Multi-ditched hilltop enclosure

The most striking discovery made by geophysical survey is a roughly circular hilltop enclosure measuring around 400 m in overall diameter, which formed the primary focus of the excavations described below. Approximately 60% of the circuit of the enclosure, which is defined by multiple concentric boundaries and encompasses an area of c. 12.7 ha, has been mapped. A large section on the west has been cut away by the quarry, while most of the north-east quadrant is on lands that were inaccessible at the time of survey. Excavation has confirmed that the three principal enclosing elements comprise substantial, partly rock-cut, ditches, 1.5–2 m deep and 3.2–3.6 m in maximum width.

The inner enclosure is demarcated by a single ditch, with a causeway entrance on the east-north-east, and has an internal diameter of 270 m (F3 on Fig. 9). Surrounding it, at a distance of c. 50 m, is an outer enclosure defined by two ditches, set 7–8 m apart (F1), inside of which, on the north-east and east, are segments of a narrower ditch or slot trench (F2). Well-defined gaps in the circuits of both ditches, on the east-north-east and west-south-west, likely represent original (causewayed) entrances in the outer enclosure. Those on the east-north-east correspond with the gap in the inner enclosure, which may likewise have had an opposing entrance on the south-west that has been cut away by the quarry. This shared alignment, coupled with the concentricity of the inner and outer enclosures, suggested them to be contemporary and – based on size and apparent morphological parallels – broadly comparable to a Late Bronze Age hillfort (450 m in diameter) excavated at Rahally, near New Inn, Co. Galway (Mullins Reference Mullins, O’Sullivan and Stanley2008; Reference Mullins2009; Reference Mullins, McKeon and O’Sullivan2014). Interestingly, these parallels extend to similarities in entrance orientation and movement through these enclosures focused on a north-east to south-west axis, the frequent occurrence of which at later prehistoric and early medieval high status sites suggests a deep rooted symbolic significance (Gleeson Reference Gleeson2012).

Possible palisaded enclosure

Traces of another possible large enclosure (F4) were identified just inside the line of the inner hilltop enclosure ditch (F3), with which it is partly contiguous. It is defined geophysically by a pair of narrow – and in places discontinuous – parallel lineations, which form a broad, irregular arc around the eastern and southern sides of the hilltop, from the inner edge of the enclosure ditch (F3), on the north-east, to the modern quarry. Excavation of a section of one of the linear features has confirmed that it is a slot trench and pre-dates the enclosure ditch by more than 2000 years (see C.110, below). Although not discernible elsewhere, this double slot trench likely represents part of a fenced enclosure that once surrounded the hilltop, the northern sector of which may have been subsumed by the hilltop enclosure ditch or truncated by later activity. The manner in which the enclosure ditch flattens out on the south seems to suggest, moreover, that a deliberate effort was made to accommodate it (and perhaps to avoid some other feature/s in this area). However, such an interpretation is difficult to reconcile with the lengthy timespan separating the putative fenced enclosure and the hilltop enclosure, which makes it highly improbable that any surface indication of the former survived when the hilltop enclosure was constructed.

Summit features

Most of the remaining archaeological features revealed by survey are concentrated on and around the relatively small, level area of the summit. Notable among them are three well-defined, circular and sub-circular features measuring c. 10–12 m in diameter. These ring-ditches are, in the main, interpreted as funerary monuments – though the uniform circularity of one example (F5) would also be consistent with a wooden structure or building. The interpretation of this feature as a possible house slot trench is perhaps supported by its location inside, and adjacent to, a semi-circular ditch that may define the northern part of an enclosure (F6). One of the ring-ditches also overlaps with the boundary of the latter enclosure which, if projected as a full circle, would have similar dimensions to a 35 m diameter circular enclosure (F7) identified a short distance to its south.

There are faint suggestions of small circular features in the interior of the latter enclosure, which is also bisected north/south by one of a pair of linear features (F8) that splay outwards across the line of the inner hilltop enclosure and continue beyond the limits of the survey. Though not visible topographically, cartographic and aerial photographic evidence suggests that these lineations may terminate at or near the outer boundary of the hilltop enclosure, raising the possibility that they formed some kind of funnel-shaped ‘avenue’ leading to the centre of the enclosure. However, the lack of a corresponding entrance in the inner enclosure ditch at this point, and their depiction as part of a woodland on the Ordnance Survey map of c. 1885 (on which the hilltop enclosure is not marked), may indicate a later date and function for these linear features. Similar uncertainties surround a large, curvilinear feature (F9) in the north-east quadrant of the inner enclosure, which is also shown on historical mapping and survives as a shallow depression, c. 1 m in width.

Settlement enclosure & field system

Geomagnetic survey over an area of relatively flat, low-lying ground on the north-east flank of the hill, some 50 m downslope of the hilltop enclosure, revealed the footprint of what appears to be a settlement enclosure and associated field system. The enclosure (F10) is roughly circular in plan and is represented by two concentric ditches measuring about 35 m and 45 m in maximum diameter, respectively. The strong magnetic response of the infilled ditches suggests they may contain significant quantities of burnt material. No clear evidence of an entrance was identified, though this may lie on the east or south where the circuit of the enclosure is less well defined. Several curvilinear features, possible pits, and other traces of activity were recorded in the interior, including the possible foundation trench of a (circular?) building c. 10 m in diameter (F11), which adjoins the enclosure boundary on the north. Based on its morphology and evidence for internal structures, this enclosure may comprise the levelled remains of a bivallate ringfort (settlement enclosure) of early medieval date.

Adjoining the enclosure on the east are several linear and arcuate ditch-type features that appear to form part of an associated field system (F12). They extend beyond the limits of the survey and are potentially related to a group of sub-rectangular and linear features (F13) just outside the north-east entrance to the hilltop enclosure.

Later agricultural activity appears to be signalled, moreover, by a series of north-east to south-west oriented linear features that extend across the northern part of the survey area. One of these (F14) seems to truncate, and therefore post-date, the boundary of enclosure F10 on the north.

Industrial activity

Lastly, a scatter of strong, amorphous magnetic responses (F15) was recorded in the vicinity of the kiln, at the southern foot of the hill. These may derive from burnt/fired material associated with the production of lime mortar in recent centuries.

Finds & samples

Dry sieving was undertaken on site using various sampling strategies to facilitate the recovery of small artefacts and ecofacts. Bulk soil samples were also retrieved from archaeological layers for subsequent processing and environmental analysis.

Bulk soil samples, charcoal, and animal bone from the excavated features were prepared and sent for specialist analysis to obtain suitable material for dating and other relevant data. The faunal assemblage, consisting of a total of 243 fragments of mammalian bone and teeth, was analysed by Fiona Beglane (Beglane Reference Beglane2018). No human remains were identified. The soil samples were processed using a flotation device and the remains recovered were sorted and identified by Ingelise Stuijts (Stuijts Reference Stuijts2018).

The finds recovered during the excavation were dominated by a lithic assemblage comprising 193 artefacts of probable Neolithic date, which were analysed by Killian Driscoll (Driscoll Reference Driscoll2018; Figs 15–17, below). Two fragments of possible metalworking waste retrieved during the excavations were examined by Paul Rondelez (Rondelez Reference Rondelez2018). A possible whetstone, a single sherd of medieval pottery, and three sherds of post-medieval pottery were also found. The results of specialist analyses are incorporated into the discussion below.

Radiocarbon dating

A total of 11 samples from seven archaeological features was submitted to the 14Chrono Centre, Queens University Belfast for AMS radiocarbon dating. The results are presented in Figure 13 and Table 1. The majority of dates were obtained from short-life samples, including charred cereal grain, animal bone, and wood charcoal from short-lived species/twigs. All calibrated radiocarbon determinations discussed in the text are cited at two sigma (95.4% probability; Reimer et al. Reference Reimer, Austin, Bard, Bayliss, Blackwell and Bronk Ramsey2020).

Table 1: Radiocarbon results from faughan hill

The dates have been calibrated using OxCal v4.4 (Bronk Ramsey 2021: http://c14.arch.ox.ac.uk/) and the INTCAL20 calibration curve (Reimer et al. Reference Reimer, Austin, Bard, Bayliss, Blackwell and Bronk Ramsey2020)

Fig. 13. Probability distributions of radiocarbon dates from Faughan Hill

Upper soil layers

The removal of a thin stratum of humic material (C.101/C.201) and the underlying sod (C.102/C.202) revealed a compact band of gravelly clay (C.103/C.203) c. 0.05 m in average thickness. This layer, which extended across much of the excavated area and was heavily stained with iron oxides leached from the upper sod horizon, likely relates to bulldozing undertaken during woodland clearance in the 1960s. The presence of a large number of lithics – mostly flakes but also including a scraper and several blades and cores – within the matrix of C.103/C.203 suggests that the mechanical spreading of soils and gravel disturbed an early prehistoric horizon that extended beyond the limits of the excavation. Related activity is indicated by C.205, a light yellowish-brown clay c. 0.35 m thick and 1.40 m in north-east to south-west extent, which probably represented a backfilled tree throw.

Beneath the topsoils, and sealing the archaeology, was a series of relatively homogeneous soil deposits that exhibited considerable variation in thickness, measuring c. 0.15 m in Trench 1 and ranging from c. 0.20 m to as much as 0.60 m at the upper (south-west) end of Trench 2. A former ploughzone was indicated in both trenches by a dark, yellowish-brown silty clay (C.104/C.204 & C.227), 0.10–0.40 m thick, containing a large number of lithics, a single sherd of medieval Dublin-type ware, several sherds of post-medieval Brownware pottery, and charcoal flecking, all of which were likely displaced and spread by ploughing. C.104 was cut by a modern stone-filled drain (C.105) and sealed all archaeological features in Trench 1. In Trench 2, the ploughzone overlay two soil layers that were only discernible in the south-western end of the cutting. C.206 was a medium yellowish-brown sandy loam that lay directly below C.204 and partly extended over one of the infilled ditches of the outer enclosure (Enclosure Ditch 2) where it was disturbed by an animal burrow. It was flecked with charcoal throughout and contained a number of lithics. Although the interface between C.206 and the underlying layer, C.224, was poorly defined, it was still possible to identify the latter as a distinct horizon characterised by a band of light yellowish-brown sandy loam averaging 0.20 m in depth. It, too, contained frequent charcoal flecking and a single chert flake, and was disturbed by animal burrows.

The variable thickness of the upper soil layers is likely due to a number of factors, including slope gradient and hillwash accumulation, as well as the spreading of soils through ploughing. Similarly, their relative homogeneity may stem from a combination of ploughing, earthworm mixing, and animal burrowing, possibly compounded by within-soil illuviation and localised colluviation. As a result, it was very difficult to recognise a distinct B-horizon, though C.224 in Trench 2 likely comprised the lower stratum of a sandy loam subsoil, albeit much affected by bioturbation processes.

In contrast, the C-horizon was well-defined. It consisted of a yellowish-grey glacial clay (C.116/C.236) with frequent gravel inclusions, which was overlaid in the south-western half of Trench 2 by a mottled, greyish-yellow sand (C.235). Where the soils were shallowest in Trench 2, the limestone and shale bedrock (C.237) was exposed at a depth of 0.20–0.30 m beneath the present ground surface. Here the bedrock comprised an irregular, eroded surface of angular and sub-angular stones (averaging c. 0.10–0.40 m in size), into which the enclosure ditches had been cut.

Trench 1

Slot trench of fenced enclosure: A stone-packed slot trench (C.110) with post settings was excavated in the upper (south-west) part of Trench 1 (Figs 18–20). Oriented north-west to south-east, this feature corresponds with the innermost of two closely spaced linear anomalies mapped by geomagnetic survey (F4 on Fig. 9), which appear to define the eastern portion of a large enclosure surrounding the summit of the hill. No trace of a second slot trench or other feature corresponding with the outer linear anomaly (which is less clearly defined geophysically and may be discontinuous) was revealed by excavation. The excavated segment of C.110 consisted of a sharp, irregular-shaped cut measuring c. 0.20 m in width and 0.15–0.20 m in depth. Four groups of 2–3 closely set, rounded post-holes (eight post-holes in total), 0.05–0.08 m in diameter and depth and aligned in a slightly angled pattern, were cut into the uneven base of the slot trench at c. 0.10 m intervals (see Fig. 20b). Based on the layout and proportions of the slot trench, it likely supported a relatively modest palisade fence, possibly consisting of angled fence ‘panels’ composed of upright stakes interwoven with wattle.

The slot trench contained a large number of sub-angular stones (C.112), presumably packing material, the largest measuring c. 0.15 × 0.20 m (see Fig. 20a). Intermixed with the stones was a moderately compacted yellowish-brown silty clay that contained carbonised cereal (wheat/barley) grain and oak charcoal. Overlying this was a medium-brown silty clay (C.111) interspersed with small stones and a few patches of heat affected clay which likely derived from elsewhere as there was no evidence of in situ burning. C.111 also contained a small quantity of charred wheat/barley as well as fragments of burnt hazelnut shell and some oak and hazel charcoal. Three flaked stone tools, all formed on chert, were recovered from the fills: a platform flake and a single platform core from C.111 and an edge-retouched platform flake from C.112.

A sample of charred wheat/barley from C.111 yielded a radiocarbon date of 3635–3380 cal bc (4754±29 bp; UBA-36120). This is the earliest date in the Faughan sequence and confirms the primary position of the slot trench/fenced enclosure with respect to the multi-ditched hilltop enclosure, as suggested by geophysical survey. The stratigraphic position of the lithics, particularly within the packing material (C.112), suggests they are broadly contemporary with the slot trench.

Enclosure ditch 1 (inner enclosure): The inner ditch (C.108) of the hilltop enclosure is defined by a broad, U-shaped cut, with a flat-bottomed, rock-cut base and moderately sloping sides (Figs 18–19 & 21). It measured c. 3.20 m in width at the top, narrowing to c. 1.50 m at the base, and was c. 1.60 m in maximum depth below the present ground surface. The upper part of the ditch was cut into the glacial clay, while the lower 0.30 m was quarried into the shale bedrock, on one side forming a step 0.25 m high above the flat base.

Six fills were recorded in the ditch (see Fig. 19). There was no evidence to indicate that the ditch had been recut or cleaned out, though the possibility of this having occurred cannot be ruled out. The primary fill consisted of moderately compacted, light orangy-grey silty clay (C.118), c. 0.20 m thick, that extended over the base of the ditch and rose c. 0.80 m along its outer (north-east) edge. It contained occasional limestone and shale fragments, a small quantity of oak and ash charcoal, and some animal bone, including fragments of mandible from a medium sized mammal and the partial root of a tooth. This layer may have derived from ditch-side weathering and slumping, possibly augmented by the inwash of sediment from an adjacent, internal rampart. Overlying C.118 was a light yellowish-brown silty clay (C.114) up to 0.35 m thick, which may likewise relate to ditch-side weathering and the partial collapse of bank material. It contained a number of small to medium sized fragments of limestone and shale (c. 0.05–0.12 m in size) and a small amount of charcoal (oak, ash, and hazel). C.114 also produced the largest quantity of animal bone in the Faughan assemblage, representing nearly 80% of the total fragment count. Almost all the identifiable remains were of cattle, apart from six deer teeth. A fragment of cattle/deer bone from this layer yielded a date of 1180–920 cal bc (2859±34 bp; UBA-36113). This broadly corresponds with the dates obtained from the inner ditch of the outer enclosure (Enclosure Ditch 2), discussed below, indicating construction/use in what is identified in Ireland as the Late Bronze Age.

Further episodes of possible bank collapse and infilling are indicated by the overlying layers, C.113 and C.109. The latter contained a small quantity of oak, blackthorn/sloe, and willow charcoal, as well as some tiny, unidentified fragments of burnt bone. A sample of charcoal from a twig-sized piece of willow from C.109 produced a radiocarbon date of 800–520 cal bc (2523±42 bp; UBA-36116), indicating activity during the final Bronze Age/Early Iron Age. This closely matches the dates obtained from corresponding levels of the outer ditches of the hilltop enclosure (see below).

The uppermost fills of the ditch, C.119 and C.120, were somewhat different in character, comprising moderately compacted layers of light to medium-brown clayey silt containing occasional small stones and flecks of charcoal. These soils may have been deliberately pushed into the ditch to level out the area for agriculture, possibly during the later Iron Age or early medieval period, or more recently.

Trench 2

Enclosure ditch 2 (inner ditch of outer enclosure): The inner ditch (C.226) of the outer enclosure has a steep, V-shaped profile and measured an impressive 2 m in depth below the present ground surface (Figs 22–24). It was 3.60 m wide at the top, narrowing to c. 0.50 m at the base. The upper part of the ditch was cut into the glacial clay, while the lower 0.50 m or so was quarried into the shale bedrock.

The ditch was found to contain a total of nine fills (Fig. 23). The basal fill consisted of yellowish-brown sandy clay (C.221) with occasional small stones. It was c. 0.20 m in maximum thickness and contained two chert flakes, a few pieces of oak charcoal, and several dozen fragments of animal bone. Meat consumption is indicated by the fragmentary condition of the bones, only three of which could be identified to species (two cattle and one sheep), and by the presence of chop marks on one fragment. A mammal long bone from this layer yielded a date of 1280–940 cal bc (2930±50 bp; UBA-36112). This calibrated range is older, and wider by approximately 100 years, than that produced by a sample from the secondary fill of Enclosure Ditch 1 (see above), but nonetheless implies that the ditches are broadly contemporary.

The basal fill (C.221) covered the base of the ditch and extended 1.10 m and 0.60 m up the inner and outer sides, respectively. It seems to have been truncated by an irregular, V-shaped (re)cut (C.219). This was filled with rounded and sub-angular stones (0.15–0.20 m in average dimensions) intermixed with dark brown silty clay (C.220) which contained some oak charcoal, a fragment of unidentified animal bone, two chert flakes, and a chert blade. This deposit was up to 0.50 m thick and may incorporate revetment material and sediment from an adjacent rampart. A sample of oak charcoal from this layer produced a date of 1260–1010 cal bc (2925±37 bp; UBA-36119), possibly suggesting that cleaning of the ditch ceased a short time after it was dug. However, given the potential for an age discrepancy due to old wood, clarity on the timing of these events is lacking.

The three overlying fills – C.238, C.215 and C.214 – were quite homogeneous in character, consisting in each case of largely sterile, light greyish-brown clay that differed only slightly in hue. Each of these layers contained frequent small stones and shale fragments, with some larger stones also recorded in the upper level of C.214. The latter also contained a chert flake and a single fragment of unidentified animal bone. An episode(s) of infilling appears to be indicated, potentially involving the spreading of bank material into the ditch.

Thereafter, the upper part of the ditch remained open for a time, allowing a lens of dark brown silty clay (C.213), c. 0.05 m thick, to accumulate. C.213 contained a small quantity of oak, blackthorn, ash and hazel charcoal, mainly derived from twig-sized wood (the hazel charcoal was indeterminate). As with the other enclosure ditches, a final Bronze Age/Early Iron Age horizon is indicated at this level, in this instance by a sample of blackthorn charcoal which produced a date of 780–540 cal bc (2499±27 bp; UBA-36115).

The uppermost fills of the ditch may represent levelling-up deposits (as in Enclosure Ditch 1), possibly introduced to even out the ground surface for agricultural purposes. These combined layers were up to 0.70 m thick, and included C.212, a dark brown silty clay, and an overlying layer of lighter brown clay (C.211). The latter contained occasional stone inclusions and charcoal flecking, as well as three chert flakes. Overlying C.211 on the north-west was a thin, gravelly band of brown silty clay (C.210) which may represent disturbed remnant bank material.

Enclosure ditch 3 (external ditch of outer enclosure): The outer enclosure ditch (C.234) was cut into the bedrock and was U-shaped in profile. It had steeply sloping sides, rising on the inner (south-west) face to form a broad ledge (c. 1.20 m wide), c. 0.70 m above the base of the ditch (Figs 22–23 & 25). The exposed bedrock on the upper, outer face of the ditch was heavily eroded. The ditch measured 3.40 m in width at the top, narrowing to 0.40–0.50 m at the base, and had a maximum depth of c. 1.50 m below the present ground surface.

A total of six fills were identified in the ditch (see Fig. 23). The basal fill, C.233, comprised dark brown silty clay with frequent small, rounded stones that accumulated along the bottom and lower inner edge of the ditch. Several larger stones (c. 0.10–0.15 m in size) also occurred near the bottom of this layer which, apart from a tiny fragment of oak charcoal, was archaeologically sterile. It is likely that this fill relates to primary erosion from the sides of the ditch and putative internal bank (see below), together with some inwash of silt, though the absence of any fine silt at its base raises the possibility that the ditch may have been cleaned. The single charcoal fragment recovered from C.233 produced a date of 2865–2510 cal bc (4107±28 bp; UBA-36121). This is highly anomalous considering the overall dating and stratigraphic evidence, as well as the similar form and layout of the three enclosure ditches, all of which suggests them to be contemporary with one another. As such, there is strong reason to suspect that the charcoal from C.233 is intrusive, and probably the result of contamination from a Late Neolithic horizon represented by several nearby features (see below).

Overlying C.233 was a layer of sterile greenish-grey clay (C.232), c. 0.30 m thick, containing frequent small- to medium-sized stones and shale fragments. The composition of this layer and the fact that it appears to have spread from the inner side of the ditch suggest that it derived from bank material. Above C.232, and likely representing collapse from the eroded outer ditch face, was a c. 0.20 m thick layer of shattered limestone and shale fragments (C.231) intermixed with sterile grey clay.

This was followed by the accumulation of a lens of dark brown silty clay (C.230), c. 0.03 m thick, containing a small amount of charcoal (oak, blackthorn, and unidentified bark) and a single cattle/deer tooth. Charcoal from a twig-sized piece of blackthorn in this fill yielded a date of 790–550 cal bc (2527±26 bp; UBA-36122), which corresponds closely with the dates obtained from mid-level deposits in the other enclosure ditches.

Thereafter, the rapid infilling of the ditch appears to be indicated by C.229, a compact, light greyish-brown silty clay, c. 0.50 m in maximum thickness, containing numerous shale fragments. This layer extended across the entire width of the ditch, sealing all the underlying deposits, and may be associated with the levelling of the internal bank. A small lump of black fayalitic slag with adhering vitrified clay, indicative of metalworking, was found close to the surface of C.229, which was otherwise devoid of archaeological material. Overlying this layer was a roughly wedge shaped accumulation of dark brown silty clay (C.228) containing frequent rounded stones averaging 0.10 m in diameter.

Linear feature (possible palisade): No evidence of a cut feature corresponding with the putative palisade trench (F2 on Fig. 9) mapped by geophysical survey inside Enclosure Ditch 2 was identified in Trench 2. Rather, this magnetic anomaly was found upon excavation to coincide with a diffuse band of dark brown silty clay (C.208), 0.50–0.60 m in width and c. 0.06 m thick, that extended roughly north–south across the cutting, approximately 7.5 m inside Ditch 2 (Fig. 26a). It contained occasional flecks of charcoal and was visible in both section faces as a discrete layer within soil horizon C.206 (see Fig. 23). A sample of blackthorn charcoal from C.208 produced a date of 1200–930 cal bc (2873±36 bp; UBA-36114), placing it firmly within the Late Bronze Age. Although evidently related in some way to the enclosure boundary, its precise nature is unclear; it may, for example, be the remains of a truncated feature, such as a palisade.

Possible pit: Approximately 0.10 m below linear feature C.208 were the remains of a possible pit that extended beyond the southern edge of the excavation cutting (see Fig. 22). The exposed remains comprised a semi-circular cut (C.223) measuring c. 0.90 m north-east to south-west and 0.10 m in maximum depth, which was filled with charcoal-stained soil (C.218). A sample of oak charcoal from this fill was dated to 3010–2880 cal bc (4304±28 bp; UBA-36118); however, as this derived from radial wood, there may be an age offset due to the ‘old-wood’ effect.

Charcoal deposit: A short distance to the north-east of C.223 was an irregular spread of dark brown silty clay (C.209) containing a significant concentration of oak and ivy charcoal (Fig. 26b). It extended over an area measuring 1.20 m in maximum (north-east to south-west) length by c. 0.08 m in thickness and was partially contained within a natural depression located 2.50 m inside the line of Enclosure Ditch 2. Ivy charcoal from this layer produced a date of 2865–2500 cal bc (4102±29 bp; UBA-36117).

Phase 1: Middle Neolithic

A stone-packed slot trench with post settings dated to 3635–3380 cal bc is the earliest recorded feature at Faughan. The size and layout of the slot trench suggest it supported a relatively insubstantial fence composed of upright wooden stakes interwoven with wattle. It is unclear whether the fence originally enclosed the entire hilltop, or its eastern and southern sides only; the geophysical evidence leaves open the possibility that part of its circuit was truncated by later activity.

Three pieces of worked chert from the slot trench are among a small number of lithics retrieved from archaeological contexts and the only ones likely to be in situ. The assemblage is composed primarily of flaked stone tools (flakes, blades, cores), with medium- to large-sized flakes and blades predominating (see Figs 15–17). Four convex scrapers – including a fine flint example (Fig. 16: d) – were also found, though none of these came from primary contexts. All but ten of the lithics were formed on chert, which may have been sourced locally. The remaining lithics comprise siltstone, sedimentary rock, flint, and volcanic rock. The chert raw material and the relatively large size of the flakes and blades from Faughan find broad parallels in the Late Mesolithic and Neolithic assemblage from Clowanstown 1 (Warren Reference Warren2009; Driscoll Reference Driscoll2018), about 20 km south-east of Faughan.

The distribution of lithics across the excavated area gives an indication of the wider extent of Neolithic activity on the hill, of which the production and use of stone tools may have formed just one component. That such activity took place in the relative shelter of the fenced enclosure is perhaps suggested by the occurrence of a single platform core, as well as two flakes, in the slot trench. It is notable, however, that most of the lithics recovered from archaeological contexts were found some distance from the fenced enclosure. Of the ten flakes recovered from the hilltop enclosure, nine came from one of the outer ditches, compared to just one example from the inner ditch (ie, adjacent to the slot trench). The most likely explanation for their presence in 13th/12th century bc and later contexts is that the lithics derive from a Neolithic surface into which the ditches were cut. Such a surface seems to be represented just inside the boundary of Enclosure Ditch 2 by a sandy loam layer (C.224) overlying the natural. A single chert flake was retrieved from this layer, which was cut by a possible pit of Late Neolithic date. A further insight into the spatial patterning of Neolithic activity on the hill is provided by the almost 100 lithics recovered from a levelling layer (C.103/C.203) associated with the clearance of woodland from the summit in the 1960s, which clearly resulted in the displacement downslope of a significant quantity of cultural material.

In addition to the lithics, the slot trench also contained a few charred wheat/barley grains and fragments of hazel shell and a small quantity of wood charcoal. Apart from the nutshell fragments and cereal grain from the slot trench, no other potential food remains (eg, animal bone) were recovered from any Neolithic features. The presence of cereals, however, may signal arable farming in the locality and places Faughan among a growing number of sites to provide definitive evidence of cultivation in the centuries following the earliest sustained appearance of cereals in Ireland around 3750 cal bc (eg, McClatchie et al. Reference McClatchie, Bogaard, Colledge, Whitehouse, Schulting, Barratt and McLaughlin2014; Whitehouse et al. Reference Whitehouse, Schulting, McClatchie, Barratt, McLaughlin, Bogaard, Colledge, Marchant, Gaffrey and Bunting2014).

Phase 2: Late Neolithic

Two of the excavated features at Faughan, a charcoal deposit and a possible pit, testify to activity during the later Neolithic. Both lay outside the circuit of the earlier fenced enclosure, near the outer boundary of the later hilltop enclosure. The first of these features (C.209) resulted from the burning of oak, including gnarled wood, and is the only substantial deposit of charcoal encountered during the excavation. Ivy was also represented in the charcoal deposit and dated this activity to 2865–2500 cal bc.

Only a portion of the second feature (C.223) was excavated and the possibility of it being something other than a pit – eg, the terminal of a ditch – cannot be discounted. It contained a single fill, suggesting it may have been dug and backfilled within a relatively short space of time. Oak charcoal from the fill produced a date of 3010–2880 cal bc. Allowing for a potential age offset due to the use of old wood, this feature could be contemporary with the above charcoal deposit, about 3 m to its north-east. These features may relate to wood clearance, among other possibilities.

There were no finds associated with either of these features and the lithics assemblage from the site contains no obvious Late Neolithic components that might shed further light on this phase of activity.

Phase 3: Late Bronze Age

Considerably more substantial than the Neolithic features at Faughan is the 400 m diameter enclosure encircling the hilltop. The enclosure can be broadly classed as a hillfort and is the only excavated example of its type in Co. Meath. Sections excavated across the concentric enclosure boundaries demonstrated that all three ditches are partly rock-cut and appear originally to have been accompanied by internal earthen banks, at least one of which (Ditch 2) may have had a stone revetment. Three dated samples from the basal/lower fills of Ditches 1 and 2 suggest a construction date in the later Bronze Age, possibly in the 13th or 12th century bc. In contrast, oak charcoal from the basal fill of Ditch 3 produced an anomalously early date (2865–2500 cal bc). The most likely explanation is that this charcoal is intrusive and derives from a local, Late Neolithic horizon represented most clearly by a charcoal deposit of identical date (C.209, see above) located c. 10 m west of the ditch.

A linear band of charcoal-flecked sediment (C.208), c. 6 m inside and concentric with Ditch 2, provided the only hint of the putative palisade recorded by geophysical survey. It contained a mixture of oak, blackthorn, ash, and hazel charcoal, nearly all of which derived from twigs less than 10 mm in diameter. This material may represent the remains of a brushwood fence that was supported at intervals by upright posts, the settings for which possibly lie outside the (2 m wide) excavation cutting. Its circuit is only evidenced geophysically in two places on the east side of the enclosure, the longer stretch running in tandem with the ditches for c. 90 m to the north of the entrance. Though its full extent and significance remain to be established, the dating of this feature to 1200–930 cal bc shows it to be broadly contemporary with the hillfort ditches.

Meat consumption is indicated by the fragmentary nature of the faunal remains from the site. These include 20 fragments that could be identified from cattle, six from red deer, one from sheep/goat, and three from cattle/deer. All but one of the 243 fragments in the assemblage came from the basal/lower fills of Ditch 1 (n=207) and Ditch 2 (n=35) and therefore represent activity associated with the primary use of the enclosure.

Phase 4: Early Iron Age

A hiatus or period of reduced activity, possibly spanning several centuries, seems to be indicated between the primary use of the enclosure and Phase 4 activity by infill layers incorporating probable bank material in all three ditches. Above these layers, short-life samples from an equivalent level in each ditch produced very similar dates: Ditch 1: 800–520 cal bc; Ditch 2: 780–540 cal bc; Ditch 3: 790–550 cal bc. The charcoal from this phase derived from the burning of several wood species, which could relate, among other things, to the clearance of trees and scrub from the partly infilled ditches (Stuijts Reference Stuijts2018). In this context, the presence of willow in the inner ditch (Ditch 1) is of interest as it may signal a transition to locally wetter conditions at this time.

A piece of black fayalitic slag from Ditch 3 suggests that metalworking may have taken place in the vicinity of the outer enclosure boundary, possibly during the 1st millennium bc. It came from an infill layer, thought to comprise levelled bank material, that overlay the 8th–6th century bc ditch fill. The slag is characteristic of waste produced by early bloomery iron smelting, but it can also occur as a result of iron smithing or copper production processes (Rondelez Reference Rondelez2018). This widens the potential date range for metalworking activity, which may be significant considering the presence of a possible high status, early medieval ringfort and field system just 50 m north-east of the hilltop enclosure.

A second possible waste fragment, consisting of shiny, grey, vitreous material, was recovered from a gravelly clay layer (C.103) associated with recent woodland clearance. It could derive from a variety of high temperature activities, including metalworking.

Neolithic beginnings

The earliest recorded activity at Faughan is represented by a fenced enclosure (3635–3380 cal bc) that coincides chronologically with the period defined as Middle Neolithic I (MNI), c. 3640–3400 cal bc (eg, Cooney et al. Reference Cooney, Bayliss, Healy, Whittle, Danaher, Cagney, Mallory, Smyth, Kador and O’Sullivan2011; McClatchie et al. Reference McClatchie, Bogaard, Colledge, Whitehouse, Schulting, Barratt and McLaughlin2014; Smyth Reference Smyth2014; Whitehouse et al. Reference Whitehouse, Schulting, McClatchie, Barratt, McLaughlin, Bogaard, Colledge, Marchant, Gaffrey and Bunting2014; McLaughlin et al. Reference McLaughlin, Whitehouse, Schulting, McClatchie and Barratt2016). It is the earliest dated hilltop enclosure in Meath and, at over 250 m in projected diameter, one of the largest enclosures of the 4th millennium bc known in Ireland. Most enclosed sites of the period were situated on hills, ridges, or in other conspicuous locations in the landscape and were defined by or included timber palisades as a boundary element (eg, Sheridan Reference Sheridan, Darvill and Thomas2001; Cooney Reference Cooney2002; Varndell & Topping Reference Varndell and Topping2002; Whittle et al. Reference Whittle, Healey and Bayliss2011; Cummings Reference Cummings2017). Palisades formed an integral component of the Early Neolithic causewayed enclosures at Donegore Hill, Co. Antrim (Mallory et al. Reference Mallory, Nelis and Hartwell2011) and Magheraboy, Co. Sligo (Danaher Reference Danaher2007; Cooney et al. Reference Cooney, Bayliss, Healy, Whittle, Danaher, Cagney, Mallory, Smyth, Kador and O’Sullivan2011), as well as the near contemporary hilltop enclosure at Hughstown, Co. Wicklow (O’Brien Reference O’Brien2017), and are represented in greater numbers by timber enclosures without accompanying banks or ditches. In most cases, the palisades appear to have been substantially more robust than the fenced boundary at Faughan, though the modest dimensions of the palisade trenches at Donegore (c. 0.5 m wide and 0.6 m deep) prompted Sheridan (Reference Sheridan, Darvill and Thomas2001, 174) to describe them as ‘more like fences than stockades, with an estimated height of only 1.5m’. A similar interpretation was put forward by Danaher (Reference Danaher2007, 95) for the concentric palisade inside the single, segmented enclosure ditch at Magheraboy. Although the excavated portion of the slot trench at Faughan measured only 0.2 m in maximum width and depth, given the high probability of truncation from later activity and ploughing, these dimensions are unlikely to provide a reliable indicator of its original size. Despite this, the estimated overall size of the Faughan enclosure (c. 5.2 ha) is considerably larger than those at Donegore (c. 3.5 ha) and Magheraboy (c. 2 ha). The floruit of the Faughan enclosure, moreover, may have overlapped in part with the primary use of these causewayed enclosures, which is estimated to have ended in the 36th or early 35th century bc (Bayliss et al. Reference Bayliss, Healy, Mallory and Whittle2011, 216–22; Cooney et al. Reference Cooney, Bayliss, Healy, Whittle, Danaher, Cagney, Mallory, Smyth, Kador and O’Sullivan2011, 572–3, 582–5).

Palisaded enclosures without accompanying banks or ditches (as seemingly evidenced at Faughan) are more numerous and include examples at nearby Knowth and (possibly) Tara, and further afield at sites such as Tullahedy, Co. Tipperary. At Knowth, two curving palisade trenches set 8–11 m apart and located immediately west of the main passage tomb, were tentatively proposed by Eogan (Reference Eogan1984) to represent successive enclosures with estimated diameters of c. 70 m and 100 m respectively. Although not directly dated, both palisades predate the main passage tomb mound and five smaller passage tombs, and the outer palisade trench cut across a subrectangular structure dating to c. 3780–3380 cal bc (Eogan Reference Eogan1984, 215–43; Eogan & Roche Reference Eogan and Roche1997, 44–5; Smyth Reference Smyth2014, 27).Footnote ⁴ As at Faughan, stone tool production appears to have taken place in the area of the palisades, though whether this represents contemporary activity is unclear.

Excavations at Tara in the 1950s likewise revealed the arc of a ditch or trench (0.6 m deep and traced for approximately 18 m) that extended beneath the edge of the cairn and mound of the Neolithic passage tomb known as the Mound of the Hostages (Duma na nGiall) (O’Sullivan Reference O’Sullivan2005, 23–7; see also Newman Reference Newman1997, 75). It, too, is suspected to form part of a large (palisaded?) enclosure, and an early date for this feature has been confirmed by three radiocarbon determinations obtained from charcoal from the fill, which cluster in the period 3500–3000 cal bc (O’Sullivan Reference O’Sullivan2005, 26–7; Bayliss & O’Sullivan Reference Bayliss and O’Sullivan2013, 32, 35–9). Bayesian modelling of the site chronology suggests that the ditch began to infill by 3325–3150 cal bc (Cooney et al. Reference Cooney, Bayliss, Healy, Whittle, Danaher, Cagney, Mallory, Smyth, Kador and O’Sullivan2011, 654; Bayliss & O’Sullivan Reference Bayliss and O’Sullivan2013, 35–9).

At Tullahedy, a single palisade traced intermittently for a length of c. 96 m partly enclosed a settlement situated on an esker that was surrounded on the east and south by water and mire (Cleary & Kelleher Reference Cleary and Kelleher2011). The settlement remains consisted of three rectangular houses with associated hearths, two further structures, and numerous pits. Dating evidence suggests the site to be broadly contemporary with the enclosure at Faughan, with the primary phase of activity at Tullahedy spanning the period 3670–3460 cal bc (Schulting Reference Schulting2011).

Several other palisaded enclosures of known or potential Neolithic date have also come to light through geophysical survey at hilltop sites that, like Faughan, acted as prominent foci within their respective landscapes from an early period. The earliest recorded hilltop enclosure at Rathcoran, Co. Wicklow, was defined by an oak palisade supported by a bank with internal ditch. Dated to 3760–3525 cal bc, the palisade enclosed an area of 7.2 ha and appears from test excavation to have been variously represented by single and double lines of posts (Hawkes Reference Hawkes2018; Reference Hawkes2020; Reference Hawkes2021). At Uisneach, Co. Westmeath, two successive, circular palisaded enclosures measuring 23 m and 35 m in diameter, respectively, occupied the highest point of the hill (181 m ASL). Part of the circuit of the smaller enclosure underlies a probable (unclassified) megalithic tomb and is likely to be the earliest recorded feature within the multi-period ceremonial and funerary complex there (Schot Reference Schot2011, 99–101). A significantly larger hilltop enclosure, defined geophysically by what appears to be a discontinuous, possibly segmented, palisade trench, was identified at Dún Ailinne (Knockaulin), Co. Kildare (Schot & Dowling Reference Schot and Dowling2008; Johnston et al. Reference Johnston, Crabtree and Campana2014). The enclosure delimits a roughly oval area some 390 m in maximum diameter and was laid out at an angle to the contours of the hilltop, ‘tilting’ to the north-east. This aspect of its design is strongly reminiscent of British, and some Irish (eg, Donegore Hill), Neolithic causewayed enclosures, which were frequently positioned to ‘face’ in a particular direction (eg, Oswald et al. Reference Oswald, Dyer and Barber2001, 91–106; Whittle et al. Reference Whittle, Healey and Bayliss2011). This may have implications for the dating of the palisaded enclosure at Dún Ailinne, which is known from previous excavations to encompass several Neolithic features (eg, a possible burial pit containing a Linkardstown bowl) and was truncated in the 1st millennium bc by an internally ditched hilltop enclosure extending over 13 ha (see Johnston & Wailes Reference Johnston and Wailes2007).

As the above sites illustrate, there is considerable variation in the size, morphology, and siting of Early/Middle Neolithic enclosures in Ireland. Such diversity is also evidenced in Britain and other parts of north-west Europe where, in addition to causewayed enclosures, there also existed a range of other sites circumscribed by ditches, banks, palisades, and/or fences arranged in various configurations (eg, Bradley Reference Bradley, Edmonds and Richards1998; Varndell & Topping Reference Varndell and Topping2002; Whittle et al. Reference Whittle, Healey and Bayliss2011; Last Reference Last2021). The polygonal ground plan and morphology of the Faughan enclosure are without direct parallel among the recorded Neolithic sites delineated, either solely or in part, by palisades. This may well be expected, for though they share distinctive attributes, the form of each enclosure was uniquely adapted to its topographic setting, cultural milieu, and the strategies of the individuals or groups that created them.

The hilltop siting and scale of the enclosure at Faughan, for example, suggest that it was a communal focus whose conception and use involved a larger social group than just a few individual families. Although it is not yet clear whether the site was occupied periodically, perhaps for specialised and/or seasonal activity, or on a more permanent basis, current evidence lends greater support to the former interpretation. A note of caution is required, however, as Neolithic settlement (and burial) remains could be represented among the array of features identified by geophysical survey, the majority of which are in the summit area, within the bounds of the fenced enclosure. Insofar as can be inferred from the limited excavation, activity on the hill during this period included the production and use of stone tools and there is also evidence to indicate cereal cultivation nearby. No animal bone, pottery, or objects other than the flaked stone tools from the slot trench were identified in Neolithic contexts, though this is most likely a reflection of the limited size of the area investigated rather than the nature or intensity of past activity.

Evidence of settlement in the wider environs of Faughan between the mid-37th and early 34th centuries bc is scant. Prior to the present investigations, the only dated features coinciding with this phase of the Middle Neolithic were a single pit (3655–3525 cal bc) in Grange townland (Kelly Reference Kelly2011, 5–6; Walsh Reference Walsh2021, 64–7), c. 1.3 km north-east of the hill, and a rectangular, trench-built house (3700–3530 cal bc) at Cookstown Great, near Kells (McLoughlin Reference McLoughlin2010a, 35–7; Walsh Reference Walsh2021, 52–9).Footnote ⁵ Both overlap chronologically with the earlier part of the date range for the fenced enclosure at Faughan (3635–3380 cal bc). Sherds of a broad rimmed bowl from Phoenixtown (Lyne Reference Lyne2010; Walsh Reference Walsh2021, 76) augment the evidence for Middle Neolithic activity in the immediate surrounds of the hill and three stone axes from Bohermeen bog might also date from this period.Footnote ⁶

The dearth of contemporary settlement evidence in the wider environs of Faughan is reflective of a broader pattern for the Middle Neolithic in Ireland. One of the dominant trends recognised during this period is a marked decrease in settlement visibility following the apparent ‘boom’ of the preceding centuries (Early Neolithic II: 3720/3680–3640/3620 cal bc), a period distinguished by a so-called ‘house horizon’ typified by rectangular, post-built structures (McLaughlin et al. Reference McLaughlin, Whitehouse, Schulting, McClatchie and Barratt2016; see also Cooney et al. Reference Cooney, Bayliss, Healy, Whittle, Danaher, Cagney, Mallory, Smyth, Kador and O’Sullivan2011; Smyth Reference Smyth2014). This Early Neolithic ‘boom’ is exemplified by the wealth of settlement sites excavated along the M3 motorway corridor between Navan and Kells, which runs through the valley between Faughan Hill and the River Blackwater, whose course it closely follows. The evidence attests to an established presence in the area from the early 4th millennium, with the greatest concentration of Early Neolithic settlement remains, including some half a dozen rectangular houses, located to the southeast of Kells, in the townlands of Kilmainham and Cookstown Great (see Fig. 3; McLoughlin & Walsh Reference McLoughlin and Walsh2008; Walsh Reference Walsh2021, 50–78).

Like Faughan, where renewed occupation in the Late Neolithic is signaled by a charcoal deposit and possible pit, many other sites in the area that acted as foci in earlier centuries also saw activity during the third millennium bc (Walsh Reference Walsh2021, 80–93). Among the most significant are two adjacent sites at Kilmainham, where a date range of c. 2900–2500 bc is indicated for several ritual structures associated with Grooved Ware, including a timber circle (Walsh Reference Walsh2011; Reference Walsh2021, 83–6; Whitty Reference Whitty2011). Grooved Ware was also found in association with a structure (2865–2485 cal bc) at Phoenixtown (Coughlan Reference Coughlan2010; Walsh Reference Walsh2021, 80), while the remains of a later structure (2575–2475 cal bc) were identified nearby at Grange (Duffy Reference Duffy2010; Walsh Reference Walsh2021, 102, 139). Several lithics from Ardbraccan (Mossop et al. Reference Mossop, Ginn and Ruddle2009) provide a further hint of Late Neolithic activity in the Faughan area, as might two examples of rock art in the Teltown complex.Footnote ⁷

Evolution of a regional centre: Faughan in later prehistory

The next major phase of activity recorded at Faughan coincides with the later Bronze Age and is represented by a hilltop enclosure that was far more formidable in scale and design than its Neolithic predecessor. The perimeter earthworks, consisting of three partly rock-cut ditches with evidence of former internal banks, were positioned along the upper hillslope and circumscribed a roughly circular area some 400 m in overall diameter, with opposing entrances on the north-east and south-west. This multivallate enclosure represents an important addition to the varied group of prehistoric earthworks commonly described as ‘hillforts’, of which approximately 100 examples are currently recorded in Ireland (O’Brien Reference O’Brien2017; O’Brien & O’Driscoll Reference O’Brien and O’Driscoll2017). This is a significant increase on the 40 hillforts identified in the early 1970s by Barry Raftery, who proposed a threefold classification comprising univallate sites (Class 1), widely spaced multivallate enclosures (Class 2) and inland promontory forts (Class 3) (Raftery Reference Raftery and Thomas1972; Reference Raftery and Harding1976; Reference Raftery1994, 38–48). Most encompass areas of 1.5–10 ha (Grogan Reference Grogan2005b, 113; O’Brien Reference O’Brien2017), with a small proportion of Class 1 and Class 2 hillforts, including that at Faughan, exceeding this size.

The dating of Irish hillforts has until recently been based on a very limited sample of sites. Investigation of the well-known Class 2 hillforts at Rathgall, Co. Wicklow (Raftery Reference Raftery1971; Reference Raftery and Thomas1972; Reference Raftery and Harding1976; Reference Raftery1994, 58), Haughey’s Fort, Co. Armagh (Mallory Reference Mallory1991; Reference Mallory, Waddell and Shee Twohig1995; Mallory & Baban Reference Mallory and Baban2014), and Mooghaun, Co. Clare (Grogan Reference Grogan2005a, 131–246), for instance, provided dates ranging from the 14th–9th centuries bc for their primary occupation (see also O’Brien Reference O’Brien2017). Subsequent excavations at Rahally, Co. Galway, suggest that the construction of the multivallate hillfort there began during the same period, possibly in the 11th or 10th century (Mullins Reference Mullins, O’Sullivan and Stanley2008; Reference Mullins2009; Reference Mullins, McKeon and O’Sullivan2014). Understanding of the chronology and development of this distinctive enclosure form has since advanced considerably as a result of a large-scale project on Irish hillforts by William O’Brien and colleagues at University College Cork, which commenced in 2004. Excavations undertaken by the project at ten hillforts (including eight Class 2 examples) across southern Ireland have doubled the total number of sites investigated and are particularly important in establishing a refined timeline for the construction and use histories of Irish hillforts. Collectively, the evidence indicates that hillforts are predominantly a phenomenon of the Middle and Late Bronze Age in Ireland, with a few examples built during the 14th–13th centuries and a significantly larger number dated to the period c. 1200–900 bc (O’Brien Reference O’Brien2017; O’Brien & O’Driscoll Reference O’Brien and O’Driscoll2017). A longer chronology of development is signalled, however, by Early Neolithic dates from Hughstown, Co. Kildare, and Spinan’s Hill, Co. Wicklow, which suggest an early ancestry for multivallate and univallate forms of ‘hillfort’ that may also encompass causewayed enclosures (O’Brien Reference O’Brien2017, 6–14). This is supported by similar early dates recently obtained from the hilltop enclosure at Rathcoran which, like the above two sites, forms part of the Baltinglass group on the Wicklow/Kildare border (Hawkes Reference Hawkes2018; Reference Hawkes2020; Reference Hawkes2021; see also O’Driscoll Reference O’Driscoll2019).

The dating of the Faughan enclosure is in keeping with the Bronze Age chronology indicated for the majority of excavated Class 2 hillforts in Ireland, with three dates from the lower ditch fills suggesting a likely construction date in the 13th or 12th century bc. Of the 20 or so hillforts investigated to date, that at Rahally is, as noted previously, the closest in form to the enclosure at Faughan. The low hill at Rahally (104 m ASL) was encircled by four concentric enclosing elements, comprising a double outer ditch and two single internal ditches which appear originally to have been accompanied by stone-revetted earthen banks (Fig. 27; Mullins Reference Mullins, O’Sullivan and Stanley2008; Reference Mullins2009; Reference Mullins, McKeon and O’Sullivan2014). A direct parallel for this juxtaposition of closely spaced and widely spaced boundary elements is presented by the double outer and single inner enclosures at Faughan, which may pre-date the hillfort at Rahally by a century or more. Evidence of occupation at Faughan prior to the construction of the hillfort, and again in the period c. 800–540 cal bc, when the ditches had partly infilled, reflects a pattern of recurrent activity also attested at other sites, including Haughey’s Fort and Rahally. Other Irish hillforts appear to have had shorter life histories, which in some cases ended dramatically. Excavations at Clashanimud (Co. Cork), Toor More (Co. Kilkenny), and Rathnagree (Co. Wicklow) demonstrated that the hillfort defenses were destroyed by fire in what has been interpreted as deliberate and highly symbolic acts of destruction most likely perpetrated by rival, external groups (O’Brien et al. Reference O’Brien, O’Driscoll, Hogan, Fernández-Götz and Roymans2018; see also O’Brien Reference O’Brien2016; O’Brien et al. Reference O’Brien, Hogan and O’Driscoll2014–2015).

Fig. 27. Rahally hillfort, Co. Galway: site plan (after Mullins Reference Mullins, McKeon and O’Sullivan2014, illus 4.5.4)

The imposing nature and siting of hillforts clearly mark them out as dominant foci and ‘visual expressions of power’ within their respective landscapes (O’Brien Reference O’Brien2017; O’Driscoll Reference O’Driscoll2017). While security and defensibility were undoubtedly important considerations in their design, the past few decades have seen a shift in emphasis from viewing hillforts as primarily military or defensive structures to interpretations that embrace a wider range of functions. Much of the recent discourse on the diverse roles these ‘central places’ had – as elite residences, centres of specialised production and trade, and as places of gathering and ceremony, among others – has been focused in Britain and the Continent where there is a considerably longer history of study of such sites (eg, Ralston Reference Ralston2006; Armit Reference Armit2007; Lock Reference Lock2011; Harding Reference Harding2012). As Raftery (Reference Raftery1994, 48–57) was among the first to note, however, the evidence from Rathgall and Haughey’s Fort, in particular, suggests a similar range of functions for at least some Irish hillforts.

The development of the hillfort during the later 2nd millennium has been linked more widely to the rise of warrior aristocracies and a strengthening of regional identities (eg, Harding Reference Harding2000, 271–307; O’Brien Reference O’Brien2017, 1–2, 53; O’Brien & O’Driscoll Reference O’Brien and O’Driscoll2017), in a climate where, to borrow from James (Reference James, Haselgrove, Rebay-Salisbury and Wellsin press) ‘social competition, conflict, coercion, and domination may well have been factors as significant as bonding, cohesion, and willing collaboration in “building communities”’. An insight into the context of hillfort use and the role of Faughan during this period is provided by several bronze weapons from the adjacent bog, which include a leaf-shaped sword of Ballintober type (NMI 1271:Wk004; Colquhoun Reference Colquhoun2015, no. 10) and a rapier (Fig. 28). Direct evidence of the communities for whom the hillfort was a central focus is scarce however, though the 14th–9th century bc dates obtained from a ring-ditch and cremation pit at Grange (Kelly Reference Kelly2010; Walsh Reference Walsh2021, 138–9) and developments at the Knockauns at Teltown (O’Brien & Waddell Reference O’Brien and Waddell2018), provide tantalising glimpses of life (and death) in the shadow of Faughan at this time. Further insights are offered by the post-built structures and related settlement evidence from Kilmainham (Bayley Reference Bayley2010; Walsh Reference Walsh2011) and Nugentstown (Lynch Reference Lynch2010a), as well as burnt mounds and associated features (c. 1200–1000 bc) at Boyerstown (Clarke Reference Clarke2008), Ardbraccan (Mossop & Ruddle Reference Mossop and Ruddle2009), Ballybeg (Lynch Reference Lynch2010b), and Cookstown Great (McLoughlin Reference McLoughlin2010b; see Walsh Reference Walsh2021 for the most recent discussion of these sites). Kilmainham, in particular, stands out as a significant locus of activity during the later Bronze Age, as in earlier periods. Conversely, knowledge of developments at Tara during the Late Bronze Age is poor; there are no radiocarbon dates and few recorded finds from the site spanning the period c. 1500–900 bc (among them a rapier and a spearhead: see Newman Reference Newman1997, 218–21), though this may simply reflect the limited scale of the excavations undertaken there to date.

Fig. 28. Bronze rapier from Bohermeen Bog (NMI 1970:215; ©National Museum of Ireland)

In terms of defining regional centres of power and collective identity, the occurrence of large multivallate enclosures of known or suspected Late Bronze Age date on other prominent heights in the wider surrounding landscape, namely the Hill of Ward (7.5 km to the south-west) and the Hill of Lloyd (10 km to the north-west), is highly significant. The enclosure on the Hill of Lloyd has not been excavated, though Late Bronze Age activity in its interior is attested by a pit or ditch-type feature containing burnt animal bone (mainly pig) dated to c. 1000 bc (Neary Reference Neary2003). Closer to Faughan, at the Hill of Ward, recent excavations produced Late Bronze Age dates (c. 1200–800 bc) for a large, sub-circular enclosure defined by three closely set ditches, originally identified by geophysical survey. The enclosure, which has overall dimensions of c. 200 × 220 m, partly underlies the upstanding, multivallate monument referred to as Tlachtga and is suggested by the excavator to have had a ceremonial function (S. Davis, pers. comm.; Davis et al. Reference Davis, Carey, Riley and Beglane2017). Although there is still much to learn about the nature and range of activities that took place at each of these sites, the evidence now emerging corroborates the view (Newman Reference Newman and Bhreathnach2005, 367–8) that Faughan and the Hill of Ward formed the principal focal points of two discrete landscapes separated by Bohermeen Bog, and that this occurred during the later Bronze Age, if not before.

The deep history and prehistoric importance of Faughan, now confirmed by a small scale research excavation, also lends further weight to the idea (Dowling Reference Dowling2015, 20–1) that it was once the dominant focus within a region or territory whose orbit extended north of the River Blackwater to encompass the Teltown area. Indeed, as previously noted, the historical and symbolic links intimated between Faughan and Teltown in early documentary sources are likely rooted in later prehistory, and it is quite possible that the emergence of Teltown as the pre-eminent assembly (óenach) site of the Uí Néill kings of Tara was predicated on the earlier significance of Faughan as a communal focus and gathering place (Dowling Reference Dowling2015, 20–1). Notwithstanding the later prominence of Teltown, moreover, Faughan remained central to the political and territorial ambitions of the Uí Néill, achieving a particular renown as the burial place of their most distinguished ancestor, Niall of the Nine Hostages.