Identifying and protecting historic landscapes.

Darvill, Timothy ; Gerrard, Christopher ; Startin, Bill 等

Six years ago, Darvill and colleagues reported (ANTIQUITY 61: 393--408) on the Monuments Protection Programme, a new English initiative to build, from a century of haphazard acts of site protection, a set of balanced judgements and priorities by which to recognize ancient places that are more precious, genuinely of a national importance. The Programme, they tell ANTIQUITY, has now completed the first-stage review of information in local sites and monuments records and is proceeding with the identification of nationally important monuments in every English county. This further paper reports on how the Monuments Protection Programme is addressing landscapes, as distinct from 'spot sites' with clear limits, where the matters of defining a 'relict cultural landscape' and judging relative value are harder.

One of the most rapidly expanding branches of archaeological research is that concerned with historic landscapes. The signs are all around: numerous books and articles (e.g. Fowler 1970; Aston 1985; Coones & Patten 1986); a Society for Landscape Studies; many survey and excavation projects with landscape-related questions at the heart of their research designs (English Heritage 1991: 37--9); and the idea of landscape study figuring prominently in the research objectives promulgated by the major period societies (e.g. Prehistoric Society 1988).

This paper sets out for discussion the approaches to the archaeology of landscape currently being developed within the Monuments Protection Programme for England. It is divided into four sections: first a background to archaeological interest in landscapes; the second deals with what are here termed 'Relict Cultural Landscapes'; the third considers the protection and preservation of historic landscapes; while the final section addresses the importance of archaeologically meaningful historic landscapes. The main purpose of the paper is to introduce, with examples drawn from the authors' experiences, the proposition that relict cultural landscapes should be defined on the basis of recorded patterns within arrangements of archaeological remains rather than simply as the identification of extensive monuments, clusters of monuments (complexes), islands of good preservation, or the setting of known sites.

Archaeologists and landscapes

Interest in the historical dimensions of the landscape extends back into the later decades of the 19th century, but did not become a significant aspect of archaeological research until the mid 1960s. Several factors prompted the development of a school of landscape archaeology, among them the ever-expanding spatial scale of archaeological investigations and theory-building, contributions from disciplines such as historical geography and anthropology, increasing availability of appropriate source material, and the influence of environmental politics on the kinds of archaeological projects carried out (Darvill 1992; Darvill et al. forthcoming).

Perhaps because of the wide range of stimuli which influenced the development of the subject, and the speed with which it has come about, there is currently no coherent body of general theory for the analysis and understanding of the archaeology of landscapes (Coones 1985), nor indeed is there any generally accepted terminology. Ask almost any archaeologist working in England to name a few historic landscapes and the answer will almost certainly include: Salisbury Plain, Dartmoor, Bodmin Moor, upper Thames Valley, the Fens, Somerset Levels and Pennines. Examined closely, all of these prove to be topographically defined areas with special characteristics of preservation, the archaeology of which has been subjected to intensive investigation in recent years. Such topographical determinism in the definition of historic landscapes fails adequately to reflect even the simplest theoretical proposition in which reconstructions of past land-use emphasize the simultaneous exploitation of different environmental zones.

If the general question just posed is followed by an invitation to describe, say, the Bronze Age landscape of Dartmoor or the Iron Age landscape of the upper Thames Valley, the answer given would almost certainly comprise statements about particular monuments (probably excavated sites) and a general environmental overview. The linking of sites to one another, to the environment and to the culturally specific conceptualization and use of space is very rarely expounded, and yet such things are widely recognized as relevant to an understanding of the past.

These examples illustrate the gap between the recognition of important ideas and concepts about the archaeology of landscape and the difficulty of handling their implications in a theoretical or practical way. Elsewhere (Darvill 1992; Darvill et al. forthcoming) we report an analysis of the main approaches used in studying the archaeology of landscapes and consider the theoretical orientations, strengths and weaknesses of each. Underlying the diverse range of approaches there are, however, three principles relevant to developing more coherent archaeological theory about the historic dimensions of landscapes. As will be shown, these in turn allow the development of more systematic approaches to the management of historic landscapes. The three underlying principles may be summarized as follows:

1 Almost any defined area of countryside contains within its fabric direct archaeological evidence relating to aspects of its former state and the processes, physical and social, which have contributed to the formation of its present appearance and use.

2 The quality and quantity of archaeological evidence that may be present in defined areas of countryside varies greatly from rich and extensive arrangements of vertically or spatially related deposits down to poor and unconnected deposits.

3 In a few cases the scale and integrity of the archaeological deposits represented will reflect either the whole history of the former activity -- and by implication land-use -- within a defined locality, or deposits reflecting the widest identifiable spatial unit or territory in which given communities operated.

The first two principles are well understood in archaeological circles. Indeed, in recent years the range of evidence recognized as being a part of the historic dimension of the landscape has increased dramatically with the inclusion of such 'hard' features as hedgerows, walls and battlefields, and 'soft' elements such as associations with historical characters and events. The third principle is less widely developed even though it represents a fundamental of archaeological research: the investigation of social, symbolic, economic, aesthetic and environmental relationships and patterning at a scale which corresponds to the spatial reference set of past communities. It is to this potentially new aspect of historic landscape studies, what are here called 'relict cultural landscapes', that we now turn.

Relict cultural landscapes

A relict cultural landscape is taken to be exactly what the name suggests: a piece of natural or artificial scenery containing remains relating to a particular form, stage or type of intellectual development or civilization which exists now in the same pattern or arrangement as in some previous age. To this may be added the imperative that sufficient archaeological evidence still exists to allow the study of socio-cultural patterning at a larger scale than is possible from individual monuments or groups of monuments.

Relict cultural landscapes do not, however, simply exist to be discovered through the imposition of a prescribed methodology. They are as much products of their theoretical conceptualization as any other defined entity used in archaeological analysis.

At one level, relict cultural landscapes form a natural extension of the well-established hierarchy of definable archaeological units, namely action-related site-specific elements (i.e. context, feature, component) and activity-related monument-specific elements (i.e. monument, group, cluster, complex). The third, additional, tier comprises patterns and regularities among groups and clusters of monuments (see Darvill 1992 and Darvill et al. forthcoming for further discussion). However, just as a monument is not simply a random selection of diverse components, so a relict cultural landscape is more than an unstructured collection of monuments. In exploring the definition of relict cultural landscapes three sets of factors are especially important.

Integrity and articulation

A relict cultural landscape is a seamless web of appropriated space variously defined, subdivided and conceptualized in social, structural, functional and cognitive terms by the communities that occupied it. Archaeological remains will rarely completely cover the area, although the spatial integrity of the landscape is only firm if the overall picture of land-use is known. In this connection land-use includes archaeologically 'blank' areas which were not the subject of intensive activity but which nevertheless existed as significant pieces within the overall pattern.

A relict cultural landscape is also likely to possess geographical or topographical coherence as a macro-region, perhaps a range of hills, a river valley, a peninsula or a plateau. Geographical integrity occurs in the sense that it is a continuous area rather than a collection of discrete unconnected topographic units, although its extent need not coincide with present land-use.

The archaeological elements of the landscape need to be articulated or connected together in a coherent way through one or more articulating features. These may be natural features such as rivers, escarpments, coastlines, and outcrops of particular minerals, or man-made such as roads, tracks, field systems and linear earthworks. The importance of articulating features is that they provide a framework which physically connects elements and delimits spaces.

Diversity and structure

Typically, a relict cultural landscape exhibits diversity of environment and diversity in the classes of monuments represented. Diversity of environment may be seen in terms of differences in soil type, relief, distribution of natural resources or aspect. Diversity in the kinds of archaeological monuments represented will often be most visible as functional variations, for example settlements, ritual structures, agricultural remains and industrial workings.

Structure appears as regularity in the occurrence of particular classes of monuments in the groups, clusters or complexes represented and/or in the way that archaeological remains relate to environmental differences. Structure may also be seen in the symbolic classification of landscape by its users: the special significance attached to rivers, lakes, hilltops, woods or indeed any other features given meaning. Ethnographic studies repeatedly suggest that societies divide up the space which they occupy, whether at the micro-scale of individual sites or the macro-large scale of whole landscapes, in terms of mythical or cosmological order (cf. Hugh-Jones 1979; Wheatley 1971). Such ordering has a direct relationship with social action and thus with archaeological remains.

Pattern and repetition

Arrangements of monuments, groups, clusters and complexes in set patterns are typical of the archaeological record, but when the patterns repeat themselves several times over it is possible to be fairly confident that the documented picture is a relict cultural landscape, as for example with a string of medieval villages each with its field system around about, pasture-land on higher ground in one direction and meadow-land beyond the field system on low-lying ground in another.

In addition to these characteristics there are three other more general considerations that must also be taken into account.

Scale

Relict cultural landscapes do not come in preset shapes and sizes and thus the scale of the analysis needed to recognize them cannot be predetermined. Fundamental sociological characteristics of defined societies, for example the degree of social complexity, the prevailing modes of production or political structure, may act as a general measure of scale for relict cultural landscapes. Thus among sedentary autonomous communities individual territorial units or areas of activity are likely to be quite small while, say, for medieval times in central southern England it is unlikely that patterns become repetitious in areas much less than 5 km across.

There will always be differences in the density of activity from one area to the next, however, and, in general, the greater the density of settlement and land-use the easier it is to recognize relict cultural landscapes because the patterning and repetition becomes clearer as its physical imprint on the countryside becomes stronger. Sometimes it will be possible to identify nested arrangements of landscapes relating to several different scales at once.

Completeness

Completeness must vary with antiquity and the impact of more recent activity. Thus today's landscape is as complete as it can be, but with distance back in time there will be gaps in our knowledge of the original pattern and discontinuity in articulating features; occasionally whole units which would otherwise allow the recognition of repetition have been lost. Accordingly, allowances have to be made for the degree of completeness that is acceptable for early periods or areas of the country that have been lived in and worked over more thoroughly than others.

Edges

There are no formal edges to relict cultural landscapes, only arbitrarily defined boundaries caused by differential survival or analytical precision. The landscape is, and always has been, a seamless canvas extending out in all directions; even if we think we have the limits of a pattern of activity in the past it is likely that some activities took place at an altogether different scale and are lost to us.

In summary, relict cultural landscapes can be conceived of as articulated sets of monuments that exist in repetitious patterns of integration and association. The recognition of such landscapes requires considerable knowledge about the archaeology of a particular area, typically mapped information. It should be emphasized, however, that just because a tract of countryside has been intensively studied (an event) there is no automatic case for claiming the existence of one or more relict cultural landscapes (an interpretation). Fielddata needs to be carefully examined in the light of theoretical propositions relating to the nature and form of relict cultural landscapes in general, and the qualities of examples in defined time and space contexts in particular, before interpretative statements can be made.

Where the level of information about an area is sufficient to permit searches for possible relict cultural landscapes the task of unpicking or sorting out the data with a view to identifying patterns can begin. Skilled interpretation is needed in order to understand and simultaneously balance up the dating, relationships, function, structure, form and meaning of the data. Just as important as the identifiable monuments is knowledge of the apparent spaces between monuments: are they the products of data-recovery, meaningful zones within which activities having no archaeological manifestation took place, or were natural features (e.g. marsh, stream etc.) present in earlier times which are not present now?

Throughout any analysis aimed at the identification of a relict cultural landscape two sets of questions must continually be asked: first, whether the remains in question are of the right scale: should the evidence in fact be interpreted as simply a large monument, a group of monuments, a cluster of monuments, or a complex; secondly, whether the evidence has integrity and articulation, diversity and structure, and pattern and repetition.

Although it is not possible systematically to classify relict cultural landscapes on the basis of present evidence it is nonetheless possible to recognize two main kinds on the basis of whether the remains belong largely to one main period of activity or to a succession of different phases of use. These may be called synchronic and diachronic relict cultural landscapes respectively.

Synchronic relict cultural landscapes

Synchronic relict cultural landscapes characteristically contain monuments of one main period. Their principal value is as evidence for the form, structure, arrangement, sub-division and appearance of a landscape at a particular horizon in its history. This is not to say that the area covered by such landscapes was not used during other periods, only that the archaeological evidence of earlier or later use is slight by comparison with the activities represented by the main set of monuments.

Synchronic relict cultural landscapes are not defined by the density of sites present, nor should they be confused with large sites, or areas of countryside containing numerous monuments. Rather they must show chronological homogeneity within and between the monuments represented, and display spatial integrity and clear signs of articulation between the main archaeological elements present. Where preserved, synchronic relict cultural landscapes are often visually impressive and parts of them may well form distinctive visual envelopes.

Several synchronic relict cultural landscapes have been documented (e.g. Fleming 1978; 1983); the following example is of relatively recent date.

Woodchester Valley, Gloucestershire

The Woodchester Valley lay at the heart of the south Cotswold woollen industry. Watermills were established on the Nailsworth Stream from at least Domesday times, but what survives today as a synchronic relict cultural landscape belongs to the 18th and early 19th centuries.

In 1850 there were 14 mills along the Nailsworth Stream north of Nailsworth itself, with others to the south and along tributary streams (FIGURE 1). Each 'mill' is actually a complex, as illustrated by Dyehouse Mill. Initially recorded in the early 16th century (Herbert 1976: 196--7), the first edition 25[inches] Ordnance Survey map shows the picture as it was in the 1880s (FIGURE 2). The main mill lay at the focus of the complex. Around it were several ancillary buildings, mostly workshops and stores. The mill-pond lay upstream with outflow leats to the north. On the east side was a circular cloth-drying kiln, now demolished. A large clothier's house (Dyehouse) lay near by to the north of the main mill. Immediately around the clothier's house were several weaver's cottages, some of which still survive. The mill owner's house lay some distance away in an elevated position, but was connected to the mill by Gydynap Lane. The area between the owner's house and the mill is known as Rack Hill and although apparently bereft of archaeological evidence was originally the place where cloth from Dyehouse Mill was hung up to dry.

[CHART OMITTED]

As a complex there are thus six or seven monuments linked together by spatial and stratigraphic associations, by functional ties, by relationships of ownership and occupancy and by the use of space and position to indicate status and order. The gaps between monuments in the complex are crucial to understanding the whole.

All the mills along the Nailsworth Stream are complexes, the stream providing the dominant articulating element, although other articulating features can also be seen: the road along the valley bottom (now the A46) constructed in the 1780s and the railway line along the valley opened in 1867. One effect of the railway was to provide new opportunities for development within the mill complexes, and at Dyehouse Mill this can be seen by the establishment of The Grove Sawmill to the southeast of the earlier focus. Closer inspection reveals yet more articulations, particularly footpaths and lanes linking hamlets, villages, farms and mills. These draw in the cultivated fields, pasture lands, commons and woods that surround the industrial areas.

In one sense the Woodchester Valley is unique, but in another it is typical of several parts of the south Cotswolds (cf. Tann 1967) and could be presented as a regional stereotype for landscapes dominated by the use of water power for industry.

Diachronic relict cultural landscapes

Diachronic relict cultural landscapes characteristically contain superimposed patterns of several main periods. Their value is the evidence they provide for changing or continuous patterns of landscape use and activity within a single area.

Diachronic relict cultural landscapes are not simply multi-phase monuments, neither are they fields full of inter-cutting and superimposed cropmarks. Rather they comprise chronologically heterogeneous arrangements of monuments and complexes which display spatial integrity, and show clear signs of stratigraphic articulation which allows the time-depth of the pattern to be documented. Good cases of repetitive spatial patterning are difficult to find with diachronic relict cultural landscapes, probably because elements in the pattern have been destroyed or because the scale of repetition at one stage or phase in the evolution of a landscape does not coincide with the scale of repetition at another.

The identification and evaluation of diachronic landscapes is based upon an ability to unpick something of the pattern in order to understand the chronological span of the landscape as a whole through an analysis of the individual monument classes represented. Such landscapes need to contain classes of monument datable to most if not all the main periods encompassed within the putative span of the complete palimpsest. Several diachronic relict cultural landscapes are well documented (e.g.. Coles & Coles 1986; Jones 1974); the following example illustrates the main characteristics of an essentially prehistoric one.

Stonehenge, Wiltshire

Recent studies of the area around this well-known monument (RCHM 1979; Richards 1984; 1990) suggest that four successive phases can be recognized. Within each phase it is sometimes difficult to find appropriate articulating features which link monuments and complexes because intensive arable cultivation and the deliberate bulldozing of upstanding monuments in the 1950s and 1960s has left many sites isolated from their neighbours. Between phases the picture is rather different. Each phase focused on a slightly different but overlapping section of the landscape which allows the chronological articulation of elements to be documented. Repetition in the patterns of activity is not especially marked within each phases, but can be seen through time.

The earliest phase (see Richards 1990: figure 157) is represented by long barrows, oval barrows, cursus monuments, shafts and occasional pit clusters. A causewayed camp lies to the northwest at Robin Hood's Ball. The overall pattern is one of dispersed activity.

The second phase (see Richards 1990: figure 158) dates to the later Neolithic. The most well-known monument is Stonehenge itself, at this stage an enclosed cremation cemetery rather than the stone-built structure of later times. The henges at Coneybury and Woodhenge, and the henge-enclosure with its massive internal timber structures at Durrington Walls, also belong to this phase. Round barrows were being built from the later 3rd millennium BC onwards and round barrow cemeteries began to take shape from the early 2nd millennium BC. Industrial sites -- flint mines and working areas -- are known near Wilsford Down. What went on between these monuments is not precisely known, although it is believed that open countryside prevailed. The pattern that can be perceived involves a central zone given over to ritual and ceremony surrounded by industrial workings and settlement areas.

In the third phase (FIGURE 3), broadly the early Bronze Age, ritual and burial monuments dominate the area. Stonehenge is modified with the addition of stone circles and the construction of the Avenue extending northeastwards from the remodelled earthwork enclosure. Round barrows were constructed in some numbers, mostly within round barrow cemeteries. Repetition here is provided by the fact that some barrow cemeteries include earlier long barrows. Most of the other monuments in use during late Neolithic times were abandoned. Overall, the area around Stonehenge seems to have become a ceremonial place arranged so that the barrow cemeteries overlooked Stonehenge itself, although the full implications of the symbolic arrangement of the landscape have yet to be explored.

[CHART OMITTED]

In its fourth phase (FIGURE 4), conventionally the middle and later Bronze Age, the area again became subject to a more diverse range of uses. Stonehenge itself continued to be modified, the concentric stone circles of Period III now being the focus of attention, although the Avenue remained significant and may have been extended eastwards. Deverel-Rimbury style urns containing cremations are known in some small bowl barrows and flat cremation cemeteries connected with round barrow cemeteries. These features again provide an element of repetition and together serve to document continuing burial and ceremonial activity.

[CHART OMITTED]

At least three possible settlements are known, and associated with them are the fragmentary remains of two or three regular aggregate field systems. Several linear earthworks are also known and as extensive monuments both the linear boundaries and the field systems provide an additional element of articulation which is as much stratigraphic in the sense of defining relationships with earlier monuments as spatial in the sense of linking contemporary monuments.

In total, these successive phases of landscape form provide a picture of changing land-use and settlement through some 25 centuries. Some of the monuments may not be typical of those found elsewhere, but many of the main features of the landscape are entirely typical.

Protecting and managing historic landscapes

Taking the three principles of landscape archaeology already outlined it is possible to identify three corresponding spheres of interest relevant to the protection and management of historic landscapes. These spheres of interest can be distinguished from one another by two factors: the degree of integration that exists between the objectives of archaeological resource management and other heritage/conservation interests; and the overall archaeological value, judged only on archaeological merit, of the resource represented.

Archaeology in the countryside

The reaffirmation of a historic or archaeological dimension to all countryside and townscape reinforces the need to integrate archaeological considerations alongside other conservation and environmental interests (cf. Darvill 1987), for example material matters such as flora, fauna, topography, geology and scenery, as well as spiritual matters such as aesthetics, artistic and literary associations, folklore and tradition. Indeed, much has already been achieved in this sphere through integrating the archaeological dimensions of the countryside and townscape with general 'heritage' designations and management initiatives as part of a more generalized concern for the total environment (e.g. HMG 1990).

At the national level, the National Parks, Environmentally Sensitive Areas and Areas of Outstanding Natural Beauty are important and increasingly concerned to integrate fully archaeological matters with conservation on a general front, although in some cases this has yet to be backed-up by appropriate clauses in the relevant legislation.

The growing recognition that most of the landscape which can be seen today is the product of human interference is nowhere more clearly illustrated than with reference to the changing definition of the Areas of Outstanding Natural Beauty (AONBs). Within these AONBs, what is known as Landscape Assessment has become the practical technique for describing, classifying and evaluating landscape quality (Countryside Commission 1987). Such assessments now include consideration of the historical and archaeological dimensions of the countryside, and this has carried through into recent statements on AONBs by the Countryside Commission (1990a: 2)

At the regional and local level, guidance from the Countryside Commission and Nature Conservancy Council on the landscape issues to be covered by District Local Plans is especially important (Countryside Commission 1990b: 14).

Archaeologically rich countryside

In some areas archaeological remains are especially prominent and deserving of preservation and management in their own right. Many such cases will be extensive monuments (e.g. field systems) or complexes and in some ways such countryside can be seen as the archaeological equivalent of the habitats and environments preserved for nature conservation interests as Sites of Special Scientific Interest or National Nature Reserves. Typical candidates for such recognition are landscapes in which a high density of archaeological monuments has been recorded. In general, however, the archaeological resource in these areas is rarely continuous, and occasionally there may no longer be any remains visible. Preserving potential (but undocumented) sources of archaeological data (e.g. an area of wetland or a piece of uncultivated Common Land) may also be considered relevant within this sphere of interest.

At the international level World Heritage Sites provide a level of recognition for archaeological monuments in the same way as they also recognize natural heritage interests. Nationally, the Register of Parks and Gardens compiled by English Heritage already recognizes one particular kind of archaeologically rich countryside, and other such registers are under active consideration.

At the regional and local level the designation of county- or district-based archaeologically rich landscapes is well under way. Wiltshire County Council, for example, has designated 39 Areas of Special Archaeological Significance (WCC 1986: 49) throughout the county, while in Somerset the County Council have designated 27 areas of the Levels and Moors as Areas of High Archaeological Potential (Brown et al. 1988: 113). The primary objective of the Somerset designations was to identify important areas and to ask private landowners to inform the County Planning Officer of any proposed works that might destroy archeological remains so that an appropriate response could be mounted. Both these archaeologically-based area designations, and similar arrangements in other counties, are voluntary schemes, but in practice serve to complement the Conservation Areas designated by local authorities mainly for the quality of their built environment (Pearce et al. 1990).

Relict cultural landscapes

The third and final sphere of interest looks forward to better understandings of landscapes as spaces within which human communities operated. In this, the archaeological remains are themselves of paramount importance; with relict cultural landscapes the emphasis is directed towards tracts of land which demonstrably contain the building blocks of an understanding of behavioural patterning, the social use of space, at a scale greater than that offered by a single site, group of sites or complex.

At present it seems that there are relatively few well-documented examples of relict cultural landscapes, but this may be a result of the fact that interest in their definition is a relatively recent phenomenon. It must also be assumed that some periods in some areas will be very poorly represented by patterns of data that can be consolidated to the level appropriate for the definition of relict cultural landscapes. Relatively recent relict cultural landscapes will always be the easiest to document, and from the mid 19th century onwards large-scale Ordnance Survey maps provide almost all the information required. The real challenge presented by a tighter conceptualization of relict cultural landscapes as archaeologically meaningful entities is to identify medieval and earlier examples.

The importance of relict cultural landscapes

The identification of relict cultural landscapes is not an end in itself. The scope for using relict cultural landscapes as the framework for the analysis of past human communities is immense, and in time will have important implications for the wider field of landscape studies. The key questions that can be addressed go beyond straightforward relationships between people and their environment to include such issues as: How was the landscape organized? What did the landscape look like at particular times in the past? How did the form and physical characteristics of the landscape influence the communities who lived in it in practical, cultural and symbolic terms? How did communities perceive and structure the space in which they operated? How did the landscape come to look as it does now? And what were the relationships between social action and material culture at different spatial scales?

All this potential hinges on there being extant data for study and analysis, and that in turn depends on the careful management and stewardship of the resource we have. But the recognition of relict cultural landscapes as identifiable and definable elements of the archaeological record worth preserving has come rather late, given the nature and extent of much recent land-use change. Single monuments (and closely associated groups, clusters and complexes) have been subject to preservation by Scheduling for over a century. Now there is also a need to study, and protect for future study and appreciation, relict cultural landscapes. Four main reasons for this can be recognized:

1 Relict cultural landscapes represent the physical remains of past patterns of human endeavour at the largest scale (i.e. sub-regional level) at which it is currently practicable to investigate such matters in detail.

2 Synchronic relict cultural landscapes are the only remaining physical manifestations of the hierarchy of monuments, groups, clusters and complexes and intervening spaces that once existed and which provide insights into communal activity at a scale greater than is possible through the study of single monuments.

3 Diachronic relict cultural landscapes contain the only remaining large-scale evidence for patterns of successive adaptation and change by human communities. They allow themes such as burial, settlement, economy, symbolism and the use of space to be explored through time while parameters of topography and geographical setting remain relatively stable.

4 Relict cultural landscapes represent the most complete sources of information relevant to the interpretation and understanding of the cultural influences that have in the past shaped the overall landscape.

All four of these points emphasize the fact that the aim of preservation is not just to protect landscapes for the particular problems of today but rather as an important archaeological data bank which preserves options for use in the future.

Acknowledgements. During the preparation of this paper many individuals have been asked for information or opinions on specific points, among them Mick Aston, Roger Mercer, Nick Johnson, Peter Rose, Andrew Fleming, Christopher Taylor, Richard Bradley, Francis Pryor, Roy Canham, Humphrey Welfare, Don Benson, David Crossley, Tom Lane, Brian Simmons and David Austin. Valuable comments and suggestions on earlier drafts of the paper have been received from Geoff Wainwright, Graham Fairclough, Dai Morgan Evans, Roberta Gilchrist and Mick Aston. To these, and any others whose contributions have escaped specific attribution in our records, we offer grateful thanks. The figures in this paper were drawn by Catriona Turner-Walker; Nicola King assisted with editorial work.

References

ASTON, M. 1985. Interpreting the landscape. London: Batsford.

BROWN, K., R. CROFT & R. LILLIFORD. 1988. Conserving the historic landscape, in M. Aston (ed.), Aspects of the medieval landscape of Somerset: 109--27. Taunton: Somerset County Council.

COLES, B. & J. COLES. 1986. Sweet Track to Glastonbury: the Somerset Levels in prehistory. London: Thames & Hudson.

COONES, P. 1985. One landscape or many? A geographical perspective, Landscape History 7: 5--12.

COONES, P. & J. PATTEN. 1986. The Penguin guide to the landscape of England and Wales. Harmondsworth: Penguin.

COUNTRYSIDE COMMISSION. 1987. Landscape assessment: a Countryside Commission approach. Cheltenham: Countryside Commission. CCD 18.

1990a. Areas of Outstanding Natural Beauty: a policy statement 1990. Cheltenham: Countryside Commission. CCP 302.

1990b. Countryside and nature conservation issues in district local plans. Cheltenham: Countryside Commission. CCP 317.

DARVILL, T. 1987. Ancient mounments in the countryside: an archaeological management review. London: English Heritage. Historic Buildings and Monuments Commission for England Archaeological Report 5.

DARVILL, T., C. 1992. Monument evaluation manual Part III: Relict cultural landscapes. London: English Heritage.

DARVILL, T., C. GERRARD & B. STARTIN. Forthcoming. Archaeology in the landscape: a review.

DARVILL, T., A. SAUNDERS & B. STARTIN. 1987. A question of national importance: approaches to the evaluation of ancient monuments for the Monuments Protection Programme in England, Antiquity 61: 393--408.

ENGLISH HERITAGE. 1991. Exploring our past: strategies for the archaeology of England. London: English Heritage.

FLEMING, A. 1978. The prehistoric landscape of Dartmoor. Part 1: South Dartmoor, Proceedings of the Prehistoric Society 44: 97--124.

1983. The prehistoric landscape of Dartmoor. Part 2: North and East Dartmoor, Proceedings of the Prehistoric Society 49: 195--242.

FOWLER, P.J. (ed). 1970. Archaeology and the landscape. London: John Baker.

HERBERT, N.M. 1976. Minchinhampton, in N.M. Herbert (ed.), The Victoria history of the counties of England -- Gloucester: 11: 184--206. Oxford: Oxford University Press.

HMG [Her Majesty's Government]. 1990. This common inheritance: Britain's environmental strategy. London: HMSO. Command Paper 1200.

HUGH-JONES, C. 1979. From the milk river. Cambridge: Cambridge University Press.

JONES, M.U. 1974. Excavations at Mucking, Essex: a second interim report, Antiquaries Journal 54: 183--99.

PEARCE, G., L. HEMS & B. HENNESSY. 1990. The conservation areas of England. London: English Heritage.

PREHISTORIC SOCIETY. 1988. Saving our prehistoric heritage: landscapes under threat. London: Prehistoric Society.

RCHM(E). 1979. Stonehenge and its environs. Edinburgh: Edinburgh University Press.

RICHARDS, J. 1984. The development of the Neolithic landscape in the environs of Stonehenge, in R. Bradley & J. Gardiner (ed.), Neolithic studies: 177--88. Oxford: British Archaeological Reports. British series 133.

1990. The Stonehenge environs project. London: Historic Buildings and Monuments Commission for England. English Heritage Archaeological Report 16.

TANN, J. 1967. Gloucestershire woollen mills. Newton Abbot: David & Charles.

WCC [Wiltshire County Council]. 1986. Wiltshire landscape local plan. Trowbridge: Wiltshire County Council.

WHEATLEY, P. 1971. The pivot of the four quarters. Edinburgh: Edinburgh University Press.