines Françaises. Collection: Les Races D’Animaux Domestiques, 3rd ed. La
Maison rustique: Paris.
Radi G. 2003. Trasano Basilicata. In Le ceramiche impresse
nel Neolitico antico in Italia e mediterraneo. Istituto
Poligrafico e Zecca dello Stato: Roma; 695–705.
Rosati A. 2000. Breeding strategies for the Mediterranean
Modicana cattle breed. In Developing breeding Strategies
for Lower Input Animal Production Environments. S
Galal, J Boyazoglu, K Hammond. (eds). ICAR Technical
Series: Rome, 503–510.
Rowley-Conwy P, Gourichon L, Helmer D, and Vigne J-D.
2013. Early domestic animals in Italy, Istria, the
Tyrrhenian islands and Southern France. In The Origins
and Spread of Domestic Animals in Southwest Asia and
Europe, S College, J Conolly, K Dobney, K Manning, S
Shennan. (eds). Left Coast Press: California; 161–194.
Salque M, Radi G, Fabbri C, Tagliacozzo A, Pino Uria B,
Wolfram S, Stäuble H, Hohle I, Whittle A, Hofmann D,
Pechtl J, Schade-Lindig S, Eisenhauer U, Evershed RP.
Int. J. Osteoarchaeol. (2014)
Calf Mortality During the Early Neolithic of NW Mediterranean
2012. Early Neolithic economy and management of animals in southern and central Europe revealed using lipid
residue analyses of pottery vessels: preliminary results.
Anthropozoologica 47(2): 45–62.
Saña SM 2013. Domestication of animals in the Iberian Peninsula. In The Origins and Spread of Domestic Animals in
Southwest Asia and Europe, S College, J Conolly, K
Dobney, K Manning, S Shennan (eds). Left Coast Press:
California; 195–220.
Saña Segui M. 2001. Els recursos animals. In El poblat lacustre
del Neolític antic de La Draga: Excavacions de 1990-1998,
Á Bosch, J Chinchilla, J Tarrús. (eds). Museo d’arqueologia
de catalunya-centre d’Arqueologia subaquàtica de Catalunya VII: Girona; 150–168.
Sherratt A. 1983. The secondary products revolution of animals in the Old World. World Archaeology 15: 90–104.
Sherratt A. 1997. Economy and society in prehistoric Europe: changing perspectives. Edinburgh University Press:
Edinburgh.
Silver A 1963. The aging of domestic animals. In Science in
archaeology, R Brothwell, ES Higgs (eds). 1st ed. W.B.
Saunders: London; 250–268.
Soysal ML, Kök S. 2006. The last survivors of Grey cattle
resisting extinction. A case study of characteristics and
sustainability of traditional systems of native Grey cattle
breeds. Options Méditerranéennes Series A 78: 55–63.
Tarrùs J. 2008. La Draga (Banyoles, Catalonia), an Early
Neolithic Lakeside Village in Mediterranean Europe. Catalan historical review 1: 17–33.
Tresset A. 1996. Le rôle des relations homme-animal dans
l’évolution économique et culturelle des sociétés des Ve-
Copyright © 2014 John Wiley & Sons, Ltd.
IVe millénaires en Bassin parisien. Approche ethnozootechnique fondée sur les ossements animaux. Thèse
de préhistoire-ethnologue-anthropologie, Université
Paris-I Panthéon-Sorbonne: 382p.
Tresset A, Vigne J-D. 2007. Substitution of species, techniques and symbols at the Mesolithic-Neolithic transition
in Western Europe. Proceedings of the British Academy 144:
189–210.
Vigne J-D. 2003. L’exploitation des animaux à Torre Sabea.
Nouvelles analyses sur les débuts de l’élevage en
Méditerranée centrale et occidentale. In Torre Sabea, un
établissement du Néolithique ancien en Salento, JJ
Guilaine, G Cremonesi (eds). Ecole Française: Rome;
325–359.
Vigne J-D, Helmer D. 1999. Nouvelles analyses sur les
débuts de l’élevage dans le centre et l’ouest
méditerranéens. In Le Néolithique du Nord-Ouest
Méditerranéen, J Vaquer (ed.). Actes du 24ième
Congrès. Préhistorique de France: Carcassonne; 26–30
septembre 1994: 129–146.
Vigne J-D, Helmer D. 2007. Was milk a ‘secondary product’
in the old world Neolithisation process? Its role in the
domestication of cattle sheep and goats. Anthropozoologica
42(2): 9–40.
Supporting information
Supporting information may be found in the online version of this article at the publisher’s web site.
Int. J. Osteoarchaeol. (2014)