[page-n-1]
18
MAINTENANCE OF IBERIAN LIVESTOCK
Benjamin Naylor
ABSTRACT
The division of the main species of Iron Age Iberian livestock into categories of
large, high-maintenance animals (equids and cattle) and medium-sized, lowmaintenance animals (sheep, goats and pigs) allows us to investigate grosso
modo Iberian livestock management. High proportions of large animals are,
unsurprisingly, correlated with greater rainfall and fewer cold months. Yet
they are also correlated with drier summers, suggesting techniques such as
growing fodder crops may have been used to support such animals.
INTRODUCTION
More than simply the range of livestock species, we want to understand the
importance and husbandry of each in order to describe their role within Iberian communities. An emerging picture from detailed zooarchaeological reporting suggests a diverse herd in most sites, the prime importance of ovicaprids (with goats perhaps more common than sheep) and a gentle rise in the
number of swine over the late Iron Age (Iborra, 2004).
Figure 1 presents the proportion of different species from Iberian (and
slightly later) sites in València and its surrounding provinces. Most of these
sites are collected in Iborra 2004, to which I have added studies by Pérez,
Alonso & Iborra 2007; Sanchis 2002, 2006; and González 2013 to expand
the range of dates. The sites are dated by century with ‘second Iron Age’
sites treated as Middle Iberian (fifth to fourth or to third centuries BC). Proportions are calculated on the basis of Minimum Number of Individuals or
Number of Remains where the former is unavailable. Note that the sample from the final century BC at Ilici includes only a handful of individuals
Interaccions entre felins i humans. III Jornades d’arqueozoologia.
Museu de Prehistòria de València (2017): 425-438
[page-n-2]
BENJAMIN NAYLOR
(González, 2013: 330-1). As can be seen, the range of livestock even from
small sites is clear. And the additional studies do not change the general
trends for the Iberian era as identified by Iborra.
There are many reasons why households and communities might prefer
different mixes of livestock. Animals might be kept for their meat, secondary
products, for work, as a reserve of wealth, an expression of status or any combination of these reasons. They may be well or poorly cared-for. And not only
FIGURE 1. The proportions of different species in the faunal remains
426
[page-n-3]
Maintenance of Iberian livestock
do settlements differ in size, population, characteristics and environments,
but so too practices differ from household to household, as shown by analysis
of the distribution of tools and animal remains in La Bastida de les Alcusses
(Bonet, Soria y Vives-Ferrándiz, 2011: 171; Iborra y Vives-Ferrándiz, 2015). We
should also consider processes of formation, preservation and recovery of archaeological remains that likely change the apparent mix of livestock, presumably privileging larger animals. And although the number of sites analysed
is growing, it is still low and mainly includes larger sites.
Accounting for the presence and importance of different livestock species
is tricky. For this reason, I seek to simplify the analysis along the lines of livestock behaviour (or agency). I divide the species into two groups. First, the
largest animals, which need the most fodder: equids and cows. Second, the
medium-sized animals that generally require less fodder: sheep, goats and
swine. We can consider these groups as high and low-maintenance animals
respectively. I exclude remains of small animals (in particular, fowl and dogs)
and wild beasts.
This division is a working hypothesis only. Sheep, pigs and even goats may
be foddered to some extent, and sheep often require dedicated provision of
salt (Iborra, 2004: 323-4). Conversely, cows and to a lesser extent equids may
be abandoned to their luck for a season or fed principally with plants that
do not compete with human consumption (Foxhall, 1998). And ethnographic examples show that cows may be left with ovicaprids when they are not
needed for work (Khuele, 1983). But as a general rule, equids and cows have
markedly higher maintenance requirements, particularly in terms of fodder
and water and particularly when needed for work (Fall, 1995).
Reconsidering Figure 1 in this light, there is not a clear chronological division in the proportions of large and medium-sized animals. The proportions
oscillate between five and twenty-five percent, although with more variation
in later centuries.
We can also use this division to consider geographical patterns, as shown
in Figures 2 and 3 (the last is a close-up area of the first). The proportion of
large animals is shown in black, the ovicaprids and swine in white. Although
maps at such large scales provide only a general impression, no clear pattern
is evident. Neighbouring sites of the same era might have proportions that
are very similar (such as El Castellet de Bernabé and El Tossal de Sant Miguel)
or very divergent (La Coronilla de Chera and El Castellar de Berrueco).
427
[page-n-4]
BENJAMIN NAYLOR
Should we expect a geographical or chronological relationship with animals that require more fodder? The latter seems less clear. Studies of the
written sources have suggested a diffusion of new fodder crops in the Italian
peninsula during the last two centuries BC, which might suggest an increased
ability to maintain large animals (White, 1970: 202). But archaeobotanical studies in the Iberian peninsula have found a range of legumes available, at least
through the Iron Age and much earlier in some cases (Alonso, 2000: 34-5).
Geography on the other hand is intuitively related to different agricultural
and pastoral strategies.
FIGURE 2. The proportions of large and medium-sized livestock
428
[page-n-5]
Maintenance of Iberian livestock
METHODS
To explore this relationship, I use precipitation data, kindly provided by the
Spanish Agencia Estatal de Meteorología. Late-twentieth century measurements will never be an exact match for the climate in antiquity, owing to regional variation in climates over time and changes in additional factors such
as the soil and vegetative cover (Riera et al., 2009). But they provide the density of data necessary to investigate the scale and patterns of variation. This
density is evident in figure 4, which shows the meteorological stations in the
provinces of Castelló, Teruel, València, Cuenca, Alacant and Albacete. Unfor-
FIGURE 3. The proportions of large and medium-sized livestock (close-up of
central València)
429
[page-n-6]
BENJAMIN NAYLOR
tunately this data does not cover the sites of Fuente el Saz, La Coronilla or Los
Castellazos, which are excluded from the relevant analyses. Each point displays the average annual precipitation in the decade 1981-90. Best practice in
climate modelling is to use three decades of data but given changes in station
operation over such a long period, this reduces the number of stations available. I accept a shorter time period in order to utilise data as close as possible
to the archaeological sites (even accepting occasional stations with less than
the full decade of data, down to a minimum of three years). As can be seen,
using this approach, there are generally multiple stations within around ten
kilometres from each site. To calculate the average precipitation for each site,
I use the average of the three nearest stations (without weighting).
This focus on nearby resources reflects an unwillingness to accept transhumance as a widespread livestock strategy in the Iberian era. This debate is
not settled but recent studies have tended to exclude the practice (e.g., Valenzuela et al., 2015), something which accords with the density and political
fragmentation of Iberian settlement patterns. Regardless, transhumance is
less relevant to this study because it is more practicable for ovicaprids than
equids and cows and so would tend to exacerbate the difficulty in maintaining larger and particularly working animals (although there are references
to its use for larger animals as well: Pliny Epistulae 2.17.3)
A further environmental measure is the ‘period of frost or cold’ calculated
by the Ministerio de Agricultura, Alimentación y Medio Ambiente (available
through their WMS). This measures the ‘number of months in which the average minimum temperature is less than 7 °C’. This is a somewhat crude measure but a useful complement to rainfall in that it provides a different type
of constraint, especially given some mountainous areas with higher rainfall.
RESULTS
Figure 5 shows that the resulting correlation with the proportion of highmaintenance livestock is positive (0,23) although not dramatically so. This is
unsurprising and provides little information about the mechanisms through
which rainfall is being used, although it does provides some confidence in
the general approach.
Figure 6 shows the relationship between high-maintenance animals and
months of cold or frost. The results are more granular than figure 5, but show
an expected negative correlation of -0,21. That is, higher proportions of large
430
[page-n-7]
Maintenance of Iberian livestock
animals are correlated with fewer cold months. This analysis includes the
three sites not covered by the precipitation data.
These two correlations appear logical although with one wrinkle. Conventionally, the Spanish interior has been associated with higher levels of
pastoralism despite its long and hard winters. And analysis of Iron Age tomb
furnishings has suggested that the eastern (Iberian) coast had lower levels of
horse-ownership than the (Celtiberian) north Meseta, although this could of
course reflect different burial practices (Quesada, 1998: 179).
Accepting these general correlations, however, leaves open the question
of their mechanism. More rainfall and less interruption to the growing sea-
FIGURE 4. Meteorological stations and average annual precipitation
431
[page-n-8]
BENJAMIN NAYLOR
son by inclement weather could be related to more high-maintenance livestock in both direct and indirect ways. Directly, it provides a global increase
in vegetation, hence providing more pasture. Indirectly, it permits greater
agricultural fertility and so facilitates a range of strategies to support large
livestock. Greater fertility increases the ability to divert crops to livestock,
pasture livestock on sprouting crops, cut some crops early as green fodder,
include more fodder plants (both in autumn and spring) in crop rotations, or
include thirstier, multi-use crops such as broad beans.
These indirect mechanisms remind us of Halstead’s (1987) suggestion
of virtuous circles in intensive agriculture, in which increased fertiliser and
traction from higher numbers of livestock supports the additional feed requirements of these animals. A crucial independent variable, however, is the
availability of water. This is not to say that rainfall determines the complex
FIGURE 5. Average annual precipitation and the proportion of high-maintenance livestock
432
[page-n-9]
Maintenance of Iberian livestock
interaction of humans, animals and plants. Indeed, it may be manipulated in
turn through irrigation, storage, the digging of wells, terracing, and a multitide of passive strategies. But rainfall clearly plays a role.
Because these indirect strategies for supporting more livestock have implications for how rain is useful, I include two further analyses. Firstly, average spring rainfall. One way to add fodder crops to the rotation is by including
spring-sown crops in ‘fallow’ fields. Many legumes can be sown at this time,
particularly chickpeas and cow peas, but also lentils, broad beans and peas
(Halstead, 2014: 24; Chapa and Mayoral, 2007: 48). Some fodder legumes
(lucerne, grass pea and bitter vetch) can also be sown in spring, although
autumn is more common (Arnon, 1972: 578; Halstead, 2014: 24). Both millets
are sown in spring, as are oats and rye (Alonso, 2000: 33; Cubero, 1999: 59).
Some types of wheat and barley are also spring-sown.
FIGURE 6. Cold months and the proportion of high-maintenance livestock
433
[page-n-10]
BENJAMIN NAYLOR
All these crops have their own requirements but most important factor
for short-cycle crops is spring rainfall. As such, we would expect that if these
crops were an important part of supporting high-maintenance herds, these
animals would abound in areas with high spring rainfall. Figure 7 suggests,
however, that there is not a strong relation between these two factors (the
correlation is only 0,04).
Interestingly, this result is supported by the scarce evidence for the inclusion of C4 plants in animal diets (pertinent to the millets, not the legumes or other cereals). From isotope analysis of sheep’s teeth, Valenzuela et al.
(2015) have suggested that occasional foddering with millets was a possibility
but uncertain in coastal Catalunya. And isotope analysis for two cows and a
sheep amongst twelve domestic animals from Castellet de Bernabé and La
Bastida de les Alcusses (València) found signs of only a small contribution of
FIGURE 7. Average spring precipitation and the proportion of high-maintenance livestock
434
[page-n-11]
Maintenance of Iberian livestock
C4 plants to their diet (Salazar-García, et al. 2010). Overall, the evidence for
fodder millets, and more broadly the role of legumes in general, is still very
unclear (Alonso, 2000: 35).
Figure 8 presents summer rainfall. Summer is a difficult period for livestock in the Mediterranean. Given mild, rainy winters, summer is the period
of greatest vegetative scarcity for livestock (aside from intermittent crop residues). Hence the models of transhumance found particularly in the Medieval
period. The correlation here is inverse (-0,22). That is, the areas with more
high-maintenance livestock also have drier summers.
How should we interpret this disconnect? Summer rainfall is sporadic
in the Spanish Mediterranean (Cortesi et al., 2013). Possibly such unreliable
(and lower in general) rainfall simply has little effect regardless of level and
as such the correlation is meaningless. Possibly cows and equids were simply
FIGURE 8. Average summer precipitation and the proportion of high-maintenance livestock
435
[page-n-12]
BENJAMIN NAYLOR
expected to go hungry through summer, surviving on crop residues and poor
pasture until the autumn rains arrived, although this does not explain the
negative correlation. Possibly fluvial water compensates for a lack of rainfall. Many of the sites with higher proportions of large animals have access
to rivers and rich fluvial soils, in particular Comte de Trénor and L’Almoina
in València city, Alter de Vintihuiteña and Ilici (soil data from the Instituto
Geográfico Nacional, available through their WMS). Clearly there are many
reasons to locate settlements near to rivers. But in this case, it may be that
irrigation to assure pastures or the use of periodically-flooded zones allowed
the maintenance of larger herds of cows and equids through the summer
(Hodkinson, 1988: 47). An attractice alternative, however, considering the
positive relationship between high-maintenance animals and higher annual
rainfall, is that the Iberians were using fodder crops to effectively ‘transfer’
rainfall from the rest of the year and have it available during the dry summer.
This possibility would accord with a more intensive, ‘integrated’ model of
Iberian agriculture, at least in some sites, although this suggestion must remain a working hypothesis.
CONCLUSIONS
Excavation of Iberian settlements is beginning to provide not just a rich picture
of the many activities of Iberian life but also some idea of regional differences.
Pushing further on how these settlements ‘worked’ – the relative importance of
different activities, the integration of different actors in each practice, the seasonal rhythms and population characteristics – will require that we test syntheses
of the (growing number of) sites against hypotheses of how these settlements
worked. Hopefully, this approach can resolve long-standing questions about the
nature of Iberian communities and provide the basis for further investigation.
This study is a somewhat tentative attempt to test strategies originally derived
from ethnographic literature against a combination of climatic and zooarchaeological data. The indirect nature of the approach and the caveats around both
sets of data mean that the results are suggestive at best, although the basic findings for annual rainfall and cold months give some reassurance over the method. Indeed, a pessimistic view might interpret them as supporting an Iberian
agriculture less capable of overcoming environmental constraints. The seasonal
division produced mixed results. The lack of correlation between spring rainfall
436
[page-n-13]
Maintenance of Iberian livestock
and high-maintenance livestock is frustrating given it is a subject which seems
crucial to unlocking the complexity of Iberian crop rotations. The negative correlation with summer rainfall, however, is a surprising finding that suggests
water ‘preservation’ through fodder crops may have been a feature of Iberian life
in communities which faced a seasonal deficit.
BIBLIOGRAPHY
Alonso, N. (2000): Cultivos y producción agrícola en época ibérica. Saguntum-PLAV
Extra-3, 25-46.
Arnon, I. (1972): Crop production in dry regions. Leonard Hill, London.
Bonet, H., Soria, L., Vives-Ferrándiz, J. (2011): La vida en las casas: Producción doméstica,
alimentación, enseres y ocupantes. En H. Bonet y J. Vives-Ferrándiz. (eds.), La Bastida de les Alcusses: 1928-2010. Museu de Prehistòria de València, Valencia, 139-175.
Chapa, T., Mayoral, V. (2007): Arqueología del trabajo: El ciclo de la vida en un poblado ibérico. AKAL Arqueología, Madrid.
Cortesi, N., González, J. C., Brunetti, M., de Luís, M. (2013): Spatial variability of precipitation in Spain. Regional Environmental Change 12 (4), 1743-1749.
Cubero, C. (1999): Agricultura y recolección en el área celtibérica a partir de datos paleocarpológicos. En F. Burillo (ed.), IV Simposio sobre los Celtíberos: Economía.
Institución Fernando el Católico, Zaragoza, 47-61.
Fall, A. (1995): Feeding and working strategies for draught oxen in semi-arid zones.
International Livestock Reseach Institute y University of Edinburgh, Edinburgh.
Foxhall, L. (1998): Snapping up the unconsidered trifles: The use of agricultural residues in
ancient Greek and Roman farming. En M. Charles, P. Halstead y G. Jones (eds.), Fodder:
Archaeological, historical and ethnographic studies. Oxbow Books, Oxford, 35-40.
González, A. (2013): La riqueza en la basura: Estudio del registro arqueozoológico de
la excavación de la muralla oeste de la antigua ciudad de Ilici. En A. Sanchis y J. L.
Pascual (eds.), Animals i arqueologia hui: I Jornades d’Arqueozoologia. Museu de
Prehistòria de València, Valencia, 323-348.
Halstead, P. (1987): Traditional and ancient rural economy in mediterranean Europe:
Plus ça change?, Journal of Hellenic Studies 107, 77-87.
Halstead, P. (2014): Two oxen ahead: Pre-mechanized farming in the Mediterranean.
Wiley Blackwell, Oxford.
Hodkinson, S. (1988): Animal husbandry in the Greek polis. En C. Whittaker (ed.), Pastoral
economies in classical antiquity. Cambridge Fhilological Society, Cambridge, 134-184.
Iborra, M. P. (2004): La ganadería y la caza desde el bronce hasta el ibérico final en el
territorio valenciano. Serie de Trabajos Varios del SIP 94, Valencia.
437
[page-n-14]
BENJAMIN NAYLOR
Iborra, M. P., Vives-Ferrándiz, J. (2015): Consumo de carne y diversidad social: La distribución espacial de los restos faunísticos en la Bastida de les Alcusses (Moixent,
Valencia). En A. Sanchis y J. L. Pascual (eds.), Preses petits i grups humans en el
passat: II Jornades d’Arqueozoologia del Museu de Prehistòria de València. Museu de Prehistòria de València, Valencia, 287-306.
Khuele, M. (1983): The integration of livestock into irrigated agricultural systems in
semi-arid tropics with particular reference to Gezira. Unpublished Thesis, University of Edinburgh, Edinburgh.
Pérez, G., Alonso, N., Iborra, M. P. (2007): Agricultura y ganadería protohistóricas
en la península ibérica: Modelos de gestión. En A. Rodríguez e I. Pavón (eds.),
Arqueología de la tierra: Paisajes rurales de la protohistoria peninsular. Universidad de Extremadura, Cáceres, 327-373.
Quesada, F. (1998): Aristócratas a caballo y la existencia de una verdadera caballería
en la cultura ibérica: Dos ámbitos conceptuales diferentes. En C. Aranegui (ed.),
Actas del Congreso internacional: Los Íberos, principes de Occidente. Universitat
de València, Valencia, 169-183.
Riera, S., Currás, A., Palet, J. M., Ejarque, A., Orengo, H., Julià, R., Miras, Y. (2009):
Variabilité climatique, occupation du sol et paysage en Espagne de l’Âge du fer
à l’époque médiévale: Intégration des données paléoenvironmentales et de
l’archéologie du paysage. En E. Hermon (ed.), Changements climatiques dans une
perspective historique et systémique des interactions société-environnement naturel dans l’empire romain. Université Laval, Québec, 251-280.
Salazar-García, D., Vives-Ferrándiz, J., Fuller, B., Richards, M. (2010): Alimentación
estimada de la población del Castellet de Bernabé (ss.V-III a.C.) mediante el uso
de ratios de isótopos estables de C y N. Saguntum-PLAV Extra 9, 313-322.
Sanchis, A. (2002): La alimentación de origen animal en la Valentia republicana. En J.
L. Jiménez y A. Ribera (eds.), Valencia y las primeras ciudades romanas de Hispania. Ajuntament de València, Valencia, 323-334.
Sanchis, A. (2006): Estudio arqueozoológico. En R. Albiach y J. L. de Madaria (coords.),
La villa de Cornelius. Administrador de Infraestructuras Ferroviarias y Ministerio
de Fomento, Valencia, 19-27.
Valenzuela, S., Jiménez, S., Evans, J., López, D., Jornet, R., Albarella, U. (2015): Analysis
of seasonal mobility of sheep in Iron Age Catalonia (north-eastern Spain) based
on strontium and oxygen isotope analysis from tooth enamel: First results. Journal
of Archaeological Science: Reports 6, 828-836.
White, K.D. (1970): Roman farming. Thames & Hudson, London.
438
[page-n-15]
18
MAINTENANCE OF IBERIAN LIVESTOCK
Benjamin Naylor
ABSTRACT
The division of the main species of Iron Age Iberian livestock into categories of
large, high-maintenance animals (equids and cattle) and medium-sized, lowmaintenance animals (sheep, goats and pigs) allows us to investigate grosso
modo Iberian livestock management. High proportions of large animals are,
unsurprisingly, correlated with greater rainfall and fewer cold months. Yet
they are also correlated with drier summers, suggesting techniques such as
growing fodder crops may have been used to support such animals.
INTRODUCTION
More than simply the range of livestock species, we want to understand the
importance and husbandry of each in order to describe their role within Iberian communities. An emerging picture from detailed zooarchaeological reporting suggests a diverse herd in most sites, the prime importance of ovicaprids (with goats perhaps more common than sheep) and a gentle rise in the
number of swine over the late Iron Age (Iborra, 2004).
Figure 1 presents the proportion of different species from Iberian (and
slightly later) sites in València and its surrounding provinces. Most of these
sites are collected in Iborra 2004, to which I have added studies by Pérez,
Alonso & Iborra 2007; Sanchis 2002, 2006; and González 2013 to expand
the range of dates. The sites are dated by century with ‘second Iron Age’
sites treated as Middle Iberian (fifth to fourth or to third centuries BC). Proportions are calculated on the basis of Minimum Number of Individuals or
Number of Remains where the former is unavailable. Note that the sample from the final century BC at Ilici includes only a handful of individuals
Interaccions entre felins i humans. III Jornades d’arqueozoologia.
Museu de Prehistòria de València (2017): 425-438
[page-n-2]
BENJAMIN NAYLOR
(González, 2013: 330-1). As can be seen, the range of livestock even from
small sites is clear. And the additional studies do not change the general
trends for the Iberian era as identified by Iborra.
There are many reasons why households and communities might prefer
different mixes of livestock. Animals might be kept for their meat, secondary
products, for work, as a reserve of wealth, an expression of status or any combination of these reasons. They may be well or poorly cared-for. And not only
FIGURE 1. The proportions of different species in the faunal remains
426
[page-n-3]
Maintenance of Iberian livestock
do settlements differ in size, population, characteristics and environments,
but so too practices differ from household to household, as shown by analysis
of the distribution of tools and animal remains in La Bastida de les Alcusses
(Bonet, Soria y Vives-Ferrándiz, 2011: 171; Iborra y Vives-Ferrándiz, 2015). We
should also consider processes of formation, preservation and recovery of archaeological remains that likely change the apparent mix of livestock, presumably privileging larger animals. And although the number of sites analysed
is growing, it is still low and mainly includes larger sites.
Accounting for the presence and importance of different livestock species
is tricky. For this reason, I seek to simplify the analysis along the lines of livestock behaviour (or agency). I divide the species into two groups. First, the
largest animals, which need the most fodder: equids and cows. Second, the
medium-sized animals that generally require less fodder: sheep, goats and
swine. We can consider these groups as high and low-maintenance animals
respectively. I exclude remains of small animals (in particular, fowl and dogs)
and wild beasts.
This division is a working hypothesis only. Sheep, pigs and even goats may
be foddered to some extent, and sheep often require dedicated provision of
salt (Iborra, 2004: 323-4). Conversely, cows and to a lesser extent equids may
be abandoned to their luck for a season or fed principally with plants that
do not compete with human consumption (Foxhall, 1998). And ethnographic examples show that cows may be left with ovicaprids when they are not
needed for work (Khuele, 1983). But as a general rule, equids and cows have
markedly higher maintenance requirements, particularly in terms of fodder
and water and particularly when needed for work (Fall, 1995).
Reconsidering Figure 1 in this light, there is not a clear chronological division in the proportions of large and medium-sized animals. The proportions
oscillate between five and twenty-five percent, although with more variation
in later centuries.
We can also use this division to consider geographical patterns, as shown
in Figures 2 and 3 (the last is a close-up area of the first). The proportion of
large animals is shown in black, the ovicaprids and swine in white. Although
maps at such large scales provide only a general impression, no clear pattern
is evident. Neighbouring sites of the same era might have proportions that
are very similar (such as El Castellet de Bernabé and El Tossal de Sant Miguel)
or very divergent (La Coronilla de Chera and El Castellar de Berrueco).
427
[page-n-4]
BENJAMIN NAYLOR
Should we expect a geographical or chronological relationship with animals that require more fodder? The latter seems less clear. Studies of the
written sources have suggested a diffusion of new fodder crops in the Italian
peninsula during the last two centuries BC, which might suggest an increased
ability to maintain large animals (White, 1970: 202). But archaeobotanical studies in the Iberian peninsula have found a range of legumes available, at least
through the Iron Age and much earlier in some cases (Alonso, 2000: 34-5).
Geography on the other hand is intuitively related to different agricultural
and pastoral strategies.
FIGURE 2. The proportions of large and medium-sized livestock
428
[page-n-5]
Maintenance of Iberian livestock
METHODS
To explore this relationship, I use precipitation data, kindly provided by the
Spanish Agencia Estatal de Meteorología. Late-twentieth century measurements will never be an exact match for the climate in antiquity, owing to regional variation in climates over time and changes in additional factors such
as the soil and vegetative cover (Riera et al., 2009). But they provide the density of data necessary to investigate the scale and patterns of variation. This
density is evident in figure 4, which shows the meteorological stations in the
provinces of Castelló, Teruel, València, Cuenca, Alacant and Albacete. Unfor-
FIGURE 3. The proportions of large and medium-sized livestock (close-up of
central València)
429
[page-n-6]
BENJAMIN NAYLOR
tunately this data does not cover the sites of Fuente el Saz, La Coronilla or Los
Castellazos, which are excluded from the relevant analyses. Each point displays the average annual precipitation in the decade 1981-90. Best practice in
climate modelling is to use three decades of data but given changes in station
operation over such a long period, this reduces the number of stations available. I accept a shorter time period in order to utilise data as close as possible
to the archaeological sites (even accepting occasional stations with less than
the full decade of data, down to a minimum of three years). As can be seen,
using this approach, there are generally multiple stations within around ten
kilometres from each site. To calculate the average precipitation for each site,
I use the average of the three nearest stations (without weighting).
This focus on nearby resources reflects an unwillingness to accept transhumance as a widespread livestock strategy in the Iberian era. This debate is
not settled but recent studies have tended to exclude the practice (e.g., Valenzuela et al., 2015), something which accords with the density and political
fragmentation of Iberian settlement patterns. Regardless, transhumance is
less relevant to this study because it is more practicable for ovicaprids than
equids and cows and so would tend to exacerbate the difficulty in maintaining larger and particularly working animals (although there are references
to its use for larger animals as well: Pliny Epistulae 2.17.3)
A further environmental measure is the ‘period of frost or cold’ calculated
by the Ministerio de Agricultura, Alimentación y Medio Ambiente (available
through their WMS). This measures the ‘number of months in which the average minimum temperature is less than 7 °C’. This is a somewhat crude measure but a useful complement to rainfall in that it provides a different type
of constraint, especially given some mountainous areas with higher rainfall.
RESULTS
Figure 5 shows that the resulting correlation with the proportion of highmaintenance livestock is positive (0,23) although not dramatically so. This is
unsurprising and provides little information about the mechanisms through
which rainfall is being used, although it does provides some confidence in
the general approach.
Figure 6 shows the relationship between high-maintenance animals and
months of cold or frost. The results are more granular than figure 5, but show
an expected negative correlation of -0,21. That is, higher proportions of large
430
[page-n-7]
Maintenance of Iberian livestock
animals are correlated with fewer cold months. This analysis includes the
three sites not covered by the precipitation data.
These two correlations appear logical although with one wrinkle. Conventionally, the Spanish interior has been associated with higher levels of
pastoralism despite its long and hard winters. And analysis of Iron Age tomb
furnishings has suggested that the eastern (Iberian) coast had lower levels of
horse-ownership than the (Celtiberian) north Meseta, although this could of
course reflect different burial practices (Quesada, 1998: 179).
Accepting these general correlations, however, leaves open the question
of their mechanism. More rainfall and less interruption to the growing sea-
FIGURE 4. Meteorological stations and average annual precipitation
431
[page-n-8]
BENJAMIN NAYLOR
son by inclement weather could be related to more high-maintenance livestock in both direct and indirect ways. Directly, it provides a global increase
in vegetation, hence providing more pasture. Indirectly, it permits greater
agricultural fertility and so facilitates a range of strategies to support large
livestock. Greater fertility increases the ability to divert crops to livestock,
pasture livestock on sprouting crops, cut some crops early as green fodder,
include more fodder plants (both in autumn and spring) in crop rotations, or
include thirstier, multi-use crops such as broad beans.
These indirect mechanisms remind us of Halstead’s (1987) suggestion
of virtuous circles in intensive agriculture, in which increased fertiliser and
traction from higher numbers of livestock supports the additional feed requirements of these animals. A crucial independent variable, however, is the
availability of water. This is not to say that rainfall determines the complex
FIGURE 5. Average annual precipitation and the proportion of high-maintenance livestock
432
[page-n-9]
Maintenance of Iberian livestock
interaction of humans, animals and plants. Indeed, it may be manipulated in
turn through irrigation, storage, the digging of wells, terracing, and a multitide of passive strategies. But rainfall clearly plays a role.
Because these indirect strategies for supporting more livestock have implications for how rain is useful, I include two further analyses. Firstly, average spring rainfall. One way to add fodder crops to the rotation is by including
spring-sown crops in ‘fallow’ fields. Many legumes can be sown at this time,
particularly chickpeas and cow peas, but also lentils, broad beans and peas
(Halstead, 2014: 24; Chapa and Mayoral, 2007: 48). Some fodder legumes
(lucerne, grass pea and bitter vetch) can also be sown in spring, although
autumn is more common (Arnon, 1972: 578; Halstead, 2014: 24). Both millets
are sown in spring, as are oats and rye (Alonso, 2000: 33; Cubero, 1999: 59).
Some types of wheat and barley are also spring-sown.
FIGURE 6. Cold months and the proportion of high-maintenance livestock
433
[page-n-10]
BENJAMIN NAYLOR
All these crops have their own requirements but most important factor
for short-cycle crops is spring rainfall. As such, we would expect that if these
crops were an important part of supporting high-maintenance herds, these
animals would abound in areas with high spring rainfall. Figure 7 suggests,
however, that there is not a strong relation between these two factors (the
correlation is only 0,04).
Interestingly, this result is supported by the scarce evidence for the inclusion of C4 plants in animal diets (pertinent to the millets, not the legumes or other cereals). From isotope analysis of sheep’s teeth, Valenzuela et al.
(2015) have suggested that occasional foddering with millets was a possibility
but uncertain in coastal Catalunya. And isotope analysis for two cows and a
sheep amongst twelve domestic animals from Castellet de Bernabé and La
Bastida de les Alcusses (València) found signs of only a small contribution of
FIGURE 7. Average spring precipitation and the proportion of high-maintenance livestock
434
[page-n-11]
Maintenance of Iberian livestock
C4 plants to their diet (Salazar-García, et al. 2010). Overall, the evidence for
fodder millets, and more broadly the role of legumes in general, is still very
unclear (Alonso, 2000: 35).
Figure 8 presents summer rainfall. Summer is a difficult period for livestock in the Mediterranean. Given mild, rainy winters, summer is the period
of greatest vegetative scarcity for livestock (aside from intermittent crop residues). Hence the models of transhumance found particularly in the Medieval
period. The correlation here is inverse (-0,22). That is, the areas with more
high-maintenance livestock also have drier summers.
How should we interpret this disconnect? Summer rainfall is sporadic
in the Spanish Mediterranean (Cortesi et al., 2013). Possibly such unreliable
(and lower in general) rainfall simply has little effect regardless of level and
as such the correlation is meaningless. Possibly cows and equids were simply
FIGURE 8. Average summer precipitation and the proportion of high-maintenance livestock
435
[page-n-12]
BENJAMIN NAYLOR
expected to go hungry through summer, surviving on crop residues and poor
pasture until the autumn rains arrived, although this does not explain the
negative correlation. Possibly fluvial water compensates for a lack of rainfall. Many of the sites with higher proportions of large animals have access
to rivers and rich fluvial soils, in particular Comte de Trénor and L’Almoina
in València city, Alter de Vintihuiteña and Ilici (soil data from the Instituto
Geográfico Nacional, available through their WMS). Clearly there are many
reasons to locate settlements near to rivers. But in this case, it may be that
irrigation to assure pastures or the use of periodically-flooded zones allowed
the maintenance of larger herds of cows and equids through the summer
(Hodkinson, 1988: 47). An attractice alternative, however, considering the
positive relationship between high-maintenance animals and higher annual
rainfall, is that the Iberians were using fodder crops to effectively ‘transfer’
rainfall from the rest of the year and have it available during the dry summer.
This possibility would accord with a more intensive, ‘integrated’ model of
Iberian agriculture, at least in some sites, although this suggestion must remain a working hypothesis.
CONCLUSIONS
Excavation of Iberian settlements is beginning to provide not just a rich picture
of the many activities of Iberian life but also some idea of regional differences.
Pushing further on how these settlements ‘worked’ – the relative importance of
different activities, the integration of different actors in each practice, the seasonal rhythms and population characteristics – will require that we test syntheses
of the (growing number of) sites against hypotheses of how these settlements
worked. Hopefully, this approach can resolve long-standing questions about the
nature of Iberian communities and provide the basis for further investigation.
This study is a somewhat tentative attempt to test strategies originally derived
from ethnographic literature against a combination of climatic and zooarchaeological data. The indirect nature of the approach and the caveats around both
sets of data mean that the results are suggestive at best, although the basic findings for annual rainfall and cold months give some reassurance over the method. Indeed, a pessimistic view might interpret them as supporting an Iberian
agriculture less capable of overcoming environmental constraints. The seasonal
division produced mixed results. The lack of correlation between spring rainfall
436
[page-n-13]
Maintenance of Iberian livestock
and high-maintenance livestock is frustrating given it is a subject which seems
crucial to unlocking the complexity of Iberian crop rotations. The negative correlation with summer rainfall, however, is a surprising finding that suggests
water ‘preservation’ through fodder crops may have been a feature of Iberian life
in communities which faced a seasonal deficit.
BIBLIOGRAPHY
Alonso, N. (2000): Cultivos y producción agrícola en época ibérica. Saguntum-PLAV
Extra-3, 25-46.
Arnon, I. (1972): Crop production in dry regions. Leonard Hill, London.
Bonet, H., Soria, L., Vives-Ferrándiz, J. (2011): La vida en las casas: Producción doméstica,
alimentación, enseres y ocupantes. En H. Bonet y J. Vives-Ferrándiz. (eds.), La Bastida de les Alcusses: 1928-2010. Museu de Prehistòria de València, Valencia, 139-175.
Chapa, T., Mayoral, V. (2007): Arqueología del trabajo: El ciclo de la vida en un poblado ibérico. AKAL Arqueología, Madrid.
Cortesi, N., González, J. C., Brunetti, M., de Luís, M. (2013): Spatial variability of precipitation in Spain. Regional Environmental Change 12 (4), 1743-1749.
Cubero, C. (1999): Agricultura y recolección en el área celtibérica a partir de datos paleocarpológicos. En F. Burillo (ed.), IV Simposio sobre los Celtíberos: Economía.
Institución Fernando el Católico, Zaragoza, 47-61.
Fall, A. (1995): Feeding and working strategies for draught oxen in semi-arid zones.
International Livestock Reseach Institute y University of Edinburgh, Edinburgh.
Foxhall, L. (1998): Snapping up the unconsidered trifles: The use of agricultural residues in
ancient Greek and Roman farming. En M. Charles, P. Halstead y G. Jones (eds.), Fodder:
Archaeological, historical and ethnographic studies. Oxbow Books, Oxford, 35-40.
González, A. (2013): La riqueza en la basura: Estudio del registro arqueozoológico de
la excavación de la muralla oeste de la antigua ciudad de Ilici. En A. Sanchis y J. L.
Pascual (eds.), Animals i arqueologia hui: I Jornades d’Arqueozoologia. Museu de
Prehistòria de València, Valencia, 323-348.
Halstead, P. (1987): Traditional and ancient rural economy in mediterranean Europe:
Plus ça change?, Journal of Hellenic Studies 107, 77-87.
Halstead, P. (2014): Two oxen ahead: Pre-mechanized farming in the Mediterranean.
Wiley Blackwell, Oxford.
Hodkinson, S. (1988): Animal husbandry in the Greek polis. En C. Whittaker (ed.), Pastoral
economies in classical antiquity. Cambridge Fhilological Society, Cambridge, 134-184.
Iborra, M. P. (2004): La ganadería y la caza desde el bronce hasta el ibérico final en el
territorio valenciano. Serie de Trabajos Varios del SIP 94, Valencia.
437
[page-n-14]
BENJAMIN NAYLOR
Iborra, M. P., Vives-Ferrándiz, J. (2015): Consumo de carne y diversidad social: La distribución espacial de los restos faunísticos en la Bastida de les Alcusses (Moixent,
Valencia). En A. Sanchis y J. L. Pascual (eds.), Preses petits i grups humans en el
passat: II Jornades d’Arqueozoologia del Museu de Prehistòria de València. Museu de Prehistòria de València, Valencia, 287-306.
Khuele, M. (1983): The integration of livestock into irrigated agricultural systems in
semi-arid tropics with particular reference to Gezira. Unpublished Thesis, University of Edinburgh, Edinburgh.
Pérez, G., Alonso, N., Iborra, M. P. (2007): Agricultura y ganadería protohistóricas
en la península ibérica: Modelos de gestión. En A. Rodríguez e I. Pavón (eds.),
Arqueología de la tierra: Paisajes rurales de la protohistoria peninsular. Universidad de Extremadura, Cáceres, 327-373.
Quesada, F. (1998): Aristócratas a caballo y la existencia de una verdadera caballería
en la cultura ibérica: Dos ámbitos conceptuales diferentes. En C. Aranegui (ed.),
Actas del Congreso internacional: Los Íberos, principes de Occidente. Universitat
de València, Valencia, 169-183.
Riera, S., Currás, A., Palet, J. M., Ejarque, A., Orengo, H., Julià, R., Miras, Y. (2009):
Variabilité climatique, occupation du sol et paysage en Espagne de l’Âge du fer
à l’époque médiévale: Intégration des données paléoenvironmentales et de
l’archéologie du paysage. En E. Hermon (ed.), Changements climatiques dans une
perspective historique et systémique des interactions société-environnement naturel dans l’empire romain. Université Laval, Québec, 251-280.
Salazar-García, D., Vives-Ferrándiz, J., Fuller, B., Richards, M. (2010): Alimentación
estimada de la población del Castellet de Bernabé (ss.V-III a.C.) mediante el uso
de ratios de isótopos estables de C y N. Saguntum-PLAV Extra 9, 313-322.
Sanchis, A. (2002): La alimentación de origen animal en la Valentia republicana. En J.
L. Jiménez y A. Ribera (eds.), Valencia y las primeras ciudades romanas de Hispania. Ajuntament de València, Valencia, 323-334.
Sanchis, A. (2006): Estudio arqueozoológico. En R. Albiach y J. L. de Madaria (coords.),
La villa de Cornelius. Administrador de Infraestructuras Ferroviarias y Ministerio
de Fomento, Valencia, 19-27.
Valenzuela, S., Jiménez, S., Evans, J., López, D., Jornet, R., Albarella, U. (2015): Analysis
of seasonal mobility of sheep in Iron Age Catalonia (north-eastern Spain) based
on strontium and oxygen isotope analysis from tooth enamel: First results. Journal
of Archaeological Science: Reports 6, 828-836.
White, K.D. (1970): Roman farming. Thames & Hudson, London.
438
[page-n-15]