An approach to prehistoric shepherding in La Gomera (Canary Islands) through the study of domestic spaces

Quaternary International xxx (2015) 1e13

Contents lists available at ScienceDirect

Quaternary International journal homepage: www.elsevier.com/locate/quaint

An approach to prehistoric shepherding in La Gomera (Canary Islands) through the study of domestic spaces
ndez-Marrero a, *, Juan-Francisco Navarro-Mederos b, Juan-Carlos Herna
-Miguel Trujillo-Mora a, Sandra Cancel c, Carmen Machado-Yanes b, Jorge Pais d, Jose Jacob Morales f, Juan-Carlos Rando e
n de La Gomera, Canary Islands, Spain
gico de La Gomera, E-38800, San Sebastia Museo Arqueolo Archaeology of Territory Research Group, Dept. de Geografía e Historia (UDI Prehistoria, Arqueología e Historia Antigua), Universidad de La Laguna, E38206, La Laguna, Canary Islands, Spain c
rico, Spain Servicios Integrales de Patrimonio Histo d
gico Benahoarita, E-38760, La Palma, Canary Islands, Spain Museo Arqueolo e Tarha Research Group, Universidad de Las Palmas de Gran Canaria, E-35001, Las Palmas, Canary Islands, Spain f Dept. de Biología Animal (UDI Zoología), Universidad de La Laguna, E-38206, La Laguna, Canary Islands, Spain a

b

a r t i c l e i n f o

a b s t r a c t

Article history: Available online xxx

The present research, included in the project “A history of shepherding on the Island of La Gomera”, investigates the historical development of shepherding on this island from prehistoric to modern times, and from a social, economic, territorial, ecological, and cultural perspective. Within the framework of the project and its multidisciplinary approach, we present the initial phase of the archaeological study, which aims to delve into the management and consumption of livestock during the island's prehistory. In order to track this “culture” from an archeological standpoint, residential contexts have been proposed as a first stage in the survey, as they have not so far been addressed as an important issue for the archeology of La Gomera. To this end, thirteen archaeological sondages have been carried out in five separate points of the island. Exhumed archaeological materials relating them to the stratigraphic context of each domestic space are analyzed. Data from the consumption of food, especially those from the livestock, has been studied. The text provides preliminary information on shepherding and management of the island territory by ancient Gomerans. © 2015 Elsevier Ltd and INQUA. All rights reserved.

Keywords: Shepherding Cave settlement Landscape La Gomera Canary Islands Prehispanic period

1. Geographical and historical frame With its 378 km2, La Gomera is the second smallest island of the seven comprising the Canarian Archipelago. However, it is also a most rugged one (Figs. 1 and 2). As with the rest of the archipelago, the origin of the island is volcanic. The basal complex has been dated at 15e20 Ma (Anguita et al., 2002; Carracedo, 2011). La

* Corresponding author. E-mail addresses: [email protected] (J.-C. Hern
andezMarrero), [email protected] (J.-F. Navarro-Mederos), [email protected] (J.-M. Trujillo-Mora), [email protected] (S. Cancel), [email protected] (C. Machado-Yanes), [email protected] (J. Pais), [email protected] (J. Morales), [email protected] (J.-C. Rando).

Gomera is the only island which has not seen any volcanic activity throughout the Quaternary, giving erosion an important role as a landscape shaper. The insular topography is characterized by deep ravines in a radial pattern around the island. A high and relatively smooth central plateau reaches an altitude of 1487 m at El Alto del Garajonay. The island is influenced by the trade winds (alisio) coming from the Azores anticyclone, which differentially affects the insular geography depending on its orientation. This climatic condition aids the development of extraordinarily diverse vegetation throughout the island: the north-eastern half with short, steep slopes, and the south-western half with longer, gentler ones. Authors normally refer to three major bioclimatic belts. On the top central area there is a dense monteverde (evergreen cloud forest or laurisilva), characterized by high humidity, winter mists, and many tree species

http://dx.doi.org/10.1016/j.quaint.2015.11.092 1040-6182/© 2015 Elsevier Ltd and INQUA. All rights reserved.


ndez-Marrero, J.-C., et al., An approach to prehistoric shepherding in La Gomera (Canary Islands) Please cite this article in press as: Herna through the study of domestic spaces, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.11.092

2


ndez-Marrero et al. / Quaternary International xxx (2015) 1e13 J.-C. Herna

Fig. 1. Location of the studied area.

Fig. 2. Map of La Gomera and selected archaeological sites for the background study.

(Ocotea foetens, Apollonias barbujana, Laurus novocanariensis, Persea indica, Morella faya, Erica arborea) (600e1487 m). Below this area is the domain of thermophilous forest with a moderate climate (200e600 m). Its composition usually presents clumps, dense thickets and undergrowth, as well as a high diversity and richness of species, namely Convolvulus floridus, Hypericum canariense, Olea cesariformes, Pistacia atlantica, Retama rhodorhizoides, Cistus monspeliensis or Spartocytisus filipes. Among these, Juniperus turbinata ssp. canariensis and Phoenix canariensis stand out. The third altitudinal belt, the basal vegetation, is considered semiarid and its soils endure erosion and drought. The lower the altitude, the more the size of the species is reduced (0e200 m). This

is the domain of Euphorbiaceae (Arozena, 1991; Gil-Rodríguez et al.,
ndez-Palacios and de Nicolas, 1995; Santos, A., 2008; 1992; Ferna
et al., 2013). Nogue It is estimated that human populations had arrived in the Canary Islands by the middle of the first millennium BC, coming from a North Africa Berber context (Navarro-Mederos, 2001). Europeans established contact with the archipelago during the 13th century, and by the end of the 15th century (1489) the conquest of La Gomera was finalized (Diaz and Rodríguez, 1990; Morales, 1994). Europeans introduced a manorial system, and the natives from La Gomera underwent a very slow and difficult acculturation process (Aznar, 1983, 1985; Lobo, 1984; Martínez, 1993; Baucells, 2014).


ndez-Marrero, J.-C., et al., An approach to prehistoric shepherding in La Gomera (Canary Islands) Please cite this article in press as: Herna through the study of domestic spaces, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.11.092


ndez-Marrero et al. / Quaternary International xxx (2015) 1e13 J.-C. Herna

The ancient inhabitants of the Canary Islands, with the exception of the island of Gran Canaria, lived mainly by shepherding. All narrative sources between the 15th and the 17th century agree that the ancient Gomerans were shepherds with a livestock composed mainly of goats (Capra hircus) and sheep (Ovis aries), followed to a much lesser extent by pigs (Abreu y Galindo, 1977; Zurara, 1978; Marín de Cubas, 1993). Not only did they practise an intense gathering activity which extended throughout the territory but they also exploited coastal and marine resources, mainly consuming limpets (Patella spp.) and other molluscs, crustaceans,
ndez, 2010). They cultivated barley, wheat and fish (Mesa-Herna and gathered fruit, stems, leaves, and rhizomes in a variety of vegetation ecosystems (Navarro-Mederos, 1992; Morales et al., 2011). 2. Introduction and background The occupation of animal husbandry dates back to the origins of the island's settlement and continues to this day, thus providing an uninterrupted link between past and present. Today, it is an alternative to the current economic drive towards an increasingly stabled livestock. The importance of this legacy is widely acknowledged. The Archaeological and Ethnographic Museums and the Insular Archive of La Gomera launched the present project in 2008, which aims to deepen the research into all of its aspects and to cover important demands on the island. Documenting the cultural diversity of the island and rating the social condition of shepherding as an historical activity which has produced a deeply embedded heritage in Gomeran society, the project has been carried out as a comprehensive and all-inclusive one. The multidisciplinary work has depended on the coordination of three distinct areas: the documentary study addresses the texts generated by ethno-history, but also any text from institutional or private sources providing data on livestock from La Gomera between the 15th and the 20th century. The work of anthropology and ethnography focuses on the 19th and 20th centuries, and much of the research is being done through interviews. The archaeological work is based on the study of the traces of animal husbandry in domestic spaces. Studies on domestic spaces or linked directly to any aspect of livestock in La Gomera are few and diachronically distant. In 1950,
). AcL. Diego Cuscoy dug the caves of “Punta Negra” (Alajero cording to the author, both caves served simultaneously as dwellings and cattle pens (Diego, 1953). The archaeological data is scarce and inconclusive. Navarro-Mederos directed the archeological excavation of “La Era de Los Antiguos” (1979, Vallermoso), a domestic outdoor hut structure (unpublished), as well as the
, 1983), specifically the excaworks at “Los Polieros” site (Alajero vation of a small rock shelter (Cave-F), which must have been the cooking area of an attached larger cave (Navarro-Mederos, 1984, 1988, 1992). In 2009, the study of paleodiet on bioanthropological materials at the site of El Accesso al Pescante de Vallehermoso was published; from this site we obtained the earliest date for La Gomera (Cal AD 120e330 and Cal AD 420e600). Analysis of stable isotopes of carbon and nitrogen, plus analysis of trace elements (barium and strontium) showed a diet with a high percentage of consumption of C3 plants (wheat or barley). It also pointed to a mixed diet of sea products, feral species and a significant consumption of livestock products such as milk or meat (Arnay et al., 2009). As for the funeral context, domestic animal bones can be found in burial caves, though not always. This was also detected where the prehistoric populations of Tenerife are concerned (Alberto, 1999). Livestock ~ a de Los Cocos or Los bones were found in burial caves at La Montan Riscos del Tabaibal (Navarro-Mederos, 1992), but not at La Cordillera

3

(Torres-Palenzuela, 1993), Los Tejeleches (Navarro-Mederos and Rosa-Arrocha, 1988), Los Polieros (Navarro-Mederos, 1984) or El Acceso al Pescante de Vallehermoso (Arnay et al., 2009). The social value of the domestic animal and/or the access to it as a product for consumption seems to be at the origin. An indication of this could be such a basic object as clothing, for which the inhabitants used all their livestock skins, in evidence throughout the whole archipelago according to the study of Gonz
alez and Rodríguez (1998). In La Gomera, there is another widely documented type of archaeological site which connects livestock to the religious sphere: the sacrificial shrine (Navarro-Mederos, 1992, 1994, 1996; Navarro-Mederos et al., 2001; Navarro-Mederos, 2002; Morales et al., 2011). It is built with stone walls and follows a circular or oval pattern. Its size is variable, containing one or more chambers where offerings were brought to sacrifice. Domestic livestock slaughter as part of a ritual display is undeniable. This staging occurs in the ceremonial centre of the ancient Gomerans par excellence, El Alto del Garajonay (Cal AD 340 to AD 600 and Cal 980 to 1060). There, the collected bones resulting from the cremation of goats, sheep and pigs (this last group in a comparably small quantity) approach the figure of 3,000,000 pieces, which appear in an extremely fragmented state (Navarro Mederos, 2005). In this context, the importance of the relationship between human beings and domestic animals is confirmed, exceeding the strict economic order to penetrate the realm of beliefs and ritual practices.

3. Objectives and methodology The smallest social unit in the daily life of the “local community”, domestic space metaphorically synthesizes both social relationships and the overall qualities of the productive system. Therefore, and for the orientation of archaeological research in La Gomera, we consider the study of the basic territorial unit a priority
ndez-Go
mez and Alberto, 2006; Herna
ndez-Marrero et al., (Herna
ndez-Marrero and Navarro-Mederos, 2012). We shall 2011; Herna preliminarily address livestock culture in the prehistory of La Gomera through its context and environment. As seen throughout multiple projects, archaeological surface sondages conducted in La Gomera (Navarro-Mederos, 1975, 1996, 1984, 1994, 2002; Navarro-Mederos et al., 2005; Torres, 1991; Hernandez-Marrero, 2014), and the information coming from ethno-historical texts (Baucells-Mesa, 2004), the Gomeran people, as the Portuguese navigator Gomes Eanes de Zurara wrote in 1453, “... don't have houses but they live in caves and huts ...” (Zurara, 1978). Native settlements on the island are linked to natural caves. The caves are shown individually or grouped in small or large units, depending on the geomorphology of the area. In the 2014 Inventory of Archaeological Heritage in La Gomera Island, a total of 947 archaeological sites were recorded: 4.22% are huts and 22.36% are residential caves, normally accompanied by burial caves forming highly cohesive archaeological zones in the landscape. Settlement models must have been combined caves and dry stone wall structures (Hern
andez-Marrero and Navarro-Mederos, 2012). Most of the open air settlements on the island have been altered by subsequent land movements, due to the scarcity of flat land suitable for cultivation. Originally being more numerous than the surface ones, many more cave settlements have been preserved (Navarro-Mederos and Clavijo (Eds.), 2011). Even today many of them continue to be in use, leading to the alteration of their archaeo-sedimentary content. Therefore, the initial difficulty was locating caves which had undergone little change and could be excavated extensively.


ndez-Marrero, J.-C., et al., An approach to prehistoric shepherding in La Gomera (Canary Islands) Please cite this article in press as: Herna through the study of domestic spaces, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.11.092


ndez-Marrero et al. / Quaternary International xxx (2015) 1e13 J.-C. Herna

4

A selection process determined the archaeological sites for study. We conducted surface sondages to select 30 settlement complexes using all the available scientific and administrative information on the island. We also made a selection based on scientific, economic and accessibility criteria (because of the difficulties that the insular geography presents), finally narrowing it down to 8 archaeological sites (Fig. 2). The locations of each site were very different: east, west, northwest, northeast, and south. Excavations were carried out in all during 2009 and 2010. The average altitude in the studied sites, 465 m, is quite close to the average of 425 m noted by Hernandez-Marrero and NavarroMederos and (2012) for the settlements at the south side of the island (Fig. 2). Almost all sites are found in the vegetation belt known as “dry termocanarian”, into the thermophilous forests and of great floral abundance and diversity. The sets of caves are located in the upper tracts of the ravines as well as in their middle reaches. Those which have been mentioned are representative of the prevalence of such a function, living spaces, for these places (Navarro-Mederos, 1992). The excavation process followed the standard procedures of any stratigraphic excavation. The initial approach at each site was to make a 1 m2 sample in the selected areas depending on the complexity of each site, having to adapt to the topography in each individual case. Stratigraphic units were dug, documented, and related to Matrix Harris. In the present text, due to the preliminary nature of this study, we have centered our attention on those results which provide more thorough information, stratigraphic complexity, a greater variety, better preservation and a greater quantity of evidence. These are the following: sondage-1 and 2 of ~ ada de La Gurona”, sondage-2 of “Las Cuevas de Herrera “La Can Gonzalez”, “El Sobrado de los Gomeros” soundage-1 and “El Lomito del Medio” soundage-1 (Fig. 2) Table 1.

The sediment was dry sieved on site, using meshes of 2 and 4 mm, and a large amount of sedimentary samples were saved for future research (72.5 k) especially those coming from fire structures. Zoo-archaeological study was carried out with normalized standards (macro and micro fauna). For charred seed remains, a small number of samples of sediment were collected and later processed through manual flotation with hand buckets due to the small volume of the sample. We used meshes of 0.25 mm to collect the plant remains, but only contents of the 2 mm and 1 mm fractions were analyzed. For wood charcoals, the items analyzed belong to the fragments that were hand-picked during dry sieving on site. Selected material (<2 mm) was studied with a reflection optical microscope (100e800 increases) and without chemical manipulation (Machado, 1994; Rando and Perera, 1994; Pais, 1996; Morales, 2010). This text also presents an analysis that uses the number of fragments from materials of different nature such as the quantification system (graphics), and it will be taken into account for each site as an orienting factor mainly in two aspects: the continuity/ discontinuity patterns of use/consumption and the regularity in the occupation of these spaces. 4. Results and discussion
lez (Tazo) 4.1. Las Cuevas de Herrera Gonza Sondage-2 was performed on a terraced area outside a group of caves (Fig. 3). The most remarkable aspect is that in stratigraphic units (SU) 3 and 4, European presence and contact with the indigenous world is detected. Manufactured items like fragments of sugar molds and imported white, orange and glazed pottery (from SU1 to SU4), and a Portuguese ceitil or ceutí coined by Alfonso V of Portugal (1448e1481) were found (SU4). Pottery shows a long

Table 1 Data on sondages with positive results, the stratigraphic units and the domestic units. Site name, sondage and total depth

Stratigraphic units (SU) depth (cm.) Approximate area of the domestic unit Sondage orientation in the domestic unit


lez” sondagee2 SU “Las Cuevas de Herrera Gonza Total depth: 166 cm SU SU SU SU SU SU “El Sobrado de Los Gomeros” sondagee1 SU Total depth: 110.5 cm SU SU SU SU SU “El Lomito de El Medio” sondagee1 SU Total depth: 87.75 cm SU SU SU SU SU SU SU SU ~ ada de La Gurona” sondagee1 “La Can SU Total depth: 18 cm SU SU SU SU SU ~ ada de La Gurona” sondagee2 “La Can SU Total depth: 48.53 cm SU SU SU SU SU

1: 2: 3: 4: 5: 6: 7: 1: 2: 3: 4: 5: 6: 1: 2: 3: 4: 5: 6: 7: 8: 9: 1: 2: 3: 4: 5: 6: 1: 2: 3: 4: 5: 6:

46 21 18 34 9 38 8 5 33.4 53 7.9 11.1 19.1 3.5 6 2.9 6.6 19 25.6 20.8 8.3 20.6 5.3 4 5.4 2.3 8.26 8.4 19.65 8.37 6.77 6.17 7.85 6.83

66 m2

WeNW

54 m2

WeSW

59.8 m2

SE

33.5 m2

EeSE

62.1 m2

EeSE


ndez-Marrero, J.-C., et al., An approach to prehistoric shepherding in La Gomera (Canary Islands) Please cite this article in press as: Herna through the study of domestic spaces, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.11.092


ndez-Marrero et al. / Quaternary International xxx (2015) 1e13 J.-C. Herna

5


lez Fig. 4. Fauna fragments (marine and terrestrial) at Las Cuevas de Herrera Gonza (sondage-2).

Fig. 3. Location of sondage-1 (1), sondage-2 (2) and sondage 3 (3) in “Las Cuevas de
lez” site. Herrera Gonza

survival of indigenous morphologies (SU3-4 to SU6), that coexists with imported forms (SU1 to SU4) and above all, with other types of pottery generically named as “traditional”, representing the evolution of the indigenous forms to others adapted to the new social and economic order (SU1 to SU4) (Navarro-Mederos, 1999). Those materials are also confirmed by the presence of rabbits (Oryctolagus cuniculus) and roosters (Gallus gallus) located on the upper levels (SU3 and 4), species brought by Europeans after the 15th century (Nogales et al., 2006). We know from documentary and archeological inferences that near this archaeological place there was a sugar mill that was in operation until at least 1505 (Moreno-Fuentes, 1988; Díaz-Padilla and Rodríguez-Yanes, 1990; Navarro-Mederos and Hern
andez-Marrero, 2006). In those “moments of contact” (Baucells, 2004) represented by SU4 and SU3, marine harvesting leaves more traces than cattle does, perhaps because European settlers occupied a part of the land previously used by the Gomrans to livestock grazing, and were forced to intensify the exploitation of other food sources. A zooarchaeological study was carried out for this site (1631 fragments and individuals). On SU3-4, there were 195 specimens of domestic animal bone, but only 78 fragments were identifiable. For the sacrificial patterns with goats, adults consist of 56.09% and younger animals 36.58%, whereas the kids account for 7.31%. For pig sacrifices, there is a large disproportion between the adult type (91.66%) and the piglet. SU4 has as many as 83 remains, out of which 57 could not be identified due to their small size. As in the previous stratigraphic unit, the bones of sheep and goats prevail over those of pigs. The sacrificial patterns of the former ones differ from those of SU3-4, as the younger type make up 50%, whereas adults reach 33.33% and the cubs 16.66%. However, the same pattern is present for pigs, where there is a prevalence of developed types (80%). More than 54.54% of sacrificed animals were adults; after these, the second largest group is that of newborns (23.63%). Younger animals represent 21.81%. Regarding sacrificial patterns for pigs, piglets prevail (44.44%), followed by young ones (33.33%) and a smaller number of adults (22.22%). The presence of a pre-molar cat at SU6 should also be mentioned. In a different direction, dog remains appear at “Los Polieros” Cave-F (SU2), excavated at 1983 by Navarro-Mederos, making up 0.4% of the identifiable remains (Navarro-Mederos, 1992). These levels (SU1, 2, 3 and 4) stratigraphically overlie SU5, which is sterile possibly due to a cleaning episode, or even one of abandonment (Figs. 4 and 5). SU6 is fully indigenous. In this SU6 level, shepherding and marine gathering activities are the most


lez. Charcoal fragments of Fig. 5. Anthracological results at Las Cuevas de Herrera Gonza each taxa have been grouped by vegetation belts (Unidentified fragments have been added).

evident. Being the closest site to the coast of all the studied settlements, it is surprising that the presence of shellfish and fish is not equal or superior to that of goat and sheep. Fish remains occur in all the stratigraphic units, but in a low proportion. Fragments of domestic animals are clearly superior in this period, representing 88% of the total set of osteofauna/malacofauna. It seems that the importance of livestock was much higher before the period of contact with the Europeans (Table 2).

Table 2
lez (sondage-2). Results of zoo-archaeological study atLas Cuevas de Herrera Gonza Species percentage

SU1 SU2 SU3-4 SU4 SU5 SU6

% Ovicaprids

% Goat

% Sheep

% Total goat þ sheep

% Pig

25.00 0.00 67.65 37.84 0.00 77.14

25.00 0.00 2.94 0.00 0.00 5.72

50.00 0.00 11.76 5.40 0.00 4.29

100.00 0.00 82.35 76.19 0.00 87.15

0.00 0.00 17.65 23.81 0.00 12.85

A profound change between SU3-SU4 and SU6 can be observed in the analysis (Table 2 and also Figs. 4 and 5). SU6 shows a greater presence of domestic animal bones (395). Most (295) could not be identified, and 104 were less than 1 cm long, which reveals the extreme degree of fragmentation these bones. The species represented in this stratigraphic unit are different from those corresponding to historical times. We can still detect the prevalence of sheep and goat bones over those of pigs, thus showing a larger lack of proportion than in the previous unit. A noteworthy aspect in comparison with the contact period is the presence in SU6 of a


ndez-Marrero, J.-C., et al., An approach to prehistoric shepherding in La Gomera (Canary Islands) Please cite this article in press as: Herna through the study of domestic spaces, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.11.092


ndez-Marrero et al. / Quaternary International xxx (2015) 1e13 J.-C. Herna

6

similar quantity of items corresponding to goats and sheep. In this unit, sacrificial patterns for these animals also differ, thus following the common practices of most of the Canarian aboriginal sites. Livestock must have been even more important than the numbers suggest, because in terms of nutrient intake, the conservation of different materials holds no direct relation to the amount of food provided by the shellfish and herd. So, if we focus exclusively on the number of fragments found at each archaeological layer, we can incorrectly assess the relevance of certain types of activity. For instance, when comparing the presence of malacofauna to the number of domestic animal remains, we should take into account the increased ratio of calories and fat from meat, compared to seafood, which provides proteins but few calories. Another element to bear in mind is that some of the most important products in pastoral societies, such as milk, or those deriving from it, do not normally leave any macroscopic traces in the sediment.
lez” Compared to other sites, “Las Cuevas de Herrera Gonza presents a difference: the lack of fire structures, although there are unstructured charcoals and ashes. It is possible that the low amount of charcoal fragments identified, but especially the high amount of unidentified fragments found this uncovered space is due to natural and cultural processes of erosion (Fig. 5). For those reasons, the floristic richness identified in the site is important (13 different taxa have been recognized). We emphasize in this site the presence of one species usually absent in the anthracological studies of the Canary Islands: Euphorbia canariensis (although the low number of fragments entails certain reservations about this item). This is an endemic succulent shrub, whose dry stalks could be used as tinder to light a fire.

Fig. 7. Sondage-1 at “El Sobrado de los Gomeros” site from the top of the natural wall.


n) 4.2. El Sobrado Los Gomeros (Geria Geri
an (Vallehermoso) was one of the most populated areas in the island's prehistory (Fig. 6). This site seems to have served as a dwelling place only in aboriginal times, which is peculiar on an island where after the European conquest there was a dense continuity in settlements or use of caves for other functions. This is especially so if we take into account that the area of Geri
an was devoted to the cultivation of cereals and to shepherding activities
n residents, personal communicauntil 1960 (Gueleica and Geria tions, 2009). Sondage-1 was the most interesting (Fig. 7). Unlike the other sites, it is distinguished by being in a peripheral area of the domestic nucleus, formed by two principal caves. The stratigraphy is composed of loose and highly mobile sediments, where the levels

Fig. 8. Fauna fragments (marine and terrestrial) and seed (fragments and individuals) at El Sobrado de Los Gomeros (sondage-1).

Fig. 9. Anthracological results at El Sobrado de Los Gomeros. Charcoal fragments of each taxa have been grouped by vegetation belts (Unidentified fragments have been added).

Fig. 6. Location of “El Sobrado de Los Gomeros” site (1), facing the ravine of Argaga and
n (2). near the big slope of Geria

were thicker than studied elsewhere (total stratigraphic sequence reached 129.7 cm). A large number of aboriginal materials are present throughout the surface of the whole site, and very little attributable to later times were detected. Outstanding examples of sheep and goat manure as scabs formed with plenty of organic matter, dried goat/sheep droppings in excellent conditions (SU2: 1 fragment; SU3: 6 complete individuals, 81 fragments; collapse SU1 to SU3: 25 complete individuals; SU4: 3 fragments), bone remains with butchering marks, and substantial archaeological material were identified (Fig. 10). Much of the SU3 was formed by this type of scab. Another outstanding feature is the absence of fire structures within the stratigraphy compared to the other sites, although some


ndez-Marrero, J.-C., et al., An approach to prehistoric shepherding in La Gomera (Canary Islands) Please cite this article in press as: Herna through the study of domestic spaces, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.11.092


ndez-Marrero et al. / Quaternary International xxx (2015) 1e13 J.-C. Herna

Fig. 10. Fragment of ovicaprid adult rib with slaughtering marks at “El Sobrado de Los Gomeros” site.

indicators are present, such as ash spots, sediments affected by heat, and abundance of charcoal. It is striking that 31 different taxa were identified from an analysis of 785 charcoal fragments (Fig. 9). Moving away from the earliest levels, SU6 has a large geological content and SU5 has very poor evidence of human occupation, much of it filtered from upper levels. SU4 has very few materials. SU3 is the most remarkable level, as it contains large quantities of manure linked to the presence of livestock (Fig. 8). The low level of pottery fragmentation and other evidence may be due to rapid formation of sediment that further preserves the material. It is important to highlight the large amount of barley grains (Hordeum vulgare), recovered from this site, remarkably high compared to other sites of La Gomera. This is consistent with the finding of a piece of hand mill on the surface of the deposit. In addition, we have recorded the presence of wild oat (Avena sp.). This is a wild grass that was also identified at the ceremonial site of “El Alto del Garajonay” (Morales et al., 2011). The seeds of this grass are edible and there is ethnographic evidence of its consumption until recent times in La Gomera (Perera, 2005). Wild oats may occupy the cultivated fields and be harvested along with cultivated cereals, thus becoming an agricultural by-product. Also relevant at the same SU3 and at “El Alto del Garajonay” is the presence of Phoenix canariensis (Canarian palm) seeds (Table 3). Ethnohistorical texts attest to the fact that the aboriginal population consumed different products of the palm, such as the dates and a type of syrup called guarapo (Perera, 2005; Morales et al., 2011).

7

Visnea mocanera is another taxon identified at this site. Although abundant archaeological and ethnohistorical information exists about the consumption of fruits by the prehispanic inhabitants (Morales, 2010; Morales and Gil, 2014), we were lacking data concerning its particular use in La Gomera during this period. This tree grows in the thermophilous forest and monteverde and it was probably used as an edible plant. The charcoal sample of this site comes from all vegetation belts and different environments on the island, above all in SU2 (Fig. 9). In “Las Cuevas de Herrera Gonz
alez”, sondage-1 was carried out in an uncovered area where taphonomic processes have reduced the amount of identifiable fragments of charcoal and increase the quantity of non-identifiable fragments. However, it is easy to appreciate how important the monteverde forest was as a general resource far above other vegetation belts (Fig. 9). Trees from the monteverde forest were used as fuel (Erica arborea is, as in other sites, the preferred species). Persea indica, Ilex canariensis, Picconia excelsa, cf. Arbutus canariensis, Morella faya, cf. Laurus novocanariensis and Viburnum rigidum are represented species. Without excluding other functions, some of the identified charcoal played a nutrition role, such as those of thermophilous forest: Phoenix canariensis, cf. Olea cerasiformis ssp. cerasiformis or Juniperus turbinata ssp. canariensis. Following the ethnographic example, other vegetation was chosen for its natural features as pasture for livestock: cf. Adenocarpus foliolosus, Retama rhodorhizoides, cf. Rubia fruticosa and Rumex lunaria (Perera, 2005). The high quantity of angiosperms and legumes point to this. A new taxon has been located for first time on the island. Ricinus communis appeared in sondage-3, SU1 (and also at El Sobrado de Los Gomeros, sondage-1, SU2); although perhaps those data should be viewed with caution because the deposit levels of the charcoals are close to the topsoil. SU3 is the first unit containing a large number of remains; marine harvest traces clearly exceed those derived from livestock. This coincides with a higher proportion of the recorded icthyofauna (Fig. 8), which is surprising because, being in steep coastal area, there is greater difficulty in reaching the sea from this site
lez”, “La Can ~ ada de la compared to “Las Cuevas de Herrera Gonza Gurona” or “El Lomito del Medio”. Maybe those statistical proportions occur precisely for this reason. The level directly above SU2 consists mainly of broken fragments of volcanic ash from the disaggregated rock wall and shows increased remains linked to shellfish and fish. Osilinus atrata fragments increase in number, while evidence of livestock remains constant. Many whole shells of Patella ssp. appear in SU3 and SU2, related to the rapid deposition of

Table 3 Macro-botanical remains collected from studied sites, with numbers of seeds unless otherwise stated. Site

~ ada de La Gurona La Can

Sondage SU Cropplants Hordeum vulgare Hordeum vulgare raquis Triticum aestivum/durum Triticum durum raquis indeterminated cereal Wild plants Avena sp. (Oat) Juniperus turbinata Neochamaelea pulverulenta Phoenix canariensis Pistacia atlantica Retama rhodorhizoides Spartocytisus filipes Visnea mocanera Total

1 6

5

2 1

1


lez El Lomito del Medio La Cueva Honda Las Cuevas de Herrera Gonza 1 3

1 5

20

4 11 16 4

10

9

1 8

2

3

7

El Sobrado de Los Gomeros

5

4

4

3

1 2

collapse

4

16

1

10

4

4

2

1

1

1

69 11 17 4 5

5

3 6 1 4 1 19 60 1 201

1

3 1

5

1 1

2

1

1 19 60 1

2

1

99

35

1

2

3

1

4

23

2

1 17

5


ndez-Marrero, J.-C., et al., An approach to prehistoric shepherding in La Gomera (Canary Islands) Please cite this article in press as: Herna through the study of domestic spaces, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.11.092


ndez-Marrero et al. / Quaternary International xxx (2015) 1e13 J.-C. Herna

8

the soil after a considerable decrease in the number of materials that indicates the end of the occupation (Fig. 8). Therefore, the operating system remains quite stable over the entire sequence of this site. Sondage-1 was done in an uncovered area, so taphonomy should be taken into account in this numerical difference (for example, the extreme fragility of Osilinus atrata respect fragmented bones), and for any conclusion from this information. 4.3. El Lomito del Medio (Casas Caídas) This is one of the studied sites which provided the most interesting results. The sondage was performed inside a cave (Fig. 11). The materials were very fragmented and eroded, which may indicate continuous transit of inhabitants and even perhaps of livestock during some periods. SU4 was a fire structure placed in a type of basin that was removed in its entirety, and has been reserved for future studies. As in the rest of the sondages carried out, the same types of materials can be found in each one of the levels (Fig. 16). The charts show clearly how all these archaeological materials maintain proportional relationships with each other, remaining stable throughout time (Figs. 12, 14 and 15). At this particular site and in all of its strata, fragments of domestic animal bones occur more frequently than shellfish fragments, clearly indicating the importance of livestock to these communities. This fact is even more significant during the first occupancy (SU9 and above all SU7). Over the whole survey, domestic animal bones represent an average of 73% of the total sum of collected materials. The unpublished zooarchaeological study from “Los Polieros” (Cave F) also showed a greater number of domestic animal bones (95.5% at SU1 and 93.83% at SU2), goat remains being much more common than those of sheep Fig. 17. In SU6, all materials decrease significantly, which might be partially explained by this being an abandonment period (Figs. 12e15). SU5 shows an intense use of space. Considering the common variables of domestic animal bones and shells of molluscs, the former represents 70% of the total. The patterns found in SU7 are repeated in this SU5 level (Fig. 12). Other type of remains, including pottery fragments and lithic pieces, follow the same guidelines as the other materials under scrutiny, and we find a similar pattern for the charcoal fragments (Figs. 14 and 15). Another element contained in the register throughout the period of human occupation is fish remains. Collection activities comprise and involve fishing practices as well, which might seem logical. Ethnographic interviews conducted by the Archaeological

Fig. 12. Fauna fragments (marine and terrestrial) at El Lomito del Medio.

Fig. 13. Anthracological results at El Lomito del Medio. Charcoal fragments of each taxa have been grouped by vegetation belts. Because its importance the taxa Pinus canariensisis has been considered individually.

Fig. 14. Pottery fragments and lithic pieces at El Lomito del Medio.

Fig. 15. Number of charcoal fragments at El Lomito del Medio.

Fig. 11. Situation of the cave settlement of “El Lomito del Medio” (1).


ndez et al., 2010) note that Museum of La Gomera (Mesa-Herna shell collecting was traditionally combined with fishing in intertidal stages. Although fish remains generally have a lower frequency occurrence than mollusc shells, in this site there is a numerical correlation: the larger the amount of malacology, the greater the amount of icthyofauna (Fig. 12).


ndez-Marrero, J.-C., et al., An approach to prehistoric shepherding in La Gomera (Canary Islands) Please cite this article in press as: Herna through the study of domestic spaces, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.11.092


ndez-Marrero et al. / Quaternary International xxx (2015) 1e13 J.-C. Herna

Fig. 16. Stratigraphic units at “El Lomito del Medio” (113.3 cm) North profile.

~ ada de La Gurona”(2) Fig. 17. Location of the sites “El Cabezo de Castilla”(1), “La Can and “La Cueva Honda”(3), all of them in the valley of Hermigua.

Identification of wheat grains (Triticum aestivum/durum) in SU10 is also interesting, as it is the first record of this crop for the prehispanic period of La Gomera. In addition, we recovered seeds of barley and juniper (Juniperus turbinata ssp. canariensis). Charred seeds of juniper have been identified in archaeological samples from prehispanic sites in Tenerife and La Palma, and most possibly they were used as food (Morales and Gil, 2014). The berries of this tree are edible despite their strong resin aroma. There is ethnographic evidence of the consumption of these fruits in the Mediterranean region, where berries are processed to produce a very energy-rich drink. It is possible that the ancient Gomerans also collected and consumed the juniper berries, although further research on this topic is necessary in order to prove this hypothesis.

9

Charred seeds and other food residues such as bones or shells increase in the upper part of the stratigraphy. Although they are present in the earlier levels, they disappear from SU4. This last unit indicates the presence of a fireplace. From this event onwards, the place no longer had the intensity of use of previous periods, nor was it ever recovered, as attested in the stratigraphic units formed later: SU3, 2 and 1, characterized by shortage of materials (Fig. 12). Moreover, there is a certain amount of dried and mostly fragmented sheep/goat droppings and organic material derived from these deposits (SU3: 98; SU4: 16; SU5: 21; SU6: 372; SU7: 3; SU9: 1). The anthracological analysis was done on 1769 charcoal fragments (35 different taxa were identified). The study shows a clear prevalence of Pinus canariensis, with a total percentage of 43.5% charcoal (especially in SU7 and SU8 from 40 identified taxa). Viburnum rigidum, Erica arborea and Juniperus turbinate ssp. canariensis also appear, although to a lesser extent. The geographical location of the site in the middle section of El Barranco de las Puertitas determines its domain as that of the thermophilous forest (Fig. 11). However, the importance of this forest appears to be less in comparison with pine and the monteverde. Pines may have been relatively abundant in the area, although perhaps not historically so, and would have been stratified between the climax community (Mayteno-Juniperium) and Morella-Erica forest (Fig. 13). The existence of cf. Dracaena draco in SU7 is another highlight of this site's stratigraphy. The dragon, which is a species endemic to the Canary Islands, typical of the thermophilous forest, is currently not known in natural populations in La Gomera. The presence of this taxon among the charcoal allows us to confirm its existence in the insular prehistoric landscape. Ethnographically, its uses are manifold: fibers for ropes, baskets or tissues. Fuel also derived from its dry bark, and its sap served for medical purposes. Fragments or splinters, slightly burned at one end, have been identified, suggesting they could be remnants of torches, though we cannot confirm it yet. The firewood which was collected and brought to the site came especially from the top of the hill, monteverde domain initially (SU10). During SU9, most fuel came from pine and another secondary source from the monteverde and thermophilous forest. The lower parts of the slope also were used where drier vegetation is found (dry thermocanarian). From SU1 to SU8 and 5, the curves of thermophilous forest and monteverde increase, following a parallel track, while pine decreases, reaching 37.8%. At the end of the occupation in SU2, the elements of monteverde and especially the thermophilous forest (50%) continue to rise. As for the pine forest, the percentage drops to 11% (Fig. 13). In “El Lomito del Medio” changes are observed in fuelling strategies, these depending on the periods of greater or lesser occupation. The exploitation of Pinus canariensis seems to coincide with periods of high occupancy, while that of monteverde and thermophilous forest increases in the opposite situations. The chart shows that charcoal figures and evolution is similar to those of other archaeological materials, only the selected type of vegetation varying (Fig. 13). There are also some lithic artifacts worked on obsidian, in older stratigraphic levels (SU7, 8 and 9), which is new for the archaeology of the island, bearing in mind that natural obsidian has not yet been found in La Gomera (Navarro-Mederos and Clavijo, 2011). Other materials common to domestic contexts are abundant, including bone residues and mollusk industry. Outside the cave there are remains of walls and terracing, and part of a wall that blocked the access. At the base of the rock wall that runs along the bottom of the cave, there is an alignment of holes for posts, which helps to establish a framework, the purpose of which would be the partition of the space, as seen in other cave dwellings. Two of these holes,


ndez-Marrero, J.-C., et al., An approach to prehistoric shepherding in La Gomera (Canary Islands) Please cite this article in press as: Herna through the study of domestic spaces, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.11.092

10


ndez-Marrero et al. / Quaternary International xxx (2015) 1e13 J.-C. Herna

with an average depth of 14.9 cm, were also identified during excavation (SU9 sub.2). Goat and sheep droppings in strata show some peculiarities: the stools increase at moments of greatest abandonment (SU6 and 3), which also coincides with greater exploitation of the thermophilous and monteverde forest. There is an inverse relationship between human presence and livestock remains. This correspondence is also supported by the evolution of lithic industry and pottery (Fig. 14). The presented data suggests a temporary change in the function, at least, of the same space inside the cave. During periods of lesser occupation, the same space was used as a corral with an undetermined seasonality. The stratigraphic relationships that we have observed show changes in the way these populations dealt with land management for the supply of firewood, and perhaps of pastures. We do not know what the reason for these changes could have been, but a close connection seems to exist between the exploitation of pine areas, monteverde or thermophilous forest, and the function assigned to caves. The change might also be due to the depletion of a resource such as Pinus canariensis. Perhaps the very scarcity of this tree, caused by overexploitation, might account for its being exclusively kept for ritual celebrations, its wood reserved for burning on sacrificial altars (Navarro Mederos et al., 2002). However, these data confirm that pine trees formed a greater part of the landscape on this side of the island than they do today, although not forming a vegetation layer itself, as La Gomera lacks the proper natural circumstances to fulfil this condition. ~ ada de La Gurona (El Tabaibal) 4.4. La Can This site, like “La Cueva Honda”, is in the middle-upper tract of the steep and wide space that surrounds the deep valley of Hermigua. It is in a disadvantageous position due to the distance to be travelled to reach it from the ground or from the top of the slope. This location involves an investment of energy that does not seem to match the choice for a permanent settlement. However, we can interpret that it had some stability as settlement. A little higher up, in the upper tract of the slope, with convenient and quick access to the mountainous area rich in pastures that crown the valley, there are large caves, some of which still bear names that allude to their archaeological nature, as “La Cueva de Los Guanches” or “La Cueva del Sobrado”. The good conservation of archaeological material in the area of the cave is surprising. Perhaps there was only sporadic transit for wild animals and goats in recent centuries, very different from what we observed in “El Lomito del Medio”, a place of controlled grazing until the 1960s (Perera, 2005). The cave is naturally divided in two spaces, where sondages were excavated. Sondage-1 presents a thin (18 cm) but interesting stratigraphic sequence, perhaps because there were cleaning events in this domestic space. Although the archaeological evidence is less abundant than at other sites under scrutiny, the entire sequence is archaeologically fertile. It shows an almost constant human presence from the initial occupation. Most remarkable is the presence of fire in the space where the excavation was carried out over a long period (SU3 to SU6) (all the sediment affected by fire events in the sondage-1 was reserved for future work, 16.5 kg of sedimentary sample). The excavated area must have represented the center, or a center, of the living space in the cave. Much archaeological evidence is present over the entire surface of the cave. In sondage-1, the materials become scarcer and smaller with depth, the droppings of goat and sheep gradually decreasing along with the remains of animal hair that appear in the middle of SU2. On the surface of this first layer, traditional pottery manufacture was found, which places this stratigraphic unit between the

modern and contemporary ages. Three bones of R. Rattus (black rat) are present in only this site in the upper layers SU1 and SU2. Black rats were introduced into most of the islands of the archipelago after European contact in the 15th century, with the exception of Lanzarote, where they were introduced much earlier, probably by the Romans (Nogales et al., 2006; Rando et al., 2012). The location of these bones in the latest levels is thus consistent with a late introduction. The activity in the cave declined in both SU3 and SU2, which could represent a level of neglect. The latest level, SU1, shows an increase of materials that perhaps denote a sporadic activity, with more whole materials as we reach the final abandonment as a dwelling place. Regarding the last level that appears in both sondages, on both sides it represents the final occupation. SU4 is a sedimentary matrix altered by fire, with large concentrations of charcoal and an abundant of ancient archaeological materials. It represents the last level influenced by fire, and the most intense occupation of the site. At this level, domestic animals are 59% of the complete list of macroscopic nourishing remains. The edible part of this type of animal is superior to the remaining bone residues, while shell has more mass and weight than the flesh it houses. Icthyofauna is also represented, but no seeds, these appearing only in the final levels of occupation. SU5 is a welldefined fire structure independent of the previous one. In this layer, we found charred seeds of Phoenix canariensis and Hordeum vulgare (Table 3). SU6 is the aureole of sediment affected by varying degrees of heat intensity of a fire structure. Both fire events rest on a smooth bowl formed in the natural rock, which appears red (in the web version) from the heat (Fig. 18). This site is close to the inhabitants' firewood source, the monteverde forest. The Erica arborea is the most common species (55.3% of the total charcoal), and to a lesser extent there is Juniperus turbinata ssp. canariensis (8.22%) and Viburnum rigidum (7.11%). We also found a small amount of charcoal from other trees (Illex canariensis, Morella-faya and Visnea mocanera), shrubs and thickets as Cistus sp., as well as other leguminous species. The presence of Pteridium aquilinum at “La Cueva Honda”, SU2, must be pointed out, and its consumption is mentioned throughout history until recent times (Perera, 2005). Sondage-2, conducted in the widest part of the cave, shows that the occupation started on the natural rock of the cave. Although the archaeological material undergoes a large drop from the initial events (Fig. 18) there are no signs of long periods of neglect. Whereas archaeological evidence related to fire is often scarce, in these levels combustion events were detected. All signs indicate a differential use of this domestic space, and perhaps to a partitioning (simultaneous?) of its functions, given the differences in the contents of the sondage-1 and 2. Fig. 11 shows a striking relationship, which must be considered with caution because we do not yet have confirmed dating, indicating diametrical opposition with the use of

~ ada de La Gurona (sondageFig. 18. Fauna fragments (marine and terrestrial) at La Can 1).


ndez-Marrero, J.-C., et al., An approach to prehistoric shepherding in La Gomera (Canary Islands) Please cite this article in press as: Herna through the study of domestic spaces, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.11.092


ndez-Marrero et al. / Quaternary International xxx (2015) 1e13 J.-C. Herna

charcoal from monteverde/thermophilous forest between the two sondages. From the first instance of occupation of this cave, domestic and marine fauna were consumed. The former represents 59% in SU6 (as in SU4 in sondage-1), but the difference decreases in the upper stratigraphic units. SU1 is third in quantity of materials and stratigraphic thickness. Similar amounts to those of SU6 are noted there. It seems that the malacofauna acquired more importance in later periods of occupation (Fig. 19). The icthyofauna is well represented at all times, which confirms it as a permanent dietary component. Industry (in this case pottery and the lithic tools) follows a parallel route to dietary elements. The numbers show a very specific proportional relationship of each kind of material, whatever its nature, and showing similar changes from top (SU6, 4 and 1) to bottom (SU5, SU3 and SU2), compared as well to the other sites studied. ~ ada de La Gurona and La Cueva Honda provide grounds for La Can debate. Although closer to the village of Hermigua and therefore to its cultivated lowlands, these caves are less exposed and better preserved than other possible large settlements located at the top of the slope, which over time have become more accessible from the paths that cross the ridge. The data (still in need of accurate dating) suggests that both sites were outlying habitats, compared to a main dwelling on the upper tract and/or base of the slope, now badly damaged by its increased accessibility. The settlement consists of the cave and its surroundings: a side entrance leads to a terrace of great interest as a space for domestic use. Directly above, there is another broad but shallow rock shelter and small storage caves and terraces that may have been used to carry out a variety of activities. A few metres away there are smaller, more exposed caves which would have provided good conditions for use as an outside kitchen or storage area, such as the one located in Cave-F at “Los Polieros” (Navarro-Mederos, 1984, 1988). 5. Conclusions The results of the current research as well as of other projects previously carried out confirm that the ancient people from La Gomera constituted a pastoralist society to such an extent that livestock was not an exclusively economic issue, but also one that influenced each and every aspect of their daily concerns and activities. This idea does not contradict the fact that these societies diversified livelihoods as an economic strategy, conveyed by the exploitation of animal and vegetable resources found all over the landscape. This production system, where grazing had a clear dominance, was perfectly embedded in the natural cycles of the island, above all, in those involving water: the dry season, rainfall and flowering. In this context the capacity for the reproduction of this insular society rested somewhat on the group's ability to

~ada de La Gurona Fig. 19. Ovicaprid bones and malacofauna fragments at La Can (sondage-2).

11

predict the water wealth of the year, which in turn meant pastures
ndez-Marrero, and therefore food (Navarro-Mederos and Herna 2008; Barrios-García et al., 2014). From an ecological point of view, archaeobotany indicates the extraordinary floral variety that the native population took advantage of, reflecting the relationship between the Gomeran people and the ecological diversity of the island's environment, from the coasts to the mountains and throughout the vegetation belts. Data shows that the ancient islanders used the species Erica
rea, Morella faya, Ilex canariensis, Adenocarpus foliolosus, arbo Retama monosperma, Juniperus turbinatassp.can. and Pinus canariensis as fuel. It is probable that some of these had other purposes related to daily life, including woodworking, fencing pasture for animals, or construction of screens or roof for outside structures, and for stables or other sort of huts. The pine tree is present in almost every anthracological spectrum and it is also very common in the south-western sites of the island. Therefore, its distribution in the past was more profuse than it is today, where it has become atypical. The relative abundance of this taxon in habitat contexts allows us to state that pine did not just have a symbolic value in religious rites (Machado, 1999). Three species appear registered for the first time in the archaeology of La Gomera: Ricinus communis, Pteridium aquilinum, and perhaps most pointedly, the discovery of cf. Dracaena draco. The dragon has been subject to intense human exploitation. Nowadays, no natural occurrences of cf. Dracaena draco appear to be present (future confirmation concerning this point, and the previous, is required). Another relevant point is the presence of today rare species of monteverde forest, Sambucus sp., Heberdenia excelsa and Sideroxylon canariense, which suggests a higher presence of these species during the Prehispanic period or/and perhaps a particular use of them, such as medicinal, as ethnography notes for Sambucus sp. (Perera, 2005). Barley (Hordeum vulgare) is the most common seed recorded in the samples, identified in all the sites analyzed. Barley has also been recorded in “El Alto Garajonay” (Morales et al., 2011). The abundance of barley suggests that this cereal was the main cultivated plant and that it could have had an important role in the diet of the ancient Gomerans. Wheat (Triticumaestivum/durum) has been also identified in the samples. This is a very important find, as it is the first record of this crop during the prehispanic period of La Gomera. Nevertheless, we must be cautious about the chronology of these seeds until direct AMS dates are performed on the grains. We have also recorded the presence of wild plants in the sites, although seed numbers are usually low. In total, we have identified 8 different taxa that could have been used as sources of food, fuel, or fodder. All the identified species are more likely to have been gathered on the bioclimatic belt of the thermophilous forest. This is the area where the sites are located, so wild plants were collected in the vicinities of the sites. Wood charcoal analysis points to the same
ndez-Marrero and Navarro-Mederos, 2012). hypothesis (Herna Non-domestic animals were also part of the ecologic frame where shepherding was carried out. Three types of birds were identified for the first time in the archaeology of the island: Columba sp., Canolectis diomedea and Falco sp. It is possible that the first two were consumed by the aborigines, but their exploitation is probable since it has been confirmed in other islands of the archipelago (Rando and Perera, 1994; Rando et al., 1996, 1997). The stratigraphic location of the bones of R. Rattus from La Gomera is consistent with a late introduction shown for other islands (Rando et al., 2012). Nevertheless, the house mouse was introduced in aboriginal times in many islands of the archipelago (Alcover et al., 2009), and its presence in deep stratigraphic units of La Gomera correlates to an early introduction to the island during the aboriginal period. The most abundant remains of microfauna are


ndez-Marrero, J.-C., et al., An approach to prehistoric shepherding in La Gomera (Canary Islands) Please cite this article in press as: Herna through the study of domestic spaces, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.11.092

12


ndez-Marrero et al. / Quaternary International xxx (2015) 1e13 J.-C. Herna

those of the giant lizard (Gallotia bravoana) present in all the sites and at most of the stratigraphic units. Currently, this is a critically endangered species (Nogales et al., 2001), but in former times it had a wider distribution in low areas on the island (Mateo, 2007). Taking into account the large size these lizards reach, it is possible that a part of these remains respond to human exploitation (Nogales et al., 2001; Mateo, 2007). The bones of the small common lizard (G. caesaris and Tarentola gomerensis) seem to be the result of natural deposition. Alternation between periods of human occupation and those of abandonment has been noticed in all the studied domestic areas, along with unknown regularity. Future dating will provide an understanding of the timing and the key reason of this temporary abandonment. This data may or may not allow us to propose a review of the concept “transhumance” as applied to the island of La Gomera and its replacement by “transterminancia” aimed at lower livestock movements. In any case, there is a sustained election of the same places as settlements throughout time. The spatial relationship between dwelling and burial areas also point to this. All the studied cave settlements present minor modifications to their roof, walls or soil, such as small holes, pockets or postholes, and remains of exterior and interior walls. These features divide the domestic space in either functional and/or symbolic dimensions. Most of the studied settlements seem to have a central core and then a periphery where different activities took place. Following the behaviour of the materials in the previously mentioned relationships, it seems that there were no significant changes in the food resources, at least not in those which are the object of this study. In terms of consumption, data indicates that animal bones crushed for the sake of fat and marrow were a very important nourishing resource, common both in prehistoric and historical times. In addition, in light of the low proportion of burned bones found at archaeological sites, ancient Gomerans tended to eat their meat undercooked. Dismemberment and flesh removal marks also appear in most stratigraphic units, especially in the ribs. The butchery was carried out in the same place where the meat was processed and consumed (remains of all animal anatomy were ~ ada de La Gurona”, “Las Cuevas de Herrera founded in “La Can Gonzalez”, “El Sobrado de los Gomeros” and “El Lomito del Medio”), probably close to any watercourse when possible. Regarding their guidelines for the slaughtering of ovicaprids, the remains of adults always accounts for more than half of the ratio. The figures for young animals are always below this percentage and the kids/ lambs are the lowest. In pigs, this average is much less. These data clearly suggest a pastoral economy based on the production of milk as the main product, whereby herds were composed mostly of adult females in dairy age. Young animals constitute a meat reserve for those times of increased scarcity of other food, perhaps the summer. In the case of kid, lamb, and piglet slaughtering, they respond to the growth control of the herd. Not all bcould be brought to full maturity, as it might have meant the annual depletion of pasturelands, which was to be avoided. Some bones, above all of sheep and goat, are practically unbroken and in a good state of conservation, suggesting a new line of research into species and the comparison between indigenous races and European ones. Up to now, the general belief has been that the shepherding system during the prehispanic period of the island was that of controlled open grazing. The vertical and horizontal (through vegetation belts) coordinated displacement regime was oriented to prevent decrease of pastures (Navarro-Mederos, 1992). This idea comes from the ethnographic observation and to a lesser extent from archaeology. The present text using archaeological data helps us to approach a more complex reality. Excavations indicate that at least part of the livestock was also kept in pens at the place of settlement, and animals could share the same dwelling caves. This

hypothesis is strengthened by the sustained presence of charcoal and seeds of leguminous shrubs which were considered fodder or bedding for the livestock, like Spartocytisus and Retama, found in most of the archaeological sites, especially in levels with plenty of goat and sheep manure at “El Sobrado de Los Gomeros” (sondage1). The existence of general herd management involving shorter or longer daily/seasonal movements is possible and we should be open to the idea that the ancient Gomerans used several methods of handling cattle; for example keeping livestock in settlements, or collectively reserving part of the territory to maintain a semi-wild cattle group that every so often was rounded up. The droppings ~ ada de la found at this site, “El Lomito del Medio” and “La Can Gurona”, should be investigated in the future to produce more accurate data on species, pastures, grazing areas, movement of cattle, and perhaps the timing of livestock management. Along with the specific importance of livestock in the economy of La Gomera, marine and vegetable gathering also played an important role, as shown by previous paleodiet studies (Arnay et al., 2009). These activities, carried out for different purposes and all through the island's landscape, serve as an example of the significant and stable behaviour taking place throughout the duration of the settlements at the sites studied. Therefore, those varied economic activities were practised simultaneously over time and throughout the seasons. Although it does not mean that the collection was exerted with the same intensity, there seems to have been no period in which these resources were not used. This fact also indicates the tendency of a long stability in the means of exploitation of the territory and its resources, which was maintained as a lifestyle up to the 15th century. Although we do not yet have data on some important aspects of livestock such as the management of the herd within the territory (daily/seasonal movements, borders, etc.), the property system (access to and circulation of products, etc.) or the production of “secondary products” (manufacturing processes, uses, etc) (Alberto, 2004), the archaeological excavation of the presented sites has proven to be essential in helping us to approach some aspects of ancient Gomeran shepherding and also in opening up new possibilities for future research. Acknowledgments We appreciate the comments and suggestions of Arnoldo Santos
pez on botanic and ecologic asGuerra and Angel B. Fern
andez Lo pects of the text, and the English review of the text by Amy Langston. References Abreu y Galindo, J. de, 1977. In: Goya (Ed.), (1632). Historia de La Conquista de las siete islas de Canaria. Santa Cruz de Tenerife.
cticas funerarias guanches. Anuarios de Alberto, V., 1999. Los animales en las pra
nticos 45, pp. 19e60. Estudios Atla
poca prehispa
nica. El Pajar. Alberto, V., 2004. De carne y hueso. La ganadería en e Cuaderno de Etnografía Canaria 18, 4e8. Alcover, J.A., Rando, J.C., García-Talavera, F., Hutterer, R., Michaux, J., Trias, M., Navarro, J.F., 2009. A reappraisal of the stratigraphy of Cueva del Llano (Fuerteventura) and the chronology of the introduction of the house mouse (Musmusculus) into the Canary Islands. Palaeogeography Palaeoclimatology Palaeoecology 277 (3-4), 184e190. ~ eiras, P., Herna
n, F., 2002. In: Rueda (Ed.), Los volAnguita, F., M
arquez, A., Castin
gica e itinerarios. Madrid. canes de Canarias. Guía Geolo
mez, A., Navarro-Mederos, J.F., Herna
ndez, J.C., Fregel, R., Yanes, Y., Arnay, M., Ga
lez, E., 2009. Dietary patterns during the early Galindo, L., Romanek, C.S., Gonza prehispanic settlement in La Gomera (Canary Islands). Journal of Archaeological Science 36, 1972e1981. Arozena Rozena, M.E., 1991. Los paisajes naturales de La Gomera. Cabildo Insular de La Gomera, Santa Cruz de Tenerife.
n de las Islas Canarias en la Corona de Castilla Aznar, E., 1983. La integracio (1478e1526). Cabildo Insular de Gran Canaria, Las Palmas de Gran Canaria.


ndez-Marrero, J.-C., et al., An approach to prehistoric shepherding in La Gomera (Canary Islands) Please cite this article in press as: Herna through the study of domestic spaces, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.11.092


ndez-Marrero et al. / Quaternary International xxx (2015) 1e13 J.-C. Herna
nsito del siglo XV al XVI. Aspectos Econo
micos. Aznar, E., 1985. La Gomera en el tra Coloquio de Historia Canario-Americana 5 (2), 403e420. Barrios-Garcia, J., Hern
andez-Marrero, J.C., Trujillo-Mora, J.M., 2014. Investigaciones
micas en La Gomera. La Cueva de San Blas y el origen del culto a arqueoastrono La Candelaria en Chipude. In: Coloquio de Historia Canario-Americana 21 (unpublished ms).
n Caja Rural de Canarias (Ed.), Cro
nicas, historias, Baucells, S., 2004. In: Fundacio
n relaciones y otros relatos: Las fuentes narrativas del proceso de interaccio cultural entre aborígenes canarios y europeos (siglos XIV a XVII). Las Palmas de Gran Canaria.
n y etnicidad. El proceso de interaccio
n entre Baucells, S., 2014. Aculturacio guanches y europeos (Sigos XIV-XVI). In: Instituto de Estudios Canarios, La Laguna.
n, edad Carracedo, J.C., 2011. In: Rueda (Ed.), Geología de Canarias I. Origen, evolucio y vulcanismo. Madrid. ~ orío en las Canarias Occidentales. La Gomera y Díaz, G., Rodríguez, J.M., 1990. El Sen El Hierro hasta 1700. Cabildo Insular de El Hierro y Cabildo Insular de La Gomera, Santa Cruz de Tenerife.
gicas en las Canarias Occidentales. Diego, L., 1953. Nuevas excavaciones arqueolo Yacimientos de Tenerife y La Gomera 1947-1951. In: Comisaría Provincial de
gicas. Informes y Memorias 14, Madrid. Excavaciones Arqueolo
ndez-Palacios, J.M., de Nicol
Ferna as, J.P., 1995. Altitudinal pattern of vegetation variation on Tenerife. Journal of Vegetation Science 6, 183e190.
gica Gil-Rodríguez, M.C., Afonso-Carrillo, J., Horoun Tabraue, R.J., 1992. Flora ficolo
n del Archipie
lago de las Islas Canarias. In: Edirca, S.L. (Ed.), Flora y vegetacio Canario. Tratado florístico, 1ª parte. Las Palmas de Gran Canaria.
lez-Marrero, M.C., Rodríguez-Rodríguez, A.C., 1998. La mirada del otro: de Gonza
mo los europeos percibieron la vestimenta de los antiguos canarios. Coloquio co de Historia Canario-Americana 12 (1), 675e696.
ndez-Go
mez, C.M., Alberto, V., 2006. Buscando a la Comunidad Local. In: En: Herna
gicos versus escenarios sociales en la Canarias preeuropeas, Paisajes arqueolo vol. 21. El Pajar: Cuaderno de Etnografía Canaria, pp. 22e31.
ndez-Marrero, J.C., 2014. Memoria te
cnica del proyecto: “Prospecciones Herna
gicas en las zonas afectadas por el incendio de 2012 en el Parque arqueolo Nacional Garajonay (La Gomera)” (Unpublished ms). Unidad de Patrimonio
rico, Cabildo Insular de La Gomera. Histo
ndez-Marrero, J.C., Navarro-Mederos, J.F., Trujillo, J.M., Cancel, S.J., 2011. La Herna
n arqueolo
gica en La Gomera: Ciencia y Comunidad. Actas del investigacio
n del Patrimonio Arqueolo
gico Arqueomac (La Restinga, Seminario sobre Gestio
n General de Cooperacio
n y Patrimonio Cultural del El Hierro), 67e88. Direccio Gobierno de Canarias, Santa Cruz de Tenerife.
ndez-Marrero, J.C., Navarro-Mederos, J.F., 2012. Arqueología del Territorio en Herna La Gomera (Islas Canarias). Tabona 19, 25e58. Lobo, M., 1984. Los gomeros en el poblamiento de Tenerife. Museo Canario 45, 55e79.
gicos en el Archipie
lago Canario, Machado, M.C., 1994. Primeros Estudios Antracolo NW. de Tenerife: Las comarcas de Icode y Daute (Unpublished doctoral thesis). Universidad de La Laguna. Machado, M.C., 1999. El hombre y las transformaciones del medio vegetal en el
lago Canario durante el período pre-europeo: 500 a. C/1500 d. C. II Archipie
s del Neolític a la Península lberica. Sagvuntum-PlaV 2, 53e58. Congre
n de los gomeros en las partes de Daute y sur de Martínez, P., 1993. Integracio Tenerife durante la primera mitad del siglo XVI. In Strenae Emmanuellae Marrero Oblatae, Pars Altera. Universidad de La Laguna, La Laguna. Marín de Cubas, T.A., 1993. In: Globo (Ed.), Historia de las Siete Islas de Canaria (La Laguna). Mateo, J.A., 2007. El lagarto gigante de La Gomera. Cabildo Insular de La Gomera, Santa Cruz de Tenerife.
ndez-Marrero, J.C., Navarro-Mederos, J.F., Gonza
lez, G., 2010. Mesa, E., Herna Archaeological shell middens and shellfish gathering on La Gomera island. Munibe (Suplemento) 20, 34e41.
nicas de La Conquista. Cabildo Insular de Gran Morales, F., 1994. Canarias: Cro Canaria, Las Palmas de Gran Canaria. Morales, J., 2010. El uso de las plantas en la prehistoria de Gran Canaria. Agricultura,
n y ecología. Cabildo Insular de Gran Canaria, Gran Canaria. alimentacio Morales, J., Navarro-Mederos, J.F., Rodríguez-Rodríguez, A., 2011. Plant offerings to the Gods: seed remains from a pre-hispanic sacrificial altar in La Gomera (Canary Islands, Spain). In: Fahmy, Ahmed G., Kahlheber, Stefanie, Catherine D'Andreaed, A. (Eds.), Current approaches to African archaeobotany. Reports in African Archaeology 3, pp. 67e78. Morales, J., Gil, J., 2014. Gathering in a new environment: the use of wild food plants during the first colonisation of the Canary Islands, Spain (2nde3rd century bce to 15th century ce). In: Plants and People. Choices an Diversity through Time. Oxford and Philadelphia (Oxbow), pp. 216e264. Moreno, F., 1988. Las Datas de Tenerife. In: Libro V de datas originales. Instituto de Estudios Canarios, La Laguna.

13


n a la Carta Arqueolo
gica de l isla de La Navarro-Mederos, J.F., 1975. Contribucio Gomera (Islas Canarias) (Unpublished bachelor memory). Universidad de La Laguna.
polis de Los Polieros, Alajero
, isla de La Navarro-Mederos, J.F., 1984. Poblado y necro Gomera. Tabona 5, 481e483.
gico de Los Polieros (Alajero
e la Navarro-Mederos, J.F., 1988. El conjunto arqueolo
gicas en Canarias 1, 133e140. Gomera). Investigaciones Arqueolo Navarro-Mederos, J.F., 1992. Los gomeros. Una prehistoria insular. Estudios Pre
n General de Patrimonio Histo
rico, Gobierno de Canarias, hisp
anicos 1. Direccio Santa Cruz de Tenerife.
cnica del proyecto “Garajonay: Arquelogía Navarro-Mederos, J.F., 1994. Memoria te ~ as” (Unpublished ms). Universidad de La Laguna. de Las Montan
cnica del proyecto “Inventario del PatriNavarro-Mederos, J.F., 1996. Memoria te
gico de los Municipios de Alajero
, Vallehermoso y Valle Gran monio Arqueolo
n General de Patrimonio, Gobierno Rey (La Gomera).” (Unpublished ms) Direccio de Canarias.
n de tradiciones Navarro-Mederos, J.F., 1999. “El viaje de las loceras”: la transmisio

ricas e histo
ricas de Africa
n en cer
amicas prehisto a Canarias y su reproduccio
nticos, pp. 61e118. las islas, vol. 45. Anuarios de Estudios Atla Navarro-Mederos, J.F., 2001. La arqueología y el poblamiento humano de La Gomera (Islas Canarias). BSCP Can Ped 25 (2), 119e129.
cnica del proyecto “Inventario Navarro-Mederos, J.F., et al., 2005. Memoria te
gico y su aplicacio
n a los programas interpretativos del Parque Arqueolo Nacional de Garajonay” (Unpublished ms). Parque Nacional deGarajonay (Expediente 20P/04).
ndez, C.M., Alberto, V., Borges, E., Barro, A., Hern
Navarro-Mederos, J.F., Herna andezMarrero, J.C., 2001. Aras de sacrificio y grabados rupestres en el Lomo del Piquillo (isla de La Gomera). Estudios Canarios 45, 317e340.
ndez, C.M., Alberto, V., Borges, E., Barro, A., Hern
Navarro-Mederos, J.F., Herna andezMarrero, J.C., 2002. El diezmo de Orahan: los conjuntos de aras de sacrificio en la isla de La Gomera. Tabona 10, 91e127.
ndez-Marrero, J.C., 2006. Evidencias arqueolo
gicas de Navarro-Mederos, J.F., Herna los primeros asentamientos europeos en La Gomera (Islas Canarias). Coloquio de Historia Canario-Americana 16, 388e407.
ndez-Marrero, J.C., 2008. El agua en la Prehistoria. La Navarro-Mederos, J.F., Herna
n de los antiguos gomeros con el agua” InLa Cultura del Agua en La relacio Gomera. Consejería de Obras Públicas y Transportes del Gobierno de Canarias, Santa Cruz de Tenerife, pp. 21e43. ~ os de arqueología canaria: Navarro-Mederos, J.F., Clavijo, M.A. (Eds.), 2011. 44 an textos escogidos de Luis Diego Cuscoy. Instituto de Estudios Canarios, La Laguna. Nogales, M., Rando, J.C., Valido, A., Martín, A., 2001. Rediscovery of a giant lizard, genus Gallotia (Reptilia: Lacertidae) from La Gomera, Canary Islands. Herpetologica 57, 169e179. Nogales, M., Rodríguez Luengo, J.L., Marrero, P., 2006. Ecological effects and distribution of invasive non-native mammals on the Canary Islands. Mammal Review 36, 49e65.
, S., de Nascimento, L., Ferna
ndez-Palacios, J.M., Whittaker, R.J., Willis, K.J., Nogue 2013. The ancient forests of La Gomera, Canary islands, and their sensitivity to environmental change. Journal of Ecology 101, 368e377.
n en la prehistoria de la isla de La Palma: la Pais, F.J., 1996. La economía de produccio ganadería. Canarias. Viceconsejería de Cultura y Deportes. Gobierno de Canarias, Santa Cruz de Tenerife. Perera, J.D., 2005. La Toponimia de La Gomera. In: Un estudio sobre los nombres de lugar, las voces indígenas y los nombres de plantas, animales y hongos de La
n Insular de Desarrollo Rural de La Gomera (CD-ROM), Las Gomera. Asociacio Palmas de Gran Canaria. Rando, J.C., Rodríguez, A.C., Pais, F.J., Navarro-Mederos, J.F., Martín, E., 1996. Los
gico de El Tendal (La Palma, Islas Canarestos de aves del yacimiento arqueolo rias). El Museo Canario 51, 87e102. Rando, J.C., Perera, M.A., 1994. Primeros datos de ornitofagia entre los aborígenes de Fuerteventura (Islas Canarias). Archaeofauna 4, 13e19.
pez, M., Jimenez, M.C., 1997. Bird remains from the archaeological site Rando, J.C., Lo of Guinea (El Hierro, Canary Islands). International Journal of Osteoarchaeology 7, 298e302. Rando, J.C., Alcover, J.A., Michaux, J., Hutterer, R., Navarro-Mederos, J.F., 2012. Late Holocene asynchronous extinction of endemic mammals on the Eastern Canary Islands. Holocene 22, 801e808. ~ a. Santos, A., 2008. Atlas y Libro Rojo de la Flora Vascular Amenazada de Espan Adenda 2008. Ministerio de Medio Ambiente y Medio Rural y Marino, Madrid. Torres, J.A., 1992-1993. Antropología del terreno: el ejemplo de la Cordillera (Valle Gran Rey, La Gomera). Tabona 8 (1), 297e316.
nica dos feitosnot
Zurara, G.E. de, 1978. Cro avei que se passaram na conquista de
por mandado do Infante D. Henrique. Academia Portugesa da Histo
ria, Guine Lisboa.


ndez-Marrero, J.-C., et al., An approach to prehistoric shepherding in La Gomera (Canary Islands) Please cite this article in press as: Herna through the study of domestic spaces, Quaternary International (2015), http://dx.doi.org/10.1016/j.quaint.2015.11.092