Integrating Zooarchaeology and Paleoethnobotany A Consideration of Issues, Methods, and Cases

A Tale of Two Shell Middens: The Natural
versus the Cultural in “Obanian” Deposits
at Carding Mill Bay, Oban, Western Scotland

László Bartosiewicz, Lydia Zapata, and Clive Bonsall


Composed largely of mollusc shells resulting from food procurement activities,
coastal shell middens have been regarded as valuable sources of information about
past human exploitation of coastal and marine resources. It is less widely appreciated
that these sites, which lie at the interface between the sea and the land, have
significant potential to inform us about the terrestrial environment and its resources.
In this chapter, an attempt has been made to integrate results of paleoethnobotanical
and zooarchaeological studies with existing archaeological knowledge concerning
Mesolithic and Neolithic environments and subsistence at a shell midden site on the
west coast of Scotland. We compare and contrast the information derived from
macrobotanical and vertebrate faunal remains from two locations at the site of
Carding Mill Bay. Although the midden deposits were also studied from the malacological
point of view, the shellfish remains are not considered here as they characterize
only the marine environment. Moreover, the terrestrial component of a
midden may tell us more about post-depositional taphonomic processes than the
marine component.
Carding Mill Bay I and II are the westernmost known sites in a cluster of
Mesolithic/Neolithic shell middens around Oban Bay (Fig. 1). Following the
discoveries made in caves in the area during the late 19th century (Anderson 1895,
1898; Bonsall and Sutherland 1992; Lacaille 1954), these sites were originally
believed to represent a discrete Mesolithic culture confined to the coastal areas of
central-west Scotland, which came to be known as the “Obanian culture” (Movius
1942). Subsequent research has cast doubt on this interpretation, and the “Obanian”
shell middens are now seen as simply one aspect of the Mesolithic maritime adaptation
in western Scotland (Bonsall 1996). Moreover, they do not belong exclusively
to the Mesolithic period (9500–3900 cal. bc); the deposition of shell middens containing characteristic “Obanian” bone tools continued in western Scotland into
the Neolithic period (3900–2500 cal. bc) and, possibly, as late as the Bronze Age
(2500–600 cal. bc) (Griffitts and Bonsall 2001). Faunal studies of several “Obanian”
sites have been undertaken, including the analysis of material from Carding Mill
Bay I (CMB I), which is located some 15 m south of CMB II. The excavation of
CMB I was undertaken in 1989, and the excavation report (Connock et al. 1993)
included a study of the vertebrate remains by Hamilton-Dyer and McCormick
(1993) as well as a report on the paleoethnobotanical remains by Boardman (1993).
Their detailed accounts permit comparisons with the animal and plant remains
recovered more recently at CMB II.
1 Materials and Methods
Carding Mill Bay II (CMB II) is located approximately 1.5 km southwest of Oban,
at latitude 56° 24¢ 26″ N and longitude 5° 29¢ 26″ W. Prior to its discovery, the site
was buried beneath the talus deposits that accumulated at the base of a raised
marine cliff. The cliff and the rock platform in front of it constitute a former marine
shoreline (known as the “Main Rock Platform”) that is particularly well developed in the Oban area. The site was discovered in 1988 when the talus deposits were
disturbed during construction work. Excavations, funded by Historic Scotland and
directed by Clive Bonsall, were undertaken between 1991 and 1993.
The site faces northwest and consists of a shallow natural recess in a near-vertical
rock face. The recess was infilled with sediments containing a variety of archaeological
remains. These could be divided into a series of lithologically distinct layers
(Fig. 2). Excavation was based on a 50-cm grid, and the deposits were removed in
horizontal unit levels (“spits”) of 5 cm thickness between layer boundaries. All
excavated materials were passed through a nest of sieves, with mesh sizes between
1 and 4 mm, while the material from one 50 × 50 cm grid square (Q5D) was treated
as a column sample and subjected to flotation and wet sieving using mesh sizes
down to 0.25 mm – the majority of the nonwood botanical samples were recovered
from this column.
Radiocarbon dates for the shell midden at CMB I range from ca 4010 to 3550
cal. bc, suggesting an Early Neolithic age, possibly extending back to the time of
the Mesolithic–Neolithic transition in western Scotland (Bonsall and Smith 1992;
Connock et al. 1993). The dating of the midden layers at CMB II is less well-documented.
Currently, no radiocarbon radiocarbon dates are available, and the artifact
inventories differ significantly from CMB I – for example, there are no bevel-
ended tools of bone or antler (see also Griffitts and Bonsall 2001) from either the lower or
upper midden of CMB II, though these were common in the CMB I midden. The
presence of pottery and bones of domestic livestock, especially caprines, in the
upper midden of CMB II indicates a post-Mesolithic date. The lower midden, however,
produced no pottery, and only a single bone from a domesticate was observed.
On the basis of the pottery typology, the upper midden is provisionally assigned to
the Late Neolithic; the lower midden could date earlier in the Neolithic or, if the
single sheep/goat (Ovis/Capra spp.) bone was not in situ, to the Mesolithic.
However, differences in the artifact and faunal inventories of the midden deposits
at CMB I and II (see below) may indicate that the respective middens belong to
different periods in the human use of the Carding Mill Bay locality.
An initial set of samples was retrieved by wet sieving. The Q5D “column” was
processed separately at a later date, primarily for the recovery of macrobotanical
remains and land snails. This yielded a vertical series of 58 individual samples, each
comprising material from a single 5 cm-thick “spit.” The paleoethnobotanical material
was recovered by flotation in the laboratory using a 0.25 mm sieve. The residues
(heavy fraction or “retent”) from the flotation process were hand sorted in order to
retrieve other archaeological components (molluscs, small bones, lithic debitage,
etc.). The plant material that was found during this sorting was included in the paleoethnobotanical
sample. Some of the residues were double-checked by Zapata in order
to assess the effectiveness of the initial sorting, which was found to be adequate.
The processed flotation samples (light fraction or “flot”) were sorted under a lowpower
reflected-light microscope. Wood charcoal was identified using epi-illuminated
light microscopy. Schweingruber (1990) was used for identification and nomenclature.
Identification was facilitated by using a reference collection of modern carbonized
wood. All fragments of wood charcoal >2 mm were examined, which is usually
considered to be the minimum size when using anatomical features as criteria for
identification. Most fragments were close to this size, so the process was quite timeconsuming
and may have increased the number of problematic identifications.
Virtually all the identifiable animal bones were found in the 1–4 mm sieve fractions,
and these fractions were critical for the recovery of the remains from
microvertebrates. The <1 mm sieve fractions provided very few animal remains.
Even the recognized pieces (e.g., incisor splinters from rodents, fragmented fin
rays, branchyostegalia, and ribs from small fish) likely originate from identifiable
specimens encountered in the less finely recovered fractions. This falls in line with
experimental evidence that bone splinters shorter than approximately 20 mm have
a 95% chance of being missed when only hand collection is practiced (Bartosiewicz
1988; see also Peres, this volume chapter 9, for further discussion). The significance
of sieving is clearly illustrated by the example of fish bones at the site, of
which only a few large fragments ended up in the hand-collected sample of bones.
Evidently, fine sieving led to an increase in the number of nonidentifiable, small
fragments, but also resulted in a more complex faunal picture, especially in the
case of fish and rodents. As a convention, “cf.” is used here to signify that the
identification is not certain, but there is a very high probability of belonging to a
particular taxon (see also Peres, this volume chapter 9). 2 Study Results
2.1 Macrobotanical Remains
The types of remains recovered in the samples include: (a) wood charcoal, the most
abundant type; (b) a few seeds, badly preserved; (c) fragments of hazelnut (Corylus
avellana) pericarp; (d) a few small fragments of possible vegetative parenchyma
(soft storage tissue) which forms the major part of organs such as roots and tubers
(Hather 1993); and (e) miscellaneous plant material that could not be identified
further.
The results of the charcoal analysis are presented in Table 1. A summary of the
results can be seen in Table 2 and Fig. 3. In order to simplify presentation, identifications
have been grouped by the most probable taxa (that is, cf. oak/durmast oak
[Quercus robur/petraea] was classified as oak/durmast oak [Quercus robur/petraea]).
The fragment described as alder/birch (Alnus sp./Betula sp.) was not considered.
The data show that in the lower midden hazel is the most important taxon
(73.4% of the fragments identified), followed by oak (12.4%), alder (Alnus glutinosa,
6%), elm (Ulmus sp., 4%), and willow/aspen (Salix/Populus, 3%). Only one
fragment of birch (Betula sp.) and another belonging to the rose family (Rosaceae)
have been identified. Remains from the upper midden show that hazel continues to
be the most important taxon, but its proportion is reduced to approximately half of
the fragments identified (52.6%). In contrast, oak increases in relative importance
(30.7%) in the sample. Alder (9.6%), willow/aspen (5.2%), and birch (one fragment)
are the other taxa identified; in these cases the proportions are low, similar to
their frequency in the lower midden. No fragments of the rose family were identified
here (but only one had been recognized in the lower midden) and most importantly,
elm wood is no longer present. Layer 4 and the Layer 5/6 transition did not
yield statistically significant results since the number of fragments identified was
very low (18 and 25, respectively). The presence/absence of species is similar to
that in the midden deposits, however, in both cases the percentages are closer to the
lower midden, with proportions of hazel wood exceeding 70%.
Other carbonized plant remains besides wood charcoal are extremely scarce and
poorly preserved. There are only a few seeds (grass [Poaceae] and stitchwort
[Stellaria sp.]) and a few fragments of unidentified, possible parenchyma tissues.
The stitchwort seed was also identified at CMB I, along with the sporadic charred
seeds of sedge/knotgrass (Cyperaceae/Polygonaceae), bedstraw (Galium sp.),
cinquefoil (Potentilla sp.), and bramble/raspberry (Rubus fruticosus seu ideaus).
Except for the latter, these weeds – indicative of a humid environment – are difficult
to interpret in archaeological terms (Boardman 1993).
A few uncharred seeds were also present in some of the samples but these are
interpreted to be modern in origin. The small number of carbonized seeds and (possible)
parenchyma does not allow any paleoecological or paleoethnobotanical
interpretation. Fragments of hazelnut pericarp occurred commonly in the midden
samples. Much has been written about the role of hazelnuts as a past food resource, especially in the context of the European Mesolithic (Mason 1996), although hazelnuts
seem to have remained very popular until at least the Bronze Age at many sites
in Britain (Moffett et al. 1990).
Several explanations have been provided for the abundance of carbonized hazelnut
shells at the Mesolithic sites. It is often considered that they were roasted to
benefit storage or shelling, or to kill insects. Heating also releases oil and changes
oil structure, thereby improving flavor and making grinding easier. Hazelnuts may
also have been used as fuel. Boardman (1993) further notes that hazelnuts are one
of the few edible wild plants likely to be very productive under both fully wooded
and more open habitats. Most experts agree that these general explanations should
be supported by further ethnographic, taphonomic, and experimental research. 2.2 Animal Remains
To inform our discussion of the zooarchaeological remains from CMB II, we compared
our results for the major taxonomic groups with the number of identifiable animal
specimens (NISP) in water-sieved samples from CMB I, identified by Sheila
Hamilton-Dyer (1993) and Finbar McCormick (1993).
2.2.1 Bony Fish
This group comprised mainly the remains from both marine and anadromous/catadromous
fish (Table 3). The main characteristics of the identified taxa will be
reviewed in terms of their occurrences at CMB and the habitat types they represent.
Common eel (Anguilla anguilla) may attain a maximum length of 140 cm
(Campbell 1989:274). In contrast to CMB I, bones of this species occurred sporadically
at CMB II. The colorless elvers of eel move into brackish water areas and start
moving upstream to live in fresh water for several years. Some of them, however,
remain in river estuaries. The older individuals move downstream during their
spawning migration by the end of summer (Angel 1977: Figure 40). Present day
otter (Castor fiber) feces from Mull (Argyll) contained up to 7.4% eel remains
(Watt 1991:24, Table 7).
Remains from salmon (Salmo cf. salar) and trout (Salmo trutta; Salmonidae)
were found in significant numbers. Most of the remains originate from fairly large,
adult individuals. Smolts of salmon (Angel 1977: Figure 36) move down river to
the sea in May and June. They spend some time in estuaries where they acclimatize
to the salinity of sea. Most of these fishes migrate to the sea at a length of 10–19 cm
(Muus and Dahlstrøm 1977:76). The maximum adult length of these fishes is
1.5 m, although none of the bones recovered at CMB II belonged to such large
individuals. Trout (maximum adult length 1 m), a species with a similar life history,
is distinguished from salmon by its plumper body. Young trout migrate to the sea
when 15–25 cm long but they stay in the vicinity of the coast for 0.5–5 years (Muus
and Dahlstrøm 1977:78). Consequently, fragments not identifiable at the species
level may originate from either salmon or trout.
With the exception of cod (Gadus morrhua), species in the cod family (Gadidae)
prefer waters of high salinity. Their remains dominated the CMB II fish bone
assemblage, both in terms of NISP and weight. The young, especially, may be
caught along the shore. Species identification was limited to the few most diagnostic
skeletal elements. Of the known Mesolithic zooarchaeological assemblages
from Scotland, cod formed the majority of the identified fish remains at the east
coast site of Morton as well (McCormick and Buckland 1997:90). In later periods,
fishing in Scottish waters has increasingly concentrated on several species in the
cod family (Barrett et al. 1999:354).
At CMB II, poor cod (Trisopterus minutus), saithe (Pollachius virens), and
pollack (Pollachius pollachius) contributed the most identifiable bones to the gadid remains. Both small and large individuals were recognized among the latter.
While 19–23 cm long, poor cods are considered economically unimportant by
modern standards (Muus and Dahlstrøm 1977:106); saithe and pollack may reach
a maximum length of 130 cm, while they are 60–70 cm long in the fifth year.
Remains of small individuals from CMB II seem to correspond to the first year age
group in the bimodal size distributions obtained for this species by Mellars and
Wilkinson (1980:21). Whiting (Merlangius merlangus) can attain maximum
lengths of 40–50 cm (Muus and Dahlstrøm 1977:106). This species was far better
represented at CMB I.
Cuckoo wrasse (Labrus mixtus) is a fish whose length varies between 30 and
35 cm. It lives in the algal zone on rocky coasts (Muus and Dahlstrøm 1977:128),
usually in waters below 10 m (Campbell 1989: 288). Among the non-gadid species,
the contribution of labrids to prehistoric faunal assemblages seems to decrease
through time (Barrett et al. 1999: Figure 4). Gray gurnard (Eutrigla gurnardus) is
a bottom-dwelling species found in waters 10–150 m deep. It may reach a length
of 45 cm (Muus and Dahlstrøm 1977:162).
Bullheads or sea scorpions (Cottidae) are predatory, bottom-living fish with no
swim bladder. They do not move far from the area in which they have grown up
(Muus and Dahlstrøm 1977:12). While these usually small fish are of no known
commercial value today, their remains made up 11.3–12.4% of the fish bone recovered from otter feces in Mull (Watt 1991:24, Table 7). The remains of similar small
individuals were found in the CMB II material, a possible indication of animal
predation rather than human activity.
The family of right-eyed flatfish (Pleuronectidae) was represented by numerous
bones from dab (Limanda limanda), commonly occurring in coastal waters. The
length of this flatfish species rarely exceeds 40 cm (Muus and Dahlstrøm 1977:182).
While most adult flatfish occur at depths of 10–15 m, the young usually frequent
shallower coastal waters (Muus and Dahlstrøm 1977:184) and may even be caught
by hand. This is consistent with the observation that bones of this fish at CMB II
originate from small individuals.
High concentrations of small fish bones, especially when flattened and distorted,
showing signs of digestion (such as some salmon remains at CMB II), would alternatively
be characteristic of offal and feces from otter holts (Cerón-Carrasco
1992:3), daily hideouts of these animals.
2.2.2 Amphibians
Only sporadically occurring long bones of frogs/toads (Anura) were recognized in
the material. Remains from these small terrestrial animals, often considered commensal
at ancient settlements, most typically represent “taphonomic gain” resulting
from active intrusion or water transport (precipitation). In contrast to warmer
climates, where these animals grow large enough to be exploited for meat (Cooke
et al. 1996), it seems unlikely that bones of small frogs at CMB would have been
introduced by human consumption into the stratigraphy.
2.2.3 Birds
The exploitation of birds was of great importance in coastal adaptations throughout
the Mesolithic of northwestern Europe (Grigson 1989:60). In comparison with
mammals, usually numerous avian species are represented by relatively few bones,
a trend characteristic of bird remains owing to both their natural taxonomic diversity
and the greater degree of fusion between elements in the bird skeleton
(Bartosiewicz and Gál 2007). The bird taxa identified are compared to the results
from CMB I in Table 4. Bones of willow tit (Parus montanus), robin (Erithacus
rubecula), finch (Fringillidae), and other small perching birds may be considered
natural deposits at CMB II. These small birds indicate that the midden was located
in the woodland/littoral ecotone. Like amphibians, small birds identified at this site
are indicative of the natural habitat, rather than human activity.
Given the long tradition of seabird exploitation in the coastal areas of Scotland
(Serjeantson 1988), it is somewhat disappointing that larger birds are represented
only by one bone of a red-throated diver (Gavia stellata) and a cormorant
(Phalacrocorax carbo) each. The seven bones of puffin (Fratercula arctica) may
have even originated from burrowing individuals. At CMB I, remains of common or herring gull (Larus argentatus seu marinus), razorbill (Alca torda), and guillemot
(Uria aalge) form a group of findings that originate from birds most commonly
encountered in marine environments. White-tailed [sea] eagle (Haliaetus albicilla) is
more typical of coastal habitats than the golden eagle (Aquila chrysaetos), and its
eyries were usually on sea cliffs and pinnacles (Barnes 1975:83). The bird bone assemblages
from the two CMB sites may be seen as complementary to one another.
2.2.4 Mammals
Mammalian remains identified at CMB I by McCormick (1993) are compared to
the recent results in Tables 5 and 6. The greatest difference between the two faunal
assemblages is the overwhelming dominance of small rodents, especially bank vole
(Clethrionomys glareolus) and field vole (Microtus agrestis) remains, at CMB II.
Bones from these species must have been fewer at CMB I, although they were
quantified only using the loose term “dozens” in that report (see Table 5). Rodent
remains occur rather uniformly across all contexts.
Pigmy shrew (Sorex minutus) contributed the smallest bones to the CMB II
assemblage. Although common shrew (Sorex araneus) is known from CMB I
(Hamilton-Dyer and McCormick 1993), it was not recorded at Morton or sites on the island of Oronsay (Inner Hebrides). Although it is impossible to recover articulated
skeletons from sieve residues, given the excellent preservation of their bones at
CMB II, it is possible that these tiny animals found their way into the cracks and
cavities of the midden by entering into archaeological deposits in search of their
prey. The possibility of water transport must also be considered.
Of the numerous rodent remains deposited in all contexts, bank vole, field vole,
and water vole (Arvicola amphibius) could be identified to species on the basis of
tooth enamel patterns. Some specimens were assigned to the latter, largest species
on the basis of size. From an environmental point of view, it is significant that bank
voles inhabit mixed woodland habitats. This species is the only vole that climbs
bushes (Mitchell and Delap 1974:56). Field vole was also considered one of the
faunal indicators of a forested environment at the site of CMB I (Hamilton-Dyer
and McCormick 1993).
Woodland is best indicated by the presence of shrew and bank vole. Field and
water voles are typical of humid, bushy areas; this does not contradict the previous
conclusion. All these species may burrow near the soil surface, but are unlikely to
go deeper than 10 cm (Zsófia Kovács, personal communication 2008). Whether the
animals died at the site or their remains were washed in from above, their presence
is yet another indication that the immediate environment of the shell midden was
forested. Some bones of red squirrel (Sciurus cf. vulgaris) and pine marten (Martes
martes) at CMB I support the reconstruction of a woodland environment. The bones
of larger mammals, especially those of domesticates found in the upper midden at
CMB II, are the most likely to have been introduced by human activity (Table 6).
In contrast to rodent remains, two bones of common hare (Lepus timidus) in the
lower midden may originate from prey items of either animals or humans, and in
fact may have been imported by humans to the site. This is even more of a possibility
in the case of large mammals. Remains of sheep (Ovis aries) or, less likely, goat (Capra hircus), usually referred
to here by the subfamily name, Caprinae, were found in relatively large numbers in
the upper midden but only one such bone was found in the lower deposit. While the
latter, single occurrence may be considered accidental, the presence of domesticates
in the upper midden corresponds with the later chronological position of that deposit.
No cattle (Bos taurus) remains were found in either of these two layers at CMB II.
Wild boar (Sus cf. scrofa) is a mammal whose domestic form (Sus domesticus) is
often difficult to recognize from its skeletal remains. No suid bones were large enough
at this site to be unambiguously identified as those of wild pig, a game animal known
from Mesolithic deposits both at the site of Morton and even on the small island of
Oronsay (McCormick and Buckland 1997: Table 6.1). Although pig domestication is
not thought to have taken place before the Neolithic, several biological traits of this
species could have facilitated the development of a special relationship between wild
pigs and humans that preceded domestication in a classic sense (Bolomey 1973:48). Like dogs (Canis familiaris), these social animals were more likely to have scavenged
on human refuse than the ancestors of other, herbivorous domesticates.
In comparison with other ungulates, roe deer (Capreolus capreolus) is underrepresented
in the CMB II faunal assemblage. In addition to the identifiable remains
of this species, many of the 1,000+ small ungulate bone splinters probably belong
to roe deer. While some of these bones could not be distinguished from the remains
of sheep (hence, the term “small ruminant”), sheep could be present only if the
lower midden were post-Mesolithic in date. Roe deer remains occurred both on the
island of Oronsay and at Morton (McCormick and Buckland 1997: Table 6.1).
Together with wild pig, roe deer is considered to be an animal of deciduous woodland
with dense undergrowth or parkland habitats, although recently increasing
adaptation to open grassland has been recorded in Hungary (Bencze 1978:56). No
antler fragments could be identified as those of roe deer.
Red deer (Cervus elaphus) remains, including some antler fragments, occur sporadically
both in the upper and lower middens. Although red deer remains were found
in relatively small numbers, non-identifiable large ruminant bones probably also originate
from this animal. In comparison with small and adaptable roe deer (15–27 kg; van
den Brink 1968:157), large bodied red deer (95–160 kg in Scotland; van den Brink
1968:154) are more dependent on the cover provided by forested habitats.
When the number of taxa are plotted against the number of identifiable specimens
by vertebrate classes (fish, bird, and mammal), the subassemblages of various
sizes from the two CMB middens show the same trend in that they are in that they are comparable regardless of the differences in sample size. CMB II was particularly rich
in fish remains. The only outlier from this main trend (not taken into consideration
when calculating the regression line, Fig. 4), is the group of microvertebrate
remains. As mentioned before, the large number of these bones was counted in
terms of NISP only at CMB II, but the subassemblage was not very diverse, containing
only four identifiable taxa.
3 Discussion
Plant and animal remains differ radically in terms of their taphonomic histories,
including exploitation by humans as a cultural factor of biostratinomy. Prehistoric
plant remains recovered from CMB II have been preserved through charring. Wood
charcoal is the most frequent type of paleoethnobotanical material. Apart from the
possibility of natural forest fires, this special form of preservation also offers
evidence of human activity: the use of fire at the locus investigated, which includes
the selection and transport of firewood to the place of use.
Zooarchaeological remains may represent different forms of deposition. In order
of increasing complexity, faunal materials may originate from (Gautier 1987):
1. Intrusive animals (bones of actively burrowing rodents, shrews, puffins, etc.)
2. Carcasses originating from in situ death, water transport, or deposited by predators
(otter feces, owl pellets, gut contents of larger fish preyed upon by otters or
humans)
3. Human food refuse (bones showing butchering marks, burning, or marrow
extraction)
4. Debitage from craft activities (waste from in situ manufacturing, bones with
skinning marks)
Given the horizontal, coastal location of shell middens and their vertical positions
relative to prehistoric sea levels, the direct role of tidal activity in accumulating fish
remains should be considered. Many sites in central-west Scotland located below
15 m above present sea level were probably destroyed by wave erosion, or at least
affected by storm waves, during the Mid-Holocene marine transgression, which
culminated in the Oban area between 5500 and 5900 cal. bc (Bonsall and Sutherland
1992). As opposed to marine tidal activity, some bones representing terrestrial
microfauna (amphibians, insectivores, and rodents) may indeed have been washed
into the midden deposit by rainwater from above, a phenomenon frequently
observed in rockshelters and caves (e.g., Andrews 1990). This type of accumulation
is a characteristic source of microvertebrate remains.
Just as burning by humans contributed to the preservation of charred plant parts,
the vertebrate remains recovered from the middens also owe their survival, at least
partially, to human activity. In the highly acidic soils of the area, the preservation of
bone and antler simply would not have been possible without the accumulation of
shell middens. The calcareous organic matrix of predominantly limpet shell deposits (Russell in Connock et al. 1993:34; Bonsall et al. 1994) creates an alkaline environment,
buffering soil acidity and thereby promoting the preservation of bone.
3.1 Vegetation
Two of the main characteristics that explain the presence and relative importance of
any type of wood at an archaeological site are its relative abundance in the vicinity
and its quality as a fuel wood. The most abundant species and those that are considered
good fuel tend to be well represented. A relevant ethnobotanical study was
carried out among farmers in the Atlantic Basque Country, in areas where mixed
oak woodland is the main arboreal community. In Table 8, arboreal species identified
at CMB II were ranked according to the mean fuel quality value established
during that study (Zapata and Peña-Chocarro 2003).
The presence and relative abundance of the different fuel woods in the CMB II
deposits seem to indicate that hazel and oak wood were abundant around the site.
In all the samples, these two taxa account for more than 80% of the identified fragments.
Taphonomic problems aside, this would mean that there was a mixed-oak
woodland close to this site. According to Table 8, both hazel and oak are highly
valued fuel woods (oak scores 4 and hazel 3.8 on the 5-point scale of fuel quality
in domestic fires). Thus, hazel must have been a very important component of the
woodland around the site (in all samples, it constitutes over 50% of the fragments
identified). This is also supported by evidence from CMB I, where hazel nutshells
dominated the charred plant remains. Even if they were highly fragmented, significantly
overrepresenting the number of nuts involved, their presence is important in
itself (Boardman 1993).
The presence of elm charcoal in the lower midden and its absence from the
upper midden is interesting in light of the general decline in elm pollen values
recorded in pollen diagrams across NW Europe around the time of the Mesolithic–
Neolithic transition, 4150–3400 cal. bc (Parker et al. 2002). Could this indicate that
the lower midden at CMB II (layer 5) pre-dates the elm decline, which in the Oban
Table 8 Average scores of fuel quality for various species established in domestic fires (from
1 = very bad fuel to 5 = excellent fuel) according to modern farmers’ opinions (Zapata and Peña-
Chocarro 2003)
Species Fuel quality
Oak 4
Hazel 3.8
Willow 3
Birch 2.5
Elm 2.25
Alder 2
Aspen 1.4


area occurs ca. 3900 cal. bc (Davies 1997)? Since cultural selection of wood must
always be considered in charcoal analysis, this ecological conclusion needs to be
confirmed by direct radiocarbon dating of terrestrial organic matter (plant macro
remains or animal bone) from layer 5.
3.2 Fauna
Animal remains from the excavations at CMB II further complement the environmental
picture outlined by the charcoal analysis and also corroborate the previous
zooarchaeological study of CMB I by Hamilton-Dyer and McCormick (1993).
Their opinion that “much of the bone in the shell midden contexts can be accounted
for by natural factors” can be substantiated. The rodent and bird species identified
at this site reflect the proximity of wooded/bushy environments in the site’s immediate
proximity.
Some differences in diversity between the two faunal samples from CMB are
evidently related to assemblage size as has already been shown in Fig. 4 by the very
close (almost 99% determination) linear relationship between sample size and
sample richness. Poorly represented species are more likely to occur when more
specimens are available for study (Grayson 1984:137; see also Peres, this volume).
This is clearly the case with fish remains, where the number of identifiable bones
from CMB II is twice as great as from CMB I. Consequently, a slightly richer
inventory of species was recognized. Large numbers of salmon and cottid remains
at CMB II are especially conspicuous. The number of non-identifiable fish remains
is also greater at CMB II.
Almost the reverse of this situation, however, was observed in the case of birds.
Only remains of various (mostly non-identifiable) perching birds were found in the
samples from CMB II, while a fairly rich range of predominantly marine birds
came from the CMB I excavation. The different representation of fish and birds in
the two assemblages may indicate that there is more than a size-related difference
between the two assemblages. The greater numbers of non-identifiable fish and
vole bones from CMB II may result from finer recovery techniques applied at this
latter site; many of the bones were found by sieving through mesh sizes of 1 and
2 mm, respectively.
4 Summary and Conclusions
Both palaeoethnobotanical and zooarchaeological evidence available from the two
sites at CMB are primarily indicative of the environment. While charcoal was present
as a result of probably human-induced fire, it must have originated from locally
gathered wood. Many of the small terrestrial vertebrates (perching birds, rodents)
also seem to be indicators of a wooded natural environment. On the other hand, remains of hazelnut and hand-collected animal bones are indicative of human
subsistence activities related to the two shell middens.
It is too simplistic to assume that the proportions of tree taxa represented in
archaeological assemblages reflect their relative abundance in the local vegetation.
However, it is usually accepted that archaeological wood charcoal is an indicator of
at least some of the plant communities formerly growing in the site area and thus a
valuable source of data for reconstructing prehistoric vegetation and environments
(Smart and Hoffman 1988). Consideration should be given to: (1) taphonomic
questions; (2) cultural selection of wood; and (3) context-related variation within
the site.
Like plant remains, cultural deposits of animal remains tend to reflect culturally
idiosyncratic preferences, rather than the fauna of the local environment. However,
as far as the archaeological interpretation of the animal remains under discussion
here is concerned, rodent remains, rare bones from amphibians, and even some
small fish are unlikely to have been deposited by humans at CMB II. Sporadic long
bone fragments from frogs /toads (Anura) tend to appear only in the largest sieved
assemblages, thus being indicative of the effect of sample size on taxonomic
richness.
Although otter was identified only at CMB I, it is an important carnivore at midden
sites along the west coast of Scotland. The number of otter bones was second
only to those of seals at the “Obanian” site of Cnoc Coig, Oronsay, where these
animals were considered to have been hunted (Grigson and Mellars 1987:274).
In contrast to the other mammals previously discussed, otters could easily colonize
island habitats. Although this species thrives in fresh water, populations adapted to
marine environments have been regularly observed (Mitchell and Delap 1974:64).
Today, however, otters are rare around all the islands of the Inner Hebrides. These
avid predators of fish were treated as “vermin” in the recent past (Mercer 1974:46).
Otters feed on fish, such as wrasse, cod and other gadids, and flatfish, as well as on
various crustaceans (Matthews 1989:246).
The few pig and deer bones, on the other hand, directly reconfirm observations
by Hamilton-Dyer and McCormick (1993:34), concerning human exploitation of
the wild fauna. Although deer bones were most probably deposited by humans,
they are also characteristic of the wooded environment than the narrow littoral
zone. The CMB II assemblage contained only a few identifiable bones from marine
birds, a group which may have been targeted by Mesolithic hunters.
Complex relationships between the factors that determined the contents of the
two shell middens are summarized in Fig. 5. Special emphasis has been placed on
the importance of human activity in the preservation of zooarchaeological and
paleoethnobotanical remains in the middens, located in the ecotone between the
forested inland habitat and the littoral zone. Although humans inadvertently
contributed to the survival of the floral and faunal remains under discussion here,
many of the findings represent natural deposits, especially in the case of vertebrate
fauna. Studies of animal and plant remains from coastal shell middens have rarely
been given equal emphasis in archaeological research in western Scotland, and even
more rarely have attempts been made to integrate the two lines of evidence to aid in site interpretation. The results of zooarchaeological and paleoethnobotanical
research at Carding Mill Bay serve to emphasize that shell midden sites are the
result of activities that took place at the interface between land and sea, and while
they are an important repository of information on human exploitation of coastal
and marine resources, they also have the potential to inform us about the local terrestrial
environment and its use by humans.

The influence of time averaging and space averaging on the application of foraging theory in zooarchaeology

1. Introduction
For more than 20 years, Americanist zooarchaeologists
have called on foraging theory [58] to provide
models of human-resource use [3,68]. Even biologists
have occasionally applied such models to zooarchaeological
materials [35,55]. During the 1990s, use of these
models in western North America resulted in the development
of a set of methods for monitoring the history of
human resource use as evidenced by the zooarchaeological
record [8–11,15,19,32–34,50,51,63]. Some of the
more interesting implications of these models concern
the fact that pre-industrial humans had significant influences
on prehistoric faunas the world over; these implications
have now been abundantly confirmed by the
zooarchaeological record (summarized in Ref. [29]). In
particular, as a result of their selective exploitation of
prey taxa that provided either or both low costs and
high returns, humans with primitive technologies often
caused changes in faunal diversity (taxonomic richness
and evenness), independent of changes in climate. Therefore,
humans sometimes had to alter what they were
exploiting as a response to a change in the availability of
animal prey that they themselves had caused.
Many significant new insights into the past have been
provided by applying foraging theory to zooarchaeological
problems. Yet such application is not always
straightforward and thus the voice of caution has
occasionally been heard. For example, many studies
conclude that changes in the list of exploited prey to
include more small, low-value prey taxa were the result
of human exploitation depressing the availability of
large, high-value prey taxa. The voice of caution
requires that this explanation be substantiated by disconfirmation
of alternative causes of depressed prey
availability, such as changes in technology, changes
in how technology was used, and environmentally or
climatically driven changes in taxonomic abundances
[30]. Further, it has been pointed out from an
Old-World context that zooarchaeological abundances
of Linnaean taxa do not always reveal changes in prey
availability with respect to particular spatio-temporal
coordinates [59]. Finally, it has been noted that the
traditional means of quantifying prey abundances—the
number of identified (faunal) specimens (NISP) per
taxon—entails various epistemological and interpretive
difficulties [66]. Importantly, analytical techniques have been and are being developed for addressing a number
of these issues [18,31,60,61,67].
One potential difficulty for which there presently is
no obvious analytical solution is that, as Grayson and
Delpech [31, p. 1119] put it, if one is to apply explanatory
models derived from foraging theory to the zooarchaeological
record, “concepts that are meant to apply
in ecological time must be translated to archaeological
time.” In particular, they underscore the fact that with
respect to prey-choice or diet-breadth models, where
either is measured as the number of taxa exploited or
taxonomic richness (NTAXA), the time period represented
by a zooarchaeological assemblage results in
NTAXA comprising the maximum number of taxa
exploited during the total duration of time represented
by that assemblage. Paraphrasing their example, if five
taxa are exploited over 99 years, but those five taxa plus
five additional taxa are procured in the 100th year, then
diet breadth rendered as NTAXA for the assemblage
resulting from that 100 years will be 10—the maximum
number of taxa exploited (see also Ref. [12]). The
significant point of this example is that variation
between the first 99 years and the 100th year will be
masked by the fact that multiple annual assemblages
have been lumped by the formational history of the
zooarchaeological assemblage. It is the effect of such
lumping, whether along the temporal dimension, the
spatial dimension, or both, that is one concern here.
A second concern is that there is another potential
difficulty not previously mentioned in the pertinent
literature. It resides in the variable typically referred to
as sample size. The influences of sample size rendered as
NISP are often acknowledged, and most zooarchaeologists
who apply models derived from foraging theory
to their data take steps to account for sample size
when taxonomic richness or taxonomic abundance is
measured with NISP (see Ref. [20] for a particularly
important method in this respect). No zooarchaeologist
I am aware of, however, has considered the potential
influence of sample size rendered as the number of
analyzed spatio-temporally distinct assemblages—what
I will term NASM—of faunal remains, which arguably
comprises a measure of sample size at a different scale
than NISP.
The third and final concern resides in the fact that
zooarchaeological assemblages comprise samples of
different portions of the spatial and temporal continua.
For discussion purposes I refer to these portions
as contexts, irrespective of size or duration. Each
assemblage of faunal remains occupies a unique context
in terms of its spatial and temporal coordinates. The
influence of the number of unique spatio-temporal contexts
included in an analysis on interpretations of
changes in prehistoric foraging is only now beginning to
be examined (see subsequently). I refer to the frequency
of these contexts generally as NCTX, the number of
spatial contexts as NCTXspace, and the number of
temporal contexts as NCTXtime. Note that sometimes
NASM = NCTX, but the more typical relation will be
NASM > NCTX as a result of analytical lumping. In
this article, I explore some of the effects that NASM and
NCTX have on analytical results produced by application
of methods derived from foraging-theory models
and used by zooarchaeologists working in western
North America.
2. The usual method
I will not review the reasoning behind the application
of foraging-theory models to zooarchaeological data
here, as there are numerous such discussions now available
[8,9,30]. It suffices to note that as used in western
North America, the typical basic model—the preychoice
model—is that human foragers will preferentially
exploit the largest prey first because these taxa are the
most valuable. The model further holds that if valuable
prey decrease in availability and thus the frequency at
which they are encountered decreases, then foragers
will turn to progressively more kinds of less valuable,
generally smaller, prey—each individual considered less
valuable than a single larger individual—to maintain a
constant level of nutrition. Thus, the ratio of large to
small + large prey will fluctuate over time, decreasing
(becoming progressively smaller than 1.0) as large prey
become less available relative to small prey and increasing
(becoming progressively closer to 1.0) as large prey
become more available relative to small prey. Typically,
prey abundances measured as NISP are rendered as an
index value between 0.0 and 1.0 that expresses the ratio
of large to small + large prey. Ugan and Bright [66,
p. 1309] refer to these index values as “relative abundance
indices (AIs),” and I use their acronym in the
remainder of my discussion. Plotting each AI against
the temporal midpoint of the time span during which the
faunal remains on which it is based were deposited—
whether stratigraphically, radiometrically, or culturally
determined—in a bivariate scatterplot, visually reveals
what are interpreted to be changes in prey return rates
that are in turn interpreted to reflect changes in prey
availability or abundance. Interpretation of the scatterplot
is sometimes aided by calculating a simple best fit
regression line through the point scatter in order to
reveal temporal trends [8,9,33].
Many applications of foraging-theory models to
zooarchaeological data concern temporally distinct
(usually stratigraphically delimited) faunal assemblages
recovered from a single site, or faunal assemblages from
several geographically propinquitous sites distributed
across several temporal periods. In a bivariate scatterplot
of data points, with the x-axis usually representing
time and the y-axis usually representing the AIs, there
are few data points. This makes both the visual and the
596 R.L. Lyman / Journal of Archaeological Science 30 (2003) 595–610
statistical detection of trends in changing faunal abundances
relatively easy and the interpretation of the
covariation between the two variables rather straightforward.
The critical issue to recognize is that if NASM
is small, then few points comprising AIs will be plotted.
For example, Butler [16] plotted AIs for eight temporally
distinct assemblages from a single site against
their ages; Broughton [9] plotted AIs for nine assemblages
from nine sites against their ages; Potter [50]
plotted AIs for six assemblages from two sites against
four time periods; Janetski [33] plotted AIs for 20
assemblages from 19 sites against their ages; Broughton
[8] plotted AIs for 18 assemblages from 14 sites against
their ages; and Broughton [11] plotted 14–16 AIs against
10 temporally distinct strata in a single site. In each of
these applications, NASM%20.
Not only does NASM tend to be small, but in each of
the examples just listed, spatial and/or temporal variation
was variously muted over multiple assemblages.
There are several ways in which spatio-temporal variation
can be analytically muted. These can be illustrated
by considering two fictional assemblages (NASM = 2),
each from a different site. Assume that the sites are
geographically close, but vary in microhabitat (NCTXspace
= 2), that the assemblages were deposited over
similar durations of time (say, 400 years), that the
assemblages are temporally sequential in age
(NCTXtime = 2), and that the assemblages belong to the
same archaeological culture based on their associated
artifacts. What are the analytical alternatives? Because
the assemblages differ in spatial and temporal coordinates,
each AI might be plotted against its age in a
separate graph or plotted in the same graph with a
different symbol denoting a different geographic position,
thereby maintaining the distinctive geographic
(and temporal) coordinates of the two. Or, the two AIs
might be plotted in the same graph using the same
symbol, thereby muting their geographic distinctiveness
(NCTXspace = 1), but maintaining their temporal distinctiveness
(NCTXtime = 2). Alternatively, the AIs
might be plotted with different symbols in the same
graph to maintain their geographic contexts, but according
to the cultural period to which they belong, thereby
muting their temporal distinctiveness (NCTXtime = 1).
Finally, the NISP per taxon of the two assemblages
might be summed, an AI calculated, and that value
plotted against either the absolute (average) age of the
two assemblages or the cultural period represented by
the two. This last possibility would mute the spatial and
temporal distinctiveness of the two assemblages (NCTX
= 1) as well as mute their distinctive taxonomic richnesses,
evennesses, and heterogeneities in favor of what
we can call a general or universal (with respect to the
two assemblages) spatio-temporal-taxonomic measure
(NASM = 1). In the first three alternatives, variation
in geographic space and/or time between the two
assemblages is muted; NCTX is reduced analytically. In
the fourth alternative, not only are both spatial and
temporal variations muted, but so are variables
of taxonomic structure and composition; NASM is
reduced.
Examples of the potential ways that one might
analytically lump contexts or assemblages are found in
the extant literature. For example, Janetski [33] plotted
(NASM = 16) AIs from a single physiographic area and
representing a single cultural period against the midpoint
of their absolute ages. The 16 assemblages were
derived from an area approximately 400150 km, but
NCTXspace was analytically reduced from 16 to 1 by
lumping. Szuter and Bayham [64] summed NISP data
from 70 sites (NASM = 70, NCTX = 70) into 11
temporally distinct sets (NASM = 11, NCTXtime = 11)
in a single area (NCTXspace = 1) and plotted the AIs
against the midpoint of the temporal period represented,
thereby muting both temporal and spatial variations.
Butler [15] summed NISP data from eight sites (NASM
= 9, NCTXspace = 8, NCTXtime<9) by cultural period
(NCTXtime = 4) for an area (NCTXspace = 1) and
plotted AIs against the appropriate temporal midpoint
of a cultural period, thereby potentially muting both
temporal variation and spatial variation.
With no explicit warrant for analytically reducing
NCTX or NASM by lumping temporally and/or
spatially propinquitous assemblages, the question of
what effect that lumping might have on results is begged.
The source of such a question resides in a fundamental
dictum of archaeology that the subject of study comprises
three dimensions of variation [43]. Those three
dimensions are form, space, and time. Form in this case
is the value taken by the AI; space comprises geographic
variation in the locations of the assemblages from which
AIs are derived; and time comprises variation in the
temporal positions of the assemblages from which AIs
are derived.
In many cases, the lumping of otherwise spatially
and/or temporally distinct assemblages is not explicitly
justified in the literature. Some possible reasons for such
lumping can, however, be suggested. It could be argued
that lumping removes the effects of seasonal, geographic,
and individual variation in resource exploitation
in the interest of determining an annual, regional,
or overall pattern; such a warrant has been characterized
as the “‘noise-filtering’ of time averaging” [47, p. 226]. In
zooarchaeology, such noise filtering would provide what
Uerpmann [65] referred to 30 years ago as the “average
diet” of prehistoric peoples across a geographic area
(rather than a single site locality) and/or across a greater
or lesser span of time (rather than a single meal or day’s
worth of meals). Paleobiologists have long been concerned
about the influence of what they refer to as “time
averaging” on ecological interpretations of faunal
assemblages [4,48,52], and they have developed various
R.L. Lyman / Journal of Archaeological Science 30 (2003) 595–610 597
methods for analytically detecting and contending with
time averaging (Refs. [14,37], and references therein).
Paleobiologists have also long been aware that some
accumulations of faunal remains represent multiple
biotic communities [53,54], or what are readily termed
‘space-averaged’ assemblages, and methods have been
developed for recognizing these [28]. Archaeologists,
too, have long been cognizant of such assemblages,
referring to them as resulting from ‘blending and smearing’
[2] that produces “coarse-grained assemblages” [6].
Such spatio-temporal palimpsests are the bane of those
seeking fine-scale resolution, but this is not meant to
suggest that spatially or temporally lumped or blended
assemblages are to always be avoided. Palimpsest assemblages
may be useful and appropriate for some questions
[47], and I return to this point later.
“Palimpsest,” “blended,” and “averaged” are reasonable
terms for those spatio-temporally unique materials
that were lumped during formation of the zooarchaeological
record or that have been analytically produced.
The important point in the context of how foraging
theory has been applied to zooarchaeological data is that
such averaging comprises the conversion of ecological
time and space into archaeological time and space.
Choosing to plot an assemblage against the temporal
midpoint of the period during which it was deposited, or
lumping data of various ages into a single temporal or
cultural period to derive a measure of diet—a more or less
continuous variable—per period results in some unknown
degree of time averaging. Similarly, lumping zooarchaeological
data from multiple sites that are variously geographically
distant from one another and that date to the
same period (if not precisely contemporaneous) in order
to derive a measure of diet results in some unknown
degree of spatial averaging. Cannon [19] has recently
addressed the possible effects of the spatial averaging and
suggested ways to analytically contend with such effects.
In the following sections, I use real data sets to explore the
potential effects of temporal averaging, spatial averaging,
and simultaneous averaging of both time and space on
interpretations. In doing so, I am less interested in the
interpretations that one might derive from a data set than
on the effects on those interpretations that result from
temporal and spatial averaging.
3. Materials and methods
For purposes of detecting the influences of spatiotemporal
lumping on analyses and interpretations, the
ideal situation would be to have assemblages meeting
three criteria. First, the assemblages would be from
various geographic locations in one physiographic
area. Such cases are commonplace in western North
America where many archaeological projects have
been undertaken under the auspices of cultural-resource
management. Second, each assemblage would be of
sufficient size in terms of NISP to allow analytical
lumping of more-or-less contemporaneous and geographically
propinquitous assemblages. This is sometimes
found in western North America, but many
assemblages there are also small, given the contingencies
of the particular project. Third, the assemblages would
be of relatively short duration—say, deposited in a
single season in temperate latitudes—in order to allow
analytical lumping dictated by research questions rather
than by the realities of the variously time-averaged
zooarchaeological record. No such zooarchaeological
record in terms of this third criterion is known to me.
Several regional archaeological projects can, however,
be analytically made to come close to meeting these
criteria.
In the 1970s and 1980s, a number of archaeological
sites located adjacent to a stretch of the Columbia River
of north-central Washington state were excavated as
part of two reservoir projects [17,21]. These sites contained
evidence of human occupation spanning the last
7000 years and representing multiple local cultural
periods. Many, but not all, of the sites produced
numerous faunal remains. I compiled data on NISP per
taxon per assemblage, assemblage age, and assemblage
location from reports resulting from the excavations.
The data comprise 31 assemblages (NASM = 31)
recovered from 18 sites (NCTXspace = 18) distributed
along approximately 420 km of the Columbia River
(Table 1). Five individual site assemblages from the
westernmost portion of the area (Wells Reservoir) were
spatio-temporally lumped for purposes of this analysis
to avoid possible effects of sample size rendered as
NISP, so NCTXspace has already been reduced to 12.
Assemblages used in my analyses are of various ages
(NCTXtime = 30) and span the last 5000 radiocarbon
years; some of them are relatively synchronic and represent
brief time intervals (ca. 200 years) whereas others
represent longer periods of duration (e.g. 2200 years).
For analytical and graphing purposes, I follow previous
researchers and use the chronological midpoint (to the
nearest 100 years) of each assemblage. This means that
varying degrees of time averaging have already been
built into the analysis.
In light of a recent study of the early historic faunal
situation in eastern Washington that suggested that
prehistoric human foragers there had depressed large
game populations [44,45], I designed three indices to
monitor changes in prey abundances over time. One
index is based on the fact that as large mammalian prey
decrease in abundance, one response of human foragers
might be to broaden the diet to include more smallmammalian
prey and more individuals of smallmammal
taxa. Because few small-mammal taxa seem to
have served as human prey throughout the Holocene
(last 10,000 years) of eastern Washington [41], and
because NISP sample sizes for these taxa tend to be
598 R.L. Lyman / Journal of Archaeological Science 30 (2003) 595–610
small, I lumped all small-mammal prey taxa into a single
category—small mammals. Similarly, because virtually
all large-mammal prey are artiodactyls, and because
deer (Odocoileus hemionus (mule deer) and O. virginianus
(white-tailed deer)) tend to dominate most collections, I
lumped the five taxa of artiodactyls (Odocoileus spp.,
Ovis canadensis (bighorn sheep), Antilocapra americana
(pronghorn antelope), Cervus elaphus (wapiti, or elk),
Bison bison (bison)) into one category—artiodactyls.
The first AI is termed the small-mammal index and is
calculated as

artiodactyl NISP/(
artiodactyl NISP

small mammal NISP)
To calculate this index, I included only taxa the
remains of which displayed evidence of human exploitation.
That evidence typically was in the form of
butchering marks, burning, bone breakage, and skeletal
disarticulation [42]. Small-mammal taxa include marmot
(Marmota flaviventris), beaver (Castor canadensis),
muskrat (Ondatra zibethicus), red fox (Vulpes vulpes),
canids (Canis latrans, C. lupus), mustelids (Mustelidae),
and bobcat (Lynx rufus). Bears (Ursus americanus, U.
arctos) and cougars (Felis concolor) were omitted from
the analysis because they are very rarely represented.
Note that the greater the abundance of artiodactyl
remains relative to small-mammal remains, the closer
the small-mammal AI will be to 1.0.
The specific response of human foragers in eastern
Washington to depressed large-game populations hypothesized
by Martin and Szuter [44,45] involved the
intensification of fishing. To determine if this in fact
happened, a measure of the abundance of artiodactyls
relative to that of fish is necessary. The second AI, then,
is termed the fish index and is calculated as

artiodactyl NISP/(
artiodactyl NISP
fish NISP)
Fish exploited by prehistoric foragers included
several species of salmonids and steelhead trout
Table 1
Zooarchaeological data from north-central Washington state by site and age
Site/location Age Age
midpoint
Artiodactyl NISP Small mammal
NISP
Fish
NISP
Small mammal
AI
Fish
AI
Nonartiodactyl
AI
Western
Wells Reservoira 4350–3800 4100 247 151 303 0.62 0.45 0.35
Wells Reservoira 3300–2200 2700 88 82 4812 0.52 0.02 0.02
Wells Reservoira 900–100 500 2 13 7235 0.13 0.00 0.00
45DO-326 300–100 200 81 21 8 0.79 0.91 0.74
45DO-326 1500–800 1100 64 52 3 0.55 0.96 0.54
45DO-326 4500–3000 3700 12 51 10 0.19 0.55 0.16
45DO-326 5000–4500 4700 45 166 67 0.21 0.40 0.16
Eastern
45OK287/288 1500–850 1200 167 1 5 0.99 0.97 0.96
45OK287/288 4400–1500 2900 426 11 4 0.97 0.99 0.97
450K287/288 4800–4400 4600 103 22 0 0.82 1.00 0.82
45OK-2 300–100 200 581 22 61 0.96 0.90 0.88
45OK-2 1300–500 900 662 22 84 0.97 0.89 0.86
450K-2 3000 3000 346 6 95 0.98 0.78 0.77
45OK-2 4000–3000 3500 506 8 117 0.98 0.81 0.80
45OK-250 2800–1000 1900 158 4 16 0.98 0.91 0.89
45OK-250 3800–2800 3300 492 37 340 0.93 0.59 0.57
45OK-250 5000–3800 4400 136 5 20 0.96 0.87 0.84
45OK-4 2000–200 1100 91 6 48 0.94 0.65 0.63
45OK-4 3200–2000 2600 744 16 1108 0.98 0.40 0.40
45OK-4 5000–3200 4100 72 2 60 0.97 0.55 0.54
45OK-258 800–100 400 1872 41 27 0.98 0.99 0.96
45OK-258 3600–2400 3000 2030 170 88 0.92 0.96 0.89
45DO-242 4000–2000 3000 537 46 42 0.92 0.93 0.86
45OK-11 4000–2000 3000 283 94 201 0.75 0.58 0.49
45DO-214 1200–1000 1100 228 27 264 0.89 0.46 0.44
45DO-211 3500–2000 2700 41 25 1026 0.62 0.04 0.04
45OK-197 800–100 400 201 52 110 0.79 0.65 0.55
45OK-197 1100–800 900 157 20 47 0.89 0.77 0.70
45OK-197 1500–1200 1300 456 19 30 0.96 0.94 0.90
45OK-197 1950–1600 1700 354 3 1 0.99 1.00 0.99
45DO-189 2950 2900 273 8 609 0.97 0.31 0.31
aData from five sites and 13 assemblages.
R.L. Lyman / Journal of Archaeological Science 30 (2003) 595–610 599
(Salmonidae), several species of cyprinid (Cyprinidae),
and several species of catastomids (Catastomidae). Most
fish bones were identified only to family. Given that the
genus and species of fish is not central to the issue of
concern here, and that Martin and Szuter’s [44,45]
hypothesis did not include expectations with respect to
which fish species would have been more intensively
exploited, finer taxonomic distinctions are unnecessary.
The greater the abundance of fish relative to
artiodactyls, the lower the fish AI.
Because the response to depressed abundances of
artiodactyls could have involved an intensification of the
exploitation of both small mammals and fish, I also
determined what I call the nonartiodactyl AI, calculated
as

artiodactyl NISP/(
artiodactyl NISP

small mammal NISP
fish NISP)
If artiodactyl abundances were progressively more
depressed over time—irrespective of why they were
depressed—then small mammals, fish, or both should
increase in abundance as human foragers attempted to
maintain their food intake levels. Again, low values of
the nonartiodactyl AI suggest that artiodactyls were of
low availability whereas high AI values indicate artiodactyls
were of high availability. Technological change,
environmental change, change in the human-population
level, or some combination of these factors might be the
causal variable driving any perceived changes in the AIs,
and although I call on these variables in the following,
they are of less concern to the discussion than the effects
of time averaging and of space averaging.
It is possible in the geographic area and temporal
period of concern that intensification of the exploitation
of river mussels (Pelecypoda) might have compensated
in part for depressed ungulate populations (Refs. [39,40]
and references therein), but shellfish remains were not
consistently identified among the 31 assemblages.
Further, so few bird remains were recovered that it is
apparent that exploitation of this taxon would not have
made any significant difference in measures of human
subsistence. I follow analytical tradition and use the
temporal midpoint of an AI’s age for graphing and
statistical purposes, and plot a simple best-fit regression
line to identify trends in prey abundances graphed in
scatterplots. I do not reproduce all scatterplots here, but
instead show only representative ones. Similarly, I do
not always discuss all possible indices.
4. Results
To examine the influence of spatio-temporal lumping
of assemblages, I first calculated the three AIs for all 31
assemblages. Here, NASM = 31, NCTXspace = 1, and
NCTXtime = 30; Fig. 1 shows the scatterplot for the
nonartiodactyl index. All three indices tend to increase
over time—the slopes of the best-fit regression lines are
positive—suggesting that the overall trend was for the
abundance of artiodactyls to increase relative to small
mammals (r0.15), fish (r0.17), or small mammals
+ fish (r0.21). The slopes are not steep, suggesting that
change in the structure of the exploited resource pool
was gradual and not very marked. The correlation
coefficients are, however, low and insignificant (P>0.25
in all), indicating that the conclusion that there was in
fact change in the exploited resource pool is statistically
unwarranted. What happens when assemblages are
lumped into a limited number of temporal periods?
4.1. Time averaging
To address the preceding question, the temporal
midpoint of each assemblage was used as its ‘age’ to
assign it to one of five time periods of 1000 radiocarbon
years duration each beginning with the period 0–1000
BP. (All included assemblages are in fact R100-year
old.) This reduces NCTXtime from 30 to 5. Values of
the nonartiodactyl AI were plotted against their time
period. The resulting scatterplot (Fig. 2) is a bit less
noisy than that produced when the midpoint age of each
assemblage is used (Fig. 1) because the ages have been
‘averaged’ to fall within a particular 1000-year period.
Not surprisingly, given the reduced variation in the
independent variable, the correlation coefficient
increases slightly, although it is still insignificant
(r0.23, P>0.1). The midpoint age was then used to
assign each assemblage to one of three cultural phases
recognized in the area [17]. These phases and their
temporal durations are Coyote Creek (0–2000 BP),
Hudnut (2000–4000 BP), and Kartar (4000–7000 BP).
This reduces NCTXtime from 30, or 5, to 3. Once
assigned to a phase, the nonartiodactyl AI for each
assemblage was plotted against the temporal midpoint
of that phase. Because all Kartar samples date between
4000 and 5000 BP, the age midpoint of 4500 BP was
retained. The resulting scatterplot (Fig. 3) is again just
what might be expected. Because temporal variation has
been removed by ‘time averaging’, the scatterplot is
cleaner visually and thus perhaps more easily interpreted
subjectively. As well, the correlation coefficient increases
a bit more, although it remains insignificant (r0.28,
P>0.1).
Although clarity remains imperfect, successively
greater time averaging affected by reducing NTCXtime
such as that represented by the progression of Figs. 1–3
seems to produce a clearer result. And, the statistical
correlation also, not unexpectedly, improves. Of course
the price paid is a loss of temporal resolution; that is,
there is a loss of the ability to detect variation across
time and also within a particular temporal span. This
600 R.L. Lyman / Journal of Archaeological Science 30 (2003) 595–610
begs the question of why the analyst might time average
in the first place. The answer is simply that time averaging
such as that represented in Fig. 3 relative to Fig. 1
might be said to be warranted by the fact that cultural
phases are often construed as periods of cultural stability
(e.g. Ref. [56]; see also Ref. [43] and references
therein), despite the fact that it has long been known
that this is seldom the case in reality [49]. Perhaps the
most important thing to note with respect to Figs. 1–3 is
that all of them involve plotting an AI value against a
single age assigned by determination of a temporal
midpoint, whether of the assemblage (Fig. 1), of an
arbitrary temporal period (Fig. 2), or of a cultural
period (Fig. 3). The differences apparent in those three
figures should give us pause when we use the temporal
midpoint of an assemblage as the plotting algorithm.
When we use temporal midpoints, we are using an
estimate of the ‘average’ age of the assemblage. Two
problems can be identified: the estimate may be to some
unknowable degree off the mark, and we have converted
a duration or period of time to a single year. The former
is an archaeological problem; the latter is a statistical
problem. Do similar problems attend the averaging of
variation in geographic space?
4.2. Space averaging
As noted earlier, the area from which the 31 assemblages
listed in Table 1 were recovered comprises a
stretch of the Columbia River valley. All assemblages
were collected from sites adjacent to the river. This
stretch of river cannot be subdivided into multiple
physiographically distinct areas on the basis of significant
vegetation differences from one portion of this
stretch of river to the next [25,27]. However, one might
divide the stretch into two subareas on the basis of the
fact that the western half of this stretch of river valley
today receives less than 25 cm mean annual precipitation
whereas the eastern half receives over 25 cm mean
annual precipitation [26]. In addition and corresponding
to this climatic difference is the fact that the western half
includes the mouth of the Okanogan River; the eastern
half includes only the mouth of the much smaller
Nespelem River. Salmon could and did make spawning
Fig. 1. Nonartiodactyl-index values for 31 assemblages plotted against ages (NASM = 31, NCTXspace = 1, NCTXtime = 30). Pearson’s r0.21 for
best-fit regression line.
R.L. Lyman / Journal of Archaeological Science 30 (2003) 595–610 601
runs up the Okanogan as well as the Columbia, but
could not proceed more than 1 km up the Nespelem due
to an impassable falls. Finally, topographic maps indicate
that the western third of the stretch of Columbia
River valley comprises broad expanses of level terrain
only a few meters above the river whereas the eastern
two-thirds of the valley are more deeply incised and level
terrain is significantly farther above the active floodplain.
Thus, topographic differences may have contributed
to geographic variation in human predation
behaviors (and perhaps prey—particularly salmonid—
availability) between the western and eastern portions of
the stretch of Columbia River that produced the 31
assemblages under consideration here.
Seven assemblages were collected from sites in the
western half of the valley and the remaining 24 assemblages
from the eastern half (Table 1). Bivariate scatterplots
for each areally designated set of assemblages
(NCTXspace = 2) were generated. The tendency is for
the western assemblages to have more small mammals
and more fish relative to artiodactyls than the eastern
assemblages (Fig. 4). Such is not noticeable if NCTXspace
= 1, as in Figs. 1–3. With spatial variation
included, the correlation coefficients are a bit stronger
and the slopes of the simple best-fit regression lines are a
bit steeper for the western assemblages (r0.09 for the
small-mammal AI, r0.11 for the fish AI) than for the
eastern assemblages (r0.30 for the small-mammal AI,
r0.20 for the fish AI), suggesting artiodactyls became
more important (relative to small mammals and fish)
over time more rapidly in the west than in the east,
although none of these differences are statistically significant
(P>0.05). The important point here is that by
considering all 31 assemblages simultaneously without
regard for geographic variation (NCTXspace = 1), as in
Figs. 1–3, the differences between the eastern and western
assemblages are undetectable because in effect the
geographic signal has been muted by spatial averaging.
4.3. Simultaneous time averaging and space averaging
Thus far, index values for individual assemblages
have been lumped by plotting them as groups according
to their particular temporal or spatial coordinates. That
Fig. 2. Nonartiodactyl-index values for 31 assemblages plotted against 1000-year period (NASM = 31, NCTXspace = 1, NCTXtime = 5). Pearson’s
r0.23 for best-fit regression line.
602 R.L. Lyman / Journal of Archaeological Science 30 (2003) 595–610
is, NCTX has been reduced but not NASM. What
happens if the NISP values for various assemblages are
combined on the basis of spatio-temporal propinquity,
and the indices recalculated, such that NASM is
reduced? Answer this question requires that the NISP
values per assemblage for each taxonomic category be
summed if they are not only similar in geographic
location, but also similar in age. First, assemblages that
fell in the same 1000-year period were lumped together
by western or eastern provenience as indicated in
Table 1 (NCTXspace = 2, NCTXtime = 5 for each
subarea, NASM = 5 for each subarea). The fish index
and the nonartiodactyl index were then recalculated for
each of the two sets of spatio-temporally lumped data
(Table 2) and the index values were plotted against the
temporal midpoint of each 1000-year period. The resulting
scatterplots (Figs. 5 and 6) are based on 5 points and
are cleaner and not as noisy as when all 31 data points
are plotted.
The scatterplots suggest that at about 2500 BP, things
were different in the two areas. In particular, whether
one considers the fish index or the nonartiodactyl index,
and whether one considers the west area (Fig. 5) or the
east area (Fig. 6), artiodactyls were exploited the least
intensively between about 3000 and 2000 BP, and this
trend seems to have begun about 4500 BP or earlier. One
might be tempted to argue that this decrease was in
response to a relatively cool–moist climatic interval
between about 4000 and 2200 BP [22,23]. Computer
simulations of anadromous fish runs in light of models
of climate and river runoff [24,46] suggest fish would
have been more abundant in the Columbia River during
this cool–moist interval than they were 4500 or 1500
years ago. This, however, leaves unexplained the marked
difference between the two areas between 1000 and 0 BP.
At that time, there seems to have been decreased exploitation
of ungulates in the west area (Fig. 5), but not the
east area (Fig. 6). There is no obvious technological
Fig. 3. Nonartiodactyl-index values for 31 assemblages plotted against cultural period (NASM = 31, NCTXspace = 1, NCTXtime = 3). Pearson’s
r0.28 for best-fit regression line.
R.L. Lyman / Journal of Archaeological Science 30 (2003) 595–610 603
difference between the two areas in terms of artifact
inventories, so I am tempted to suggest that the geographic
difference is the result of fish being more available
(or perhaps more technologically accessible) in the
western area where the Okanogan River is found.
Whatever the explanation might be, the influence of
spatio-temporal lumping can be shown by comparing
the results in Figs. 5 and 6 with the results of lumping all
assemblages irrespective of geographic provenience.
Now, NCTXspace = 1, NCTXtime = 5, and NASM = 5.
Fig. 4. Nonartiodactyl-index values for 24 assemblages (filled symbols) from the eastern area plotted against their ages (NASM = 24, NCTXspace
=1, NCTXtime = 23). Pearson’s r0.15 for upper best-fit regression line. Nonartiodactyl-index values for seven assemblages (open symbols)
from the western area plotted against their ages (NASM = 7, NCTXspace = 1, NCTXtime = 7). Pearson’s r0.38 for lower best-fit regression
line.
Table 2
Zooarchaeological data from north-central Washington state by area and 1000-year temporal period (derived from Table 1)
Location Age Age midpoint Artiodactyl NISP Small mammal NISP Fish NISP Fish AI Nonartiodactyl AI
Western 5000–4001 4500 292 317 370 0.44 0.30
4000–3001 3500 12 51 10 0.55 0.16
3000–2001 2500 88 82 4812 0.02 0.02
2000–1001 1500 64 52 3 0.96 0.54
1000–0 500 83 34 7243 0.01 0.01
Eastern 5000–001 4500 311 29 80 0.80 0.74
4000–3001 3500 998 45 457 0.69 0.67
3000–2001 2500 4639 351 2147 0.68 0.65
2000–1001 1500 1454 60 364 0.80 0.77
1000–0 500 3473 157 329 0.91 0.88
604 R.L. Lyman / Journal of Archaeological Science 30 (2003) 595–610
The small-mammal index (Fig. 7) suggests artiodactyls
were not very important relative to small mammals at
the end of the mid-Holocene climatic interval generally
referred to as the Altithermal. The Altithermal period
ended about 5000 BP and is locally characterized as
comprising warmer, wetter winters than at present;
stream flow was reduced as were salmon runs, and
grasslands were replaced by shrub–steppe habitats [23].
These environmental differences suggest a possible cause
of apparently low artiodactyl abundances at about 4500
BP. The nonartiodactyl index (Fig. 7), which is very
similar to the fish index, suggests something quite different.
Similar to Figs. 5 and 6, the lumped data suggest
much fluctuation in the abundance of artiodactyls over
the last 5000 years (Fig. 7). As indicated in the preceding
paragraph, explanations might comprise elements of
environmental fluctuation. They might also include evidence
of fluctuation in the size of the human population
[22] and changes in fishing technology such as increased
use of nets [1]. Whatever the case, the explanation would
be complex.
Explanatory complexity largely disappears if the
index values are calculated on the basis of cultural
period and irrespective of geographic position. Here,
NCTXspace = 1, NCTXtime = 3, and NASM = 3. The
plot of the small-mammal index per cultural period
indicates that artiodactyls were not frequently exploited
relative to small mammals at the end of the Altithermal,
but subsequent to that point small mammals were
significantly overshadowed by artiodactyls (Fig. 8). This
trend aligns nicely with the standard conception of the
Altithermal and subsequent climates and habitats and
human adaptations [5,38]; small mammals were more
economically important during the Altithermal than
subsequently. Similarly, the fish index also aligns with
standard explanations [13,62] that the exploitation of
fish intensified after the Altithermal as a result of
increased run sizes and improved technology (Fig. 8).
These results align with inferences produced by
archaeologists who were working 35 years ago and were
interested in region-wide economic trends, a point I will
return to subsequently. Of more concern here is the
Fig. 5. Fish-index (open squares) and nonartiodactyl-index values (x) for seven summed assemblages from the western area plotted against 1000-year
period (NASM = 5, NCTXspace = 1, NCTXtime = 5). Upper best-fit regression line is for fish index (r0.18); lower best-fit regression line is for
nonartiodactyl index (r0.14).
R.L. Lyman / Journal of Archaeological Science 30 (2003) 595–610 605
following simple fact: interpreting a bivariate scatterplot
of many points representing relatively precise
spatio-temporal data (Fig. 1) is a rather different matter
than interpreting a bivariate scatterplot of few points
representing the same data in spatio-temporally lumped
form (Fig. 8).
5. Discussion
The general significance of the example discussed
above for applications of models derived from foraging
theory (or any explanatory theory) to zooarchaeological
data should be clear. On the one hand, the
manner in which zooarchaeological samples are
lumped by space or by time—whether taphonomically
or analytically—may reveal regionwide trends precisely
because that lumping masks spatio-temporal variation
within the region. Lumping data without consideration
of the included spatio-temporal variation, on the
other hand, could mask what may otherwise comprise
significant differences in resource exploitation
patterns.
Notice that I said “may comprise significant differences.”
Does spatio-temporal lumping in fact mask
significant differences? That depends on the spatiotemporal
scale of the question being asked, and this is
the most critical point of this discussion. Paleobiologists
who have examined time averaging argue that it only
occurs when the process or property one seeks to
measure occurs or was created, respectively, over a
shorter time span than is represented by the fossil record
under study [28,36,47,57]. Thus, time averaging is a
problem when two or more analytically interesting
events of distinct ages appear to be contemporaneous. A
similar argument is made for spatial averaging. If the
property of analytical interest is a biological community’s
structure in terms of taxonomic richness,
evenness, or diversity, and if the fossil record under
study comprises organisms from multiple communities,
Fig. 6. Fish-index (open squares) and nonartiodactyl-index values (x) for 24 summed assemblages from the eastern area plotted against 1000-year
period (NASM = 5, NCTXspace = 1, NCTXtime = 5). Upper best-fit regression line is for fish index (r0.55); lower best-fit regression line is for
nonartiodactyl index (r0.66).
606 R.L. Lyman / Journal of Archaeological Science 30 (2003) 595–610
then that record is spatially averaged. The critical issue,
as paleobiologists note, is one of ascertaining the spatiotemporal
scale of the fossil record and asking questions
of the same scale.
In terms of applying explanatory models derived
from foraging theory to the zooarchaeological record,
the critical issue is the same as that in paleobiology.
Asking if prehistoric hunters depressed prey populations
by exploiting them is so general as to have no spatiotemporal
coordinates. Adding the phrase “in area X
during cultural periods A, B, and C” to the question
provides coordinates. The zooarchaeologist must,
however, determine if the assemblages included in the
analysis fall within those coordinates and thus not be
spatio-temporally averaged. To make this determination
demands that explicit spatio-temporal parameters be
built into the research questions we ask. If they are
not, then zooarchaeologists will be open to criticisms
that their analyses mask potentially significant
spatio-temporal variation at finer scales.
6. Conclusions
The observation of Grayson and Delpech [31] that
the application of models derived from foraging theory
requires the translation of ecological time into archaeological
time must be expanded to include the notion that
those applications must also translate ecological space
into archaeological space. Such translations have, to
date, not been accomplished by literally converting or
rewriting the ecological model in terms of archaeological
variables. Rather, the models have simply been applied
to archaeological data, and the latter interpreted in
terms of the former. The influence on interpretations of
the resultant muting of temporal and/or spatial variation
should be clear and is the central point of the example
Fig. 7. Small-mammal-index values (open squares) and nonartiodactyl-index values (x) for 31 summed assemblages plotted against 1000-year period
(NASM = 5, NCTXspace = 1, NCTXtime = 5). Upper best-fit regression line is for small-mammal index (r0.78); lower best-fit regression line is
for nonartiodactyl index (r0.10).
R.L. Lyman / Journal of Archaeological Science 30 (2003) 595–610 607
discussed here. However, lumping might be affected—
decreasing NASM by summing assemblages, ignoring
temporal variation by decreasing NCTXtime, ignoring
spatial variation by decreasing NCTXtime—it mutes,
smooths, and averages what might otherwise be significant
differences. This demands a decision regarding
what comprise significant differences. Such a decision
can be made at the initiation of analysis, but may have
to be modified once the degree of spatio-temporal averaging
of the assemblages at hand has been determined.
The lessons here, then, are two. We must first be
explicit about the values of the spatio-temporal variables
within the questions we ask of the zooarchaeological
record. What should be abundantly clear is that spatiotemporal
palimpsests are averaged, blended, or coarsegrained
relative to some preconceived and generally
implicit size of spatio-temporal unit [7]. And second, we
must be cognizant of the influences of spatial and
temporal averaging on our analytical results in lieu of
rewriting the ecological models we apply to the zooarchaeological
record in terms of the spatio-temporal
scale of that record.

THE PREHISTORY OF BRITAIN AND IRELAND pt.5

THE END OF PREHISTORY
FOREGROUND AND BACKGROUND
English common law is based on precedent. The results of any court case
establish a principle that must be followed in the future. In theory the legal
system should become increasingly rigid, but that is not what happens. Rather
than considering themselves bound by previous rulings, judges institute subtle
changes to the law. They do so by arguing that the details of particular cases are
different from one another, so that what once appeared to be a point of general
application actually applies in very specific circumstances (Figs. 5.1 and 5.2).
Prehistorians have taken a similar approach to the Three Age Model. This
was devised in the nineteenth century when archaeologists believed that
progress was based on technological innovation, so that stone gave way to
bronze and bronze to iron. Since that time they have retained the terminology
but have changed its connotations. These period divisions are not used in a
consistent way. The Neolithic, for instance, is defined by its material culture
but also by its economy and systems of belief. Recent work has tended to
play down the distinction between that phase and the beginning of the Bronze
Age. There are other problems as well. In this book two of the chapters begin
halfway through the periods into which the sequence was organised. This
procedure reflects the timing of fundamental changes in the prehistoric landscape.
Archaeologists rely on precedent when they use the vocabulary of the
Three Age system, but the distinguishing features of those different phases have
changed completely.
The Iron Age is no exception, but this is not always apparent from the way
in which it is studied. At the risk of oversimplification, there have been two
main approaches to the period. One treats it as a self-contained entity, giving
way to the Roman occupation of Britain. It can even emphasise a certain
continuity between those two phases. A good example is Barry Cunliffe’s Iron
Age Communities in Britain, now in its fourth edition (Cunliffe 2005). The other
approach envisages a longer sequence of change and interprets the distinctive features of the Iron Age as the outcome of processes which had started late in
the second millennium BC. That is illustrated by Kristian Kristiansen’s review
of Europe Before History (Kristiansen 1998). What is called the Bronze Age
‘background’ in one account is central to the interpretation put forward in the
other book.
That second approach is perhaps more satisfactory, and that is why this
chapter begins with the closing years of what can still be called the ‘Bronze
Age’. It will end as these islands first came into contact with the peoples of
the Classical world, since Martin Millett (1992) and John Creighton (2000,
2006) have shown that the Late Iron Age is better studied in relation to the
Roman period, at least in England. The earliest use of iron was not a sudden
event, nor did it depend on the technological progress favoured by Victorian
writers. In fact it was probably precipitated by a shortage of metals of any
kind. Indeed, it may be that the period around 800 BC is more aptly characterised
by a reduced supply of bronze than by the adoption of an unfamiliar
material.
Such developments are not well understood, but it seems as if copper of
increasingly poor quality had been extracted as the supply of the most suitable
Continental ores ran out (Sperber 1999). Some of the Alpine mines ceased
operation, and new ones took their place, but the situation did not improve significantly.
The metal supply came under yet more pressure and greater attempts
were made to recycle the material that was still in circulation. These difficulties
began about 1100 BC and became especially severe during the ninth century.
Other developments were taking place at about the same time. In the
Mediterranean, complex societieswere forming in Greece and Italy and needed
regular metal supplies of their own. Theywere already using iron when a period
of rapid social and economic change began around 800 BC. In Italy itself, the
rise of new elites put pressure on the supply of metalwork from the north, and
bronze seems to have been imported across the Alps (Kristiansen 1998: chapters
5 and 6). The artefacts of this period are very similar to those in Central
Europe. This would have reduced the supply to other regions.
These developments in the Mediterranean did not take place in isolation, for
the late eighth century BC also saw the establishment of a Phoenician trading
network extending as far west as the Iberian peninsula, where it exploited
established contacts along the Atlantic coast of Europe to obtain raw material
(Aubet 2001). This axis assumed a greater importance as the circulation of
metals became more difficult. The last metalwork of the Bronze Age was
obtained by a different route from most of the objects of that period and placed a
greater emphasis on the Atlantic coast (Ruiz-G´alvez Priego 1998). This change
of geographical alignment proved to be short lived, for soon the process was
curtailed, and the deposition of artefacts in hoards and rivers declined. In the
end the system collapsed. The clearest indication that established practices had
been abandoned is provided by the working of iron. These developments affected both Britain and Ireland and can be illustrated
in several different ways. There is evidence that bronze was being distributed in
standard forms so that some of the latest axeheadswere probably treated as ingots
(Briard 1965: chapter 13). They contained so much lead that they could hardly
have been used as tools, and in Brittany, where many of them had been made,
it seems as if they conformed to standard weights. Discarded artefacts were
increasingly recycled. This is most obvious from metal analysis, but it may also
be reflected by the large number of ‘scrap hoards’ that were deposited during
the ninth century BC. They certainly include fragments of many different
kinds of objects, including tools, ornaments, and weapons, and some of them
also contain traces of slag (Bradley 1988). Colin Burgess (1979) suggests that
people were dumping surplus artefacts; by withdrawing so much material from
circulation they might have forced its value to rise. The reasoning behind this
hypothesis is anachronistic for it suggests a degree of coordination which would
be possible only in the modern economy – in many ways the best comparison
is with the operation of the stock exchange. It is equally unsatisfactory to
suppose that supplies of bronze were concealed and lost during a political
crisis, for the main evidence of that crisis is provided by the hoards themselves.
That is a circular argument which leads to even more confusion. In fact it may
be wrong to interpret these collections entirely in practical terms. Perhaps they
were really offerings associated with the transformation of the raw material.
That is certainly consistent with ethnographic accounts of metalworking in
traditional societies (Budd and Taylor 1995).
One clue is provided by changes in the circulation of bronze artefacts which
emphasised the importance of the Atlantic seaways. It is not clear whether
there really was an ‘Atlantic Bronze Age’ ( Jorge ed. 1998), but no one would
deny that a growing number of metal items were exchanged or copied along
the western rim of Europe, from Scotland and Ireland in the north, to Spain
and Portugal in the south. Some of these connections even extended into
the Mediterranean (Ruiz-G´alvez Priego 1998). Such links had always been
important, especially in prehistoric Ireland, but now they seem to have assumed
a greater prominence than before, and ultimately this network reached across southern and eastern England into Northern Europe. The latest bronze hoards
seem to reflect the new alignment (R. Thomas 1989). Such evidence suggests
that the inhabitants of these islands were drawing on different sources of supply.
Any relief was short-lived, however, and towards the end of the Late Bronze
Age there are indications that people had started to experiment with iron
production, using ores that could probably be discovered locally; only the
techniques of working them were foreign. It is not clear quite when this process
began, or how far iron was intended to take the place of bronze, but iron slag
has been found in at least thirty contexts dating from the ninth century BC
(Lawson 2000: 202), and there is evidence from a recently excavated settlement
at Hartshill Quarry in the Kennet Valley that production may have begun even
earlier (Collard 2005). At the same time, the new technology could be used to
make composite tools in combination with bronze or could even be employed
to produce traditional forms of artefact in a different material. Thiswas certainly
the case with socketed axes (Manning and Saunders 1972).
The deposition of metalwork became an important issue as bronze was
supplemented, and to some extent replaced, by iron. The latest bronze hoards
date from the Llyn Fawr industrial phase which runs from approximately 800
to 600 BC, and the quantity of river metalwork decreased sharply at that
time. It happened throughout Britain and Ireland, but it can hardly suggest
a shortage of both kinds of metal since exactly the same pattern is found in
Continental Europe, where iron weapons were commonly deposited in graves
(Gerdsen 1976). Indeed, it seems to have been important that the votive deposits
should contain foreign material. That might apply to the styles of the objects
themselves but more often reflects the metal from which they were made.
Copper and tin did not occur in many of the areas with these artefacts, and
this may have provided the source of their power. In that case locally produced
iron would not have been an acceptable substitute. Although prehistorians treat
bronze and iron as if they were equivalent to one another, people in the past
may have categorised them in a very different way (Bradley 1998a: 150–4).
FROM POSSESSION TO DISPOSSESSION
Access to foreign metal had been of fundamental importance in Later Bronze
Age society, so that these changes would have had drastic consequences. The
crucial point was made by Gordon Childe more than sixty years ago (1942:
182–3). The production of bronze artefacts involved a combination of copper
and tin, both of which had restricted distributions in prehistoric Europe. One
problem lay in bringing these materials together, and another concerned the
circulation of the finished products. Some of the densest concentrations of Late
Bronze Age metalwork were in areas which lacked metal sources of their own.
They included southern and eastern England. There are other regions where
metalwork could have been produced locally but seems to have been imported. Among them are Ireland, Scotland, and, most probably, Wales. Because highquality
artefacts were being introduced from distant areas, it would have been
possible to control their circulation, and the same applies to the raw material
out of which they were made. Access to such objects could have been restricted in two ways. It involved a process of long-distance travel by sea which might
have been in the hands of a specialist group of traders, or of a social elite who
controlled their activities. The skills needed to make the finest objects were
not generally available, and again the activities of smiths may have come under
political control. Some of their work was so demanding that they must have
depended on a patron for support (Rowlands 1976). At the same time, it would
be necessary to accumulate suitable commodities with which to participate
in exchange. Although this process may well have involved the circulation
of marriage partners, it could also have included the movement of textiles,
hides, and, in Ireland, gold. For Childe, the long-distance movement of Bronze
involved the creation of alliances and was a source of power.
Iron, on the other hand, is widely available and could have been locally
produced in many parts of these islands. Unless the process took place on a
large scale – and that is not evidenced before the production of ‘currency bars’
in the third century BC (Hingley 2005) – it would be difficult to control.
Thus the virtual collapse of the long-distance circulation of bronze might have
undermined the influence of an elite, but the first adoption of ironworking
would not have offered an equivalent power base. For Childe, Iron Age society
became more ‘democratic’. It is possible to quarrel with the idea that the adoption
of iron improved the quality of life, but there is certainly some evidence
of social change.
This takes two main forms. There are the new developments that happened
during the period of transition, and there are signs that older practices were
abandoned. Both happened simultaneously between about 800 and 600 BC,
but it will make the argument clearer if they are treated separately here.
This account starts with the new developments. One of the most dramatic
discoveries of recent years is the identification of a series of large middens
in southern Britain whose chronology appears to span the last years of the
Bronze Age and the period of transition (Needham and Spence 1997). Ironically,
the first of these to be investigated, All Cannings Cross in Wessex, was
once treated as the type site for the Early Iron Age (Cunnington 1923). That
was not because of the structural evidence from the excavation, which was
meagre, but because of the extraordinary abundance of artefacts. The same
applies to the sites recognised more recently, whose distribution extends from
East Anglia to south Wales. They attract attention for several reasons. First, it
is most unusual to find such enormous accumulations of cultural material, for
in normal circumstances such deposits would have been removed from settlements
and spread on cultivated land; this is confirmed by excavation of the
eroded ploughsoil found on valley floors (Bell 1983). Indeed, the material was
not left where it had accumulated but was allowed to build up into considerable
mounds. That is quite exceptional. It happened at the recently published site
of Potterne on the chalk of southern England (Lawson 2000) and at another site, East Chisenbury, which was identified as a standing earthwork (McOmish
1996). Similar deposits have been recognised at some of the Late Bronze Age
sites discussed in Chapter Four, including the occupied islands at Wallingford
and Runnymede Bridge (Cromarty et al. 2005; Needham and Spence
1996).
The middens themselves have an unusual composition. The sediments
at Potterne include an extraordinary quantity of cattle dung, but there are
also considerable amounts of bronze metalwork. The upper levels contain a
small amount of iron slag (Lawson 2000: 166–73). There is an abundance of
fine pottery and animal bones and small quantities of unburnt human bone.
Runnymede Bridge has produced a rather similar assemblage, but here there
is evidence that some of the faunal remains were carefully organised within
the filling of the midden (Needham 1992). There are indications of craft production,
including bronze working, the working of antler, spinning, and the
production of textiles. In each case the sheer abundance of bones suggests that
feasts were taking place, and this is particularly likely at Runnymede Bridge
where the midden is located on an island in the River Thames (Needham
and Spence 1995). Detailed study of the ceramics from Potterne suggests that
the people who used that site had a wide range of contacts with other places
(Lawson 2000: 166–73). The same is true of the metalwork from Runnymede
Bridge.
Lastly, not all these sites show much evidence of structures contemporary
with the accumulation of the midden. There is considerable uncertainty, for
pits and postholes are difficult, if not impossible, to recognise within the dark
soil of these deposits; the same problem affects research in the late levels of
Roman towns in England (Courty, Goldberg, and Macphail 1989: chapter 15).
It is certainly true that a range of timber buildings was identified at Runnymede
Bridge and that the midden at Potterne overlay a Bronze Age occupation site.
Even so, the only structural evidence found within some of these accumulations
consists of small patches of cobbling, often rectangular in outline, various
hearths, and the remains of ovens. It may be that in their final form such places
were not settlements at all. They are readily accessible from the surrounding
area, and it is tempting to compare them with the Middle Saxon ‘productive
sites’ which raise many of the same problems (Pestell and Ulmschneider eds.
2003). These were often located in places that were easy to reach beside Roman
roads, and they eventually became local centres, but their main characteristic
is the exceptionally large collections of metalwork which are found together
with evidence of craft production. Again their roles are very difficult to define.
Some may have been high-status settlements, whilst others could have been
the sites of markets or fairs. Given the unusual character of the excavated evidence,
the same ideas provide the most plausible explanation of the prehistoric
middens. Occasionally they can be associated with other features. The midden at East
Chisenbury overlay one side of an earthwork enclosure (Fig. 5.3; McOmish
1996). A similar deposit at Wittenhem Clumps in the Upper Thames was
just outside an early hillfort notable for its deposits of human remains (T.
Allen pers. comm.), and at Balksbury on the chalk ofWessex another midden,
which included large amounts of cattle dung, was within a defended enclosure
(Ellis and Rawlings 2001: 83). Again there was perhaps some evidence of
feasting.
Balksbury is one of a group of earthwork enclosures which were built in
southern England during this period. They are closely related to the Late
Bronze Age hillforts discussed in Chapter Four, but some of them may be
slightly later in date and are usually assigned to an Earliest Iron Age that extends
from about 800 to 600 BC. They share certain features in common (Cunliffe
2005: 378–83). They are of considerable extent and often occupy the tops of
particularly prominent hills. They are enclosed by surprisingly slight earthworks,
and excavations inside them suggest that they were not intensively
occupied. They provide evidence for a very limited number of dwellings, but
the main structures were small square buildings which are usually interpreted as
granaries or storehouses. They do not produce many artefacts. In the absence
of more detailed information, it has been suggested that these places were used
intermittently and perhaps on a seasonal basis. In that respect they have features
in common with the midden sites.
In each case the evidence suggests that certain places were serving as focal
points for a wider population in a way that had not happened since the Earlier
Bronze Age. Indeed, they could have played a part in public events at which
feasting was particularly important. In the early first millennium BC there was
probably an added element, for these seem to have been places where people
congregated for the purpose of production and exchange. The great accumulations
of manure might not be there because large numbers of animals were
collected for slaughter; surely there were also gatherings at which livestock
were changing hands. Similarly, there is evidence that artefacts were being
made, including fine metalwork. This is especially interesting as it had normally
taken place at more secluded locations, including Irish crannogs, Welsh
hillforts, and some of the English ringworks, where it might have been easier to
exercise control over its circulation. The new sites may have been places where
people transacted communal business, but if so, this was a novel development
and one which suggests that political organisation was changing. Perhaps the
community was assuming greater authority at the expense of a weakened elite.
If these were new developments, other features of the landscape seem to have
gone out of use. One characteristic of the Late Bronze Age landscape was the
evidence of fortified ringworks (Needham and Ambers 1994). In most cases
activity ended during the Bronze Age/Iron Age transition, and the same applies to the building of Irish crannogs (Fredengren 2002). That may be significant
as they have similar contents and both may have been of high status. Just as the
deposition of fine metalwork in water seems to have diminished, these locations
lost their special significance, and the same is true of most of the platforms and
islands discussed in Chapter Four. It is a trend that applies to other sites in the
English landscape, for the co-axial field systems which were such a feature of
the Later Bronze Age were largely abandoned during this phase (Yates 2001).
The significance of this change will be considered in due course; what matters
here is that the associated settlements were normally deserted or moved to new
positions at this time.


HOUSES AND ENCLOSURES
There are other signs of dislocation in the settlement pattern. Two features are
particularly important.
The first is a gradual shift in the distribution of prehistoric activity. During
the Later Bronze Age there was an extraordinary density of occupation sites
along some of the rivers discharging into the North Sea. That is where many
of the field systems were created and accounts for the siting of most of the
ringworks. Perhaps more important, it is in this area that the principal deposits
of fine metalwork occur: there are weapons and ornaments in the Fenland,
and more artefacts are found in the Thames. The previous chapter suggested
that these features were directly related to one another: elites based in the
ringworks and perhaps at other settlements controlled the flow of metalwork
and engaged in long-distance exchange. They were able to do so because of
the surplus provided by cattle raising and crop cultivation, and it even seems
possible that these people were commemorated in death by the bronze objects
deposited in water.
During the Early Iron Age, however, the distribution of weapon deposits
contracted until it was practically confined to the Middle Thames ( Jope 1961).
There is much less evidence for occupation sites in the areas that had played
such a prominent role before, and instead there are more signs of activity in
other regions. In the south, these included the midlands, the upper reaches of
the Thames, and the southwestern peninsula. In the north, there was a similar
increase in settlement sites, especially in the uplands on either side of the
modern border between Scotland and England. This may also have happened
inWales, although the evidence is limited, but in Ireland, Iron Age settlements
of any kind are very difficult to find. That is partly because pottery was not
used during this period (Raftery 1995). The best chance way of discovering
the ‘missing’ sites is by dating the ore roasting pits and grain drying kilns which
are commonly found in fieldwork. A few of these date from this period. On
the other hand, there was probably a reduction in the cultivated area or in the
intensity with which the land was exploited. That is certainly suggested by
pollen analysis (Weir 1995).
On one level the new settlements took distinctly regional forms, just as the
ceramics of this period can be divided into a series of mutually exclusive style
zones (Cunliffe 2005: 90–7). To some extent these contrasts may also be deceptive
as Chapter Four showed that Late Bronze Age settlements are difficult to
identify except by large scale excavation. That is because so many of them were
open sites and may be hard to recognise by other means. That is particularly
true outside the distribution of regular field systems. Iron Age settlements, on
the other hand, were sometimes enclosed by a palisade or more often by a bank
and ditch (R. Thomas 1997). That was certainly the case in England, Scotland,
and Wales, and it may have happened in Ireland, where aerial photography is revealing a variety of crop mark enclosures which do not conform to the
standard types of field monuments (G. Barrett 2002). Despite some promising
indications from rescue archaeology, they remain largely undated.
Roger Thomas (1997) has offered an interesting discussion of this phenomenon.
He makes two important points. The process of enclosure began
at about the same time as the decline in the supply of bronze entering these
islands from the Continent. The new sites were usually associated with agricultural
production, and there is every reason to suppose that Iron Age activity
had an even greater impact on the landscape than that of the previous period.
Excavations at these enclosures support this interpretation. They provide evidence
for stock raising and cereal growing on a substantial scale and include
a whole range of agricultural facilities within their area, in particular the sites
of raised granaries or storehouses, and silos for keeping grain over the winter.
Such sites can produce large collections of carbonised cereals, and faunal
remains, almost entirely of domesticates (Fowler 1983). Thomas suggests that
with the decline in the circulation of prestigious metalwork peoplewere placing
a greater emphasis on food production. Such evidence is so widely distributed
that it is hard to disagree.
At the same time the building of boundaries may have had a further significance.
Like their Late Bronze Age predecessors, individual houses seem to have
been subdivided on the axis of the porch, and particular activities took place
at specific locations within the building (Fitzpatrick 1997). The landscape was
also divided according to some simple conventions, but in this case it was occupied
by a number of communities who seem to have emphasised the differences
between them by monumentalising the outer limits of their settlements. To
an increasing extent they also imbued those boundaries with a special significance
by the deposition of human bones and other items. This is a special
feature of the enclosure ditch and the area around the entrance (Hill 1995:
chapter 8). Few of these enclosures could be defended against attack. Roger
Thomas (1997) suggests that the inhabitants of these places were increasingly
self-sufficient and that this might even be related to concerns over succession
and the inheritance of land from one generation to the next. It is certainly true
that, once founded, many of these settlements were occupied and rebuilt over
a considerable period of time.
The forms of these enclosures have attracted less attention than their
chronology, and yet they support a similar interpretation (Fig. 5.4). The dominant
feature of the Late Bronze Age ringworks had been large round houses
like those at Mucking, Springfield Lyons, and Thwing. These were generally
located within a circular enclosure whose defences could be built on an extravagant
scale. Such buildings were generally aligned on the gateway, but could be
separated from it by a screen. Like the henge monuments of the Later Neolithic
period, the entire structure gives the impression of one enormous house, and
this is even more obvious at a transitional site like West Harling, where the earthwork perimeter abutted the outer edge of such a building (Clark and Fell
1953).
Houses of similar size occur widely during the Early Iron Age, which runs
from about 600 to 400 BC, but they are comparatively rare after that time.
That is not to say that this was the standard form of dwelling. Rather, like the
Late Bronze Age buildings that they resemble in so many respects, they were
the largest examples in a wide range of timber structures. Their associations changed, too. Ringworks with their timbered ramparts were no longer built
during the Early Iron Age, but large circular buildings could still be the dominant
feature inside the new earthwork enclosures. Others occurred in some
of the early hillforts discussed in a later section of this chapter.
What is less often considered is the importance of the circular or subcircular
ground plan. This is not confined to the major houses but quite often extends
to the earthwork perimeter as well. Again it is a format that applies to several
different classes of monument: to palisaded enclosures, earthwork enclosures,
and even to the organisation of many hillforts. It is by no means ubiquitous,
but at its simplest it has two distinct elements. The perimeter is generally
curvilinear, and it is frequently breached by an entrance to the east or south (Hill
1995: chapter 8). In both respects it resembles the plan of a round house. There
can be some elaborations of this format. A second entrance may be provided on
the opposite side of the circuit, but even this feature is shared with a number
of large timber buildings, especially in northern Britain (D. Harding 2004:
fig. 2.6). It is often argued that such buildings have their doorways to the south
or east to allow the morning light to illuminate the interior, but the relationship
between these houses and the position of the sun may have a cosmological
significance as well, for such a practical argument can hardly explain why
the same principle extended to the gateway to the settlement (Oswald 1997).
Not all the enclosures adopted this particular layout, nor were all settlements
defined by a ditch or a palisade, but circular enclosures are far too common
to have developed fortuitously. Others were roughly square or rectangular, but
these variations usually follow regional lines. Thus circular enclosures are common
inWessex, or southwest England, for instance; rectilinear compounds are
more often found in the east midlands and the Welsh Marches. One reason
for these distinctions concerns the overall pattern of settlement. Like roundhouses,
circular enclosures are usually set apart from one another. Rectilinear
enclosures, however, can easily be joined together. Although many of them maintain their isolation, sometimes they are found in groups, as also happens
in the Iron Age of northwest France (Arbousse Bastide 2000).
If there were practical advantages to a rectangular ground plan, why did so
many communities in Britain choose to build circular enclosures? Was it to
evoke connections with domestic architecture (Fig. 5.5)? The word ‘house’
usually refers to a residential building, but it can be used as a metaphor as
well (Helms 2004). Thus the royal family in Britain is known as the House of
Windsor. In this sense it relates not just to the place where they live but also
to the social bonds that exist between them and distinguish them from other
people. The term refers both to the members of an extended family and to a
line of descent. One reason why the domestic building may have been such an
important symbol during the Early Iron Age was that it stood for the integrity
and independence of different communities. Perhaps that was less important in
those regions where other kinds of enclosure were built. The question requires
more research.
It is impossible to discuss the Irish evidence, but in most parts of Britain it is
clear that large domestic buildings were gradually replaced by smaller circular
structures. These are a particular feature of the Middle Iron Age, which ran
from about 400 to 150 BC. For the most part they followed the same conventions
as the earlier houses, but now it seem as if there were few, if any, significant
differences of size between the individual dwellings. Indeed, they present such
a uniform appearance that it seems almost as if any overt distinctions between
them were suppressed. Like the buildings of the Late Bronze Age and Early Iron
Age, they were normally replaced in exactly the same positions. Where this
did not happen, successive structures overlapped, implying a set of conventions
that were respected over the generations. Important thresholds within them
could be marked by offerings of artefacts, animal bones, or even human remains
(Bradley 2005b: 50–7). Again there seems to have been an emphasis on the continuity
of the domestic group of the kind suggested by Roger Thomas (1997).
At the same time, certain of the enclosures were reconstructed on an increasing
scale. It is possible that the social distinctions that had been signified by
domestic buildings now applied to the settlement as a whole. Perhaps the main
differences of status were between the inhabitants of different sites and were
expressed by the size and elaboration of the earthworks that enclosed them.
THE ORGANISATION OF THE LAND
Iron Age farming has always provided a major topic for research. Food remains
are well preserved and many settlement sites have seen large-scale excavations.
Surviving earthworks in England, Scotland, andWales have all been the subject
of field survey, and even the agricultural implements of this period have been
studied in detail. The results of this endeavour have been tested by a series
of experiments. These have been concerned not only with the practicalities of constructing Iron Age buildings but with raising livestock and cultivating
crops. The Butser Ancient Farm on the chalk of southern England has been
claimed as a working model of the prehistoric landscape (P. Reynolds 1979),
and similar recreations of life in this period have provided the material for
television programmes (Percival 1980).
As a result it might seem easy to characterise the archetypal Iron Age landscape,
and exactly this assumption lay behind a series of important excavations
on Fyfield and Overton Downs on theWessex chalk (Fowler 2000). The results
of that work were published recently and provide some indication of how far
knowledge of this subject has changed. When the project was devised in the
1960s, it seemed important to investigate a series of Iron Age settlements, land
boundaries and ‘Celtic’ fields which were located not far from a major hillfort.
When the results of this project became available, those elements proved to
be less closely related to one another than had originally been expected. The
field systems which seemed to unite these different features actually dated from
the Late Bronze Age and, like many others, had gone out of use in the Early
Iron Age. One of the settlements was certainly integrated into this system of
boundaries, but by the time it was established few of the plots were being used.
Instead an open settlement was fitted into the surviving earthworks. Although
it was accompanied by a small area of arable land, the system as a whole was
derelict and may have been used as pasture. The project did not investigate the
hillfort, Barbury Castle, but comparison with similar sites in the same region
suggests that it would probably have included the remains of houses, granaries,
and storage pits; this is indicated by its earthworks and by geophysical survey
(Bowden 2005). Such monuments were closely integrated into the process of
crop production, but may not have been linked with the use of regular field
systems. How, then, was Iron Age agriculture organised?
Recent work by David Yates (2001) has shown that co-axial fields were
established in lowland England during the Bronze Age. Chapter Four argued
that they were first created on Dartmoor towards the end of the Earlier Bronze
Age and that the same form of land organisation was adopted widely on the
chalk and the river gravels of southern England during the later part of that
period. Not all the systems initiated during the Middle Bronze Age remained
in use for long, and new ones were certainly established in the Late Bronze
Age. What is striking is that their distribution was limited to the region with
the greatest density of metalwork deposits. It was also related to the areas with
ringworks. Except in the Upper Thames, these systems went out of use during,
or soon after, the transition to the Early Iron Age, and many of the associated
settlements were abandoned. A second group of co-axial field systems was
created towards the end of the Iron Age and covered a much larger area. That
process continued into the Roman period and is not considered here (Fulford
1992). Not only were some of the relict field systems reused at that time, new
ones were created. There is little sign of a similar development in Ireland. In fact those changes should not have seemed so troubling, for some of
the best recorded prehistoric landscapes are without surviving fields of any
kind, even though there is evidence that crops were being grown. Perhaps the
most obvious example is provided by the YorkshireWolds (Stoertz 1997). The
problem needs to be approached in another way.
Certain points are clear. Environmental evidence shows that, unlike the
situation in Ireland, the British landscape was largely open during the Iron
Age and that it was being exploited on a large scale. There are signs of soil
erosion caused by cultivation and excessive grazing, and carbonised cereals
are virtually ubiquitous on excavated sites. So are the remains of domesticated
animals (Fowler 1983). Many settlements in the south contain grain storage pits
which sometimes preserve traces of their original contents, and where these
do not occur there may be the foundations of timber granaries. The problem
is not whether Iron Age people engaged in intensive mixed farming: it is how
that activity was carried out on the ground. If Celtic fields went out of use,
what does this imply for the nature of Iron Age society? If the older boundaries
were no longer important, how was farming organised? That discussion must
take in other elements: settlements, boundaries, and hillforts.
Chapter Four discussed the origin of the linear earthworks which divided
up large tracts of the later prehistoric landscape. Many were poorly dated, but
they seem to have originated during the early first millennium BC, and in
central southern England they often cut across groups of coaxial fields in a
way that would certainly have put them out of use (Bradley, Entwistle, and
Raymond 1994; McOmish, Field, and Brown 2002: 56–67). The distribution
of these boundary ditches was more extensive than that of field systems, and a
second group of earthworks of Bronze Age origin has been studied in northeast
England. One of its focal points was the great ringwork at Thwing (Stoertz
1997: fig. 42). Unlike the co-axial fields, these boundaries remained in use in
the Iron Age. Indeed, more of them were built, and they occur across a much
greater area. The simple banks and ditches that seem to characterise the earliest
examples were increasingly supplemented by other kinds of feature, including
multiple dykes and pit alignments.
Such changes happened in many areas, but they have been studied in particular
detail on Salisbury Plain (Bradley, Entwistle, and Raymond 1994; McOmish,
Field, and Brown 2002). Here large areas of co-axial fields had been established
during the Middle and Late Bronze Ages. Some of their boundaries were
aligned on older round barrows, and the original layout seems to have been
associated with a number of small enclosures of the kind discussed in Chapter
Four. During the Late Bronze Age a series of long linear earthworks was constructed
which cut obliquely across many of the existing plots, meaning that
some of themmust have gone out of use. These ditches could also be orientated
on older mounds. The newboundaries extended from the river valleys onto the
high chalk plateaux and defined a series of elongated blocks of land, not unlike
the parishes of the early medieval period. Each territory contained at least one open settlement, and beyond the limits of this system there were burnt mounds,
sources of flint, and the findspot of a major metal hoard (Bradley, Entwistle,
and Raymond 1994: 130–1). During the Early and Middle Iron Ages some of
the earthworks were rebuilt, often on several occasions, and deposits of human
bone and animal skulls were placed within them; similar material was associated
with the boundaries of settlements. Certain earthworks were probably
extended during this period and others were levelled (Bradley, Entwistle, and
Raymond 1994). Curvilinear and rectilinear enclosures were built within these
land blocks, whilst some of the points at which the separate territories converged
became the sites of Iron Age hillforts. It was not until such monuments
had gone out of use in the Late Iron Age that field systems were re-established
on Salisbury Plain (ibid.: chapter 8).
A comparable process happened on the Yorkshire Wolds, where some of
the boundaries were defined by lines of pits rather than ditches. Excavation at West Heslerton has shown that the land divisions long recognised
on the chalk hills extended down into the Vale of Pickering to their
north (Powelsland 1988). At least two of these territories were dominated by
palisaded enclosures situated on prominent summits, whilst a large open settlement
with round houses and raised granaries was identified on the lower
ground. Another unenclosed settlement has been investigated in the valley
known asWetwang Slack (Dent 1982). This contained a similar range of structures
distributed on one side of a prominent earthwork boundary which was
maintained throughout the Iron Age. It is doubly important because a major
cemetery developed there.
In Yorkshire and to some extent inWessex, these linear earthworks coexisted
with pit alignments, and there are even cases in which a discontinuous boundary
was eventually replaced by a more substantial feature. In Lincolnshire and
Yorkshire there was also a predilection for constructing several earthworks side
by side (Boutwood 1998; Stoertz 1997). Such features are often very difficult to
date as linear ditches might be recut many times. Pit alignments, on the other
hand, had a more restricted currency and are easy to identify on air photographs
(Wilson 1978). Their distribution extends across large areas of the English river
gravels, into the Welsh borderland to the west and northwards into Scotland.
It is unfortunate that so few of the boundaries have produced any artefacts, but
what little dating evidence is available suggests that the earliestwere built around
the Bronze Age/Iron Age transition and the latest were probably constructed
in the Middle Iron Age (Fig. 5.7). They are occasionally associated with human
and animal burials and with deposits of quernstones and other artefacts. As so
often, the chronology is more secure in regions with a well-defined pottery
sequence and less precise in those areas where ceramics show less change or
were little used. At present it is uncertain whether a similar boundary system
developed in Iron Age Ireland.
Although these land divisions could have formed over a long period of time,
they seem to have been employed in the same ways across a considerable area.
They normally defined large blocks of land. Often the boundaries ran parallel to
one another and at right angles to a river. Although the layout may have changed
over time, the effect was to enclose substantial territories with a mixture of
different resources extending from the floodplain onto the higher ground. A
number of these units show some evidence of subdivision, but normally they
are of similar extent to the land of a small modern farm. There are few signs
of field systems within these enclosures, and what excavated evidence exists
suggests that either they predate these boundaries or were created after those
features had gone out of use ( J. Taylor 1997). It does not seem likely that these
different elements were contemporary with one another.
At the same time the large areas of land defined by these divisions might
include the positions of one or more settlements, which varied considerably in
their layout and their history. Some were established beside these boundaries and could have been contemporary with them; others were located some
distance away, but were situated within one of the blocks of land which they
defined. It was unusual for any settlement to extend across such boundaries,
and, when this happened, it was often because that feature had gone out of
use. The settlements themselves might be open or enclosed and in many cases
oscillated between these forms in the course of their history. Several settlements could be established within a single land unit, and where they were attached
to the same boundary they have been described as ‘clothesline enclosures’.
Sometimes their use was quite short-lived, so that one occupation site replaced
another, but just as often they coexisted. It is clear that the boundaries usually
show a greater stability than the rather fluid pattern of activity within them.
There is some evidence to suggest that the individual settlements practised
a specialised economy, but this is largely confined to a few regions, including
Wessex (Cunliffe and Poole 2000a) and the Upper Thames (Lambrick and
Allen 2004), and it is unlikely that these ditched territories had a single role. It
is not possible to argue that they were reserved for grazing land. Rather, they
seem to be associated with a mixed farming regime in which cultivation and the
raising of livestock were both important. Nearly all the excavated settlements
include large collections of faunal remains and are associated with carbonised
cereals, raised granaries, and grain storage pits.
The main difference between the Bronze Age and Iron Age systems is that
now there seem to have been no fixed boundaries within the separate land
units; Celtic fields had virtually disappeared, although the sharp edges to the
distribution of storage pits on sites in the Thames Valley suggests that more
ephemeral boundaries did exist. They could have been established on the temporary
basis, but they have left no trace behind. It is sometimes suggested
that the landscape was broken up by hedges (Pollard 1996), but even these
would have been bedded in a low earthwork if they were to flourish (Pryor
1998b: 87). At the same time, arable land needed to be protected from predators
and the movement of livestock had to be controlled. Maybe this was
achieved using a series of temporary divisions that could be changed on a regular
basis.
A possible model is provided by Caesar’s account of the early Germans
(Thompson 1965: 10–11). Instead of being dominated by a social elite, decisions
were made at a communal level and leaders were elected to serve on a
temporary basis. Access to land was allocated every year, so that differences of
wealth and power were reduced to a minimum. The distribution of agricultural
produce may have been administered in the same way. This method had
certain advantages, for it would be difficult to accumulate and control a surplus
in the way that seems to have happened during other periods. This is simply
an analogy since there was no direct connection between the inhabitants of
Early Iron Age Britain and the people described by Roman writers, but it is
very similar to George Lambrick’s carefully researched reconstruction of the
Iron Age landscape around Gravelly Guy in the Thames Valley (Lambrick and
Allen 2004: chapter 12). It would certainly account for the surprising rarity
of Early and Middle Iron Age field systems when they were used before and
after that time; and it may also provide a reason why the houses within the
settlements developed such a uniform character. If the Later Bronze Age had
seen the growth of social distinctions, this phase illustrates their decline. It may also provide the background to the development of early hillforts in
Britain; it is uncertain whether such monuments were in use over the same
period in Ireland. They are widely distributed across the south and west. Barry
Cunliffe dates them to the period between about 600 and 400 BC, and in some
respects they follow a similar trend to the earthwork enclosures considered
in the previous section (2005: 384–7). Some of the oldest examples, such as
Crickley Hill (Dixon 1994) orWinklebury (K. Smith 1977), include unusually
large round houses rather like those associated with Late Bronze Age ringworks,
but the later buildings are smaller and are of roughly uniform size. If the first
of these structures were used to express social distinctions, those differences
had either disappeared or they were no longer emphasised. A second link
concerns the ‘defences’ of these sites. They were built on a variety of different
scales, and the more impressive examples often included ramparts that were
reinforced with timber in the manner of an older site like Springfield Lyons
(Buckley and Hedges 1987). The less-monumental examples were perhaps no
different from other ditched enclosures. They have been identified as hillforts
by their position in the landscape or because their earthworks have not been
levelled by the plough, but, like the round houses of the Early Iron Age, such
earthworks really form a continuum. That is not to deny that some of these
places could have been attacked. A number of hillforts seem to have been
destroyed by fire, particularly in western and northern Britain. These sites
include Crickley Hill on the Cotswolds where the defences are associated with
deposits of sling stones (Dixon 1994: 105 and 115–16). Although Cunliffe distinguishes between ‘early’ and ‘developed’ hillforts,
this is not based on major differences in their forms and associations but on
their chronology and distribution. In general terms, the early hillforts were
widely distributed, and some of them provide only limited evidence for the
activities taking place inside them. The developed hillforts, whose history runs
down to the second century BC, were fewer in number and in some areas may
have been spaced at roughly equal intervals across the landscape. These sites
generally provide more evidence of occupation (Cunliffe 2005: 388–96).
There were also some differences in their outward appearance (Fig. 5.8). The
ramparts of the early hillforts were more likely to have a vertical outer face,
supported by a stone wall or a setting of timbers, whilst later fortifications were
often defined by a deep V-profile ditch and a steeply inclined bank so that a
potential attacker was confronted by a continuous scree slope (Avery 1993). In
the same way, the earlier sites often had two opposing entrances, whilst those
used in later phases sometimes had only one which might be reinforced by a
series of outworks (Cunliffe 2005: 365–74). Certain enclosures also increased
their area at a late stage in their development.
How were these places related to the exploitation of the wider landscape? It
has always been tempting to think of them as the power bases of a social elite, but
it is difficult to find much evidence for that idea. In southern England a few of
these sites, like Sidbury and Quarley Hill,were established at the meeting points
of several of the territories defined by linear ditches (Bradley, Entwistle, and
Raymond 1994; Cunliffe and Poole 2000a: chapter 4; Bowden 2005), and for
the most part the artefacts associated with these places have a similar character
to those found at other settlements in the same area. Perhaps these hillforts
represented a more public expression of the same concerns as those sites. The
main difference between these two categories is that in the hillforts certain
activities were performed in a more ostentatious manner (Hill 1996). Thus the
defences were sometimes constructed on an altogether larger scale than the
domestic enclosures, yet there seems to have been as much emphasis on their
appearance as there was on military architecture, and it is not always easy to
see how they would have provided much defence against attack (Bowden and
McOmish 1987). In the same way, hillforts in both the categories recognised by
Cunliffe include an exceptionally large number of raised storehouses, which
were probably used to hold grain (Gent 1983). These buildings were often
the most conspicuous features of these sites, and in several areas, including
the Wessex chalk and the Welsh Marches, they could be laid out in rows or
organised on a grid. It may have been as important to display the harvested
crop as it was to protect it (Fig. 5.9).
A number of recently excavated hillforts also contain specialised structures
which have been interpreted as shrines, but their chronology varies from site to
site. At Danebury they were used throughout the occupation (Cunliffe 1984:
83–7), yet at Cadbury Castle a similar building was not constructed until the use of the hilltop was largely over, some time after the Roman Conquest (Barrett,
Woodward, and Freeman 2000: 170–3). Before then, the same area had been
used for the production and deposition of bronze metalwork and for a series
of animal burials. Even these specialised buildings have parallels on other kinds
of site.
The animal burials at Cadbury Castle recall the wide range of offerings that
are associated with Iron Age settlements, althoughmuch of the evidence comes
from a small number of regions: central southern England, the south midlands,
East Anglia, northeast England, the Hebrides, and Orkney. Their main association
seems to be with food production and the agricultural cycle, although
they extend from the burial of selected artefacts to the treatment of the dead
(Whimster 1981; Hill 1995; M. Williams 2003). Such deposits are found in a
series of different contexts which may have changed their locations over time,
but they are especially common in hillforts. Some were associated with the
perimeters of these sites, and others with earthwork enclosures, land boundaries,
and the edges of open settlements. Such deposits were directly linked
with the storage of food, so they are found in grain silos in southern Britain or
in souterrains in Orkney. The human remains reveal some striking patterns. In
certain cases there were important differences between these groups, so that in
the hillfort at Danebury the remains of adults were kept separate from those of
the very young who were associated with the positions of the houses (Cunliffe
and Poole 1991b: chapter 8). In the open settlement at Glastonbury adult skulls
were distributed along the edge of the settled area, and the remains of children
were found in the platforms where people lived (Coles and Minnit 1995: 17–24
and fig. 8.10). More might be learnt from excavation outside the settlements
themselves as recent excavations in East Anglia,Wessex, and the Thames Valley
have located small inhumation cemeteries (Hey, Bayliss, and Boyle 1999).
What distinguishes the sites known as hillforts from the other earthwork
enclosures of the Early and Middle Iron Ages? Apart from their size and the
labour devoted to their construction, there seem to be three important features.
These monuments include a greater density of storage facilities than other
sites. That applies not only to the raised granaries mentioned earlier but also
to the evidence of pits. Given the comparatively limited number of houses
inside the excavated hillforts, it seems unlikely that so much food was produced
by the occupants of these sites (Gent 1993). Although the raised storehouses
could have displayed food for consumption, the pits most likely held seed corn.
At the same time, there is more evidence for ritual practices in hill forts than
there is in the settlements of the same period (Hill 1996). These focused on
the process of crop production, and the main focus was the grain storage pit.
They contained a series of deposits but may have been selected because of their
links with the cycle of death and regeneration associated with the harvesting
and keeping of grain. Such notions of fertility and renewal may have extended
to the treatment of the dead. Lastly, the houses inside some of the hillforts have a different character
from many of those at other sites. They are often more ephemeral structures
with light stake walling, and some were even built of turf. They were often
rather smaller than their counterparts on other sites and may not have been
inhabited on a permanent basis. That is also implied by the creation of building
platforms inside the hillforts at the Wrekin (Stanford 1984) and Dinorben
(Guilbert 1981), for they are not associated with the remains of any recognisable
structures. The same idea is suggested by the insects preserved by a pond within
the developed hillfort of the Breidddin on the Welsh border, for despite the
presence of houses and granaries within this enclosure, the use of the site had
little impact on the local environment (Buckland et al. 2001). Similarly, some
of the storage pits inside the Wessex hillfort of Winklebury seem to have been
left open over the winter, trapping a number of wild animals which must have
been living on the site (K. Smith 1977: 111). In other cases there may have been
more houses and other features outside the defences than there were within
them, again suggesting that this space may have played a specialised role. That is
certainly the implication of recent fieldwork at three of these sites:Wittenham
Clumps in the Upper Thames valley (T. Allen pers. comm.), Wandlebury in
eastern England (French 2004), and Cadbury Castle in the southwest (Tabor
2004).
The relationship between hillforts and ordinary settlements remains quite
problematical. The defended sites contain rather slighter traces of domestic
buildings, and yet they have a higher density of storehouses and pits. A few of
them seem to have included shrines, whilst the proportion of special deposits
exceeds that at other sites. They were closely associated with the agricultural
cycle and the keeping of grain (M. Williams 2003). Large numbers of people
obviously made use of these places, but it is not clear that they settled there on a
permanent basis (Lock, Gosden, and Daly 2005: chapter 4). In many ways these
were public monuments which reflected the concerns of people during this
period and emphasised them on an impressive scale (Bradley 2005b: 168–77).
Indeed, the very form of some of these hillforts still seems to echo the basic
principle of the round house. They can adopt a roughly circular ground plan
with an entrance that faces the rising sun; Danebury provides a good example.
Such places may have been conceived as the houses of an entire community
who could have used them in much the same way as an early medieval assembly
(Pantos and Semple eds. 2003). Perhaps they were where communal business
was transacted and important decisions were made. Again Caesar’s description
of the early Germans could be relevant to the argument. Such an arrangement
would not have been unprecedented for it is also the interpretation suggested
for the middens which are such a striking feature of the Late Bronze Age/Early
Iron Age transition. It may have been through periodic meetings in such places
that land and its products were distributed among the population. That did not
require a permanent elite. In fact the very term ‘hillfort’ is probably a misnomer. At times they may
have served as fortifications, as places where resources were protected from
attack, but they were also production sites and even a kind of theatre at which
public events took place and the concerns of a farming people were played out
in ritual and ceremonial. That may have changed only towards the end of their
period of use.
If so, then it may be possible to suggest how some of the same activities
could be organised at other kinds of site. In southwest England, for example,
the wetland settlements of Glastonbury and Meare are associated with an
extraordinary abundance of artefacts, in spite of their rather marginal location
(Coles and Minnit 1995; J. Coles 1987). The authors of the most recent
accounts of these two sites suggest that they acted rather like medieval fairs,
as seasonal meeting places where goods were made and exchanged and where
social transactions took place. The same interpretation might apply to other
areas that were occupied on a discontinuous basis or where large numbers of
people gathered for only part of the year, like the East Anglian Fens (Evans
2003) or the Severn Estuary (Bell, Caseldine, and Neumann 2000). A further
possibility is that much older monuments were brought back into use by the
wider community. One candidate is the henge monument of the Devil’s Quoits
in the Upper Thames Valley (Fig. 5.10). George Lambrick has suggested that
it was surrounded by a large area of open pasture which was shared by the
occupants of a number of different settlements (Lambrick and Allen 2004:
chapter 12). Another example might be the Ferrybridge henge in northeast
England. In this case the area outside the Neolithic monument was enclosed
by a series of pit alignments associated with human burials, and a sword scabbard
was deposited inside the ancient enclosure (Roberts 2005: 229–31). The
activities that took place at the most famous hillforts may have happened in
other places too.

VARIATIONS ON AN ORIGINAL THEME
If these features represent the main currents in the archaeology of the Iron
Age, in some parts of Britain they were expressed in a very different way.
Virtually nothing is known about Irish settlements of the late first millennium
BC (Raftery 1994), but it is clear that those in western Britain possessed
a distinctive character. This is not a unique feature of insular prehistory, for
there are similar sites in a number of regions along the Atlantic seaboard of
Europe, including Finist`ere, Galicia, and northern Portugal. In each case the
settlement pattern was dominated by a dense distribution of small fortified
enclosures (Cunliffe 2001a: 336–539). These were often circular constructions,
but along the coast they could be supplemented by a distribution of promontory
forts. These in Brittany are often compared with examples in southwest
England. There are other links between the regions along the western limit of the
Continent. In the Iberian peninsula the circular enclosures or castros were
normally associated with round houses – it was only during the Roman period
that they were replaced by rectangular buildings – and, where more impressive defences were built, they could be supplement by a chevaux de frise: a setting
of upright stones so called because it was thought to impede attackers on
horseback. Their distribution is revealing for, like that of the round houses,
it focuses on Britain and Iberia. Examples are found on sites in north and
west Wales, in eastern and southwest Scotland, and along the west coast of
Ireland (Harbison 1972). Their overall chronology is uncertain, and individual
examples could be earlier or later than the period considered here.
Several distinctive features characterise the earthwork enclosures of western
Britain, including those of the Isle of Man. Most of them are fairly small,
but individual examples can be massively defended. In southwest Wales the
entrance is sometimes screened by monumental outworks (Fig. 5.11; Cunliffe
2005: chapter 13). They contain a relatively small number of circular houses
as well as raised storage structures, but their relationship to one another is not
consistent from one site to the next. At Walesland Rath the putative granaries
lined the inner edge of the rampart, and the houses were in the middle of
the enclosure (Wainwright 1971); at Woodside they were to the right of the
gateway, and the dwellings were towards the rear of the enclosure; whilst at Dan
Y Coed the positions of both groups of buildings overlapped (Williams and
Mytum 1998). Another way of providing secure storage was by constructing
a kind of cellar. In Cornwall, it took the form of a stone-lined trench, roofed
by a series of lintels, and in this case it could be associated with an individual
house (Christie 1978: 314–33). These features are called souterrains and occur
also in Brittany. There is no obvious difference between the structures found
on these sites and those in the larger enclosures which are usually described as
hillforts. Although there are some signs of open settlements during the Early
and Middle Iron Ages, enclosures are densely distributed across the landscape
and may have been largely self sufficient.
Apart from differences of size – most of the enclosures are small – the main
distinctions between them concern the scale of the surrounding earthworks.
In west Wales and southwest England many of them conformed to a precisely
circular ground plan, as if to echo the same principle as the round house. It was
a tradition that was to last into the post-Roman period in Britain (Quinnell
2004), when it had its counterpart among early ring forts in Ireland. The
latter normally contain circular buildings (M. Stout 1997), but these structures
were quite similar to one another. The early medieval Irish laws show that it
was the scale of the perimeter earthwork that was the main way of displaying
status. It was carefully controlled, so that the number of concentric earthworks
enclosing a settlement site might have been related to the social position of
its occupants (Edwards 1990: 33). Perhaps a similar model would explain the
evidence from western Britain in the pre-Roman period.
There were other regions with a dense distribution of small circular enclosures.
These include the uplands of northern England and southern Scotland,
where the field evidence can be exceptionally well preserved. The pattern extends further up the North Sea coast, but here there may have been a higher
proportion of open sites (Armit and Ralston 2003). It is clear that the basic
pattern goes back to the beginning of the Iron Age and that on either side of
the border between England and Scotland individual sites might be rebuilt on
an increasingly impressive scale, so that an enclosure could be defined first by
a palisade, and then by a low earthwork (Fig. 5.12); some sites were eventually
defended by a substantial rubble wall. The main feature inside these enclosures
was the round house (Hingley 1992). These structures occurred in varying
numbers, from a single example to a dense distribution of buildings, but some
of them were as large as any dwellings occupied during this period. Individual
settlements can be associated with plots of cultivated land which may have been
worked by hand (Topping 1989), but there is no sign of any specialised storage
structures. The upland enclosures have their counterparts on the lowland soils,
but they are more difficult to interpret as few artefacts or food remains survive.
The results of commercial excavation show that open sites also existed there.
In central and northeast Scotland houses within these settlements were often
associated with souterrains (Armit and Ralston 2003).
It is difficult to say whether the upland enclosures which have dominated
the discussion were inhabited all year round. Many of them were in exposed
places on the high ground and might have been inhospitable or inaccessible in
winter. Moreover, the houses rarely show much sign of maintenance or repair,
suggesting that they had not been occupied for long (Halliday 1999). The same
problems affect the large hillforts of highland Britain (D. Harding 2004: 58–
66). They are strongly defended, they were built in dominant positions, and
they enclose the sites of many round houses, but it is difficult to see how they
could have been inhabited continuously. They might have been used during
the summer months when climatic conditions were more favourable, but in any
case the sheer density of internal buildings is not consistent with the character
of the local environment which might not have been capable of supporting a
large population. Again it is tempting to suggest that these were aggregation
sites, used on an occasional basis and possibly in the course of summer grazing.
A number of these monuments adopted a curvilinear ground plan.
The tendency to build self-contained circular enclosures reached its apogee
in western and northern Scotland, and especially in the Hebrides, Orkney,
and Shetland (Fig. 5.12). Unfortunately some of the strongest patterning has
been obscured by disagreements about terminology and chronology (Armit ed.
1990; Armit 1992; Parker Pearson and Sharples 1996: chapter 12; D. Harding
2004: chapter 5).
Again the circular archetype was very important and extended from individual
dwellings to more monumental walled enclosures. All these features were
conceived on an impressive scale. They vary from the crannogs built in open
water to small circular compounds, and from relatively insubstantial dwellings
to massive domestic buildings, the most impressive of which – the brochs of the
Scottish mainland, the Western and Northern Isles – resemble towers (Armit 2003b). Some of these structures are isolated but quite densely distributed and
were surely designed to impress, whilst others can be found inside defended
enclosures which contain a variety of other buildings. Many of them were
distributed along the coast where there were a number of promontory forts
(Armit 1992). Here is another case in which large round houses may have been an important
settlement form from an early stage of the Iron Age. Ian Armit (2003b) has
argued that structures ancestral to brochs were built as early as 600 BC and
that during the Iron Age stone houses in Atlantic Scotland became increasingly
complex in design. True brochs seem to have emerged about 200 BC. They are
interpreted as defended high-status dwellings, characterised by such features as
internal staircases and guard cells. They must have had more than one storey,
and there can no longer be any doubt that they were roofed.
Some of the structural principles that characterise the brochs extend to other
forms of defensive architecture: to some of the circularwalled enclosures known
as duns and even to the monumental gateways of a number of the promontory
forts in Shetland (Hamilton 1968; D. Harding 2004: 137–50). Another
key element is the way in which domestic structures were organised. In some
brochs there seems to have been a communal space with a hearth located in
the centre of the building. It was ringed by a range of compartments which
were divided from one another by partitions projecting from the interior wall.
In Shetland and the Hebrides, this principle was expressed on a smaller scale in
the distinctive dwellings known as wheelhouses (Parker Pearson and Sharples
1999: chapter 12). A number of these were built after the broch themselves,
and it has yet to be worked out how far their histories overlapped. In any case
there are important contrasts between them. Wheelhouses were sometimes set
into the ground, whereas brochs were conspicuous monuments, and on certain
sites the domestic accomodation was probably at first floor level (Armit 2003b:
chapter 3). Wheelhouses were sometimes associated with souterrains, but the
connection between storage structures and individual dwellings is entirely different
from the more centralised system illustrated by hillforts in southern
Britain.
In many ways the Scottish sites combine the two of the key elements discussed
so far, the enclosure and the round house, and sometimes they fused
them together in a single structure (Parker Pearson and Sharples 1996). But
a very different tradition remains to be defined. If the Atlantic round houses
are conspicuous features of the northern landscape, the settlements of eastern
England have left little trace, and most are known from air photography or from
small-scale excavation. Others have been found as scatters of surface artefacts.
In his study of Iron Age Britain Barry Cunliffe suggests that the region
between the Thames and the Humber was dominated by ‘villages and open
settlements’ (2005: figs. 4.3 and 21.6). That pattern extended westwards as far
as the zone of hillforts that follows the modern border between England and
Wales. The division is not clear-cut, for earthwork enclosures are common
in both the midlands and the Thames Estuary, and some hillforts occur in
both those areas. Moreover, open settlements are among the largest occupation
sites on the YorkshireWolds (Dent 1982; Powelsland 1988). Such regional
divisions only describe broad tendencies in the evidence, but there is considerable
variation. At the heart of this zone is East Anglia. Enclosed sites are not particularly common here, and not all of them need have been settlements, as
it is possible that they played a specialised role (Martin 1988). Although some
standing earthworks have been identified as hillforts, there is little evidence
that they were used intensively (Davies et al. 1991; Evans 2003). Instead, occupation
sites extend over considerable areas, sometimes changing their centre
of gravity over time (Fig. 5.13; Hill 1999). There were few fixed boundaries,
although individual houses or small groups of houses might have been located in compounds within a more extensive living area. These houses were interspersed
with granaries and storage pits. In some cases rectilinear compounds
were built onto one another as the settled area increased: a process that would
have been more difficult in a landscape of circular enclosures.
It is not clear how many of the structures were contemporary with one
another, making it difficult to decide whether these sites were really villages.
Nor is it obvious how much of the occupied area was in use at any one time.
J. D. Hill (1999) has made an interesting comparison between some of these
places and the ‘wandering settlements’ of the same date in Northern Europe,
where houses and other structures were often abandoned and replaced after
a limited period of use (Gerritsen 1999, 2000). Their positions often shifted,
but over a restricted area so that the dwellings occupied during successive
phases had mutually exclusive distributions. In eastern England that process
could easily have generated the large areas of buildings, pits, and ditches that
now survive. Some sites may be as much as ten hectares in extent. Another
possibility is that certain of the settlements resulted from the amalgamation of
smaller units as larger numbers of people elected to live together in one place
and to give up such independence as they had once possessed. That would be
the opposite of the situation on the chalk of southern England, and it is certainly
suggested by the chronological evidence from East Anglia, where there were
few large occupation sites during the Early Iron Age. The emergence of these
settlements seems to be a later development, and the relationships between
their separate elements may be expressed by the ways in which compounds
and even individual dwellingswere joined on to one another over time (Bradley
1984: 140). This may have been one way of defining the relationships between
different groups of people. If the occupation sites served large communities in
this way, it may explain why there was less need to coordinate activities through
the use of hillforts.
In some respects this chapter has followed a conventional sequence, starting
with some of the larger enclosures and hillforts, then turning to smaller sites
in the north and west before finally considering the large open settlements of
eastern England. That reflects the extent of current knowledge, where certain
regions have been intensively excavated at the expense of others, but it may
not reflect the actual situation in prehistory. For all the labour invested in their
construction, the hillforts could have been a rather peripheral phenomenon,
on the margin of a more prosperous and perhaps more expansive system with
its emphasis on the North Sea coast. Perhaps it would be better to think in
even broader geographical terms, so that the large open settlements of eastern
England could be treated as the local equivalents of similar sites in Northern
Europe (Hill 1999), whilst the enclosed sites of southern England might be
compared with their counterparts in the north of France. The tradition of
building small but monumental enclosures along the west coast of Britain could
then be accepted as part of a still-wider phenomenon extending southwards into Atlantic Europe (Cunliffe 2001a). That mental adjustment would involve
turning the map of Britain on its side so that the North Sea became the
dominant axis (Fig. 5.14). That would provide a corrective to accounts of this
period which are conceived on too small a scale, but any new scheme will be
incomplete until more is discovered about the Irish Iron Age.
AN END OF ISOLATION
It may be difficult to envisage such broad alignments, but they seem to be
implied by the convergence of British and Irish archaeology with the prehistory
of the European mainland. This involved two related processes. The first were
changes in local patterns of settlement, whilst the second saw the renewal of
relationships with the Continent. As part of that process activities in Ireland
assumed a growing importance.
Cunliffe’s interpretation of the hillforts of southern England suggested that
there were gradual changes in the ways in which they were used. Although the
details of his argument have proved controversial, there seems no doubt that
some of his developed hillforts did assume new roles towards the end of their
history (Cunliffe 2005: 388–96). One finding is of particular importance. Two
of the largest projects carried out at these sites involved fieldwork in and around
Danebury and Maiden Castle, each of which contained houses, pits, granaries, and a shrine. These places might have been occupied continuously or they
could have been the setting for communal assemblies, but recent fieldwork
suggests that each of these interpretations might apply to different phases in
their development. Thus there is little sign of any Middle Iron Age enclosures
or settlements in the vicinity of Maiden Castle (Sharples 1991: chapter 3).
Danebury, on the other hand, was located amidst a series of settlement sites,
most of which were used concurrently with the hillfort, but towards the end
of its occupation they were apparently abandoned (Cunliffe and Poole 2000a:
181–2). It seems possible that people might have moved there for protection.
There is certainly skeletal evidence that some of its inhabitants had suffered
violent deaths, and other sites provide similar indications of warfare at this
time. Cunliffe suggests that the same sequence took place around the hillfort of
Barbury Castle, mentioned earlier in this chapter (Cunliffe 2005: 393; Bowden
2005).
A number of developed hillforts contain a distinctive assemblage. Sheet
bronze artefacts were being made at Danebury and Maiden Castle during
their later phases (Northover 1984). The same happened at Cadbury Castle.
This is only one of a series of new associations. Swords are also found at these
places (Piggott 1950), and from the third century BC, so are iron currency
bars which are recorded from no fewer than seventeen of these sites (Hingley
2005). Another clue to the changing role of the last hillforts has been suggested
by John Creighton, who draws attention to the increasing importance of the
horse, both as a symbol of power displayed on the earliest pre-Conquest coins
and for its use in warfare (2000: 15–17). Bronze horse bits are another artefact
associated with the last use of southern English hillforts ( Jope 2000: 152–60).
Perhaps the roles of these sites had changed. Recent excavations have also found
the remains of horses in a number of specialised deposits in eastern England.
Until the Middle Iron Age there was no sign of a regular burial rite in Iron
Age Britain. In Ireland, the situation is equally confusing. Ring barrows may
still have been used there, but few of their contents can be dated (Raftery 1994:
189). In southern England the remains of the dead were treated in other ways
and were often associated with settlement boundaries or with storage pits (Hill
1995).
It was in the fourth century BC that formal cemeteries developed in at least
two regions of England (Whimster 1981: chapters 3 and 4); others may have
existed outside the settlements of this period, but they have not been recognised
until recently (Hey, Bayliss, and Boyle 1999). Each of the burial traditions
possessed a distinctly regional character, although in both cases bodies were
laid out in the ground according to the same conventions as those in pits. On
the other hand, each of these traditions was also linked with a different region
of Europe.
The first of these groups is found in southwest England and has some features
in common with Iron Age cemeteries in Brittany and the Channel Islands (Whimster 1977; Whimster 1981: chapter 3; Burns, Cunliffe, and Sebire 1996).
All these sites were located near to the sea and contained inhumation burials,
which, unlike their Continental counterparts, assumed a crouched position.
They were in use between the fourth century BC and the first century AD.
There was one body to a grave, and they were usually associated with personal
ornaments such as brooches, bracelets, and pins. One of the best known of
these sites was at Harlyn Bay and may have been associated with a timber round
house, but it is not clear whether this formed part of a settlement or whether
it had played a more specialised role connected with the treatment of the dead.
Three of the brooches from Harlyn Bay were probably imported from Spain
(Whimster 1981: 60–9).
The distinctive Iron Age burials of the Yorkshire Wolds are much better
known and have often been discussed because of their striking similarity to
those found in Champage, the Ardennes, and the Middle Rhine (Dent 1982; Stead 1991). Their currency is roughly similar to that of the cemeteries in
southwest England. They probably originated during the fourth century BC,
but may have gone out of use during the first century BC when settlements
expanded over the sites of some of the graves.
These burials have been attributed to a unitary ‘Arras Culture’, but in some
respects they are surprisingly diverse. The best known are a small series of vehicle
burials, which may have occasional counterparts elsewhere in eastern and
southern Britain (Fig. 5.15); another has recently been excavated at Newbridge
on the outskirts of Edinburgh (Carter and Hunter 2003). They resemble their
Continental counterparts in many ways, but again the details of the funeral rite
are subtly different and so are the associated artefacts. The vehicles were usually
dismantled and the bodies were laid out in the crouched position that characterises
other mortuary traditions of the British Iron Age. Even the associated
artefacts are British versions of European prototypes (Stead 1991).
The same applies to the barrow cemeteries in northeast England. They
consist of dense concentration of square mounds, defined by a shallow ditch
which allows these distinctive monuments to be recognised from the air. They
developed over a long period of time but probably increased in number during
the late third and second centuries BC. Again many of the burials were laid
out according to a specifically British tradition, and they were accompanied by
a small selection of personal ornaments which were generally insular versions
of Continental forms (Stead 1991).
There is a further subgroup of Arras Culture graves which needs to be considered
here, for in many respect it differs from the norm. These contained
extended inhumation burials accompanied by weapons – swords and spears –
and are a particular feature of the Makeshift cemetery at Rudston (Stead 1991:
6–15 and 185–208). They seem to be a late development within this local tradition
and are probably related to a series of isolated ‘warrior burials’ found
in other parts of these islands (Collis 1973). There are only a small number of
them, and their distribution extends across lowland England with a distinct concentration
in Wessex. Outlying examples of burials with swords are recorded
from central Scotland, north Wales, and the east coast of Ireland (Fig. 5.16;
Hunter 2005). They are important because they suggest that the symbolism
of conflict and warfare was becoming more important during the later part of
this period. That is surely confirmed by a series of small chalk sculptures found
on the Yorkshire Wolds which depict warriors carrying swords (Stead 1988).
Like the changing character of southern and western British hillforts, this evidence
suggests that the egalitarian ethos of the earlier Iron Age was breaking
down.
The most extensively excavated barrow cemetery in Yorkshire is at Garton
and Wetwang Slacks (it has two names because it crosses a modern parish
boundary). It was established beside a linear earthwork and was located on
a site which already included a series of mortuary monuments of earlier prehistoric date (Brewster 1981; Dent 1982). It is important because the cemetery
developed alongside a large open settlement which contained the usual
mixture of round houses, raised granaries, and storage pits. What is especially
interesting is that there are few convincing signs of social divisions within the living area, even though the cemetery itself contained some exceptionally rich
burials. There are certain anomalies – a few of the houses departed from the
usual easterly alignment (Parker Pearson 1999); some of the buildings contained
large deposits of animal bone of a kind more familiar in southern England –
but it would be impossible to postulate major differences of status from the
buildings in the settlement. Moreover, it is surely significant that the one
region of Britain to posses a tradition of weapon graves does not seem to have
included any hillforts of the same date. Even though the skeletal evidence from
the cemeteries shows that certain individuals had engaged in combat (Dent
1983), there is no indication of anything that might be interpreted as military
architecture.
In fact the evidence from the Yorkshire Wolds is quite anomalous, and
nowhere more so than in the combination of square barrows of Continental
inspiration with round houses of an entirely insular kind. It is hard to understand
their relationship. There are cases in which square mounds were built over the
positions of circular dwellings and others where these two types respected one
another, suggesting that the houses of the dead were replaced by mortuary
monuments, but both halves of the same complex seem to have maintained
their distinctive character over time (Fig. 5.17; Dent 1982). Perhaps this is most
obvious from the concentration of infant burials associated with certain of the
houses at Garton Slack (Brewster 1981). Similar deposits are widespread in
Iron Age Britain, and yet the burials of adults were carried out with greater
formality in the cemetery that was located alongside the living area. These
curious relationships require much more research.
Although the burials of the Arras Culture are rather different from their
Continental counterparts, there was obviously a close relationship between
them: so much so that it is perfectly possible that the Yorkshire Wolds were
settled from overseas. The distinctive artefacts associated with some of the
graves pose a more general problem, for they clearly form part of a much wider
network that extended across large parts of Europe and takes its name from
the Swiss site of La T`ene (Raftery 1984; Jope 2000). The forms of the British
artefacts were clearly inspired by those produced on the European mainland,
but there is little to suggest that they were imports. Rather, they were made
according to specifically local techniques and often show regional distinctions
of their own.
If this observation applies to some of the grave goods of the Arras Culture,
it is even more relevant to the archaeological sequence on the other side of
the Irish Sea (Raftery 1984, 1994). Because pottery went out of use at the
end of the Bronze Age, settlement sites have been very difficult to identify.
So few have been investigated by excavation that it is understandable that
most attention has focused on portable artefacts, particularly metalwork and
querns. This approach is inescapable but it poses many problems. Most of the
metalwork comes from votive deposits in rivers, bogs, and lakes, and a smaller amount is found in hoards. It is rarely associated with burials, and only a few
graves contain brooches or other ornaments which recall the La T`ene style
of decorated bronzes. The querns, on the other hand, are domestic artefacts
and seem to be virtually indestructible, but even these could assume a special character. Again some of them seem to have been placed in bogs, whilst others
were decorated in the same style as late pre-Roman metalwork: a style that also
extends to a small series of decorated standing stones (Raftery 1984: 290–303).
This evidence presents two rather different problems. There are very few
finds of metalwork between the end of the Irish Bronze Age and the adoption
of newstyles inspired by developments on the Continent from perhaps the third
century BC. Either fine artefacts were no longer being made, or the tradition of
depositing them had lapsed (Bradley 1998b: chapter 4). The latter seems more
likely for it is a trend that can be recognised across much of northwest Europe.
It is certainly preferable to the attempts that once were made to extend the
Irish Late Bronze Age to cover any apparent gap in the sequence (Champion
1989). The second problem is that when the circulation of fine metalwork – or,
more accurately, its use as offerings – began again, the evidence was confined
to only part of Ireland and was entirely missing from the south. The same
applies to the distribution of rotary querns which should date from the same
period, although they are absent from an even larger area (Raftery 1994: 124).
How are such anomalies to be explained? Perhaps the best solution is to focus
on the supposedly mundane artefacts. They were obviously used for grinding
grain and seem to have replaced the less efficient saddle querns, but the reasons
for this change are rarely considered. Did this happen simply because they
were more effective, or was it really because the new technology allowed the
processing of larger amounts of food? In that case they could have played
an important social role. Maybe they were most important in contexts that
required large scale consumption, and that may be why a few of them were
lavishly decorated. Indeed their distribution may overlap with that of La T`ene
metalwork because both were associated with the activities of an elite. In the
south, the sequence may have continued as before. There could have been less
role for conspicuous consumption and here a simpler way of treating cereals
continued.
The metalwork of Iron Age Ireland raises similar issues to the artefacts associated
with Arras Culture burials; and there are stylistic links between them. Very
little of this material could have been imported directly from the Continent,
and yet it conforms to the same basic styles of decoration as the metalwork
found across large parts of Western Europe. The quality of these objects is
quite exceptional and includes not only small personal items such as brooches
and pins, but swords with decorated hilts and scabbards, spearheads, cauldrons,
and horse harness. This must have been produced by specialists working for
a patron and surely provides evidence for the growth of a new elite, whose
concerns included riding, feasting, and warfare (Creighton 2000: chapter 1).
Much of this material may have been used to communicate social status, but in
the end it was employed in votive deposits. Irish archaeologists have sometimes
asked themselves whether the appearance of so much wealth was the result of
settlement from overseas, but the distinctly insular style of this material argues against that idea. Rather, it suggests that Ireland was increasingly integrated
into a social network that extended into Continental Europe.
Much the same can be said about the fine metalwork produced in Britain,
which Martyn Jope (2000) considered was made with increasing frequency
from the third century BC. Unfortunately this has not played a major role in
studies of the Iron Age. Instead many of the objects have been treated as ‘Celtic
art’ and regarded as a specialist field in themselves. Apart from the relatively rare
objects found in graves, these artefacts lack an obvious archaeological context
for, like so many of their counterparts in Ireland, they are generally found in
rivers rather than dry land (Fitzpatrick 1984). This evidence occurs in most
parts of Britain and Ireland, but perhaps the best documented group comes
from Fiskerton in the Witham Valley, where a timber causeway was associated
with a large number of tools and weapons dating from this period (Field and
Parker Pearson 2003). In certain respects such deposits recall Bronze Age sites
like Flag Fen, but this case is subtly different, as similar material is sometimes
found in graves.
The finest metalwork of the Iron Age is too often studied in isolation. That
is unfortunate as it loses sight of some of the connections that have already
been suggested in this chapter. In particular it overlooks the evidence for the
production of these objects which has been identified during recent years.
This shows that some of the finest metalwork – weapons, ornaments, and
specialised equipment for feasting and horse riding – was produced not only
in certain of the developed hillforts but also at a series of other settlements of
the same period (Foster 1980). It provides a clue to the sheer scale on which
such items were made, for it is obvious from the clay moulds that survive that
many more fine objects were being manufactured than found their way into the
archaeological record. That is hardly surprising, for exactly the same problem
has been identified in the Later Bronze Age.
The La T`ene style extended into other spheres and is represented by decorated
pottery in several areas of Britain, including parts of East Anglia and
the east midlands (Cunliffe 2005: 106–13), but it is not found in the same area
as the Arras Culture burials (Rigby 2004). That is interesting, for the distribution
of square barrows seems to extend into both these regions, although
it avoids the southwest, where another ceramic style, Glastonbury Ware, has
similar decoration. It is not clear whether these vessels played a specialised role,
although they were particularly common in an enclosure at Weekley which
seems to have been attached to a larger settlement with a different pottery
assemblage. This enclosure was associated with a round house, with finds of
three iron spears and with evidence of metalworking ( Jackson and Dix 1987).
A small selection of other artefacts, often personal ornaments, have been
identified in excavations, but many more as known as chance finds. Their distribution
is very striking indeed ( Jope 2000). The most elaborate metalwork,
which is often described as parade armour, is mainly associated with the rivers discharging into the North Sea. The principal concentrations of surface finds
echo this distinctive pattern and seem to be concentrated in the zone of
large unenclosed settlements in eastern England. Although there are many
exceptions, the best provenanced artefacts often come from sites of that kind,
although very few examples have been excavated. They also include artefacts
made of gold, whose use had lapsed since the Late Bronze Age. It seems increasingly
likely that the large open settlements in this area provided the power base
of an elite, but again there is a need for more research.
SUMMARY
It is difficult to sum up such a wide range of evidence, particularly when
it comes from so many sources: the later use of hillforts in some areas; the
evidence of burials in others; large open settlements with a range of distinctive
surface finds; and the fine metalwork found in rivers. Even so, they lead to a
similar conclusion.
This discussion began by considering the social impact of a diminishing
supply of bronze from overseas. Access to this material had been one source
of social power, and food production may well have been reorganised so that
local leaders could participate in exchange. As those connections fell away, their
political power was threatened and, with it, their ability to participate in long
distance alliances. There followed a period of comparative isolation, in which
internal social differences were less apparent and the community itself may have
played a more important role. The house became the dominant symbol of this
period, and its characteristic layout may have influenced the organisation of
enclosures and fortifications. Control over agricultural land and its products
seems to have been of central importance.
That egalitarian ethos was not to last, and from about the fourth century
BC in parts of Britain there was a greater emphasis on certain individuals,
expressed through the development of new burial rites. They were closely
related to practice in Continental Europe, and that link was emphasised even
more strongly from about the third century BCby the adoption of a newsuite of
fine metalwork of foreign inspiration, even if most of the products were of local
manufacture. Although these special artefacts had rather different associations,
a similar process affected much of Britain, and in particular the east coast, as
well as the northern half of Ireland. The circulation of these objects suggests
some new concerns: the display of personal wealth and adornment, feasting,
horse riding and the use of wheeled vehicles, and, above all, the importance of
armed conflict. The latter is surely expressed by the later burials of the Arras
Culture and perhaps by the changing character of the major hillforts, and yet
it was around the open settlements of the North Sea coast that much of the
new wealth was concentrated. That may have been partly because there was
greater scope for the expansion of population into this area, but perhaps it was also because the more open landscape presented fewer impediments than the
dense network of enclosures and boundaries that was found further inland.
Still more important, this was a zone with ready access to the sea and easy
communications with Continental Europe. In that respect the new system that
developed had much in common with the political geography of the Late
Bronze Age.
Between the fourth and second centuries BC both Britain and Ireland were
more closely integrated with the European mainland than had been the case
for many years, and it is no accident that it was in this period, around 320 BC,
that Pytheas made the famous voyage described in Chapter One. It seems clear
that by then these islands already played a part in long-distance trade. It is
sometimes claimed that it was the northward expansion of Roman power that
provided the catalyst for changes in the insular sequence (Cunliffe 1988), but the
chronological developments considered here suggest that it merely accelerated
processes that were already under way. British society was very different when
Julius Caesar first invaded lowland England in 55 BC, and between his second
expedition and the Claudian Conquest nearly a century later a more profound
economic and political transformation took place (Millett 1992; Creighton
2000, 2006). These processes tend to overshadowthe more subtle developments
that have been considered here, but it is to see events through the eyes of
Classical writers to believe that it was the influence of Rome that was the only
important issue. As Chapter One has demonstrated, those authorities knew
little about the societies that they were describing, and could not understand
the geography of the countries where those people lived.
That is not to underestimate the importance of trade with the Roman world,
but it deserves a book in itself, and so do the intricate political relationships that
developed between different areas before and after the Conquest of AD 43.
Fortunately, such work is already available and rightly advances the view that
the archaeology of the Late Iron Age is best studied in relation to a much
longer process of interaction with Rome which did not even end with the
withdrawal of her army from Britain in AD 410 (Millett 1992). There was
considerable variation during this period of cultural contact, and it took too
many guises for it to be considered here. Over four thousand years the archaeological
sequence had assumed very different forms, but now in quick succession
societies in lowland Britain experienced the development of the state and even
their incorporation into a foreign empire. These two islands were no longer
inconceivable places on the very edge of the world. They had lost something
of their mystery.
THE END OF PREHISTORY
Four brief vignettes will serve to illustrate the course of events, two of them in
what was to be the heart of the Roman province of Britannia, and the other
two well beyond its limits. The first of these examples falls into the latter category. This is Navan Fort
in northern Ireland (Waterman 1997). It is best known as the legendary capital
of the kingdom of Ulster and the home of the gods of the underworld, but
the archaeological sequence has much to add to what is known of its legendary
associations. This was one of the Late Bronze Age ringworks discussed
in Chapter Four, but, unlike the other examples, it remained in use throughout
the pre-Roman Iron Age. During that period it consisted of a circular earthwork
enclosure with a sequence of substantial round houses at its centre. It was
not until the beginning of the first century BC that the scale of the monument
increased significantly, and it may be no accident that during the later pre-
Roman Iron Age there is unusual evidence for long-distance contacts along
the Atlantic coastline. Excavation of the monument between 1961 and 1971
brought to light the skull of a barbary ape, which must have been introduced
to the site from its natural habitat in Gibraltar or North Africa. Its authenticity
is confirmed by radiocarbon dating. At the same time the landscape to the
south of Navan Fort was subdivided by massive linear earthworks, part of a
system of such features that seem to be contemporary with the royal centres in
the north of the island (Raftery 1994: 83–97). A great timber road was built at
Corlea during this phase (Raftery 1996: 418–22), and the entire hill at Navan
was enclosed by an earthwork which resembled a Neolithic henge. Its Iron
Age date was established by a recent excavation (Mallory 2000).
At the centre of this enclosure there was what may have been the largest
round house ever constructed, a massive wooden building thirty-seven metres
in diameter with a gigantic post at its centre (Fig. 5.18). According to dendrochronology
the timber was felled in 95 or 94 BC. Unlike a domestic structure,
it seems to have been entered from thewest, the position of the setting sun.
It seems likely that this was actually roofed, but it contained no hearth, nor were
any domestic artefacts found inside it. No sooner had it been constructed than
it its interior was packed with rubble and the outer wall was set on fire, then,
once it had collapsed, it was buried beneath an enormous mound. Richard
Warner (2000) is surely right when he suggests that the site was regarded as an
entrance to the underworld and that this building was the house of the deities
who dwelt there. The purely archaeological evidence is extraordinary, and this
monument surely took the symbol of the circular building to its limits.
The second example is the broch of Gurness on the Mainland of Orkney
(Figs. 5.19 and 5.20; Hedges 1987: Part 2). Its chronology is less well understood,
but again it would be easy to overemphasise its geographical isolation.
Not only did the Romans make a treaty with the inhabitants of Orkney, even
that distant archipelago includes finds of Roman imports (Fitzpatrick 1989).
Gurness is one of a series of monuments in the Northern Isles which have been
interpreted as enclosed villages (D. Harding 2004: 116), for the broch itself is
at the centre of what would once have been a circular enclosure with two
concentric banks and ditches (part of the monument has been lost to coastal erosion). Both the enclosure and the broch have their entrances to the east,
and these are actually aligned on one another. In between the central structure
and the earthwork perimeter was a whole series of other buildings, whose
interpretation is far from clear; nor is the site record good enough to establish
their sequence of development (McKie 1995). Even so, the plan has such a unitary character that it was obviously conceived as a single design. Like other
brochs, the monument at Gurness was divided between a central space with a
hearth, and a series of compartments arranged radially around the inner wall.
Within this building there was also a curious feature which was interpreted as
a well. It was a complex structure reached down a flight of steps, but it may not
have played an entirely practical role, as recent excavation at Mine Howe on
the same island has investigated a similar feature. This was built at the centre of
a large earthwork enclosure and was associated with evidence of metalworking
and with deposits of human bone (Card and Downes 2003).
One feature of Gurness is quite remarkable, although a similar argument
might apply to other broch villages in Orkney and Shetland. The basic plan
of the site is a circle set within a larger circle. The interior of the central
monument is subdivided by a series of radial divisions, and the same is true
of the area in between its outer wall and the defensive earthworks which are
occupied by a whole series of less imposing buildings. The effect is rather like
that of a set of Chinese boxes, one inside the other, so that the entire complex
could be interpreted as a single enormous dwelling. Like the timber building at Navan Fort, it adopted the principle of the circular house and reproduced
its features on a monumental scale.
Both those sites were beyond the periphery of the Roman Empire, but the
last two examples were more closely integrated with Continental Europe. The
first was Hengistbury Head, a defended promontory on the coast of the English
Channel commanding the sheltered waters of Christchurch Harbour (Cunliffe
1987). The site was occupied throughout the Iron Age but played a major role
in two distinct phases (Fig. 5.21). At the beginning of the period it seems to
have been one of a series of coastal settlements which were actively engaged in
overseas trade during the Bronze Age/Iron Age transition. During this phase
the excavated part of the site was dominated by an unusually large round house,
comparable to those further inland on the Wessex chalk. Hengistbury Head
became even more important during the period in which southern Britain was
again involved in regular contact with the mainland, and at this stage it seems
to have acted as both a seaport and a production site. It enjoyed a wide range of
contacts extending from Northern andWestern France to southwest England,
and it was through this site that metals were exported to the Continent and
a variety of exotic commodities, chiefly wine, were introduced. Only a small
part of the site has been excavated, but it seems possible that the internal area
was reorganised at this time. There may have been a grid of square enclosures
on Hengistbury Head, and the same is evidenced much more clearly at a
nearby site that took its place, Cleavel Point on the shore of Poole Harbour
(Sunter andWoodward 1987: 44–124). Not far away at Green Island a massive
pier or jetty was built (Markey, Wilkes, and Darvill 2002). Southern Britain increasingly formed part of a wider network which extended well beyond that
island.
The final example is Silchester, a royal capital of the Late Iron Age which was
eventually replaced by a major Roman town (Fig. 5.22; Fulford and Timby
2000). It was bounded by a large earthwork enclosure. It enjoyed a wide
range of outside contacts extending to the Continent and was engaged in
craft production, including the minting of coins. These are features that occur
on other major sites of this period, some of which are interpreted as royal
residences. There were two main groups of Iron Age buildings, deeply buried
beneath the Roman levels. The first consisted of the fragmentary remains
of what were apparently circular houses of the kind that is discovered on
practically any settlement of this period. The second group, which probably
replaced them, consisted of rectangular buildings that were arranged in an
orderly layout and conformed to a grid of metalled roads. Therewere also traces
of a considerable palisaded enclosure which followed the same orientation. It
dates from the early first century AD. The buildings were associated with a
wide variety of artefacts which showed the close connections between the
inhabitants of Silchester and the Roman world. These extended to particular
forms of social behaviour, including the consumption of wine, to imported
foodstuffs and to Classical notions of hygiene and bodily adornment. Similar buildings have been found at other sites in lowland England and seem
to have been influenced by Roman prototypes, but Silchester is remarkable in
two other ways. The site was little used before the Late Iron Age and has
even been interpreted as a deliberate creation: a very different process from the gradual expansion of settlement that happened over much of southern
England. At the same time, the best parallels for the rectangular houses and the
grid of streets are not found in Britain but in the north of France (Pommepuy
et al. 2000). It seems as if new practices and new forms of communal life were
developing and that older ways of living were rejected.
That is surely the point at which to end this account. Prehistory and Roman
archaeology flow into one another, but a transformation of this kind requires
a new kind of study. It would be a fitting subject for another volume in this
series.