Middle and Late Neolithic

Introduction

The Middle and Late Neolithic in the West Midlands: previous research

The Middle Neolithic (c 3400-2800 BC) and Late Neolithic (c 2900-2100 BC) are distinguished from the Early Neolithic primarily by new artefact categories, the appearance of new monument forms (some built on a massive scale) and the development of large ceremonial centres and ‘sacred landscapes’. The Middle Neolithic is associated, in particular, with cursus monuments, oval barrows and Peterborough Ware ceramics, and the Late Neolithic with henge monuments, stone circles and avenues, timber circles, palisade enclosures and Grooved Ware ceramics. The latter half of the Late Neolithic is also marked by the appearance of single grave funerary traditions associated with Beaker pottery and the earliest copper and bronze metalwork.

Research work devoted specifically to the Middle and Late Neolithic periods in the West Midlands has been limited, although recent large-scale fieldwork projects have made a significant contribution to the study of monument complexes and depositional practices in several parts of the region (especially the Avon valley in Warwickshire and the Middle Trent valley in Staffordshire: Ray 2007, 54-6; Woodward 2007, 187-92). Existing regional and county-based syntheses of the evidence (the same as those listed for the Early Neolithic) are limited in scope and mostly out of date, and there have been very few landscape surveys. In this context, the wide range of papers entirely or partly concerned with the Middle and Late Neolithic prepared for the Regional Research Framework seminar (Garwood (ed) 2007d) provide an important new basis for investigating these periods in the region. These reveal the distinctive character of the West Midlands evidence and provide important insights into the nature of regionality and long-term change in the British Neolithic (Ray 2007, Barfield 2007, Garwood 2007c). The present discussion of the evidence is based partly on these papers, additional research relating to the spatial organisation and character of monument groups and occupation sites, and a wider review of current interpretative frameworks and debates in British Middle and Late Neolithic archaeology.

Current research agenda in Middle and Late Neolithic archaeology

There has been no recent attempt to establish a comprehensive research framework for British Neolithic archaeology at a national scale (already discussed in relation to Early Neolithic studies) and there is a lack of consensus regarding temporal boundaries and the extent to which these mark cultural, social and economic changes. Even the basic descriptive terminologies that are used to characterise material culture assemblages and monument types have undergone significant revisions and chronological shifts in the last 30 years. The idea of a British Middle Neolithic, for example, has become fashionable again only recently, now that more precise dating of long barrows, causewayed enclosures, cursus monuments and ceramic types has clarified the material and cultural contrasts between the earlier and later parts of the 4th millennium BC. Even so, it is still apparent that long mound/long enclosure structures and late activity at causewayed enclosures span the Early/Middle Neolithic boundary (usually set at c 3500/3400 BC), whilst pit circles, the earliest ‘henge’ sites and the use of Peterborough Ware span the Middle/Late Neolithic boundary (usually set at c 2900/2800 BC).

There are also problems with defining and dating the diverse monument types, artefact categories and social practices of the 3rd millennium BC. The period 2500-2000 BC, in particular, has been prone to terminological confusion, being variously described as ‘Late Neolithic’, ‘Final Neolithic’, ‘Early Bronze Age’ and combinations of these, such as ‘Late Neolithic/Early Bronze Age’. One of the reasons for this is the appearance from c 2500 BC of Beaker burials and increasing numbers of round barrows that are often assumed to be emblematic of the ‘Bronze Age’. The presence of metal artefacts from c 2500 BC has also sometimes been seen as a way of defining the beginning of the ‘Early Bronze Age’ (eg Parker Pearson 1999), even though this would place henges and most stone and timber circles, traditionally regarded as the archetypal monuments of the Late Neolithic, within a hugely extended Early Bronze Age spanning a period of a thousand years or more. Given the rapid changes in monument building and ceremonial practices that took place during the 3rd millennium BC there is clearly a need for more detailed and more precise chronological schemes, although attempts to construct these (eg Burgess 1980, Needham 1996) have not met with general agreement. In this context it is important that archaeologists working on material belonging to the late 4th and 3rd millennia BC discriminate carefully between alternative chronological and interpretative frameworks, and define as precisely as possible the temporal range of the evidence under discussion, preferably with reference to absolute age ranges based on calibrated radiocarbon dates.

In the West Midlands, the boundary between Early and Middle Neolithic is marked by the widespread occurrence of large-scale monument construction (especially cursuses) for the first time and by the development of ceremonial centres. The end of the ‘Neolithic’, in contrast, is less clearly marked (eg the cessation of henge construction is not a helpful threshold because these monuments are largely absent from the region), but even so a major change is evident from c 2100 BC when round barrow construction increased very rapidly, especially in areas around the margins of the region (Garwood 2007b, 154; cf Garwood 2007a, 37-46). This is associated with the appearance of new material culture types and may be associated with wider evidence for woodland clearance and agriculture.

Although there is presently no agreed national framework or consensus with regard to research priorities in Middle and Late Neolithic archaeology, some key research questions and themes relevant to the West Midlands have been highlighted in several recent books (eg Bradley 1998, 2000, 2002, 2007; Thomas 1991, 1999; Whittle 1996, 1997b) and a wide range of interpretative studies:

1. Social and cultural change. Evolutionary models, which describe a transition from minimally ranked societies in the Early Neolithic, to increasingly hierarchical chiefdoms (Renfrew 1973) or ‘prestige goods’ societies (eg Braithwaite 1984) in the Late Neolithic, have been questioned from a variety of perspectives (eg Shennan 1982, Barrett 1994). Alternative explanatory frameworks have not, however, been forthcoming and descriptions of increasingly complex social organisations and political institutions appear still to have value (eg Whittle 1997b, 139-70). The interpretation of change over time in the structures and spatial scales of different kinds of social agency, and of group identities and relationships (cf Richards 1998, Fleming 2004) remain key research themes.
2. Agriculture and economy. The current debate concerning the relative importance of farming in the Early Neolithic (discussed above) is also relevant to Middle and Late Neolithic studies, with the same kinds of opposed views (eg contrast Richards and Hedges 1999 and Rowley- Conwy 2003, with Jones 2000, Robinson 2000 and Thomas 1993). There are, however, additional research questions concerning the scale of agrarian production and cereal consumption, practical and organisational changes relating to the adoption of plough and traction technologies, and the economic and cultural significance of pastoralism (Sherratt 1981, Entwistle and Grant 1989, Legge 1989).
3. Environmental change. The impact of Middle and Late Neolithic communities on the environment, although less widely debated than in Early Neolithic studies, is an important research issue given the evidence for increasingly widespread and sustained woodland clearance, extensive grasslands and cereal cultivation (eg Allen et al 2004; Entwistle and Grant 1989; Legge 1989; Smith 1984; cf Knight and Howard 2004, 51-3, 70).
4. Monuments and burials. Monuments and the practices associated with them still tend to dominate interpretations of Middle and Late Neolithic society. There have been excellent recent surveys of several major classes of monuments, notably cursuses (Barclay and Harding 1999, Loveday 2006), henges and timber circles (Gibson 1998, Harding 2003) and palisade enclosures (Whittle 1997b, Gibson 2002b). Less well understood are Middle Neolithic oval barrows and ‘long enclosures’ (Loveday 2003), Middle/Late Neolithic pit circles and round barrows (Kinnes 1979) and Beaker funerary monuments (Garwood 1999a, 288-89). Research has focused mainly on the following themes: (i) chronology; (ii) monument scale and construction methods, (iii) architectural designs, especially in relation to the symbolic ordering of space, cosmography and ceremonial practices (eg Richards and Thomas 1984; Barrett 1994, 9-69; Richards 1996; Bradley 1998); (iv) the significance of ‘structured’ deposition (eg Richards and Thomas 1984; Pollard 1995; Bradley 1998, 2000; Thomas 1999, 34-88); and (v) the significance of conceptions of time and history in the design, alteration and use of monuments (Bradley 2002). There have also been contrasting approaches to the funerary evidence: in particular, while studies of Middle Neolithic and non-Beaker Late Neolithic mortuary practices are rare (see Kinnes 1979 for a review of the evidence; cf Thomas 1999, 151-56), early Beaker graves have attracted a great deal of attention (eg Thomas 1991; Mizoguchi 1993; Barrett 1994, 86-108; Garwood 1999a, 281-84; Thomas 1999, 151-62), with particular emphasis on the symbolic significance of burial deposits and artefact assemblages.
5. Ceremonial centres/landscapes. The development, spatial organisation and social significance of monument concentrations, usually characterised as ‘ceremonial centres’ or ‘sacred landscapes’, are key research themes in Middle and Late Neolithic studies, attracting a diverse range of interpretative approaches (especially well represented in studies of the Stonehenge landscape: eg Cleal et al 1995; Allen 1997; Darvill 1997a; Parker Pearson and Ramilisonina 1998; Bradley 1998, 116-31; Thomas 1999, 163-83; Exon et al 2000). Recent research has focused, in particular, on cosmographic schemes (eg Darvill 1997b; Field 2004), the phenomenology of architectural forms and landscapes (eg Tilley 1994), and relationships between monument groups and settlement organisation (eg Barnatt 1998, 2000). Less prominent in recent studies are interpretations of change within such landscapes; for example, in relation to the appearance of Beaker funerary monuments within established Late Neolithic ceremonial landscapes (cf Thorpe and Richards 1984; Thomas 1984; Parker Pearson and Ramilisonina 1998).
6. Settlement. The opposed views that characterise Early Neolithic settlement studies also pervade discussions of the Middle and Late Neolithic evidence. Interpretations that emphasise short-lived occupation episodes, residential mobility and the rarity of durable houses (eg Thomas 1996; Whittle 1997a; Pollard 1999, 2000), can be contrasted with those that focus on a few well-preserved buildings and problems of site preservation and visibility (eg Darvill 1996, Gibson 2003). It is important, however, not to assume ‘continuity’ of settlement forms and functions throughout the Neolithic simply because of similar contrasts between durable/visible monuments and insubstantial/invisible settlements. The very different forms, scales and spatial arrangements of monuments at different times during the Neolithic, and evidence for economic and environmental changes, suggest that settlements in the Middle and Late Neolithic existed in very different social, economic and cultural landscapes to those of the Early Neolithic. Key research themes, in this context, include the relationship between settlements and monuments, especially within and around monument concentrations, and the changing nature of settlement (see Barnatt 1996, 2000; Allen 1997; Smith 1984; Pollard 2000).
7. Material culture and depositional practices. Research issues in Middle and Late Neolithic artefact studies are similar to those defined for the Early Neolithic (eg Bradley and Edmonds 1993, Edmonds 1995, Lithic Studies Society 2004; Hamilton 2002, Woodward 2002b). Perhaps the most significant area of debate concerns the nature of depositional practices: especially in relation to pits and middens (Thomas 1999, 62-125; cf Pollard 2002, Woodward 2002c) and ‘structured deposition’ at henges and timber circles (Richards and Thomas 1984, Pollard 1995). The three principal Middle and Late Neolithic ceramic types have attracted varying attention: there has been little general analysis of Peterborough Ware except for a reassessment of chronology (Gibson and Kinnes 1997); while Grooved Ware and Beakers, in contrast, have been subject to a good deal of recent study (see especially: Cleal and MacSween 1999; Kinnes et al 1991; Case 1993, 1995, 2001; Boast 1998; Needham 2005).
8. Regionality and cultural diversity. There is growing research interest in the large-scale spatial structuring of Neolithic cultural practices and the creation of distinctive regional identities (eg Harding 1995; Barclay 2000; Armit et al 2003). Widespread recognition of significant regional diversity in the Middle and Late Neolithic has major implications for investigating the development of ceremonial centres and interactions within and between prehistoric cultural and economic regions (eg in relation to long distance exchange).

Research assessment: current knowledge and understanding of the evidence

Middle and Late Neolithic sites in the West Midlands, like those of the Early Neolithic, are mostly concentrated around the margins of the region and finds densities are again generally low in comparison with distributions in neighbouring areas (Ray 2007, 52-3). It is possible, however, to recognise some intensification and expansion of settlement during this period. The most intensively studied area is the Avon valley in Warwickshire, where several important excavations of monuments and other sites have taken place, notably at Barford (Oswald 1969; Loveday 1989; Woodward 2007, 188-9), Charlecote (Ford 2003), Wasperton (Hughes and Crawford 1995), King’s Newnham and Church Lawford (Palmer 1999; Palmer 2007). Elsewhere in the region, significant Middle and Late Neolithic sites have been excavated at Catholme and Whitemoor Haye, Staffordshire (Bain et al 2005; Coates 2002; Woodward 2007, 189-91; Buteux and Chapman forthcoming), Kemerton, Worcestershire (Dinn and Evans 1990), Wellington, Herefordshire (Jackson 2007, 114- 16), and Meole Brace, Shropshire (Hughes and Woodward 1995). It is important to note, however, that other large-scale fieldwork projects in the region have produced very little Middle or Late Neolithic evidence (eg the M6 Toll route: Powell et al 2008).

Environment, landscape change and subsistence economy

Knowledge of environmental conditions in the West Midlands during the late 4th and 3rd millennia BC is very limited. The pollen sequences from Wellington, Herefordshire and Cookley, Worcestershire, indicate woodland clearance, grassland and farming in the late 4th millennium BC, although extensive clearance is not evident at either site until the 3rd millennium BC or later (Greig 2007, 45). More widely in the Severn valley, lowland Shropshire and Worcestershire, and the upper Trent valley and its tributaries, there is little evidence for large-scale clearance until at least the mid 2nd millennium BC (Barber and Twigger 1987, Bartley and Morgan 1990, Brown 1982, Buteux and Hughes 1995, Shotton 1978). It is notable that significant alluviation in the larger river valleys, probably related to the effects of clearance and farming on drainage patterns, also appears to date to the Late Neolithic or Bronze Age (eg at Beckford, Worcestershire; Greig 2007, 44). Overall, a gradual increase in open grassland areas during the 3rd millennium BC is suggested, perhaps indicative of a pastoral emphasis in the subsistence economy, and there is evidence for limited arable farming and some continuing reliance on hunting and gathering (Grieg 2007; cf Moffett 1999, 211). This is very similar to the pattern evident in the Thames valley (Allen et al 2004; Barclay and Hey 1999, 70-1), and other major river valleys in southern Britain such as the Trent (Knight and Howard 2004).

Monuments

Middle and Late Neolithic monuments are rare in most parts of the West Midlands; only in the Avon valley and around the Trent-Tame confluence are there significant concentrations of sites of this period (Fig 2.7). Although over 20 possible cursuses have been identified in the region, these are mostly unexcavated and several are doubtful (Barber 2007, 84, 89). The most convincing examples (see Fig 2.8 for comparative site plans) are those at Catholme, Staffordshire (Woodward 2007, 189-91; Buteux and Chapman forthcoming); Fladbury in the lower Avon, Worcestershire (Ray 2007, 61-2, fig 5.6); Barford (Loveday 1989), Sherbourne and Charlecote (Ford 2003, fig 1) in the middle Avon valley, Warwickshire; and the Walton cursus, Powys, the east end of which is in Herefordshire (Ray 2007, 61-2; Gibson 1999). These sites generally have rectangular termini, are often oriented south-southwest/north-northeast, and are all at the smaller end of the size range for cursus sites (less than 400m in length). Monuments of this kind have been characterised by Loveday (1999, 55) as ‘local centres’. These are contrasted with larger ‘regional centres’ organised around monuments such as the Aston cursus, Leicestershire (shown in Fig 2.8), and ‘cult centres’ focused on exceptionally large monuments such as the Dorchester-on-Thames cursus (ibid; cf Harding 1995, 124- 27). There are also several much smaller rectilinear or oblong enclosures in the region (‘long’ or ‘mortuary enclosures’, in some cases probably eroded mound sites) that may be Middle Neolithic in date (Loveday 2003; Barber 2007, 85). Only the small rectilinear enclosure at Charlecote (site A; see Fig 2.8) has significant dating evidence, indicating construction in the late 4th millennium BC (Loveday 2003, 37).

The enclosure at Wasperton, Warwickshire (Hughes and Crawford 1995), which has a single circular ditch circuit c 150m in diameter, with few causeways, is similar to the slightly smaller enclosures at Flagstones, Dorset (Healy 1997), and Stonehenge 1, Wiltshire (Cleal et al 1995), and like them is probably Middle Neolithic in date (Oswald et al 2001, 133-34). Surprisingly, given the wider range of Middle and Late Neolithic evidence and the large number of recorded ring ditches in the region, there is only one definite pit circle, Barford Site A, Phase 1, Warwickshire (Oswald 1969). This complex multi-phase site consists of a sequence of ring ditch and pit ring structures, very similar in several respects to some of the Middle/Late Neolithic multi-phase ring ditch and pit circle sites at Dorchester-on-Thames (especially Site XI; Whittle et al 1992) and at Etton in the lower Welland valley, Cambridgeshire (Etton Landscape Site 2; French and Pryor 2005). A similar site is known from air photographs at the National Memorial Arboretum, Staffordshire (Coates 2002). Smaller pit circles that may be Middle or Late Neolithic have been excavated at Wasperton, Warwickshire, and at Winforton and Wellington Quarry, Herefordshire, but in each case they lack precise dating evidence (Ray 2007, 66-7).

Of the classic Late Neolithic monument ‘types’ of southern Britain, henges and stone and timber circles are almost completely absent from the West Midlands (Fig 2.7). Although several possible henge monuments have been identified from air photographs, few have been investigated and many are doubtful on typological grounds (Barber 2007, 85-6, 90-2; Ray 2007, 64-5). There are only two sites that may be henges on the basis of their distinctive surviving earthworks: at Eardisley in the Wye valley, Herefordshire and at Piles Coppice, Binley, Warwickshire (ibid; Planas and Wilson 2006). There are also a number of ring ditches with single, opposed or multiple entrances, or beaded ditch circuits, that may belong to the diverse range of Late Neolithic enclosures sometimes characterised vaguely as ‘hengiform’ monuments. A flat-based penannular ring ditch with a probable external bank at Stapleton in the Lugg valley, Herefordshire, is probably a site of this kind although definite dating evidence is presently lacking (Ray 2007, 65, fig 5.10). The ring ditches excavated recently at Bredon’s Norton, Worcestershire (ibid, 56, 65), and Catholme, Staffordshire (Area A, F127; Area B, F100: M Hewson, pers comm), and in the early 1980s at Fatholme, Staffordshire (apparently associated with Grooved Ware deposits: unpublished; see Losco-Bradley 1984; Ray 2007, 68), may also be Late Neolithic although both dating and structural evidence is imprecise or ambiguous. The range of architectural forms and material deposits, and in some cases the complex multi-phase histories represented, suggests great diversity in the purpose, practical use and cultural significance of these enclosures, both from one site to another and over time.

Timber circles and related structures, which are often assumed to be closely associated with henges (ie as structures built within henges, or as separate ‘open’ ceremonial arenas) and which mostly date to the Late Neolithic (Gibson 1998), are also virtually non-existent in the region, with the notable exception of the multiple post circle and the ‘sunburst’ pit arrangement investigated recently at Catholme in the Trent valley, Staffordshire (Bain et al 2005; Woodward 2007, 189; Buteux and Chapman forthcoming). There are, however, several standing stones and stone circles (which may be the stone-built counterparts of timber structures) recorded in the upland areas around the western and southern fringes of the region. These include a group of sites at Stapeley Hill on the Shropshire- Montgomeryshire border (Fig 2.9), including Mitchell’s Fold stone circle (Fig 2.10), the Carreg-y-Big monolith near Selattyn, Shropshire (Ray 2007, 69), the Queen’s Stone monolith at Symond’s Yat in the Wye valley, Herefordshire (ibid, 69-70) and the Rollright Stones, straddling the Warwickshire-Oxfordshire border (Lambrick 1988).

At present, Late Neolithic wooden-walled enclosures are absent from the West Midlands, although the exceptionally large Hindwell 2 palisade enclosure and the Walton post-pit enclosure are located just outside the region in the Lugg Valley, Radnorshire (Gibson 1999; Gibson et al 2001). The crop mark site at Staunton-on-Arrow, Herefordshire, once thought to be a palisade enclosure, has now been shown to be ditched rather than palisaded and is probably Iron Age in date (White 2003, 25-8). Another site tentatively identified from air photographs as a possible palisade enclosure, near Milton Cross, also in the Arrow valley, Herefordshire, has not yet been investigated (Ray 2007, 65-6).

There is very little evidence for Middle and Late Neolithic funerary monuments in the West Midlands. It is possible that some of the Peak District monuments in Staffordshire (Barnatt and Collis 1996) and one or more of the mounds and ring ditches at King’s Newnham, Warwickshire (Palmer 2007, 123-6), belong to this period, but the dating evidence is far from certain. Early Beaker funerary monuments and burials are also very rare and restricted spatially to the south-western part of the region. An important early burial, possibly a ‘flat grave’, has been excavated recently at Wellington Quarry, Herefordshire (Harrison et al 1999). This was probably an adult male, with a funerary assemblage consisting of a European Bell Beaker, tanged copper dagger, stone wristguard fragment and 16 flint objects, including four barbed-and-tanged arrowheads (Fig 2.11). There are other early Beaker burials in the Olchon valley, Herefordshire (Marshall 1932), and at Bredon Hill, Worcestershire (Thomas 1965). The finds contexts of early Beakers found at Normacott, Staffordshire, and Meriden, Warwickshire, are uncertain and they may not have been associated with burials (Clarke 1970, cat. 832, 1014).

Ceremonial landscapes

Despite the overall rarity of Middle and Late Neolithic monuments in the West Midlands, it is possible to recognise the development of several distinctive ‘ceremonial landscapes’ in the region during this period (Ray 2007, 54-6; Woodward 2007). These consist of relatively large clusters of monuments and pit groups, sometimes with evidence for complex spatial organisation and the deliberate deposition of artefacts and other materials (eg at Wasperton and Barford; Woodward 2007, 187-9). The most convincing examples (Fig 2.7) can be summarised by county as follows:

Shropshire

Stapeley Hill: located on the Shropshire-Montgomeryshire border. This group of megalithic monuments consists of two surviving stone circles, at least one more destroyed circle just across the Welsh border, possibly two more destroyed circles or monoliths and several round cairns (see Figs 2.7 (1), 2.9, and 2.10; cf Burl 1976, 264-66; Ray 2007, 69).

Herefordshire/Radnorshire

Walton/Hindwell: located in the Lugg valley, mostly on the Radnorshire side of the border but just extending into Herefordshire. The main sites include the largest known palisade enclosure in Britain (Hindwell 2), the Walton post-pit enclosure, two cursuses, a probable henge and several ring ditches (Fig 2.7 (2); Gibson 1999; Gibson et al 2001).

Worcestershire

Fladbury: at least four small cursuses or rectilinear enclosures and a possible ‘hengiform’ enclosure and other crop mark sites on the other side of the rover to the east (see Figs 2.7 (3), 2.8; cf Ray 2007, 61-2).

Warwickshire

Wasperton/Charlecote: located on the east side of the Avon about 3.5km south of Barford, consisting of a circular enclosure, pit circle, long enclosure, ring ditches, pits and other features (Fig 2.7 (4); Hughes and Crawford 1995, Ford 2003, Loveday 2003, Woodward 2007, 187-8).

Barford: located on the east side of the Avon south of Warwick, with a cursus, possible ‘long enclosure’, multi-phase ring ditch/pit circle site, pit groups and ring ditches (Fig 2.7 (5); Oswald 1969; Loveday 1989; Woodward 2007, 188-9). The cursus nearby at Sherbourne on the west side of the Avon (Ford 2003, fig 1) may form part of the same monument complex, similar to the spatial arrangement of monuments at King’s Newnham/Church Lawford.

King’s Newnham/Church Lawford: located on either side of the upper Avon just to the west of Rugby, consisting of enclosures, pit groups and possible Middle Neolithic mounds (Fig 2.7 (6); cf Palmer 2007).

The Rollright Stones: located on the Warwickshire-Oxfordshire border, with a stone circle, monolith, round cairns and round barrows (Fig 2.7 (7); Lambrick 1988).

Staffordshire

Catholme/Whitemoor Haye: located at the confluence of the Trent and Tame, with at least three cursus monuments, a multiple timber circle, ‘sunburst’ pit arrangement and numerous ring ditches (Fig 2.7 (8); Woodward 2007, 189-92; cf Bain et al 2005, Coates 2002; Buteux and Chapman forthcoming).

Settlement and occupation sites

Direct evidence for Middle and Late Neolithic occupation in the form of settlement sites and houses is very rare throughout southern Britain, and the West Midlands is no exception. It is possible that some of the features recorded at Barford in Warwickshire may have been buildings: a sub-rectangular enclosure and posthole group, associated with a Peterborough Ware bowl, has been interpreted as a Middle Neolithic ‘house’ (Site C, Enclosure 4; Oswald 1969, 19-27), while several less well dated groups of stakeholes, postholes and pits have been interpreted as Neolithic huts, the most convincing of which is an oval stakehole structure (Site B, Hut 13; ibid, 16-19, fig 8). A possible Middle Neolithic ‘post and wall-slot’ building has also been excavated at Stretton-on-Fosse, Warwickshire (Site 5: Darvill 1996, 106; Gardiner et al 1980, 9-13). Evidence for Late Neolithic buildings is even more scarce, although recent excavations on the line of the Rotherwas Access Road near Dinedor Camp, Hereford, have produced evidence for at least one four-post structure associated with sherds of Beaker pottery, in an area with apparent settlement traces including pits and further postholes associated with Peterborough Ware and Grooved Ware pottery (Ray 2007, 68).

Generally, however, occupation sites of this period appear to be represented mostly by pits, other non-structural features such as gullies’ and surface scatters of lithic artefacts. The purpose of Middle and Late Neolithic pits and pit deposits is uncertain, although in many cases the deliberately placed nature of the deposits suggests practices that were ‘special’ rather than a matter of everyday routine (Thomas 1999, 64-74). Sites in the region with Peterborough Ware pit deposits include King’s Newnham, Warwickshire (Palmer 2007), Wellington, Herefordshire (Jackson 2007, 114-5), and Meole Brace (Hughes and Woodward 1995) and Brompton, Shropshire (Woodward 2007, tbl 12.1). A large assemblage of Peterborough Ware was also found in a gully at Whitemoor Haye, Staffordshire (ibid; Coates 2002). Middle and Late Neolithic sites with both Peterborough Ware and Grooved Ware pit deposits include Church Lawford (Palmer 2007) and Wasperton, Warwickshire (Hughes and Crawford 1995). Late Neolithic sites with only Grooved Ware pit deposits include Barford Site B (Oswald 1969; Woodward 2007, 188) and Broom Area E in Warwickshire (Palmer 1999, 22-37), and Aston Mill Farm, Kemerton, Worcestershire (Dinn and Evans 1990). The recent discovery of several Grooved Ware pit deposits at Severn Stoke, just south of Worcester, including one consisting of a large pottery assemblage and several stone axes (Ray 2007, 68), is especially important as this is the first well-dated Late Neolithic site to be found in the lower Severn valley.

Some of the Beaker pit deposits in the region may date to the later 3rd millennium BC (eg at Longmore Hill Farm, Astley, Worcestershire; Dinn and Hemingway 1992, 111-17) but without radiocarbon dates it is difficult to be certain. Early Beaker pottery is sometimes found in pits that also contain late Beaker material (eg Whitemoor Haye Area P, Pit F122; Ann Woodward pers comm). There are also occasional finds of early Beaker ceramics redeposited in later features (eg in probable Iron Age pits at Bromfield, Shropshire; Stanford 1982, 287-89). This material was probably derived from surface spreads of occupation debris or middens that have since been destroyed by ploughing.

There are occasional finds of Middle and Late Neolithic material in other contexts. For example, Peterborough Ware has been found at several round barrow sites (eg at Burton Hastings 1, Warwickshire; Garwood in prep) and in ring ditch fills (eg Wasperton, southern ring ditch; Hughes and Crawford 1995). Although it is possible that some of these are Middle Neolithic sites, in most cases it is likely that this material was redeposited, either accidentally in the course of mound construction or deliberately in order to incorporate ancient cultural materials within new monuments. It is also important to note the presence of Middle and Late Neolithic sites found beneath alluvial and colluvial deposits, notably at Wellington, Herefordshire (Jackson 2007; Dinn and Roseff 1992). This raises the possibility that settlements in valley locations may be far more numerous than currently recognised (ibid; Knight and Howard 2004). The recent discovery in Herefordshire of two late Beaker middens close to streams (Ray 2007, 68) also suggests that midden sites of Neolithic date may exist more widely in the region, as in other parts of southern Britain (cf Allen et al 2004).

In addition to the Middle and Late Neolithic lithic scatters found throughout the West Midlands, which indicate widespread if generally low intensity occupation, some localities around the fringes of the region have produced exceptionally large finds assemblages, notably the Clun district in Shropshire (Barfield 2003; Chitty 1963), the Golden Valley in Herefordshire, the Staffordshire Peak District and Wolvey in Warwickshire (Barfield 2007, 99-103). The distributions of flint axe heads, barbed-and-tanged arrowheads (ibid) and stone axe heads (Woodward 2007, 184-7) follow this general pattern, although the latter has a more even distribution that includes finds in the central part of the region and along the Avon valley. Stone axe heads in the West Midlands derive primarily from Welsh sources (Groups VII, VIII and XXI; especially in Worcestershire, Herefordshire and Shropshire), a Cumbrian source (Group VI; especially in Staffordshire, Warwickshire and Shropshire) and a Cornish source (Group I; especially in Warwickshire). In contrast, shaft-hole implements such as battle-axes and axe-hammers, which are Late Neolithic or Early Bronze Age in date (ibid), are mostly derived from more local sources in North Warwickshire (Group XIV) and the Shropshire-Montgomeryshire border (Group XII), or from a south Cumbrian source (Group XV) (see Fig 2.15).

Regionality and cultural diversity

There are clear contrasts between the central and outer parts of the West Midlands in the distribution of Middle and Late Neolithic sites and finds (Barfield 2007, 103- 4; Garwood 2007c, 200-1; Ray 2007, 52-3). All of the definite cursuses, enclosures, funerary monuments and major lithic artefact concentrations are found around the fringes of the West Midlands, in areas such as the Peak District and middle Trent valley in north Staffordshire, the Wolvey district, the Avon valley and the Cotswold ridge in Warwickshire, the Golden valley in south-west Herefordshire, and the Clun area and other places along the Shropshire-Wales border. In contrast, there are no monuments and very few significant artefacts groups in the central part of the region, including east Shropshire, the Birmingham area, south and west Staffordshire, the lower Severn valley and the area between the Severn and Avon. The rarity of Middle and Late Neolithic finds in large-scale research and developer-funded projects in these areas suggests that the overall spatial pattern does reflect relatively sparse and/or low intensity occupation of the central West Midlands (Barfield 2007, 103-4; Garwood 2007c, 201, 202).

This seems to sustain the overall pattern of occupation and finds densities evident in the Early Neolithic. It is also possible, however, to recognise some expansion in the distribution of both monuments and pit deposits (eg in the Severn valley), as well as greater clustering of monuments and pit groups in some areas. This suggests that certain kinds of social practice and organisation, represented by more intensive occupation and ceremonial activity, not only became geographically more widespread but gave rise to the development of ceremonial centres in especially significant and favoured locales (Garwood 2007c, 202-4).

It is also evident that Middle and Late Neolithic monuments and artefact concentrations located around the periphery of the region, like those of the Early Neolithic, were parts of wider distributions that lie mainly outside the West Midlands (Garwood 2007c, 200- 1). The forms and spatial distributions of cursus monuments in the Avon valley, for example, are paralleled most closely in the upper Thames valley to the south (Barclay and Hey 1999, Loveday 1999) and the Ouse valley to the east (eg see Malim 2000). It is also notable that monument concentrations in central Britain that include exceptionally large and/or elaborate monuments are all located outside the region, including the Welshpool and Walton monument groups in Wales (Gibson 1994, 1999), the Aston monument complex in the middle Trent valley, Leicestershire (Loveday 2004) and the Dorchester-on-Thames complex in the upper Thames valley (Loveday 1999). Although the Catholme/Whitemoor Haye and the Wasperton/Charlecote monument groups, in the Trent and Avon valleys respectively, have monument concentrations comparable with those outside the region, they do not include especially large or elaborate monuments. In this context, the West Midlands as a ‘cultural region’ seems to be distinguished by the rarity of Middle Neolithic and especially Late Neolithic monuments, and by the development of relatively small-scale ceremonial centres.

Research agenda and specific research questions

Social and economic change

Although there appears to have been an expansion of clearance, settlement activity and monument building in the West Midlands in this period, in common with many other parts of southern Britain (eg Knight and Howard 2004, 70; Thomas 1999, 188), there has been very little fine-grained analysis of Middle and Late Neolithic landscapes on which to base specific interpretations of social and economic change. The relative rarity of monument groups and the limited evidence for large-scale construction events raise important questions concerning the extent to which monument building became more complex or larger in social scale over time. If such ‘complexity’ is seen as an index of relative social organisational complexity (or hierarchy), it is apparent that this seems to have developed to a lesser extent in the West Midlands in comparison with parts of southern Britain where very dense concentrations of large and/or elaborate monuments can be found. Late Neolithic Beaker burials with grave goods, which are sometimes used to suggest prestige goods exchange and hierarchy, are also exceptionally rare in the region. The West Midlands evidence, therefore, has considerable potential for investigating regional variation in the character and direction of social change, and forms of social organisation different to those suggested for other regions such as Wessex.

Monuments and landscape

There is a clear need in the West Midlands to enhance our understanding of ceremonial monuments, their spatial arrangement and aggregation as extensive ‘ceremonial landscapes’, and the sequences and tempos of constructional and depositional events (Ray 2007, 54-6). There have been some significant recent developments in this area of study, especially in relation to the spatial organisation and long-term development of large monument complexes such as Catholme/Whitemoor Haye, Barford, and Wasperton (Woodward 2007; Buteux and Champman forthcoming; Bain et al 2005, Coates 2002; Loveday 1989; Hughes and Crawford 1995). Other monument groups, however, have received far less attention, notably the cluster of megalithic monuments in western Shropshire. There is, above all, a need for more intensive landscape projects, especially in areas that have been subject to little concerted investigation such as the upper Trent valley west of Catholme.

At a larger scale, an interpretative synthesis of the evidence from the Avon valley in Warwickshire and Worcestershire is a clear research priority. This is the only part of the region that has significant numbers of Middle and Late Neolithic monuments with several distinct ceremonial foci. It is also the only area that has been subject to extensive survey and large-scale excavations of monuments and other sites. Comparisons with river valleys elsewhere in southern Britain such as the upper Thames and Great Ouse suggest some broad similarities in terms of site densities, spatial patterns and change over time, with similar potential for investigating ceremonial complexes, territoriality, and social and economic practices. Even so, the spacing of monument groups c 5-8km apart along the Great Ouse (Malim 2000), and cursuses at 5-10km intervals along the Thames and its tributaries south of Oxford (Barclay and Hey 1999, 68), contrasts with the wider spacing of monument groups c 20km apart along the Avon between King’s Newnham/ Church Lawford, Barford/Wasperton/Charlecote and Cropthorne/Fladbury (although the spacing of cursuses within the Barford/Wasperton/Charlecote group (Ford 2003, fig 1) is more in line with the upper Thames and Great Ouse pattern). If not a reflection of uneven fieldwork, the Avon evidence suggests differences in the structuring of cultural landscapes from one part of southern Britain to another, with significant implications for future research and fieldwork projects.

Specific monument categories also deserve particular attention in regional research terms. A key aim, for example, must be to gain a fuller picture of the distribution, scale and use of Middle Neolithic monuments such as cursuses, pit rings, ring ditches and ‘mortuary enclosures’. At present, these sites seem to be concentrated only in parts of the West Midlands and very little is known about how they were used, the spatial organisation of monument groups or their relationships with contemporary settlements. The presence of cursus monuments in the Severn valley at Welshpool, Montgomeryshire (Gibson 1994), and in the Lugg valley at Walton, Radnorshire (Gibson 1999), suggests that such monuments may also be found along the river valleys in the western part of the region. Similarly, the Trent valley sites at Catholme, and the major cursus monument further downstream at Aston, Leicestershire (Loveday 2004), show the potential for future discoveries of Middle Neolithic monuments in the Trent valley.

The almost complete absence of henges, and the great rarity of ‘hengiform’ sites, ring ditches and pit circles in the region (Barber 2007, 90-2; Ray 2007, 64-8), is especially surprising given the presence of Middle Neolithic monuments which in other parts of Britain were often foci for the development of Late Neolithic ceremonial complexes (eg at Dorchester-on-Thames; Loveday 1999). The lack of henges and related sites in the Avon valley is especially striking as this is not only the most intensively investigated part of the region, with numerous Middle Neolithic monuments, but is also an area in which air photographic survey has been especially effective. It is also noticeable that where probable Late Neolithic enclosures have been identified in the region (eg in recent excavations at Stapleton, Herefordshire, Bredon’s Norton, Worcestershire and Catholme, Staffordshire), they are very small in scale and found in river valley locations.

At present, therefore, it does appear that Late Neolithic ceremonial sites in the West Midlands were rare, mostly small and architecturally unambitious. This may suggest smaller populations, less centralised social and religious institutions and/or less concern with collective ceremonial practices in comparison with areas such as Wessex (cf Harding 1995, 131). However, as so few crop mark sites have been investigated and many areas are not conducive to aerial photography, it is certainly too soon to assume this to be the case. More extensive, research-driven aerial survey and targeted investigation of possible Late Neolithic enclosures are clear research priorities (Barber 2007, 90-4). The presence of timber circles and palisade enclosures within and just outside the region at Catholme (Woodward 2007, 189; Bain et al 2005), Sarn-y-Bryn Caled (Gibson 1994), and Walton (Gibson 1999, Gibson et al 2001), suggest that there is potential for new discoveries of similar sites in the West Midlands (cf Barber 2007, 94; Ray 2007, 64-7).

The rarity of early Beaker burials and monuments, and early Beaker ceramics in general, also deserves more attention. The south-western distribution of early Beaker burials in the region may suggest local variation in the ways that ‘single grave’ funerary practices (and their symbolic associations) were culturally valued and the extent to which they were adopted or rejected. It is especially noticeable that early Beaker graves are all located in areas that appear locally to lack Middle and Late Neolithic ceremonial monuments, which may indicate avoidance of these areas by communities adopting new practices and new kinds of cultural representation in the later 3rd millennium BC, or perhaps the exclusion of Beaker-associated practices from existing monument complexes by those with interests in ‘orthodox’ religious traditions (cf Thorpe and Richards 1984, 75-80).

Settlement and landscape

Research questions relating to the nature of settlement in Middle and Late Neolithic archaeology are in many respects similar to those in British Early Neolithic studies. These focus on residential mobility, the relative permanence and scale of occupation sites, the relationship between ceremonial sites and settlements, and social organisation. There is clearly considerable scope in the West Midlands for investigating the character of settlement in areas with ceremonial monuments, and to compare these with areas in which durable and/or prominent architectural structures are absent. Recent work in many parts of southern Britain has also highlighted the importance of riverside occupation sites and the enormous but under-explored potential of these for investigating Neolithic settlement in general (eg Allen et al 2004, Knight and Howard 2004, French and Pryor 2005). The long-term project at Wellington serves to illustrate the importance of such locales in the Neolithic landscape, their exceptional research potential, and also the considerable practical challenges involved in investigating them (Jackson 2007).

There is very little evidence for Middle and Late Neolithic settlement architecture or long-lived occupation sites anywhere in the West Midlands. To a large extent this can be explained by the insubstantial nature of house structures (Darvill 1996, Gibson 2003), dispersed settlement patterns, residential mobility and low discard rates of inorganic cultural material (Thomas 1996, Whittle 1996), giving rise to thinly-stratified and spatially discontinuous occupation sites that are especially vulnerable to destruction by ploughing and natural erosion processes. In favourable preservation conditions, however, it is possible for settlement evidence of this kind to survive (as at Barford). The Late Neolithic buildings discovered just outside the region at Trelystan, Montgomeryshire (Britnell 1982, Gibson 1996), and Upper Ninepence, Radnorshire (Structure 1: Gibson 1999, 36-7), in both cases protected by later barrow mounds, and at Willington in the Trent valley (Darvill 1996, 102), highlight the potential for similar sites to exist in the West Midlands. The investigation of well-preserved Middle and Late Neolithic settlements, especially where these have been buried beneath later monuments or sealed by colluvial or alluvial deposits, should clearly be a research priority.

In the absence of architectural remains, Middle and Late Neolithic ‘settlement’ evidence is usually ephemeral and ambiguous. The interpretation of isolated pits and pit groups as settlement sites is especially problematic (cf Thomas 1999, 62-74). In some cases these may be the outcomes of ordinary daily routines and tasks that took place within or around short-lived occupation sites. Yet the deliberate and selective nature of many pit deposits suggests that these resulted from the deliberate placement of objects and materials as part of more formal social practices (ibid). Indeed, these depositional events may have been significant not only as cultural media in themselves but may have taken place at ‘special’ locales. The complexity and diversity of these practices are especially well represented by the pit groups excavated recently at Church Lawford and King’s Newnham, Warwickshire (Palmer 2007), which suggest repeated, structured depositional events spanning the entire Neolithic and Early Bronze Age. At present, however, it is not possible to discern the duration and frequency of these practices at a scale of less than half millennia, nor accurately define their spatial extent and organisation (ibid). Without some understanding of the temporal and spatial conditions of the evidence, it is very difficult to suggest specific interpretations of the kinds of social agency and signification embodied in pit deposits or changes in these over time.

Middle and Late Neolithic pit groups excavated in the West Midlands are directly comparable with those in other parts of Britain in terms of their complexity, diversity and the material assemblages they contain. They are clearly important in research terms, especially as they often provide the only evidence for both settlement-related and non- routine activities across great swathes of the British landscape. Yet the nature of the social practices and the cultural meanings represented in pit-digging and deposition remain little understood. In the West Midlands case, Ray (2007, 71-2) has suggested that in the absence of a monumental focus in many areas, pits are perhaps the defining feature of the Neolithic: detailed contextual and comparative analyses of pits and their landscape settings should therefore be undertaken with great care wherever such evidence is encountered, and at the most extensive spatial scales possible.

Other site categories that seem to represent residential or specialised economic and social activities are little represented in the region. Midden deposits, for example, which are now recognised as a significant component of Neolithic occupation practices (Pollard 1999, 2000, 2005) have been recorded only at Wellington and Staunton-on Arrow, Herefordshire (Ray 2007, 68), in both cases as linear deposits adjacent to former stream courses. The rarity of middens may be due in part to destruction by modern ploughing, but it is likely that some will survive beneath later sediments and in areas of pasture. In addition, while most of the burnt mound sites investigated in the West Midlands are Middle Bronze Age in date (Ehrenburg 1991, Hodder 1990), the presence of Late Neolithic burnt mounds in the lower Trent valley (Knight and Howard 2004, 57) and possibly at Harborne, Birmingham (Hodder 1990, 108), suggest that Neolithic burnt mounds may also be encountered more widely across the region in the future.

The relationship between occupation sites and monuments is of central importance in research terms, especially with regard to the supposed distinction between ‘sacred’ and ‘secular’ landscapes (cf Gibson et al 2001, 108-9). Whilst there is broad agreement that occupation sites existed within and/or around monument groups, it is far less clear whether such inhabitation was continuous or discontinuous, and whether it differed in character from occupation in areas without monuments. Indeed, very little is known about how occupation sites of any kind were organised, either in terms of the spatial arrangement and scale of settlement, or in terms of the temporal rhythms and durations of occupation episodes. In the Walton Basin, settlement sites and monuments were situated in close proximity but may have occupied different parts of the landscape, with settlements on higher ground overlooking the monument complex to the south (ibid). A similar spatial pattern may be apparent around Avebury, with repeated occupation of hillside locations overlooking major monuments (Pollard 2005, 109-10; Holgate 1988, 91-7), although both here and at Stonehenge it is evident that some occupation areas were interspersed among the monuments (Thomas 1999, 174-77).

It is also possible that the organisation of the later Neolithic landscape was more variable or changed more radically over time in areas where monuments were dispersed rather than concentrated. In the upper Thames valley, for example, it has been suggested that settlements and monuments at Stanton Harcourt were closely integrated (Barclay et al 1995, 112), yet not far away at Yarnton it is argued that distinct ceremonial, funerary and occupation areas were maintained over a long period (Thomas 1999, 190; cf Hey 1997). To some extent these contrasts may be less pronounced, or may be interpreted rather differently if landscapes are studied at a larger scale: in the Abingdon to Dorchester-on- Thames area, for example, changes in the nature and intensity of activity at one monument group may have been complemented by changes at others (Garwood 1999a, 292-98; Thomas 1999, 195). This suggests that occupation practices may be best understood at a very large spatial scale, and that ‘separate’ monument groups and settlement foci were in fact closely interrelated within extensive social and economic landscapes (cf Whittle 1997a). These themes have barely been addressed in the West Midlands, although the potential for investigating the spatial structuring of activity areas within monumentalised landscapes is apparent in recent studies of the Barford, Wasperton and Catholme/ Whitemoor Haye ceremonial complexes (Woodward 2007).

The distribution of lithic artefacts perhaps provides some idea of wider settlement patterns in the West Midlands, although programmes of systematic surface artefact collection in the region have been rare and localised in comparison with much larger-scale surveys in areas such as the Thames valley (eg Holgate 1988). Even so, the presence of exceptionally dense, extensive concentrations of lithic artefacts in parts of the region (Barfield 2007, 99-103), mostly in areas without major Middle and Late Neolithic monuments, raises important research questions concerning the character and overall spatial patterning of settlement, as well as particular problems of social interpretation (ibid). There is a clear need to re-assess these lithic assemblages, and to investigate the socio-political, economic and/or cultural significance of the locales in which they have been found.

Material culture

Many of the research priorities in Middle and Late Neolithic material culture studies are the same as those identified for the Early Neolithic. There is a need to produce more reliable artefact chronologies, identify functional and technological aspects of tool use, and investigate the sourcing, production, exchange and deposition of specific artefact types (Barfield 2007, 106; Ray 2007, 74). There are especially important research questions concerning the scales and forms of technology-reproducing social groups, including the organisation of stone extraction and tool manufacturing in relation to Group XII and Group XIV stone sources, and the social relationships realised in exchanges marked by the widespread use of flint derived from primary flint sources outside the region (Barfield 2003; 2007, 106). Another key research issue is the extent to which widespread trans- regional artefact categories such as Grooved Ware styles and early Beaker types were produced locally or were accumulated through exchanges and/or ‘collecting’ practices in the course of residential movements or special journeys (eg pilgrimages to cult centres; cf Loveday 1999). At regional and inter-regional scales of study, there is clearly a need for more detailed analyses of raw material sourcing (including pottery clays and tempers), production technologies and artefact ‘biographies’.

Spatial patterns and regionality

The key research issue at a regional scale is the extent to which the known distribution of monuments is a real reflection of occupation and monument building practices or a consequence of previous research limitations (Barber 2007; Ray 2007, 52-3, 72-3). If monument building really was as rare in the West Midlands as the present evidence suggests, then this raises fundamental questions about why settlement patterns and social organisations in the region were so different from those in areas where monuments and other sites were more densely clustered (ibid; Garwood 2007c). There is clearly also a need to gain a far more detailed understanding of the few areas with relatively greater evidence for monument building and the development of distinct ceremonial landscapes (notably at Catholme/Whitemoor Haye and along the Avon valley), and to compare these with other monumentalised landscapes in southern Britain. In these cases there is potential for discussions of political and territorial organisation (eg in relation to the spacing of monument groups along the Avon valley) and interactions between what appear to be distinct communities (eg the groups using the Catholme/Whitemoor Haye ceremonial centre and their neighbours just to the north in the Peak District).

The West Midlands may also be an especially suitable regional context for considering the nature of regional cultural diversity in the Middle and Late Neolithic (cf Harding 1995, Barclay 2000). This is not because of any kind of intrinsic or coherent cultural identity, but because of the lack of such an identity, and the evidence instead for considerable variation in cultural practices (Ray 2007, 72; Garwood 2007c, 194-5). The central part of the region seems to have been thinly occupied and certainly lacks monuments, while the diversity of the cultural forms, practices and local sequences of change around the periphery relates to the cultural repertoires and activities of social groups that lived mainly outside the region. The ‘West Midlands’ is thus an arbitrary unit of study in cultural terms, but embraces parts of what in the Neolithic were many culturally distinct areas that can be studied both comparatively and in terms of the cultural and economic interactions that took place between them. At a larger scale, the region is geographically central to, and traversed by, a multiplicity of routes across southern Britain: between the Welsh mountains and east midlands’ plains, between the south-west peninsular and the Yorkshire Wolds and Moors, and between the chalk and limestone hills and river valleys of southern England and the Pennine and Cumbrian uplands. The distribution of sourced stone axes in the West Midlands, and across southern Britain as a whole, is striking testimony to the scale and complexity of the networks of exchange and cultural interaction that reached or extended across the West Midlands in the Middle and Late Neolithic (eg see Clough and Cummins 1988, maps 1-23; Shotton 1988).

Research aims and methods

The research agenda and key research questions outlined above have important implications for methods of resource assessment, curatorial practices, fieldwork methods, and networks of communication and data gathering in the region. In many respects, the methodological and practical requirements of research-led archaeological work already identified for Early Neolithic studies are also applicable to the Middle and Late Neolithic: these are summarised below, with additional points where relevant.

• There is an urgent need for significant enhancement of HER resources in the West Midlands, especially in relation to the description, dating and full listing of sites and finds (Barfield 2007, 97-9, 106).

• Reassessment of the existing air photographic record and further aerial survey work (especially in major river valleys, areas which have received less attention in the past, and areas where crop marks are rarely seen) is needed in all parts of the region (Barber 2007, 94). Sample excavations of crop mark sites should be a high priority (ibid).

• The identification and investigation of Middle and Late Neolithic monuments, especially cursuses, pit circles, henges and palisade enclosures, is a clear research priority in West Midlands archaeology. There is a particular need to recover evidence relating to architectural design, artefact deposition and chronology. Opportunities for investigating possible Middle and Late Neolithic monuments in development archaeology contexts should be grasped immediately and arrangements made for full resourcing of excavation and post-excavation work (ibid; Ray 2007, 73-4).

• The dispersed character of social activity in the Middle and Late Neolithic, and the structured organisation of social practices at large spatial scales (Darvill 1997b; Pollard 2002), must be recognised in preparing fieldwork designs.

• Fieldwork methods should be appropriate to the study of Middle and Late Neolithic landscapes. In particular, site identification methods such as surface collection and test pit survey require narrow transect intervals, and site evaluation methods should operate at 6-10% sampling levels (Hey and Lacey 2001).

• The development of effective predictive modelling and site prospection methods in sub-alluvial and sub-colluvial contexts offers new opportunities for studying prehistoric settlement areas that cannot be investigated using surface survey and shallow excavation techniques (Jackson 2007, 120-1; cf Challis and Howard 2003).

• Extensive open-area excavation is essential for investigating the wider culturally-ordered spatial contexts of particular social practices. The ‘strip, map and sample’ process (Hey and Lacey 2001, 55-7) seems to be the most effective method for initial investigation of prehistoric sites and for planning effective excavation strategies.

• Where well-defined Middle and Late Neolithic sites are identified, it is important that they are totally excavated at least within the areas to be affected by development, to identify overall spatial designs, constructional features, and the structuring of depositional practices, artefacts and materials (using 100% artefact recovery and three-dimensional recording methods). The recovery of dating evidence, especially radiocarbon sample materials, is a research priority.

• A key area of concern in Middle and Late Neolithic archaeology is recognition and interpretation of the structured, purposeful deposition of artefacts and other materials. It is worth noting that this approach to British prehistoric material culture was developed initially in Late Neolithic studies, as a means of analysing and explaining the distinctive nature of artefact deposits at henges and other sites (eg Richards and Thomas 1984; Pollard 1995; Thomas 1999, 62-88). In fieldwork terms, this demands context-sensitive and precise recording methods, 100% recovery of artefacts and other materials from single depositional contexts, and rejection of inappropriate practices such as 50% sampling (eg half-sectioning) of pits and other features.

• Middle and Late Neolithic material culture studies in the West Midlands need significant attention, especially in relation to chronology (which requires radiocarbon dating programmes and typo-chronological study), function (there is a need for residue and microwear analyses), raw material sourcing and the social organisation of production (Ray 2007, 74; Barfield 2007, 106; cf Lithic Studies Society 2004).

• A major weakness in Neolithic material culture studies in the region is the extreme scarcity of specialists to undertake artefact analyses and related research work. There is an urgent need to develop systems of training and support for new specialists in artefact studies, especially within field archaeology organisations (Ray 2007, 74).

• It is important to emphasise that recognition of the potential of Middle and Late Neolithic sites depends on the familiarity of curatorial and field archaeologists with the material evidence, current interpretative frameworks and research priorities. Ray (2007) has suggested that there is a need to develop expertise in these areas, and to encourage a more dynamic and interactive research culture among the archaeologists working in the region.

Conclusion

Of the earlier prehistoric periods in the West Midlands, the Middle and Late Neolithic has perhaps attracted the most visible and consistent attention in both regional and national research literature. This in part reflects the distinct and relatively substantial nature of the material evidence, and in part the especially prominent research profile that Middle and Late Neolithic studies have had in recent prehistoric archaeology in Britain. On closer inspection, however, much of this attention in the West Midlands has focused on a very small number of sites and landscape areas, above all the cursuses, ring ditches and long enclosures at Barford, Wasperton and Charlecote in the Avon valley. In contrast, other significant sites and sources of evidence have either been neglected (such as the important monuments and lithic concentrations in western parts of the region, including Stapeley Hill and Clun in Shropshire), or have yet to reach a wider audience (especially recent work at Church Lawford in Warwickshire, and Catholme and Whitemoor Haye in Staffordshire).

In many respects, Middle and Late Neolithic monuments, ceremonial complexes and landscapes in the West Midlands are similar to those in other regions, and are directly comparable in terms of their research significance and potential. Nonetheless, the relative rarity of monument groups, their wider spacing in the landscape, limited evidence for large-scale construction events and the lack of evidence for continuing development of ceremonial centres during the Late Neolithic, raise important questions concerning the relative scale and complexity of social and political communities and the nature of social change in the region. Evidence for economic practices and settlement (in the broadest sense) is also similar to that found in other regions, with very much the same research potential. Given the general absence of architectural remains, the investigation of pit deposits is especially important, whether these were the outcomes of routine daily tasks around occupation sites or more formal social practices at ‘special’ locales in the landscape. At the same time, there is clearly considerable potential in the West Midlands for pursuing research questions that focus on the relationships between monuments, settlements and economic activities, especially through landscape-scale studies.