Printer Friendly

Radiocarbon dating and talayots: the example of Son Ferrandell Oleza.

In a recent paper in ANTIQUITY, Webster (1991) discussed the construction and labour requirements of the dry-stone built towers known as nuraghi on the island of Sardinia. The equivalents of these monuments in the Balearic islands during the Bronze and Iron Ages were talayots (e.g. Mascaro 1968; Pericot 1972; Rossello 1973). Recent research on these monuments has focussed on settlement patterns and hierarchies (e.g. Alvaro 1983; Lewthwaite 1985; da Mota Santos 1988), as well as social and economic reconstruction (e.g. Gasull et al. 1984), but radiocarbon chronologies of monument construction, use and abandonment are still rare. In this note we publish a series of radiocarbon dates from the talayotic settlement of Son Ferrandell Oleza, in the north of Mallorca, and discuss their implications for the site and the island.

The Son Ferrandell Oleza project

Son Ferrandell Oleza is located c. 2 km to the west of Valldemossa, at the foot of a terraced, limestone ridge and overlooking a cultivated plain. A Pretalayotic Beaker settlement (beginning c. 4000 b.p.) was succeeded by an undefended, linear, Talayotic settlement in the 3rd millennium b.p. (Waldren 1984; 1986; 1987). This settlement consisted of five talayots and their associated structures (in some cases directly attached to the monuments), extending over a distance of at least 250 m. Excavations by Waldren on talayots 1, 2 and 3 (T1-3), and in some of the external structures, have yielded cultural materials, radiocarbon dates in the 3rd millennium b.p. and evidence for activities such as metal- and leather-working.

Excavations in and around talayot 4 (T4) took place mainly in 1984-5, as part of a project designed to complement Waldren's fieldwork and to study four problems: these related to the layout of the settlement as a whole (did it expand linearly from T1 to T4, as suggested initially by Waldren?), the chronology and use of talayots and their surrounding structures, and the subsistence of the talayot builders. Reports on the project and on the subsistence data are being published elsewhere (Chapman & Grant in press a and b), while an analysis of the formation processes of T4 and their implications for interpretations of talayot use is already in print (Chapman & Grant 1989).

Excavation inside T4 led us to distinguish seven alternating phases of use and disuse, from the construction of the monument until its final collapse, in a depth of 2.4 m of stratified deposits (Chapman & Grant 1989: 59; Chapman & Grant in press a). In spite of the effects of erosion and terracing, traces of two apsidal structures were uncovered in the excavations outside T4 (Chapman & Grant 1989: figure 3). Deposits were shallow, and divided into three phases relating to the construction, use and abandonment of the two structures (Chapman & Grant 1989: 62). There was no stratigraphic linkage between the deposits inside and outside T4, but the cultural materials suggest broad contemporaneity.

The dating of Talayot 4: the cultural evidence

The majority of the pottery inside T4 was of Post Talayotic type, which, according to Waldren's chronology, is divided into three phases: early (c. 2750-2550 b.p.), middle (2550-2350 b.p.) and late (c. 2350-2050 b.p.). The earliest pottery is Talayotic and is represented by a handful of sherds from phases 1, 2 and 4 and may pre-date 3000 b.p. Outside T4, the majority of the pottery was of middle and late Post Talayotic type, but again a very small number of sherds of Talayotic type were found, although they too did not necessarily occur in the earliest contexts.

Punic pottery was found inside T4 in phases 4 and 7, and in the latter it was accompanied by Samian and an Augustan lamp. Given the fact that the final collapse of the monument left the interior filled with large stones with voids in between them, then much later material such as the Roman pottery could have fallen in, or been washed in by heavy rainfall. Outside T4, Punic sherds were found in phases 2-3 and in superficial contexts. Thus the Punic pottery occurs in the later phases and was introduced to this area of Son Ferrandell Oleza after the construction and primary use of T4 and its associated structures. The traditional mid 7th-century BC (654 BC) dating for the Punic colonization of Ibiza has been supported by recent research (Fernandez et al. 1984), but it is a subject of debate as to how quickly Punic material spread through the other Balearic islands. Mayoral Franco (1984) argues for the arrival of the first Punic pottery in Mallorca from c. 450 BC, with increasing frequencies and local production from c. 350 BC, while at Son Fornes (Gasull et al. 1984), the first Punic; amphorae are dated to the 3rd century BC. In contrast, Waldren (1986: 156) argues that Punic pottery appears in Mallorca at the end of his Middle Iron Age (= Middle Post Talayotic) c. 2350 b.p., or slightly later than the mid 5th-century BC date proposed by Mayoral. What this all means is that the Punic material from T4 is definitely younger than the mid 7th century BC, but its initial appearance could range from the 5th to the 3rd centuries BC.

A socketed bronze axe of Monteagudo's (1977) type 43C and a fragment of a bronze sword blade were found together in phase 6 inside T4 (Chapman & Grant in press a). The axe type mainly occurs in the Balearic islands and is dated by Delibes and Fernandez-Miranda (1988) to the period c. 750-700/650 BC, while the earliest examples may go back to the 9th century BC. The sword fragment is also a Mallorcan type (see Delibes & Fernandez-Miranda 1988: 93-9), though of a form which probably did not appear before the 10th century BC. It is reasonable, therefore, to date the T4 fragment to the 10th-9th centuries, perhaps even the 8th century BC. Taken together with the axe, we can propose that these objects were withdrawn from circulation probably in the 9th-8th centuries BC (we are indebted to Peter Northover for this dating). But if the single Punic sherd from phase 4 is in situ, then the deposition, of the bronzes would have been later in date, possibly some two to three centuries after the date range given above.

Iron fragments were found in late contexts and do not relate to the primary use of the monument and its surrounding structures. The fragment from inside T4 comes from the same context as the Punic and Roman pottery (phase 7), while the iron from outside T4 comes from the same phases (2-3) as the Punic pottery. Mayoral (1984) dates the introduction of iron into Mallorca at c. 650-450 BC, but calibrated radiocarbon dates from Son Matge (Waldren 1986: 75) extend this back to the 9th century BC, when small-scale iron-working was already under way in central Italy (Hartmann 1989). However, the much later chronology for Punic materials makes this early dating for T4 difficult to sustain.

Given the problems posed by the existing Post-Talayotic periodization, this could put the dating of T4's construction and earliest phases before the mid 7th century BC, or before the mid 5th to 3rd centuries BC. The occurrence of a small number of Talayotic sherds may suggest that the construction of T4 goes back to the beginning of the 1st millennium BC, assuming that the material is not residual. The context of the latest cultural material is ambiguous, but we suggest that the monument was finally abandoned within the second half of the 1st millennium BC.

The dating of T4: radiocarbon samples

Details of nine radiocarbon dates for T4 are given in TABLE 1. Calibration of the dates follows the high-precision curve (Pearson & Stuiver 1986; Stuiver & Reimer 1986), and the calibrated dates in TABLE 1 are given at the 2 sigma range. The calibrated dates are also plotted in a bar diagram and in a probability distribution diagram date b.p. date BC context phase
inside T4
IRPA-907 2815|+ or -~60 1160-832 126 phase 1
IRPA-1043 2910|+ or -~50 1267-940 126 phase 1
IRPA-1016 2540|+ or -~45 809-526 112 phase 2
IRPA-881 2580|+ or -~60 895-519 115 phase 3
IRPA-1042 2790|+ or -~50 1067-830 120 phase 4
IRPA-880 2680|+ or -~60 987-791 109 phase 4
IRPA-1015 2475|+ or -~40 782-416 103 phase 7
outside T4
UtC-1155 2810|+ or -~70 1208-825 033 pre-S2
UtC-1154 2490|+ or -~80 795-412 002 S2
TABLE 1. Radiocarbon dates for talayot 4 (charcoal) and
structure 2 (animal bone), with calibrations given at the
2-sigma interval.

The dates for phases 1 and 7 (that is, the monument's construction/initial use as opposed to its final abandonment) are statistically distinguishable, as appear to be the dates for phase 1 as opposed to phases 2-3. On the other hand, the dates for phases 2-3 and 7 are indistinguishable, while the two dates for phase 4 are inconsistent with their stratigraphic context and show more overlap with the date range for the construction and initial use of the monument. No contextual evidence can be cited to account for the anomaly of the phase 4 dates, but the sequence of phases of use and disuse within the monument, including hearths cutting into earlier deposits, might be expected to produce some dated samples which are older than their stratigraphic contexts. The overlap between the date ranges for phases 2, 3 and 7 might be attributed to a comparatively rapid accumulation of deposits (as predicted by the condition of the animal bone assemblage: Chapman & Grant 1989), but as the ranges span one of the 'problem areas' of the calibration curve at c. 2500-2400 b.p., the nature of the chronology seems as convincing an explanation.

Given the date ranges for phases 1 and 7, we can propose that T4 was constructed and in primary use within the period c. 1200-900 BC, and had been abandoned and collapsed by c. 600-400 BC. This would give an overall maximum 'life-span' of c. 500-800 years. The construction dating agrees with that predicted by the presence of small numbers of Talayotic sherds. The date range for phase 7 is incompatible with the presence of Roman pottery, and lends support to our suggestion that it was deposited in the monument well after its final collapse.

Outside T4, the earlier date is on animal bone excavated from beneath the east wall of structure 2 (Chapman & Grant 1989: figure 3), and pre-dating its construction. The later date is on animal bone found within the same wall, and is taken as a construction date for structure 2. This date is statistically distinguishable from the pre-wall date and from the construction/primary use dating of T4 itself, while it occurs within the same date range as that for the fill and abandonment of T4 (phases 2, 3 and 7). Although we are faced again with the problem posed by the calibration curve at this time, we propose that structure 2 was constructed after the primary use of T4 and during the time that secondary deposits were accumulating within the monument. As structure 2 was added on to structure 1, there remains the possibility that the latter was contemporary with the primary use of T4. Overall, the dating evidence is consistent with the argument, based on the analysis of sherd size, that larger materials were being dumped inside T4 during its phases of disuse (Chapman & Grant 1989).

How do the calibrated radiocarbon dates relate to the chronology proposed on the basis of the cultural material found inside and outside T4? The Punic pottery from inside T4 came from phases 4 and 7. The calibrated date range for phase 4 is incompatible with the mid 7th-century dating for the earliest material in the Balearic islands, but the date range for the phase 7 sample covers the 8th to the 5th centuries BC (that is, not as young as the 3rd-century BC dating proposed for the first Punic materials at Son Fornes or the mid 5th- to mid 4th-century BC dating of Mayoral Franco for Mallorca as a whole). The problems posed by the calibration curve, and by the sequence of use of the interior of T4, have already been discussed. Thus, it is possible that the Punic sherd from phase 4 is not in its primary context, and it is equally possible that the dates for phase 4 are on older material. Outside T4, the Punic pottery occurs in phases 2-3, which must post-date UtC 1154, which has the same time range as phase 7 inside T4.

The similarity in the date ranges for samples from phases 2, 3 and 7 does not allow precise dating of the bronze axe and sword fragment belonging to phase 6, although the date ranges as a whole are consistent with the expected dates and allow the possibility of their continuation in circulation for at least another hundred years after the mid 7th century BC. An iron blade fragment also occurred in phase 7, making it younger in date than the earliest iron from Son Matge, but within the time range proposed by Mayoral Franco (c. 650-450 BC). The iron objects found outside T4 (in phases 2-3) would appear to be contemporary with, or later than, the fragment from phase 7.

Radiocarbon dating of the Son Ferrandell Oleza talayots

How do the radiocarbon dates from T4 compare with those from the other talayots at Son Ferrandell Oleza? At the time of writing, there are twelve dates relating to talayot 1. There is an overall similarity in the sequences of use and disuse between T1 (Waldren 1986: 97-9) and T4: after the initial occupation (phases 1-2), there was a major collapse/abandonment (phase 3 = T4 phase 2), a re-use with the construction of a subdividing wall (phase 4 = T4 phase 5) and a final abandonment (phase 5 = T4 phase 7). The earliest dates for T4 fit within the overall range of the dates for phase 1 in T1, but the latter extends back by a further century. The date range of c. 1400-800 BC for phase 1 in T1 is consistent with the presence of Talayotic sherds, some of early type, and it should be noted that this was the closest of the monuments to the earlier, Pretalayotic settlement. Also, it should be remembered that the context of the Phase 1 material in T1 was, strictly speaking, preconstruction. Consequently there is no secure basis at present for arguing that the construction and primary use of T1 was earlier than that of T4. The model of linear expansion of the settlement from east to west, beginning with T1, is now less plausible than one of the contemporaneous construction of a series of monuments surrounded by domestic structures.

The dates for the reuse and final destruction and abandonment of T1 (phases 4-5) are statistically indistinguishable and cover the same date range as that for phases 2, 3 and 7 of T4, thus supporting the inference that both monuments were abandoned around the same time. However the occurrence of these date ranges within the period of the wiggles in the calibration curve (c. 2500-2400 b.p.) allows the possibility of differences in this dating within a period of 300-400 years. The date range for the quicklime burial (IRPA-778) is consistent with its later insertion in the collapsed monument. Given the current limitations of the dating methods, it would appear that T1, like T4, was in use over a period of c. 500-800 years (given that phase 1 of T1 pre-dated its construction) and that the construction and primary use of both monuments was over by c. 800 BC. date b.p. date BC phase
IRPA-1041 2970|+ or -~55 1388-1036 phase 1
HAR-3413 2910|+ or -~100 1400-900 phase 1
QL-1531 2910|+ or -~40 1258-1002 phase 1
IRPA-813 2830|+ or -~100 1270-810 phase 1
QL-4190 2730|+ or -~30 973-819 phase 2
HAR-3458 2540|+ or -~60 819-421 phase 2
IRPA-1012 2560|+ or -~80 842-410 phase 4
QL-1533 2500|+ or -~40 797-489 phase 4
IRPA-986 2520|+ or -~50 805-487 phase 5
IRPA-989 2490|+ or -~50 794-416 phase 5
BM-1842 2430|+ or -~230 Not calibrated phase 5
IRPA-778 2100|+ or -~45 353-2 phase 5
TABLE 2. Radiocarbon dates for talayot 1, with calibrations
given at the 2-sigma interval.

No radiocarbon dates are available for the construction and primary use of talayot 2. Instead there are five dates relating to the burning and collapse of the renovated structure and its final abandonment: 2625|+ or -~60 b.p. (IRPA-1044), 2580 |+ or -~30 b.p. (QL-4098), 2500 |+ or -~50 b.p. (IRPA-1045), 2460 |+ or -~80 b.p. (HAR-3459) and 2400 |+ or -~60 b.p. (I-5398). Thus there is broad contemporaneity with the abandonment dates for both T1 and T4.

On the basis of these dates from three of the five talayots at Son Ferrandell Oleza, we propose that their construction and primary use centred on the late 2nd and the early 1st millennia BC, ending by c. 800 BC. The majority of the deposits which today fill these monuments accumulated during the four centuries after this date. That human activity in the settlement continued after 600-400 BC is shown by re-use of the monuments (the lime burial in T1), by the occurrence of some later materials (e.g. Roman pottery), and in particular by the occurrence of such materials in the settlement's domestic structures. Thus the phase of talayot construction and primary use may only have accounted for a third of the total duration of the site's occupation.

Radiocarbon dating and talayots: the wider context

The radiocarbon dates from talayots 1 and 4 at Son Ferrandell Oleza comprise the largest numbers published for individual talayots anywhere in the Balearic islands (Waldren 1986). Of particular importance in the future will be studies which develop upon this work to use carefully selected, short-life samples (thus countering the 'old wood' effect, which has not been adjusted for in this discussion) and error multipliers (e.g. Baillie 1990) to provide more realistic date ranges for the construction, use and abandonment of talayots. Arguments based on monument architecture and typology have been used to propose dates for such events, but it would be fair to state that we have little independent evidence for rates of construction of talayots, and the degree of regional and local variation in such rates, let alone comparative data on the length of use and abandonment of the monuments.

Such data are important to us in our attempts to try and understand the processes of change in the Balearic islands during the late 2nd and 1st millennia BC. We have to consider questions of manpower mobilization in the construction of monuments, social differentiation as expressed in the ability to mobilize manpower, and the degree to which social competition may be expressed through the construction and use of monuments. Questions about talayot construction, use and abandonment are, in fact, political and social questions. How widespread in Mallorca is the suggested pre-800 BC dating noted for the construction and primary use of talayots at Son Ferrandell Oleza? What would the existence and extent of such a horizon signify in political and social terms? Is the rarity of talayot dates for the period c. 3400-3000 b.p. (that is, the early part of the Talayotic Bronze Age -- Rossello 1973; Waldren 1986) in Mallorca as a whole attributable to the limited number of dating programmes, or might the sites which have yielded dates in this period (e.g. S'Illot) indicate regional differences in monument construction? Finally, the abandonment of the Son Ferrandell Oleza talayots, proposed here by 600-400 BC, is close enough to the traditionally accepted beginnings of colonial contact with the Carthaginians to invite questions as to the nature and effects of interaction between colonizers and colonized. Although the change from primary to secondary use of these monuments may have occurred by c. 800 BC, what kinds of changes were taking place in Mallorcan society which made local populations receptive to colonial contact?

Radiocarbon dating, associated with multisite based projects such as those undertaken in the Valldemossa region, is only now beginning to be used more systematically in the study of monument-using West Mediterranean island cultures in the thousand years preceding the expansion of the Roman Empire. The work reported in this paper is a contribution to this study.

Acknowledgements. The Reading University excavations at Son Ferrandell Oleza were funded by the National Geographic Society, the British Academy, the Society of Antiquaries, the Prehistoric Society, and the University of Reading. The continuing research programme at Son Ferrandell Oleza is part of the work of the Deya Archaeological Museum and Research Centre, supported by Earthwatch and the Center for Field Research, while radiocarbon dating is undertaken through the Koninklijk Instituut voor het Kunstpatrimonium in Brussels and the AMS facility at Utrecht. In addition to these institutions, we thank Peter Northover for his study of the bronze objects from T4.


ALVARO, E. 1983. Analisis espacial de los poblados talayoticos en la isla de Menorca, in Homenaje al Prof. Martin Almagro II: 193-203. Madrid: Ministerio de Cultura.

BAILLIE, M.G.L. 1990. Checking back on an assemblage of published radiocarbon dates, Radiocarbon 32: 361-6.

CHAPMAN, R. & A. GRANT. 1989. The talayotic monuments of Mailorca: formation processes and function, Oxford Journal of Archaeology 8: 55-72.

In press a. Talayot 4, Son Ferrandell Oleza: problemas de los procesos de formacion, funcion y subsistencia, Revista d'Arqueologia Prehistorica 1.

In press b. Subsistence and prehistoric monument construction on Mallorca, paper presented to conference on 'New Perspectives in Western Mediterranean Archaeology', Tufts University, Boston, October 1991.

DELIBES, G. & M. FERNANDEZ-MIRANDA. 1988. Armas y utensilios de bronce en la prehistoria de las Islas Baleares. Valladolid: Universidad de Valladolid.

FERNANDEZ, J.H., C. GOMEZ & R. GURREA. 1984. La premiere periode de la colonisation punique a Ibiza, in Waldren et al. (ed.) vol. 3: 785-96.

GASULL, P., V. LULL & M.E. SANAHUJA. 1984a. Son Fornes 1: la fase talayotica. Oxford: British Archaeological Reports. International series S209.

HARTMANN, N. 1989. Society and technology in the Villanovan Iron industry, in M.L. Stig Sorensen & R. Thomas (ed.), The Bronze Age-Iron Age transition in Europe: 93-9. Oxford: British Archaeological Reports. International series S483.

LEWTHWAITE, J.G. 1985. Social factors and economic change in Balearic prehistory, c. 3000-1000 bc, in G. Barker & C. Gamble (ed.), Beyond domestication in prehistoric Europe: 205-31. London: Academic Press.

MASCARO, J. 1968. Prehistoria de las Baleares. Palma de Mallorca: Graficas Miramar.

MAYORAL FRANCO, F. 1984. La fase posttalayotica Mallorquina: periodizacion y dinamica economico-social in Waldren et al. (ed.) vol 4: 1299-1313.

MONTEAGUDO, L. 1977. Die Beile auf der Iberischen Halbinsel. Munich: Prahistorische Bronzefunde. IX: 6.

MOTA SANTOS, P.J. DA. 1988. Torres and Torrentes: a new approach to the settlement pattern of Mallorcan talayots. Unpublished MA dissertation, Reading University.

PEARSON, G.W. & M. STUIVER. 1986. High-precision calibration of the radiocarbon time scale 500-2500 BC, Radiocarbon 28: 805-38.

PERICOT, L. 1972. The Balearic Islands. London: Thames & Hudson.

ROSSELLO, G. 1973. La Cultura Talayotica en Mallorca. Palma de Mallorca: Ediciones Cort.

STUIVER, M. & P.J. REIMER. 1986. A computer program for radiocarbon age calibration, Radiocarbon 28(2B): 1022-30.

WALDREN, W.H. 1984. Chalcolithic settlement and Beaker connections in the Balearic Islands, in Waldren et al. (ed.) vol. 4: 911-65.

1986. The Balearic pentapartite division of pre-history. Oxford: British Archaeological Reports. International series S282.

1987. A Balearic Beaker model. Ferrandell-Oleza, Valldemossa, Mallorca, Spain, in W.H. Waldren & R.C. Kennard (ed.), Bell Beakers of the Western Mediterranean: 207-66. Oxford: British Archaeological Reports. International series S331.

WALDREN, W.H., R. CHAPMAN, J. LEWTHWAITE & R.C. KENNARD (ed.). 1984. The Deya Conference of Prehistory: early settlement the western Mediterranean islands and their peripheral areas. Oxford: British Archaeological Reports. International series S229.

WEBSTER, G.S. 1991. Monuments, mobilization and Nuragic organization, Antiquity 65: 840-56.
COPYRIGHT 1993 Antiquity Publications, Ltd.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1993 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Chapman, Robert; Strydonck, Mark Van; Waldren, William
Date:Mar 1, 1993
Previous Article:Mapping La Milpa: a Maya city in northwestern Belize.
Next Article:Green fluorescence from the hair of Lindow Man.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters