Bogucki P., Crabtree P. Ancient Europe 8000 B.C.-A.D. 1000: Encyclopedia of the Barbarian World. Volume 2
Подождите немного. Документ загружается.
Fig. 1. Aerial of Danebury showing the 1978 excavations in progess. PHOTOGRAPH BY BARRY CUNLIFFE. COURTESY OF THE DANEBURY
TRUST. REPRODUCED BY PERMISSION.
dates to the Late Bronze Age (c. 1000–700 B.C.);
it is joined by a linear earthwork boundary that has
been traced eastward for several miles toward the
valley of the River Itchen.
Excavations at Danebury began in 1969 and
continued annually until 1988. During the twenty
seasons of work the entrances were examined, the
earthwork circuits were sectioned, and 57 percent
of the interior of the main fortified area was totally
excavated. This work established that within the
Late Bronze Age enclosure, defined by the outer
earthwork, the first defense, probably a palisaded
enclosure, was erected in the sixth century
B.C. This
first enclosure was replaced a century or so later by
the inner earthwork, built originally as a massive
timber-faced rampart fronted by a deep ditch. At
this stage there were two gates. The earthworks and
gates underwent various phases of modification, the
most significant coming around 300
B.C., when the
rampart was heightened and reconstructed to have
a steeply sloping outer face fronted by a deep V-
sectioned ditch. From the bottom of the ditch to
the top of the rampart measured about 6 meters (20
feet). At this stage the southwest entrance was
blocked, and the east entrance began to be massive-
ly extended. In this later stage of its life the hillfort
was intensively occupied. The end came some time
in the first half of the first century
B.C., when the
gate was destroyed by fire, and there is some evi-
dence to suggest the slaughter of the inhabitants.
After this the enclosure continued to be used for an-
other fifty years or so, but activity was at a low level
and may have been linked to the continued use of
a temple complex in the center of the old settle-
ment.
Throughout its life from c. 500 to c. 50
B.C. the
hillfort was occupied. From an early stage a system
of roads was established with a main axial street run-
ning between the two gates. Even after the south-
west gate was blocked the street remained the main
axis. Other streets branched out from just inside the
main entrance and ran roughly concentrically
6: THE EUROPEAN IRON AGE, C. 800 B.C.– A.D. 400
230
ANCIENT EUROPE
around the crest of the hill. Amid the streets were
arranged circular houses, rectangular post-built
storage buildings, and a large number of storage
pits. Toward the center of the site, occupying a
prominent position directly visible from the en-
trance, was a cluster of rectangular buildings that
were probably the main shrines of the settlement.
There is, throughout the occupation, a sense of
order in the layout of the various buildings and ac-
tivities. In the early stage, when both gates were in
use, the main occupation zone lay to the south of
the main street, whereas the area to the north was
used mainly for storage. After the southwest gate
was blocked the order was reversed, suggesting that
a major conceptual change had taken place.
In the last two centuries or so of the settle-
ment’s life a rigorous order seems to have been im-
posed. The rows of four- and six-post storage build-
ings arranged along the streets were rebuilt many
times over on the same plots, whereas immediately
behind the ramparts—where the stratigraphical evi-
dence is particularly well preserved and the circular
houses cluster—it is possible to distinguish six
major phases of rebuilding. In this area individual
building plots can be distinguished. Although each
had a different structural history, their discrete spa-
tial identities were maintained, suggesting continu-
ity of ownership over a long period of time. Ar-
rangements of this kind indicate a high level of
centralized control.
The most frequently occurring structures with-
in the fort were storage pits, of which more than
one thousand have been examined. For the most
part they were probably used for the storage of seed
grain in the period between harvest and the next
sowing. Experiments have shown that, so long as
the pits were properly sealed and airtight, the seed
remained fresh and fertile. Evidence from many of
the pits indicates that propitiatory offerings were
made once the grain was removed, presumably to
thank the chthonic (earth) deities for protecting the
seed and in anticipation of a fruitful harvest. The of-
ferings vary but include sets of tools, pots, animals
complete or in part, and human remains.
Activities carried out within the fort included
ironsmithing, bronze casting, carpentry, wattle
work and basketry, the weaving and spinning of
wool, and the milling of grain. Additional evidence
points to the existence of complex exchange systems
involving the importation and redistribution of
goods, including salt from the seacoast, iron ingots,
and shale bracelets. The presence of a large number
of carefully made stone weights is clear evidence
that a system of careful measurement was in opera-
tion. In all probability the hillfort, in its developed
state, was a place where the central functions of re-
distribution were carried out to serve people living
in a much wider territory.
The excavation of a number of Iron Age settle-
ments in the landscape around Danebury showed
that, although a number of farms existed during the
early phase of the fort’s existence, after the major re-
construction c. 300
B.C. farmsteads for some dis-
tance around were abandoned. This coincides with
an increase in the density and intensity of occupa-
tion within the fort, the implication being that the
rural population coalesced within the defenses. Al-
though this may have been a response to a period
of unrest, it could equally be explained as a feature
of socioeconomic change resulting in a greater de-
gree of centralization.
See also Hillforts (vol. 2, part 6).
BIBLIOGRAPHY
Cunliffe, Barry. Danebury Hillfort. Stroud, U.K.: Tempus,
2003.
———. Danebury: An Iron Age Hillfort in Hampshire. Vol.
6, A Hillfort Community in Perspective. Council for
British Archaeology Research Report 102. London:
Council for British Archaeology Research, 1995.
B
ARRY CUNLIFFE
DANEBURY
ANCIENT EUROPE
231
THE EUROPEAN IRON AGE, C. 800 B.C.– A.D. 400
IRON AGE IRELAND
■
FOLLOWED BY FEATURE ESSAY ON:
Irish Royal Sites . . . . . . . . . . . . . . . . 239
■
Iron Age Ireland suffers from a paucity of sites and
serious dating problems, which makes it difficult to
construct a coherent framework within which to at-
tempt interpretation. Thus, the Iron Age lingers in
the long shadow of medieval Ireland; the abundant
and varied medieval literature and the rich and pro-
lific material culture of the medieval period have
strongly affected the interpretation of Iron Age ar-
chaeology. Increasingly, however, Iron Age archae-
ological research is being generated by archaeolo-
gists, formulated in archaeological terms, and
conducted using an array of archaeological meth-
ods, including aerial photography, geophysical sur-
vey, and underwater and wetland (i.e., peat bog) ex-
ploration. These research agendas do not ignore
medieval textual and archaeological evidence; rath-
er, they reflect increasing confidence that a coherent
framework for Iron Age archaeology can be con-
structed.
CHRONOLOGY
To begin with a note about terminology, “medi-
eval” is used here to distinguish the period from the
fifth century to c. 1500. In Irish writing, archaeolo-
gists normally employ the terms “early Christian”
for the fifth century
A.D. to A.D. 800, “Hiberno-
Norse” or “Viking” for
A.D. 800–1169, and “medi-
eval” starting with the Anglo-Norman invasions of
A.D. 1169–1172. For our purposes, we can think
of the Iron Age in terms of three periods bounded
by the Late Bronze Age, which ended c. 700
B.C.,
and the early Christian period. There is almost no
available data for the Early Iron Age, which spanned
c. 700–300
B.C. The Middle Iron Age, or La Tène
Iron Age, lasted from 300
B.C. into the first century
A.D. It was a time that saw major construction at
many sites and the appearance and development of
La Tène art, which flourished into the early Chris-
tian period. In the Late Iron Age, or Roman Iron
Age, contacts with the Roman world, especially
with Britain, began, as indicated by imports of vari-
ous goods. The earliest evidence of writing dates to
this time. The period ends with the first recorded
Christian missions, about
A.D. 431/432.
Archaeologists still depend heavily on conven-
tional dating by stylistic analyses and comparisons,
so this discussion will start there. The closing phase
of the Late Bronze Age, the Dowris phase, ended
c. 700
B.C. The first subsequent datable object is an
imported gold torc (neck ring) from Knock, Coun-
ty Roscommon, decorated in La Tène style and with
close parallels in the Rhineland from c. 300
B.C. A
hoard from Broighter, County Derry, includes a
gold torc with spectacular La Tène decoration,
which is dated approximately by another item in the
same collection, a gold necklace of Mediterranean
232
ANCIENT EUROPE
Selected sites in Iron Age Ireland.
origin from the first century B.C. or the first century
A.D. As the Roman Empire expanded into Gaul (in
the mid-first century
B.C.) and Britain (in mid-first
century
A.D.), increasing contact with the Roman
world resulted in the appearance in Ireland of well-
dated Roman goods, such as coins and pottery.
Coins are not plentiful, though, and most come
from isolated hoards, unrelated to sites, while
Roman pottery is rare.
Radiocarbon dating has been applied to the
Iron Age, of course, but for much of the period the
tree-ring samples used for calibration show little dif-
ference in amounts of residual radiocarbon over sev-
eral centuries. In consequence, dates are corre-
spondingly imprecise. Fortunately, however, the
dendrochronological sequence for Irish oak makes
it possible to date the felling of a tree accurately,
often to the exact year. The waterlogged conditions
necessary for the survival of wood, which are com-
mon in this region, make this technique applicable
to many Irish archaeological sites. The contrast in
precision between radiocarbon dating and dendro-
chronology is well illustrated at Navan, County Ar-
magh, where the base of a phase 4 central post has
survived. The radiocarbon date for this post is 380–
100
B.C., a range of 280 years. Dendrochronology
provided a felling date for this post of 95
B.C. (or
possibly early 94
B.C.).
IRON AGE IRELAND
ANCIENT EUROPE
233
SITE IDENTIFICATION
There are two major reasons why so few Iron Age
sites are known. The first, paradoxically, is the sheer
number of sites. The issue of ringforts, or raths, is
particularly important here, for there is hot debate
as to whether these enclosed farmsteads are all of
early medieval date or whether some may be of the
Iron Age. Of those that have been excavated and
that can be dated (many cannot), almost all are in-
deed early medieval. There are, however, some thir-
ty thousand ringforts, of which only about 1 percent
have been excavated—hardly a statistically adequate
sample. Moreover, there are other types of circular
sites of the same general size (e.g., henges, ring bar-
rows, and small monasteries) that are easily con-
fused with ringforts unless closely inspected.
The second reason is that field-walking survey
cannot be employed in this context. This method is
put to effective use in many parts of the world and
simply involves walking over plowed land, looking
for scatters of artifacts, typically, potsherds. In Ire-
land, however, a high percentage of farmland is
under pasture, and other large areas are covered by
blanket bog. Moreover, the Iron Age is virtually
aceramic, which means that there is virtually no
chance of finding diagnostic ceramics and little like-
lihood of finding diagnostic metal artifacts.
EARLY IRON AGE (C. 700–300
B.C.
)
Hardly any artifacts can been attributed to this peri-
od, and only two sites merit discussion. The first is
the crannog of Rathtinaun, County Sligo, where ex-
cavation showed a two-phase occupation. Phase 1
contained only Late Bronze Age Dowris-type arti-
facts, but phase 2 held both Dowris-type artifacts
and a few iron objects. Rathtinaun, then, appears to
bridge the Bronze Age and Iron Age and should
date to the eighth to seventh centuries
B.C. Radio-
carbon dates, however, indicate that the site was oc-
cupied no earlier than the fifth through second cen-
turies
B.C.
Second, there is site B at Navan. As at Rath-
tinaun, phase 3 artifacts were from the Dowris
phase, with only a few small iron objects. Phase 3 ra-
diocarbon dates, however, range from the fourth
century
B.C. into early A.D. times; since the end of
phase 3 was followed immediately by phase 4, dated
precisely to 95
B.C. (from dendrochronology), it is
virtually certain that phase 3 lasted until about 100
B.C. The problems posed by these two sites cannot
be resolved at present and so, by the same token, the
Early Iron Age remains singularly elusive.
MIDDLE IRON AGE (C. 300
B.C.
TO C.
A.D.
100)
The date of c. 300 B.C. for the start of this period
is based, as noted, on the first appearance of the La
Tène art style. Nearly all the Iron Age La Tène dec-
orated objects in Ireland are found on the northern
half of the island. The development of La Tène art
in this area owes much to close contacts with Wales
and northern Britain, just across the Irish Sea. Irish
craft workers, however, were not mere imitators, for
they produced their own variations of British types
as well as some artifact styles unique to Ireland, such
as Y-shaped objects, Monasterevin disks, Petrie and
Cork crowns, and the so-called latchets. As else-
where in Europe, La Tène art was displayed mainly
on high-status personal metalwork. There are also
numerous bronze horse bits, several in pairs, sug-
gesting that the two-horse chariots so well known
from Iron Age Britain and the Continent were used
in Ireland as well. Some of the enigmatic Y-shaped
pieces also occur in pairs and may be components
of chariot harnessing. Iron spearheads are known, as
are fine bronze spear butts.
To judge by several beautifully decorated
bronze scabbards, however, swords were the war-
riors’ pride. Stylistically, they derive from Continen-
tal swords of the third through second centuries
B.C.
The Irish ones are much shorter—the blades rang-
ing from 37 to 46 centimeters; one wonders how
they could be used, except as long daggers. Of all
the scabbards and swords, only one sword comes
from a securely dated context—the excavation at
Knockaulin, probably from the first century
B.C. or
first century
A.D.
Although most of La Tène art finds expression
on metal items of personal equipment or adorn-
ment, there are five La Tène decorated stones; the
one at Turoe, County Galway (fig. 1), is embel-
lished most adeptly. There are also numerous
querns (grindstones) with La Tène decoration.
Many carved stone heads are attributed to the Iron
Age, but they bear only the vaguest stylistic resem-
blance to Iron Age human representations else-
where.
6: THE EUROPEAN IRON AGE, C. 800 B.C.– A.D. 400
234
ANCIENT EUROPE
Almost all decorated metalwork has been dis-
covered accidentally, much of it taken from bogs
and lakes. The practice of votive deposits also is
known in Britain and on the Continent. In those
places, decorated metalwork also appears in burials,
however, providing good associations and dating
evidence. In Ireland few burials contain such arti-
facts, and they are virtually absent from the few ex-
cavated sites, which makes it doubly difficult to date
them or to relate them to other aspects of Iron Age
life (and death).
The major sites of the Middle Iron Age are the
so-called royal sites. Their commanding locations
and large sizes imply that they were the most impor-
tant sites of the Middle Iron Age, dominating ritual
and ceremonial life over considerable areas. Despite
their prominence, they have yielded no deposits of
high-status valuables. Such items seem to have been
reserved for watery places. Significantly, four bronze
trumpets with La Tène decoration (and, reportedly,
human skulls) were found in the nineteenth century
in Loughnashade, a small lake just below Navan.
One remarkable exotic import was discovered in a
late phase 3 context at Navan (site B), however.
This was the skull of a Barbary ape (with a radiocar-
bon date of 390–20
B.C.), which certainly had trav-
eled a very long way from its homeland in north-
western Africa.
The Dorsey, County Armagh, is a very large, ir-
regular enclosure about 30 kilometers south of
Navan. Parts of it run across bog, which preserved
timbers from its construction. Dendrochronologi-
cal dates from these timbers show two phases of
building, the first between 159 and 126
B.C. and the
second between 104 and 86
B.C. The Dorsey lies
close to a section of the Black Pig’s Dyke, a series
of linear earthworks running east to west across Ire-
land. This set of earthworks may have marked the
southern boundary of Iron Age Ulster, for one sec-
tion of the dyke is dated by radiocarbon to 390–70
B.C. Other linear earthworks in Ireland may be of
the Iron Age also, but none are dated. Trackways
constructed across bogland have been dated to the
Iron Age by dendrochronology. The best known of
these is Corlea, County Longford, where excavation
uncovered two stretches of road over 2 kilometers
long, with dates of 156 ± 9
B.C. and 148 B.C. Con-
struction required two hundred to three hundred
mature oak trees, besides other species.
Fig. 1. Turoe Stone, County Galway, Ireland. A superb
example of La Tène art on a granite boulder. COURTESY OF
BERNARD WAILES. REPRODUCED BY PERMISSION.
Hillforts are a prominent feature of Iron Age
landscapes over much of western Europe, so the
sixty to eighty hillforts in Ireland conventionally
have been assigned to this period. Of the few exca-
vated so far, however, most appear to be Late
Bronze Age rather than Iron Age. Moreover, they
are very diverse in size and form. Some are so com-
pact that they could be seen as substantial ringforts
or cashels on hilltops, some are large and rambling
in plan, and some have ramparts so small (as little
as 1 meter high) that probably they were not forts
at all. Whether there are really Iron Age hillforts in
Ireland is moot. Of the estimated 250 known coast-
al promontory forts, a few have been excavated, but
only Dunbeg, County Kerry, has any dating evi-
dence—a radiocarbon date from the first few centu-
ries
A.D., probably Late Iron Age or even early me-
dieval, rather than Middle Iron Age.
IRON AGE IRELAND
ANCIENT EUROPE
235
Residential sites are very scanty indeed. One site
under a ringfort at Feerwore, County Galway, pro-
duced a few artifacts for which dating to the second
to first century
B.C. has been suggested. Two coastal
shell-midden sites have radiocarbon dates placing
them in the Middle Iron Age, as do two crannogs
at Lough Gara, County Sligo. There is one small
ringfort known for the period, at Lislackagh, Coun-
ty Mayo, where internal circular structures were ra-
diocarbon dated to 200
B.C. to A.D. 140. A handful
of other sites have dates overlapping both the Mid-
dle and Late Iron Ages. Despite the limited evi-
dence for daily life in the Middle Iron Age, it is clear
that major constructions were undertaken, which
implies the mobilization of substantial groups of
skilled labor. Particularly noteworthy is the practi-
cally simultaneous construction of phase 4 at Navan
(95
B.C.) and the later phase of building at the Dor-
sey (104–86
B.C.). The proximity of these two sites
suggests that one authority might have directed
construction at both.
LATE IRON AGE (C.
A.D.
100 TO
C. 550
A.D.
)
There is no obvious demarcation between the Mid-
dle and Late Iron Ages. Roman material began to
appear during the first century
A.D., possibly as early
as the first century
B.C. It is not until the late first
century
A.D., however, that evidence appears of
close (though not necessarily intense) contact with
the Roman world, so an arbitrary date of c.
A.D. 100
seems suitable. The main issue for consideration is
the extent to which interaction with the Roman
world promoted changes in Irish society.
J. Donal Bateson has reviewed Roman materials
in Ireland in detail, and the total is surprisingly
small, considering Ireland’s proximity to Roman
Britain and Gaul. Clearly, Roman goods were not
reaching Ireland in anything like the quantities that
reached, say, Germany and the southern Baltic dur-
ing the same period. Roman imports into Ireland
fall into two chronological groups, the first through
second centuries and the fourth through fifth centu-
ries. There is very little third-century Roman mate-
rial, perhaps reflecting the widespread economic
contraction of the period, demonstrated, for exam-
ple, by the contraction of trade from the Continent
to Britain. The material in the earlier category con-
sists mainly of coins and fibulae (brooches) and very
small amounts of Gaulish Samian (terra sigillata)
pottery. The objects in this group and their contexts
are reasonably consistent with trade and small-scale
contacts. The later group, of the fourth through
fifth centuries, also includes coins but has significant
quantities of silver in the form of ingots and hack-
silver (silver artifacts cut into pieces). These items
look suspiciously like the result of successful raiding,
and we know from Roman sources of this period
that the Irish (or Scotti) participated in the frequent
barbarian raids on Roman Britain.
There are a very few burials in Roman style. A
cremation in a glass container at Stonyford, County
Kilkenny, from the first or early second century
A.D.,
and an inhumation cemetery at Bray, County Wick-
low, from the second century
A.D. both show famil-
iarity with Roman burial practices of the time. Pre-
sumably, these are the burials of either Roman
immigrants or emigrants returned from the Roman
world. Grave goods from the small inhumation
cemetery on Lambay Island, County Dublin, show
close affinities with items from northern Britain in
the late first century
A.D., and the people may have
been British refugees from the Roman conquest.
Inhumation burial with the body extended appears
to have become increasingly common through the
Late Iron Age, and some such burials are in long
cists (graves lined with stone slabs). Because extend-
ed inhumation burial began to replace cremation
from about the second century
A.D. in the Roman
Empire, the same shift in Ireland may reflect Roman
practice. Dating Irish burials is seriously hampered
by the general lack of grave goods, however.
Two other disparate examples of Roman con-
tact come from Golden, County Tipperary, and
Lough Lene, County Westmeath. At Golden there
was a small Roman oculist’s stamp of slate, inscribed
along one edge, and at Lough Lene part of a flat-
bottomed boat of Mediterranean construction was
found. It is assumed to be of Roman date, although
its radiocarbon date is 300–100
B.C. (This, of
course, dates the growth of the wood and not nec-
essarily the boat’s construction.)
There are few remains of residential sites from
the Late Iron Age. Traces of occupation from be-
neath two ringforts have been radiocarbon dated to
the third through seventh centuries
A.D., whereas
dates from several structures on Mount Knock-
narea, County Sligo, range from the first century
6: THE EUROPEAN IRON AGE, C. 800 B.C.– A.D. 400
236
ANCIENT EUROPE
B.C. to the seventh century A.D. A sherd of Gaulish
terra sigillata pottery of the first century
A.D. was
plowed up at the large coastal promontory fort of
Drumanagh, County Dublin. This find has fueled
suggestions that this site may have been a trading
station, and the proximity of Lambay Island, with
its cemetery of possible British refugees, lends cre-
dence to the theory.
At Tara, County Meath, the Rath of the Synods
has yielded intriguing evidence. The finds suggest
that the site had four phases of occupation: the first
and third were small cemeteries, while the second
and fourth were probably residential. Artifacts in-
cluded some items of Gaulish terra sigillata of the
first to second centuries
A.D., a lead seal, glass beads,
and iron padlocks. All the datable objects fall within
the first to fifth centuries
A.D. It is striking that al-
though several objects certainly or probably are im-
ports from the Roman world, none are definitely of
Irish manufacture. This, then, is the most “Roman”
site known in Ireland, but it assuredly does not con-
form to any type of actual Roman site. The location
of the Rath of the Synods at a royal site must surely
be significant, but how this site should be interpret-
ed is unclear.
Toward the end of the Late Iron Age, perhaps
in the fourth century
A.D., the first indications of na-
tive Irish literacy appear in the form of ogham in-
scriptions, in which letters of the alphabet are de-
noted by different combinations of vertical or
oblique strokes. The model for an alphabetic script
presumably was Roman, and its employment on
memorial stones also echoes Roman usage. There is
no space here to debate the vexed issue of when
the Irish language first entered Ireland, but these
ogham inscriptions are the earliest written evidence
for the language. The script also demonstrates the
presence of Irish settlers in western Britain, where
ogham inscriptions (many duplicated in Latin) date
to the fifth and sixth centuries, particularly in Wales
and southwestern Britain.
DISCUSSION
The picture of Iron Age Ireland sketched here is one
dominated by a welter of unassociated objects from
chance discoveries, which can be organized into a
somewhat murky picture only with difficulty. It is
striking that the only really coherent archaeological
evidence of Iron Age Ireland comes from larger-
scale excavations, such as those of wetland areas and
royal sites. Even so, it is still virtually unknown
where and how people lived. It is no wonder that
the abundant historical and archaeological evidence
of early medieval Ireland, highly visible and largely
comprehensible, still casts such a long interpretative
shadow over the Iron Age.
The traditional or “nativist” view sees Iron Age
Ireland essentially as a pagan version of Christian-
ized early medieval Ireland. Thus, the society de-
picted in the medieval law tracts, for example, pro-
vides a template for Iron Age society: the higher
ranks, supported by clients and slaves, lived in ring-
forts, crannogs, and cashels and spent most of their
time planning cattle raids. This view is epitomized
by Kenneth Jackson’s Oldest Irish Tradition: A
Window on the Iron Age, an analysis of the Táin Bó
Cúailnge (“Cattle Raid of Cooley,” the central tale
of the Ulster Cycle of stories). The Táin is an ac-
count of the raid, organized by Queen Medb
(Maeve) of Connacht, to capture the famous brown
bull of Cooley in Ulster. In this epic, war chariots,
druids, single combat between champions, and cat-
tle raiding are prominent. Jackson argued that these
elements of the tale identified a genuine Iron Age
oral epic, eventually written down in the eleventh
century
A.D. Moreover, Medb and her counterpart,
the king of Ulster, lived at identifiable sites—
respectively, Cruachain (Croghan) and Emain
Macha (Navan)—which seems to add authenticity.
The nativist position has come under revisionist
fire from both historians and archaeologists. Fur-
ther textual analysis of the Ulster Cycle shows that
it was largely a medieval composition by writers fa-
miliar with Latin literature, Greek epics, the Scrip-
tures, and writings of the early church fathers. Simi-
larly, increasingly fine-grained analyses of the
aforementioned law tracts show that they were al-
most certainly composed by monks with a Christian
agenda, rather than by secular scholars perpetuating
traditional pre-Christian law. The excavation of two
of the royal sites since Jackson’s work was published
shows that there are no satisfactory grounds for re-
garding them as the royal residences portrayed in
the Táin. More specifically, Mallory has pointed out
that the swords described in the Táin were long, re-
sembling medieval swords not the very short swords
of Iron Age Ireland.
IRON AGE IRELAND
ANCIENT EUROPE
237
The revisionists contend that the country un-
derwent a major transformation through the centu-
ries of contact with Rome, culminating in conver-
sion to Christianity and the consequent intro-
duction of literacy. In this scenario the Iron Age is
seen as a depressed period when agricultural and
pasture lands contracted, as shown by an increase of
tree pollen in several pollen diagrams from different
parts of Ireland. This contraction began in about
the seventh century
B.C., perhaps intensified around
200
B.C., and continued until about the third centu-
ry
A.D., when woodland clearance recommenced.
This renewed clearance has been attributed to the
introduction of the plow with iron share and coulter
and of dairying, through contact with Roman Brit-
ain. It is thought that productivity of both tillage
and livestock thus improved considerably, which in-
creased the wealth of the upper classes and enabled
them to invest in clients and to buy slaves. In this
way, so the hypothesis has it, the rural economy and
society that were so well documented in the early
medieval period were triggered by innovations from
the Roman world.
We have no satisfactory dating for the appear-
ance of the iron share and coulter, however, and the
introduction of dairying is the subject of controver-
sy. Pam Crabtree has argued that the mortality pat-
tern of cattle bones from Knockaulin, probably dat-
ing to the first century
B.C. or the first century A.D.,
is consistent with dairying. Finbar McCormick dis-
puted this analysis and went on to propose the hy-
pothesis that dairying was introduced through
Roman contacts (i.e., later than the Knockaulin as-
semblage). In addition, he argued that ringforts—
those typical enclosed homesteads of the earlier me-
dieval period—were developed specifically to pro-
vide protection for valuable dairy cattle. Milk
residues have been identified, however, in British
prehistoric pottery. Since this pottery is as old as the
Neolithic (fourth through third millennia
B.C.), it is
plausible to propose that dairying was introduced to
nearby Ireland in prehistoric times. Clearly, this de-
bate will continue.
The nativist and revisionist positions are not
completely incompatible: the former does not deny
that the conversion to Christianity promoted sub-
stantial changes in Irish society, nor does the latter
deny some continuity from Iron Age to early Chris-
tian Ireland (e.g., La Tène art). As archaeological
evidence gradually accrues, and textual analysis is
pursued, interpretations will improve.
See also Milk, Wool, and Traction: Secondary Animal
Products (vol. 1, part 4); Trackways and Dugouts
(vol. 1, part 4); Bronze Age Britain and Ireland
(vol. 2, part 5); Irish Bronze Age Goldwork (vol. 2,
part 5); La Tène Art (vol. 2, part 6); Irish Royal
Sites (vol. 2, part 6); Early Christian Ireland (vol. 2,
part 7); Raths, Crannogs, and Cashels (vol. 2, part
7).
BIBLIOGRAPHY
Bateson, J. Donal. “Further Finds of Roman Material from
Ireland.” Colloquium on Hiberno-Roman Relations
and Material Remains. Proceedings of the Royal Irish
Academy 76C (1976): 171–180.
———. “Roman Material from Ireland: A Reconsidera-
tion.” Proceedings of the Royal Irish Academy 73C
(1973): 21–97.
Copley, M. S., R. Berstan, S. N. Dudd et al. “Direct Chemi-
cal Evidence for Widespread Dairying in Prehistoric
Britain.” Proceedings of the National Academy of Science
100, no. 4 (February 18, 2003): 1524–1529.
Crabtree, Pam. “Subsistence and Ritual: The Faunal Re-
mains from Dún Ailinne, Co. Kildare, Ireland.”
Emania 7 (1990): 22–25.
Fredengren, Christina. “Iron Age Crannogs in Lough
Gara.” Archaeology Ireland 14, no. 2 (2000): 26–28.
Jackson, Kenneth H. The Oldest Irish Tradition: A Window
on the Iron Age. Cambridge, U.K.: Cambridge Univer-
sity Press, 1964.
Kelly, Fergus. A Guide to Early Irish Law. Early Irish Law
Series, no. 3. Dublin, Ireland: Dublin Institute for Ad-
vanced Studies, 1988.
McCone, Kim. Pagan Past and Christian Present in Early
Irish Literature. Maynooth Monographs, no. 3. May-
nooth, Ireland: An Sagart, 1990.
McCormick, Finbar. “Cows, Ringforts, and the Origins of
Early Christian Ireland.” Emania 13 (1995): 33–37.
———. “Evidence of Dairying at Dún Ailinne?” Emania 8
(1991): 57–59.
McManus, Damian. A Guide to Ogam. Maynooth Mono-
graphs, no. 4. Maynooth, Ireland: An Sagart, 1991.
Mallory, James P. “The World of Cú Chulainn: The Archae-
ology of Táin Bó Cúailgne.” In Aspects of the Táin. Ed-
ited by James P. Mallory, pp. 103–159. Belfast, North-
ern Ireland: December Publications, 1992.
———. “The Sword of the Ulster Cycle.” In Studies on
Early Ireland: Essays in Honour of M. V. Duignan. Ed-
ited by Brian G. Scott, pp. 99–114. Dublin, Ireland: As-
sociation of Young Irish Archaeologists, 1981.
6: THE EUROPEAN IRON AGE, C. 800 B.C.– A.D. 400
238
ANCIENT EUROPE
Megaw, M. Ruth, and J. V. S. Megaw. Celtic Art: From Its
Beginnings to the Book of Kells. Rev. ed. New York:
Thames and Hudson, 2001.
Mitchell, Frank, and Michael Ryan. Reading the Irish Land-
scape. 3d ed. Dublin, Ireland: Town House, 1997.
Raftery, Barry. Pagan Celtic Ireland: The Enigma of the Irish
Iron Age. London: Thames and Hudson, 1994.
Thomas, Charles. Celtic Britain. London: Thames and
Hudson, 1986.
Waddell, John. The Prehistoric Archaeology of Ireland. Gal-
way, Ireland: Galway University Press, 1998.
B
ERNARD WAILES
■
IRISH ROYAL SITES
The Irish “royal sites” are so called because medi-
eval Irish scholars believed them to have been the
capitals of pre-Christian high kings of four of the
five ancient provinces of Ireland. Croghan
(Cruachain) was the royal site of Connacht, Navan
(Emain Macha) of Ulster, Tara (Temair) of Meath,
and Knockaulin (Ailenn, Dún Ailinne) of Leinster.
No early source identifies a royal site for Munster.
Various medieval texts refer to the royal sites as for-
mer royal residences and burial grounds; venues for
major assemblies, including the inauguration of
kings; and centers of pagan ritual. Although these
sites were invoked as symbols of kingship in medi-
eval Ireland, there is no evidence that they actually
were used during the Middle Ages, and the retro-
spective nature of medieval references to these sites
demands caution in assessing their original func-
tions or significance. Archaeology can provide a
firmer understanding, and Knockaulin, one of the
two extensively excavated sites (with Navan), can
serve as an exemplar.
At Knockaulin an oval earthwork encloses c. 13
hectares, with the entrance on the east side of the
site. Despite the hilltop location, it was not a defen-
sive site, for the bank is outside the ditch. Geophysi-
cal survey showed substantial anomalies only
around the center of the site, where subsequent ex-
cavation produced the following (simplified) se-
quence:
Flame (latest): Low mound of burned material,
including many animal bones, which sug-
gests periodic feasting
Dun: Central tower and circle of posts disman-
tled; stone slabs and earth laid over the re-
stricted area of Emerald-phase burning
Emerald: Perimeter wall of Mauve phase dis-
mantled, but central tower and inner circle
of posts left standing, despite intense local-
ized burning
Mauve: Double-walled, circular timber struc-
ture, c. 42 meters in diameter, enclosing a
circle, 25 meters in diameter, of freestand-
ing posts and, at the center, a heavily built
timber structure, c. 6 meters in diameter
and with buttresses, that may have been a
wooden tower
Rose: Figure-eight, triple-walled timber struc-
ture with a larger circle, c. 35 meters in di-
ameter, and an elaborate, funnel-shaped en-
tranceway; structure dismantled to make
way for Mauve structures
White: Circular, single-walled timber structure,
c. 22 meters in diameter; dismantled to
make way for Rose structures
Tan (earliest): Neolithic trench and artifacts
(fourth millennium through third millenni-
um
B.C.)
None of the Iron Age structures (White
through Mauve) show evidence of residential or fu-
nerary use and must be interpreted as ritual or cere-
monial in nature. The White, Rose, and Mauve en-
trances are oriented toward sunrise around 1 May,
the festival of Beltane, the beginning of summer.
Radiocarbon dates (Rose through Flame) cluster
between the third century
B.C. and fourth century
A.D., while stylistic parallels for metalwork are main-
ly of the first century
B.C. to the first century A.D. An
8-meter-wide roadway runs through the site en-
trance toward the timber structures at the center of
the site. A radiocarbon sample from sod buried be-
neath one of the banks at the site entrance suggests
that bank construction took place in the fifth centu-
ry
B.C.
The other royal sites share several characteristics
with Knockaulin. First, all are on prominent elevat-
ed locations with commanding views. Second, all
have large enclosures. Those at Navan (c. 5 hect-
ares) and the Ráith na Ríg (Rath of the Kings; c. 6
hectares) at Tara both have internal ditches and ex-
ternal banks. Geophysical survey at Croghan shows
a circular anomaly enclosing nearly 11 hectares,
IRISH ROYAL SITES
ANCIENT EUROPE
239