From The Ground Up

The Publication of Archaeological Projects

a user needs survey


2 Archaeological project publication: an historical review

2.1 Introduction

A review of the history of fieldwork publication, with its underlying principles and policies, will help to place the present survey in context.

The literature on publication is diverse. It includes governmental and society documents outlining publication policy (eg Cunliffe 1983; English Heritage 1991a; Historic Scotland 1996), commentaries on these official policies and alternative approaches to preservation by record and dissemination of that record (eg Alcock 1978; Thomas 1991), and more theoretical or polemical debate (eg Hodder 1989; Tilley 1989). The subject matter is comparably broad, ranging from the principles underlying publication, to more specific issues of style, content and media. It is useful to begin by discussing the traditional, orthodox publication model, which remains influential. This is followed by a consideration of the development of publication policy over the last three decades, and variation in current policy in Britain and Ireland. Later sections deal with theoretical commentaries and debates, publication outlets, media and costs, and, finally, approaches to the hitherto-neglected issue of user needs.

2.2 Publication as preservation

2.2.1 Traditional principles: the `Cranborne Chase tradition'

`A discovery dates only from the time of the record of it, and not from the time of its being found in the soil.' This classic sentence of Pitt-Rivers proclaims fairly and squarely the ultimate moral and scientific duty of the field-archaeologist. It may be amplified by the familiar corollary that the unrecorded excavation is the unforgivable destruction of evidence; and the more complete and scientific the excavation, the greater the measure of destruction. (Wheeler 1954, 182)

Between the publication of Cranborne Chase (Pitt-Rivers 1887-98) and the 1960s, at least, it was generally accepted that excavation is destructive, a single act that simultaneously reveals and obliterates unique data, and that archaeologists have a corresponding responsibility to publish their sites and finds in extensive, if not exhaustive, detail:

Everyone who by excavating destroys the evidence contained in the ground undertakes the responsibility of restoring that evidence on paper . . . fully but concisely. (Curwen 1937, 6)

Thus, throughout the first half of the twentieth century, publication was seen as an integral part of the excavation process. In much of the literature of that period, however, there was little mention of archiving: the only record considered effective was full publication - in effect the published report and the archive were regarded as one and the same thing. The published record was in its turn conflated with the process of excavation: `discovery' takes place only from the `moment of record', and evidence that has been destroyed through excavation is `restored' on paper. In short, it was assumed that the entire site and its discovery through excavation could be encompassed by the published record, which in turn would facilitate its re-construction and re-interpretation by others, thus compensating for its destruction (Atkinson 1946, 178; Curwen 1937).

These premises have had colossal impact on the nature and form of publications. To satisfy the expectation that a published record would restore evidence on paper, it was also assumed that a publication could, and should, provide an objective record of evidence encountered in the ground. This is why most commentaries have insisted on the importance of maintaining a distinction between the description of the `facts' and the author's subjective interpretations. Atkinson, for instance, argued that detailed discussion of the structural finds and stratification,

... should be confined strictly to the description of the finds, that is, to the presentation of the evidence; discussion of the meaning of the evidence should be reserved for the following section. For it should be borne in mind that the excavator's judgements upon his discoveries, while in the nature of things likely to be better informed than those of others, are yet not final or infallible. It is only right, therefore, that the facts should be kept separate from the excavator's personal interpretation of them. (Atkinson 1946, 180 [emphasis in original])

It followed that

... the first duty of the excavator is to publish the facts; purely speculative considerations must take second place in the report, and for the sake of economy in space and expense should not be unduly elaborated. (ibid)

With these pre-requisites in mind, the ideal model for the structure and form of an excavation report became established. It embraced:

2.2.2 The influence of the traditional model

The impact of this traditional model and the principles that fed it cannot be overemphasised. A glance at a sample of post-war excavation reports reveals that, despite individual idiosyncrasies, the majority contain the constituent elements discussed above, and maintain considerable differentiation between description and interpretation (eg compare Frere & Wilkes 1989; Piggott 1962; Rodwell 1988; Wainwright & Longworth 1971; Whittle 1997). The influence of the traditional model can also be traced through more recent commentaries on archaeological fieldwork. For instance, although more detailed, the approach promoted by Grinsell et al (1974, 63-5) contains the same key elements as those outlined above, namely a summary; an overview of various aspects (although further differentiated); a detailed description of each phase or structure - `the core of the report' (ibid); and finally interpretation and discussion. The importance of a distinction between description and interpretation is emphasised, as it is in other more recent manuals dealing with fieldwork and publication (eg Barker 1982, 232-3; Joukowsky 1980, 459).

Despite a shift towards more selective and synthetic publication, and an emphasis on the archive as the primary record (see Section 2.3), elements of the traditional model have remained influential in the publication policies of government bodies. It is not overstating the case to suggest that the rhetoric, or `discourse', surrounding the publication of fieldwork projects continues to be profoundly influenced by the orthodox model. As Lavell pointed out in her review of the history of publication in British archaeology:

... there is embedded in the national archaeological consciousness the feeling that the `Cranborne Chase tradition' is still the proper, the moral way to publish. (Lavell 1981, 95 [emphasis in original]; see also Derricourt 1984, 54)

Recent practical constraints may have tainted this ideal, but for many it still provides a, arguably the, powerful archetype.

2.3 The development of selective publication and research strategies

2.3.1 Pressures, influences and crises: an outline

Across much of Britain, at least, the 1960s and 1970s saw a publication crisis characterised by an increasing intensity of archaeological activity, a growing backlog and soaring costs (Ancient Monuments Board for England 1975, 1; Grinsell et al 1974, 18; Lavell 1981, 97-100; Mytum 1978, 43). In large part the backlog resulted from the continuing expectation of full publication against a background of an escalation in the numbers of sites being excavated. From the mid-1970s the acceleration was fuelled by a large uplift in government funding for rescue archaeology, and was accompanied by an increase in the size of the excavations, the scale of recording involved, the volumes of material being collected (environmental archaeology, for instance, advanced rapidly during that period) and the number and size of resultant specialist contributions. Beyond this, some of those involved already had publication backlogs, whilst alongside the expansion in excavation stood the ascending expectations of an emerging profession. At the same time, the costs of full publication for growing numbers of projects of increasing scale were rising sharply (see Section 2.6), while the older assumption that `writing up' was a gentlemanly pursuit for which no one would necessarily be paid died hard. In parts of the UK the Government itself exacerbated the problem by funding excavations - including its own - without earmarking resources for publication. Bearing in mind the received view that publication was part of the excavation process, not an add-on, this was a remarkable, if not irrational, position at which to have arrived.

Several working parties were convened to face the crisis during the 1970s and 1980s, and their recommendations are considered below. However, field archaeology is widely perceived to have experienced a further publication crisis during the 1990s (Cunliffe 1990; Thomas 1991; Carver et al 1992). Somewhat in anticipation of developer-funding, Cunliffe (1990, 668) was the first to identify this, arguing that unless developers are required to fund publication in full, as well as excavation and recording, `a vast amount of data will be lost to the discipline' (for an earlier review of similar problems in the American context see Renfrew 1982).

The publication in England of PPG16 on Archaeology and Planning in 1990, followed by policy guidelines for Wales (a Planning Policy Guideline in 1991, replaced by Planning and the Historic Environment: Archaeology in 1996) and Scotland (NPPG5 1994) brought a welcome insistence that the publication of results, as well as excavation and recording, should be the responsibility of the developer. Nevertheless, primary emphasis was placed on `appropriate and satisfactory provision for the excavation and recording of remains' (PPG16: para 25). Although publication is specified, its nature and level are undefined, leaving it exposed to minimalist interpretation in the competitive world of contract archaeology. Furthermore, as Thomas (1991, 823) has pointed out, developer-funding is by definition site-specific and concerned with recording rather than analysis or synthesis which would put the record in context (although the subsequent Scottish NPPG5 has included analysis). In result, a growing number of sites are being published as `grey' literature, available on demand rather than through traditional outlets, while it appears that less and less material is being adequately analysed or synthesised. Both trends have given rise to concern (eg Olivier 1996).

It is important to note that these publication crises did not affect all parts of the British Isles equally or at once. Wales and Scotland did not experience the same rapid increase in the intensity of rescue excavation during the 1960s and 1970s, and received only a fraction of the rescue budget (Barclay 1997, 10-11). In part this was because those countries contain fewer deeply stratified Roman and early medieval towns, and rescue funding was elsewhere initially channelled more towards sites at risk through urban development than to counter the effects of forestry, agriculture, or environmental change. In result, Wales and Scotland did not face the dual problem of publication backlog and increased publication costs to anywhere near the same extent, or as early, as in England (ibid, 16). In Ireland, likewise, the scale of excavation did not result in the massive backlog that accumulated in parts of England, although the situation has changed in more recent years with the expansion of commercial development in Ireland. However, the subsequent rise of developer-funding throughout Britain and Ireland has resulted in anxiety about publication, and the proliferation of grey literature.

2.3.2 The Frere Report: `Principles of Publication in Rescue Archaeology'

In reaction to the English publication crisis in the 1970s, the Ancient Monuments Board for England convened a committee under the chairmanship of Professor Sheppard Frere. Its report (AMB 1975) naturally enough became known as the Frere Report. The Frere Report fully endorsed the traditional view that archaeologists are under an obligation to produce a full record of their excavations, to restore and preserve the evidence which they have destroyed. However, Frere departed from orthodoxy in arguing that, given the crisis, `publication in printed form of all the details of a large modern excavation is no longer practicable' (AMB 1975, 2; our emphasis). The Committee accordingly felt it necessary to elaborate on the functions of fieldwork publication, particularly in relation to the discipline. Publication is necessary, it was argued, to

The Frere Report advocated a rationalisation of recording and publication. Four levels of recording were held to characterise the successful completion of an excavation (ibid, 3):

Level I - the site itself and the excavated finds
Level II - records produced on-site
Level III - full illustration and description of structural, stratigraphic and artefactual/ environmental data
Level IV - a synthetic description with supporting illustrations

Hitherto, full publication at Level III had been the accepted norm - at least in theory - but given the problems now arising, refined publication at Level IV was recommended. The adoption of Level IV publication was, however, only deemed acceptable if a Level III report was produced for archiving and was readily available on request. It was recognised that selectivity at Level IV would require a higher standard of archiving than was often practised, with all excavation records being properly organised, curated and accessibly housed. Consideration was also given to other, cheaper forms of dissemination at Level III, on request (see Section 2.6). In essence, the Frere Report responded to the publication crisis by advising a reduction in the amount of material that would go into print in monographs and journals, coupled with an improvement in the organisation and curation of archives.

The Frere Report was the first coherent attempt by a state body to address systematically the principles and practice of fieldwork publication. In retrospect it can be argued that Frere did not constitute a radical departure from traditional orthodoxies. Underlying principles remained largely unchanged, if elaborated, as did expectations regarding the constituent elements in the narrative and graphical record. All that the Report advocated was an uncoupling of that record (dubbed the Level III report) from the process of formal publication (Level IV). It was a pragmatic response to the costs of formal publication and the pressures on publication outlets.

Frere's recommendations on Level III and IV reports were implemented in England by the Inspectorate of Ancient Monuments in 1978 (DoE Advisory Note 16), and reinforced in 1980 and 1981 (DoE Advisory Notes 25 & 27). The distinction between Level III and Level IV reports entered archaeology's vocabulary, and debate about publication. Subsequently, however, a number of commentators voiced doubt as to whether the distinction had had a significant impact on publication practice. Lavell (1981, 103), for instance, argued that what seemed to have resulted were `Level III and a half' publications. Derricourt (1984, 54) said that the Frere Report seemed to have been `observed more in quotation than action'. Nevertheless, during the 1980s a number of fieldwork units (for example, the Department of the Urban Archaeology in London: Schofield 1987, 430) did incorporate the distinction into their publication and archiving policy during the 1980s.

2.3.3 The Cunliffe Report: `The Publication of Archaeological Excavations'

Although Frere's recommendations went some way to reduce publication costs and to relieve pressure on publication outlets, they failed to address the difficulties which arose from the large amount of time, and therefore cost, which were inevitable concomitants of increasing numbers of large projects. Indeed, the high standard of preparation required by Level III meant that in many cases more time was required for post-excavation work than had been allocated before.

Frere could not, in any case, crack the two problems which commonly underlay backlogs: that an archaeologist could not be publishing one site if (s)he had immediately proceeded to the excavation of another, and the historical underfunding of post-excavation work. The `backlog problem' thus remained unresolved, and a joint working party of the Council for British Archaeology and the Department the Environment was convened under the chairmanship of Professor Barry Cunliffe to consider it anew.

Whereas the Frere Report had focused almost exclusively on the management of records and the process of publication, Cunliffe's recommendations addressed the process of excavation and the generation of data (Cunliffe 1983, Introduction). Greater efficiency was the target, and was to be achieved through the introduction of phases of critical review. It was recommended that a research design should be produced prior to excavation, to ensure a balance between what is excavated and what can be adequately archived: `An indiscriminate pursuit of the policy of excavating more than can be processed inevitably leads to chaos and may defer publication indefinitely.' (Cunliffe 1983, 3, section 2.4). Subsequently, a post-excavation design should be produced to distinguish between work necessary for the preparation of an archive and publication synthesis on the one hand, and forms of research that were defined as non-essential to the `direct description' of the site on the other (Cunliffe 1983, 4, section 3.4). By such means, the minimum research required for adequate dissemination of knowledge could be identified (ibid, section 3.5). Other comparative or synthetic work not carried out for the project publication could then be undertaken (and published) as part of wider-reaching research programmes at a later date (ibid, section 3.6).

Selectivity in excavation and post-excavation was intended to contribute to more rapid and concise publication. Considerable emphasis was placed on the site archive, which was to provide a comprehensive record. In contrast, it was recommended that the publication should be a synthesis primarily derived from the research archive (ibid, 6, section 4.11). It was also argued that the report should be split into two parts (in most cases to be distributed together): (i) a report digest produced in conventional printed form; (ii) more detailed material concerning the structural, artefactual and environmental evidence to be disseminated as microfiche or fiche print out (the latter available on demand). The report digest (the conventional published element of the report) should constitute `an introduction, guide, and summary to the report' (Cunliffe 1983, 7, section 4.13).

Cunliffe's recommendations increased the discrimination required in excavation, analysis and publication. In its emphasis on the importance of a well-curated and accessible archive, and on targeted research and publication, the Report marked a departure both from the traditional model, with its ideal of full excavation and full publication, and the Frere Report, which had confined the latter to Level III. While the main constituents of the report digest were not dissimilar to the traditional model, the traditional `core of the report' – a detailed description of the evidence – was to be reduced to a summary, with detail confined to microfiche.

The Cunliffe Report had considerable impact (see Section 2.3.4). In implementation, however, it was problematic. The management structures required for delivery of its aims were not outlined in sufficient detail to enable smooth introduction of the phases of critical review upon which the Report's approach rested (English Heritage 1991a, 1, section 1.3). Furthermore, there was resistance to the emphasis on microfiche. Although the Report was accepted by the Department of the Environment, it was rejected by the CBA's own Council (see also Section 2.6 and Section 2.7).

2.3.4 `Management of Archaeological Projects'

Publication was next addressed in Management of Archaeological Projects (MAP), issued by English Heritage in 1989 and revised in 1991. Discussion here relates to the revised edition - commonly known as MAP2 (English Heritage 1991a).

MAP2 drew upon the Frere and Cunliffe Reports, especially the latter, an aim being to clarify the mechanisms required for the iterative reviews which Cunliffe envisaged. This was assisted by a refined terminology, which distinguished between the site archive, the research archive, summary publication, and full publication.

MAP2 pointed out that while an initial project design will influence the final publication, the review and design prior to post-excavation analysis will play the more pivotal role. MAP2 stated that data should be identified for further analysis on the basis of their potential to contribute to local, regional or national research, and that `such work should be directed towards the final product of a project, the publication' (English Heritage 1991a, 18, section 6.16 [our emphasis]). This phase of review will distinguish, in particular, between material essential for interpreting the site, which should be published in the site report; material of intrinsic archaeological value which may be published elsewhere; and data which are considered to be of no present overriding significance and will be excluded from post-excavation analysis. Large emphasis is placed on selection at the post-excavation phase:

Analysis should be planned with the publication firmly in view, and the research archive should only contain data which derive from the analysis of material intended for publication. The urge to accumulate data not specified in the updated project design as part of the research archive or publication must be resisted. (English Heritage 1991a, 19, section 6.20)

Publication will therefore be explicitly selective.

MAP2's specifications for a final publication were more detailed than those of Frere and Cunliffe. The basic text should be derived from the analytical reports in the research archive and should provide an interpretative site narrative concerning its structural and stratigraphic history, along with artefact and environmental reports (English Heritage 1991a, 37-8). All of these should be related to the site records held in the archive. In addition, it was argued that the publication of an archaeological project should, as a minimum, always include:

Further criteria highlight, amongst other things, the need to discuss ambiguities in the evidence, to deal adequately with the site's social, political and historical context, and to cross-reference constituent parts (English Heritage 1991a, 39-40).

MAP2 was the last document produced by a public body in England which gave detailed consideration to project publication. It is currently used by English Heritage as the touchstone for all its archaeological projects (see Section 2.4). There is, however, one further document to consider: the outcome of a working party convened by the Society of Antiquaries of London in 1991.

2.3.5 `Archaeological publication, archives and collections: towards a national policy'

Archaeological publication, archives and collections: towards a national policy (Carver et al 1992; hereafter referred to as the SoA Report) can be loosely considered in conjunction with the publication crisis of the 1990s which was associated with the introduction of developer-funding (see Section 2.3.1). The association is loose partly because hindsight reveals that this crisis was more diffuse in character than its predecessors, and partly because to some extent the SoA Report anticipated difficulties rather than simply reacted to them. Nevertheless, its preamble was stark enough. Even before the impacts of PPG16 and its Welsh and Scottish derivatives were fully felt, publication was `falling behind', there was `uncertainty' about the future of archaeological archives, and museums were finding it `increasingly difficult' to cope with the material being offered to them. The context of the Antiquaries' report was `a fast deteriorating situation'.

In some respects the SoA Report shows considerable continuity with its predecessors. MAP2 is accepted as the starting point, and there is the traditional recitation of the principle `that it is the duty of all archaeologists to ensure that a full record of all excavations and other fieldwork projects is created and maintained' (Carver et al 1992, section 2.1.1). In other respects, however, the authors were more radical. They differentiated between the concept of a fully accessible archive and formal publication: the `record must be made available for study by others as an ordered archive and, where appropriate, in a published form.' (ibid [our emphasis]).

More fundamentally, and in contrast to its predecessors, the SoA Report explicitly addressed the nature of the archaeological record, and departed by several steps from Pitt-Rivers's axioms in doing so. It contested the notion that excavation can yield a record which is either complete or entirely objective, and rejected the concept of `preservation by record' (ibid: section 2.2.1). The Report argued that an excavation archive is the `result of a complex and continuous process of selection'. It must therefore follow that

since the record is selective and therefore incomplete and post-excavation analysis must also, of necessity, be selective, the excavation report can only be a contemporary statement reflecting on aspects of the site: it cannot be an immutable and complete truth. (Carver et al 1992, 2.2.1)

The SoA Report used these arguments to press a policy of selective publication. It recommended that appropriate dissemination be achieved through the creation of an `accessible site archive' and the publication of a `summary report', the definition for which was derived from the minimum requirements for publication outlined in Appendix 7.1.1 of MAP2 (English Heritage 1991a, 39; Carver et al 1992, 2.3.1). The basic requirement for a project publication was thus reduced to an outline of the research objectives and circumstances of the work along with a summary of the results and the contents of the archive (English Heritage 1991a, 39, section 7.1.1). Beyond this, the SoA Report recognised that, following post-excavation assessment, `it may be agreed that, in addition [to the summary report], the project or some aspects of it, deserve analysis leading to full publication' dealing with data from one or several sites synthetically (Carver et al 1992, 2.3.3). Alongside these steps, the Report advocated the production of a new annual publication which would contain the summary reports of completed projects, the locations of archives, a list of all archaeological fieldwork undertaken within the year, and commissioned papers providing an overview of selected themes (ibid, 2.3.5).

In place of full publication on a project-by-project basis, the SoA Report thus urged a shift towards publication which was extremely selective and highly differentiated. Amid growing numbers of planning-driven developer-funded projects, the Antiquaries proposed that

The principle…is that the form and scale of the publication should be commensurate with what the results have to offer rather than a mechanistic process which is applied regardless of the quality and potential of the data concerned. (Carver et al 1992, 2.3.3)

This, too, marked a departure from earlier orthodoxy.

An aspect of the SoA Report which has since been rather overlooked was its contention that the proposals would lead to rapid dissemination of summary information about project results without precluding more extensive publication, where justified, later on. The annual compendium would offer an immediate awareness of work in progress and provide a prestigious publication outlet for the summary reports. Much debate at the time centred on the viability of the annual compendium, and so rather missed the spirit of what the SoA Report was saying, namely that the speedier appearance and circulation of the results of new fieldwork would enrich archaeology's research culture. In large part, what the Antiquaries were addressing was delay.

The Antiquaries' group consulted on their proposals. Many were critical of them. Much of the reproach stemmed from suspicion about the way in which the document had been produced and circulated. Criticism of what the Report actually said was more complicated, and displayed a diversity of standpoint which has persisted down to the present. Not everyone agreed that the archaeological record is inherently selective, while some felt that the interpretative nature of the discipline had been underplayed. Others argued that the proposed methods of publication would reduce flexibility, and stressed the need for a number of different levels of publication rather than just two. Summary publication was criticised on the basis that post-excavation research in general, and specialist finds research in particular, would suffer if fuller reports came to be seen as optional extras rather than primary necessities. It was further argued that the many user groups which lacked the means to access archival records would be disadvantaged by summary publication.

For various reasons, the Society of Antiquaries' initiative was shelved. Hence, although this was the most recent and, in some respects, radical, approach to the subject, it remains unagreed and unimplemented.

2.4 National/Regional diversity in policy and practice

The organisation of archaeology in Ireland and Britain has always varied between regions and countries (see Coles 1972, 243). In a recent review, Morris (1995, 4) described state-funded archaeology, legislation and policy as a `crazy paving' – uneven, disjointed, and of incomplete coverage. It is thus unsurprising that publication policy and practice should be correspondingly diverse.

The four advisory reports discussed above were initiated by English bodies in reaction to problems arising in the context of post-war England. Even in England, however, these problems did not arise everywhere at the same time, or to the same intensity, while elsewhere approaches to the questions of how much to publish, where, and at what cost had varied widely. Hence, while the English reports stimulated debate elsewhere, their direct implementation was largely restricted to state-funded projects in England, with wider influence more piecemeal and sometimes muted. Variation remains evident inside England: MAP2 is more influential in development-led or EH-sponsored work than in, say, the academic sphere.

2.4.1 State-funded archaeology

The first step towards a policy of selective publication in England was taken with the implementation of the Frere Report by the Inspectorate of Ancient Monuments in December 1978 (DoE Advisory Note 16) for DoE funded rescue projects. This was further amplified in July 1980 when it was strongly recommended that microfiche, distributed with the printed report, be used for the presentation of detailed evidence (DoE Advisory Note 25). Implementation of the Cunliffe Report was hindered by widespread resistance to microfiche. Nevertheless, its recognition of publication as part of the overall research process (rather than simply as a delivery mechanism to be determined in an afterthought) and its concept of iterative review, was adopted by English Heritage and developed in MAP2.

In Wales, there is no recent public document which outlines Cadw's publication policy. However, discussion with Cadw officers confirms that the organisation supports the traditional premise that excavation is destructive, and that evidence identified during excavation must accordingly be recorded, archived and placed in the public domain. Formal, comprehensive publication is not considered to be the only means of doing this. It is argued that full project publication is no longer realistic as an across-the-board expectation, and that all reports for publication must represent a synthesis of results. The submission of a project design detailing the production of a final synthesis report is thus a requirement for funding. At present, therefore, Welsh governmental publication policy echoes English practice. Whether this will change under the influence of devolution remains to be seen.

By contrast, Historic Scotland retains faith in `fairly full publication', for which archiving alone is regarded as an unacceptable substitute (Historic Scotland 1996, 2). As a result the organisation will not generally finance the production of a primary archive report alongside a summary publication, equivalent to Frere Levels III and IV (ibid, 7). The premise underlying this stance is that relatively few Scottish sites have been excavated and published to modern standards:

Sufficient examples of various classes of site . . . must be excavated and the findings written up in adequate detail to allow the reader to assess the evidence supporting the conclusions . . . Summary reports will not be enough until we know very considerably more than we do at present. (Historic Scotland 1996, 2)

The influence of thinking about selectivity is nevertheless apparent. Historic Scotland's paper on publication continues:

On the other hand, the day is past where every feature and every find should be described in detail, or where every `fact' about a site has to be conveyed in print. (Historic Scotland 1996, 2)

Historic Scotland's position, in a nutshell, acknowledges the acceptability of selectivity in the production of a well-synthesised report, but has not adopted a framework of selective publication backed up by archiving along the lines of either the Frere Report or MAP2. `Fairly full' publication is the norm.

Policy in Ireland stands at the opposite extreme. In the Republic of Ireland anyone wishing to undertake archaeological fieldwork has to apply for a licence from Dúchas: the Heritage Service (formerly, National Monuments and Historic Properties Services). This licence requires the production of a full stratigraphic report for open-access archiving, as well as a summary report for publication in the annual Irish Excavations Bulletin. More extensive publication may or may not follow, depending on the nature of the project. Small rescue projects carried out in a particular locality over a number of years are often brought together in one volume.

Northern Ireland is the only part of the UK where excavation must be licensed, and a system similar to that in the Republic has been in use since 1937, when the licensing provision of the Republic's Ancient Monuments Act was adopted (Hamlin 1993, 173). As in the Republic, the Northern Irish system requires the publication of a summary report, although there are no specifications regarding the production of a full archival report. The publication pattern in Northern Ireland is thus closer to that of the Republic than to the rest of the UK. Interestingly, this approach in some respects resembles the model advocated by the SoA Report (Section 2.3.5), although the two evolved independently.

2.4.2 The broader archaeological community

Policies also vary in their influence. In Ireland, since any excavator must obtain a licence, the publication requirements (although not extensive) are universally applied. Elsewhere, by contrast, the policies of state agencies are only directly imposed on employees and those carrying out projects with agency funding. In the 1970s and 1980s this extended to most of the fieldwork units operating in England, Scotland and Wales, as at that time the bulk of rescue archaeology was publicly funded. However, following the publication of PPG16 and its derivatives, contractors competing for projects in the commercial sector are not necessarily constrained by the policy of their respective agencies. As noted above, while all the planning guidance documents state that the developer should provide for publication of results, the extent of what this might involve, and who should be the arbiter of that decision, is left unelaborated. In some areas publication thresholds are policed, or at least patrolled, by local authority archaeologists, but it is the case that an increasing proportion of projects are either not being published or only reach the stage of a grey literature report which is available on request (see Olivier 1996, 31).

Amateur and academic archaeologists enjoy considerably more freedom in what they publish, and for many the traditional model, despite the influence of the Cunliffe Report and MAP2, that model still provides the baseline. However, unless they finance publication themselves there is likely to be some restraint placed on the level of publication, either by the funding/grant-giving body, or the publisher. Exhaustive publication down to the last potsherd has largely been abandoned, and over the last decade some authors have experimented with alternative approaches (see Section 2.5).

Finally, a factor influencing fieldwork publication within all sectors is the publication policy of individual local and national societies. Societies determine what they will or will not accept in their journals, and authors are obliged to adhere to such requirements if their work is to be accepted. Some of the policies adopted by societies have been influenced by, or parallel, recommendations put forward in governmental advisory reports. For instance, the Society of Antiquaries of Scotland adopted a two-level system of simultaneous publication of excavation reports, similar to that advocated by the Cunliffe Report (Shepherd 1983, viii), while in England external funding for particular reports may be contingent on the adoption of MAP2 principles. Equally, many societies are influenced by their members' desire to read more synthetic and interpretative material. In the society sector, too, there are large pressures. The publication in society journals of large or complicated technical reports requires much time, effort and devotion from the honorary editors upon whom societies rely.

2.4.3 Summary publication and indexing

Patterns of broader information management vary across Britain and Ireland. Catalogues and bibliographies (whether in period/area specific journals or distinct volumes) provide the main means of notifying work carried out. Local and national society journals, as well as period journals, often list recent work relevant to their subject or nation/region. However, these usually provide only a list of projects rather than summary abstracts of their nature and results. Publications providing a forum for summary reports are, however, found in Wales (CBA Wales's Archaeology in Wales), Scotland (CSA's Discovery and Excavation in Scotland) and Ireland (the Excavations Bulletin and Archaeology Ireland). In Wales and Scotland, those in receipt of state funding for fieldwork are required to place a note in the relevant national publication, whereas in Ireland those in receipt of a licence to excavate are required to place a summary report in the Excavations Bulletin. The introduction of an annual publication which catalogues and summarises all work that has taken place was advocated by the SoA Report (see Section 2.3.5), and for development-driven work in England something along these lines is now being implemented through the Archaeological Investigations Project at Bournemouth University, funded by English Heritage. Sporadic gazetteers and catalogues have also been produced retrospectively (see Butcher & Garwood 1994; Darvill 1994). English Heritage's Archaeology Review provides an annual summary of their own projects. The only annual bibliographic publication encompassing the whole of the British Isles is the British and Irish Archaeological Bibliography (formerly British Archaeological Abstracts and British Archaeological Bibliography) (King 1994 provides a one-off bibliography). These latter are a valuable source of information, but they provide abstracts only for existing publications, rather than a rolling summary of what has been happening in the field.

2.5 Principles of publication and the generation of archaeological knowledge: theoretical issues

We have seen how successive, mainly English, governmental working parties recommended a shift towards selective and synthetic publication (Section 2.3). However, these changes did not fundamentally challenge many of the traditional premises surrounding fieldwork publication. The notion of `preservation through record' was shifted to the archive, but was not intellectually challenged until the SoA Report confronted it. Likewise, the premises that archaeologists must not be divorced from their main source of raw material, and that they should be able to reconstruct and evaluate a site on paper, were left largely intact. Despite the advent of greater selectivity, policy has also maintained the traditional format, in particular the distinction between description on the one hand (`the facts') and interpretation and discussion on the other. In the realm of specialist studies, likewise, orthodox taxonomies (wood, ceramics, metalwork, glass and so on) have largely been retained.

In broader theoretical literature, by contrast, these premises, and the modes of publication which result from them, have come into question. One of the first critiques came from Leslie Alcock (1978). Responding to the Frere Report, which he considered too superficial, Alcock argued that the idea of preservation through record was flawed (ibid, 2), and challenged three further arguments which others tended to deploy in favour of full publication: (i) that it should be possible to reconstruct the site on paper; (ii) that it should therefore be possible to re-interpret the site from the published record; (iii) and finally it should be possible to judge the quality of the excavation and authenticate the excavator's observations from the published record. These arguments, Alcock said, are spurious. Given a skilled pictorial and written record:

there is no way whereby a reader can assess and verify the skill of the excavator in recognising, dissecting and recording the primary data. It is the inevitable limitation of excavation as a means of recovering evidence that what is destroyed unnoticed is gone for ever. In simple logic we can never know what the excavator has failed to recognise, or what he fails to tell us about. (Alcock 1978, 3)

It was thus necessary to rethink the reasons for publication. Alcock concluded that the main purpose is to convey information, and that the needs of the reader should therefore be paramount. He further suggested that most of those who produce syntheses need only a summary published account, and that the exhaustive report could accordingly be confined to archive. Thus, while the conclusion of Alcock's argument resembled that of Frere, it was based on different tenets.

These issues were taken up by Barrett (1987) who argued that the publication crisis extended beyond the production of reports to the processes of critical evaluation in which archaeologists engage when they use them. Although it can be impossible to assess an excavator's general competence from a published account, it is possible `for the reader to undertake a critical analysis of the internal logic of the report, examining the linkages between the assumptions employed, the stated record of observations, and the interpretative account' (ibid, 410). Through a detailed re-reading of the Glastonbury report, Barrett demonstrated a form of internal source criticism that takes account of the historical context of the excavation, the nature of the record and the form of publication, all of which, he argued, were neglected by Clarke in his classic re-interpretation of that site. Such a re-reading, Barrett argued, does not result in re-interpretation, but is essential to clarify uncertainties against which newly-collected observations can be placed (ibid, 422). It also highlights the fallacy of claims to absolute objectivity and suggests that self-critical evaluation of the way observations were collected is more useful than neutral reportage.

Concerns about objectivity and modes of writing became a common subject of debate in theoretical literature of the late 1980s and 1990s. Hodder (1989), for instance, surveyed the changing style of excavation reports since the late eighteenth century, and observed that during the twentieth century:

The writing has become increasingly distant, objective, impersonal and universal. We have become blind to the fact that we are writing. It appears as if self-evident data are simply described in neutral terms. (Hodder 1989, 271)

For many (eg Hodder 1989; Tilley 1989; Shanks & Tilley 1992; Richards 1995), such writing seemed to be a negative trend, whereby supposedly neutral description had itself become the goal and purpose of excavation, the product of a misplaced faith in preservation through record. Their counter-view argued that since excavations involve uncertainty, differences of opinion and inconsistencies, that ideas change as a site is excavated, and that the process is interpretative from start to finish, such contingent factors which lead to the `final' interpretation should, as far as possible, be written into the report rather than kept out of it. In other words, there should be greater integration between description and interpretation.

Much of the theoretical debate has not been limited to fieldwork, but rather forms part of a broader reconsideration of the relationship between material evidence and the production of knowledge in which fieldwork publications epitomise wider problems. The impact of such thinking on publication practice has not been extensive, although some of the same authors have attempted to develop alternative approaches in media, style and structure (see for example Barrett, Bradley & Green 1991; Bender, Hamilton & Tilley 1997, for two different approaches).

Although, as far as publication policy is concerned, the impact of academic thinking has been slight, there are two areas where recent developments do echo it. First, as fieldwork publication becomes more selective, increasing value has been attached to explicit research frameworks (see Barclay 1997; English Heritage 1991a, 1991b; Olivier 1996). This reflects growing consensus on the need to attribute value to archaeological material (in terms of its contribution to knowledge) in relation to what is selected for excavation, analysis and publication. However, whereas the theoretical literature emphasises the need for plurality and dialogue in the abstract, national bodies have attempted to document and define a range of regional and national frameworks through consultation. Obviously, the process of formalising research frameworks runs the risk of solidifying them, but for the most part the organisations involved have stressed that they recognise the need for flexibility and dynamism.

Secondly, the report produced by the Society of Antiquaries of London (Carver et al 1992) reflected some of the theoretical debates in its discussion of the nature of the `archaeological record' (see Section 2.3.5). In particular, the SoA Report asserted that an objective and complete record is an impossibility, because all excavations and the archives which result from them are the outcome of continuous, co-varying selectivities. Such choices may be large, small, conscious or unconscious; they may be exercised explicitly, or they may result from implicit or habitual choices, but whatever the reason it is the case that certain things will be recorded, and others not.

This was the first time that such a policy document had flown in the face of earlier doctrine, and it may be this which explains the variety and strength of the reactions that followed. For example, a number of consultees misunderstood the point about selectivity, interpreting it negatively or supposing it to imply personal bias or technical inadequacy (Clark 1996, 6). On the other hand were those who accepted the Report's characterisation of the archaeological record, but argued that the interpretative nature of the discipline was being underplayed in a document which appeared to be offering little more than a series of techniques for managing antiquities (ibid).

2.6 Publication outlets, media and costs

2.6.1 Traditional print publishing

Print has been the traditional medium for the publication of fieldwork projects (for a detailed survey see Lavell 1981). In the late nineteenth and early twentieth century, monographs were either funded by private means, as in the case of the Cranborne Chase volumes (Pitt-Rivers 1887-98), or through societies, as in the case of the Research Report Series of the Society of Antiquaries of London, established in 1913. The Royal Commissions on Historical Monuments began their inventories in 1909, but it was not until after World War II that the HMSO monograph series was set up. Commercial publishers have seldom invested in the publication of fieldwork monographs - the market is so small - although there are exceptions (for example, certain DUA monographs in the 1980s; Barrett, Bradley & Green 1991).

Archaeological journals supporting fieldwork publication can be divided into three groups (Atkinson 1946: 178; Coles 1972: 244-5; Lavell 1981: 92-3): journals of

Traditionally, societies supported publication in their journals through membership fees and voluntary editorial services, but since the war these resources have sometimes been supplemented by specific publication grants, particularly for government- and agency-funded projects. Although commercial publishers such as Routledge have taken on archaeological journals, the titles usually focus on synthesis and issues of method and theory (eg World Archaeology), although more recently some publishers have begun to explore the commercial possibilities of international marketing.

Even before the escalation in rescue archaeology which began in the 1960s it was recognised that:

The circulation of all these [publication outlets] is relatively small, and the cost of their publication proportionately high, nor are the available funds large. Moreover, space in the `national' journals is limited, and should be reserved if possible for the material whose importance merits the widest publicity. (Atkinson 1946, 178-9)

Pressure on publication outlets was soon recognised as a factor which necessitated selectivity (Webster 1963, 132). Nevertheless, print remained the only medium for fieldwork publication into the 1970s, when the boom in rescue archaeology precipitated the crisis examined above. Two of the main problems initially identified were soaring costs and the absence of any increase in the range of outlets:

We are concerned that, due to greatly increased costs...opportunities for publication of excavation reports in monograph form have become seriously restricted...There is a danger that very shortly availability of these traditional and long-established publication media will become restricted because of sheer inability of the national and local societies to support the balance of the cost through membership subscriptions and voluntary editorial services. (AMB 1975, 7, section 4.3)

The Frere Report's response has been discussed (see Section 2.3). In addition, however, a number of variant production methods were examined. For smaller projects the use of offset-litho typescript was advocated, following the example set by British Archaeological Reports (AMB 1975, 7, section 4.5). The pros and cons of publishing major reports through series of self-contained fascicules were also considered, and in several cases adopted (as by the York Archaeological Trust).

2.6.2 Microfiche

By the late 1970s combined text-fiche publication was being advocated as a solution to problems of cost and space (Mytum 1978). This usually consisted of a two level publication - a full report on microfiche and a summary report in print - but microfiche was also used as a medium for appendices and specialist reports. In 1980 two county journals - Northamptonshire Archaeology and Research Volumes of the Surrey Archaeological Society - became the first to use simultaneous text-fiche reports (Lavell 1981, 106). The Society of Antiquaries of Scotland introduced a text-fiche policy for its Proceedings in 1983, at the same time as the Cunliffe Report was advocating such a format in England, although the policy has since been discontinued and fiche has not been used in the Society's Proceedings since 1996.

Microfiche was obviously a practical solution to the financial difficulties of publishing, and a number argued strongly in its favour (eg Lavell 1981; Mytum 1978). In her review, Lavell (1981, 106) estimated that one fiche of 98 pages cost about the same as two pages of conventional print. But there was also strong consumer resistance (surveyed by Lavell 1981, 106). Numerous critiques appeared in journals (eg Adkins 1985; Manning 1985), and much was made of the difficulties of usage. The Cunliffe Report was rejected by the CBA's Council not least because some members disliked its emphasis on microfiche. In response the CBA embarked upon a survey of user behaviour and attitudes, with specific regard for the possibilities of new technology (Dixon 1986; for further discussion see Section 2.7).

2.6.3 The new electronic information technology

By the 1990s debate about microfiche subsided as the potential of emerging electronic technology, in particular the Internet and CD-ROM, was becoming recognised. The Internet offers a cheap alternative to traditional print, and a means of rapid communication which has already revolutionised the dissemination of information. The production costs of publishing on CD-ROM are also low, although it has been argued that users have higher expectations of software development and therefore that the development costs are high (McAdam 1995). Advocates of electronic technology as a publication medium point to its advantages over traditional print (see Champion 1997; Cunliffe 1996; Heyworth et al 1995; Thomas 1995). For example, multimedia capabilities allow publication of ample colour images, video clips, and dynamic 3-D visualisations. CD-ROM and the Internet can also accommodate complete excavation databases and access to the original software used by the author, which (in the case of the Internet) can be continually up-dated. Finally, electronic technology provides a far more interactive publication medium than traditional print, and the use of search tools and hypertext links can enable the reader to use a publication in a non-linear fashion. In result of all this it is often argued that the Internet does not simply provide a cheaper medium for the dissemination of material that might otherwise remain in archives, but that it also offers the basis for resolving the problem of integrating large amounts of data with an interpretative narrative (see especially Thomas 1995). It has also been suggested that electronic technology's capacity to facilitate fragmented, non-linear, virtual texts, which promote interaction and reflexivity on behalf of author and reader, could enable a convergence between archaeological theory and practice (Hodder 1999, 180-4).

Despite the enthusiasm, a number of serious concerns have been expressed. Some argue that electronic media, like microfiche, have the potential to exclude those who do not have access to the appropriate technology, or remain unfamiliar with its use. They are not therefore quite as democratising as advocates claim (Hodder 1999, 185). Concern has been expressed about the quality of the material published on the Internet, much of which is not peer-reviewed as it usually is in printed publications. A related worry centres on the academic standing of Internet publication, which many feel will not achieve the same status as a peer-reviewed printed journal or monograph. Finally, there is anxiety about the long-term security of material published on the Internet and the fragility of digital data. All these concerns are being addressed (for example, see Heyworth et al 1997), but they still exercise large influence on authors and funding bodies.

While CD-ROM has yet to take off as a medium for archaeological publication, the Internet is already widely used (Champion 1997). Books - mainly syntheses and textbooks - have only appeared in the last three years, but Internet journals have existed since 1993. Some, such as Electronic Antiquity and Archaeology Online, are entirely electronic, while others are Web counterparts of print-published magazines such as Archaeology Ireland and British Archaeology. Also available are sites which provide information about the activities of organisations and societies, and discussion lists for the daily exchange of views and information

Many Internet offerings are concerned with fieldwork, ranging from brief particulars of work in progress to interim and summary reports, to be found on web sites set up by fieldwork units such as the University of Birmingham's Field Archaeology Unit or the Trent and Peak Archaeological Trust. A particularly significant development was the establishment in 1996 of Internet Archaeology - Britain's first electronic archaeological journal of record (see, which aims to publish excavation reports and finds studies, including relevant databases. Nevertheless, as yet there are few full fieldwork publications of the type familiar in print, and those projects which are available on the Internet tend to be published in traditional print as well. Furthermore, no government agency yet supports Internet publication as the sole or principal means of publication for state-funded projects. English Heritage publishes its annual Archaeology Review, which disseminates information about its own current funded projects, in print and on the Internet. Otherwise, whilst signalling interest in the potential of the medium, the agencies remain cautious, largely because of the concerns listed above (for example, see Historic Scotland 1996, 6, 10). However valid these concerns may be, it is also fair to point out that electronic technology already is a significant means of communication within the discipline (see Vince 1997 for an analysis of the intensity of use of Internet Archaeology). Indeed, Historic Scotland had begun to investigate the use of the Internet to disseminate the results of archaeological fieldwork in Scotland through a series of Scottish Archaeological Internet Reports (see

2.7 The audience and the role of project publications within the discipline

Commentaries on fieldwork publication have hitherto not placed much weight upon readers' needs. The production of a record and the process of publication were not distinguished from one another (see Section 2.2). Consequently the primary role of publishing fieldwork projects was that of `preserving' data that had been destroyed through excavation, in the interests of the discipline, the broader public, and posterity. Commonly, the only real concession to the reader in terms of structure and content was the practice of including a summary (cf Atkinson 1946, 180). Otherwise, comments on readers' needs were restricted to the importance of writing for an audience that extended beyond the specialist reader, and not reproducing unnecessary detail:

... just as in a newspaper, one does not want or expect to read solidly from front to back in order to discover the salient news and views of the day, so in a well-balanced excavation-report the student may properly expect to discover something of the wood without a prolonged, tedious, and exasperating hunt amongst the trees. (Wheeler 1954, 186)

With the onset of the publication crisis of the 1970s, the needs of the reader remained subordinate to considerations of practicality, cost and distribution. However, from the Frere Report onwards, when the act of publication began to become detached from the concept of preservation, a trend was initiated whereby the interests of the `audience', and the development of knowledge, began to assume more prominence. For instance, the Frere Report argued that:

Publication must cater for all aspects of scholarship. A comparatively brief statement of the results of the excavation itself is adequate for the needs of no more than a few groups of potential readers. Archaeological research must not be cut off from its main source of raw material…[and] Archaeology is an international discipline. International study would become impossible without publication designed to meet its needs. For this reason mere preservation of significant results of excavation in non-published form in a local or even national archive is no solution to the problem. (AMB 1975, 2, sections 2.3 & 2.4)

Such statements stimulated debate. Here is Leslie Alcock:

What is at issue here is readers' time [in contrast to the excavator's time]: the time of the interested layman; but more especially of those professionals who have to use the results of excavation to produce syntheses . . . As the number of excavations increases, and as the expected length, detail and complexity of excavation reports grow exponentially, so does it become inevitable that archaeologists must read a decreasing fraction of the field; and that increasingly uncritically . . . Yet, if excavation is to have any intellectual meaning, if it is to be anything more than a fun pursuit, then the creation of syntheses must keep pace with field work and excavation. (1978, 2)

Alcock went on to argue for a thorough review of publication policy, in a way which challenged a number of traditional premises (see Section 2.5). However, such sentiments were not taken up in later policy-related documents. For reasons we have seen, the Cunliffe Report and MAP2 were both primarily concerned with the processes of critical review and selection that stand behind a project publication.

Even so, audience needs did not entirely disappear from the agenda. Attention shifted towards practical concerns, such as ease of access to information and equipment relating to the new media. In addressing these concerns a number of studies have been undertaken, which can broadly be divided into surveys concerned with information handling, and with new media and formats.

(i) Information handling
Surveys in this group concerned British Archaeological Abstracts (BAA) and the British Archaeological Bibliography (BAB) (British Library 1977; Heyworth 1992; Wills 1995). The first of them consisted of a three-day seminar organised by the CBA, with funding from the British Library, in 1976. General problems of information handling, and potential solutions, were discussed in relation to the increasing intensity of archaeological fieldwork and the implications of the Frere Report (British Library 1977). It was agreed that a bibliographic and abstracting system was essential. Study groups highlighted archaeology's sectoral diversity, each sector with its own needs.

A second survey consisted of a mail questionnaire to ascertain how archaeologists were using BAA, and what improvements they would like to see (Heyworth 1992). This indicated that a high proportion of respondents used BAA for current and general awareness, whereas a smaller proportion used it for retrospective searches (74%). However the low number of responses (200) perhaps indicated that many archaeologists were still unaware of BAA, or perhaps saw no need for an archaeological abstracting service at all (ibid, 17).

A third survey (Wills 1995), based on semi-structured interviews, confirmed the general pattern of use, with more people reporting use of BAB (formerly BAA) for current awareness than for retrospective searches, and a number of other methods of obtaining information in use.

(ii) New media and formats
Following the CBA's rejection of the Cunliffe Report, two consecutive mail questionnaires were carried out by Dr Philip Dixon in 1986 on behalf of the CBA. Their main focus was the actual usage of fiche, but other aspects of the archaeological community's reading and buying habits were investigated (Dixon 1986, 1). Reaction to fiche was generally negative: over a third of respondents had made no use of fiche in the previous year, only 17.5% had used it more than five times, and those who had used fiche found a number of its aspects unsatisfactory.

As for electronic media, as yet few studies have set out to assess access to CD-ROM and the Internet within the archaeological community, although the last two BAA/BAB surveys (Heyworth 1992; Wills 1995) did request such information in relation to their specific aims. While these two surveys are difficult to compare directly, they did indicate a general increase in access to or ownership of computers and, in general, found a favourable reaction to the possibility of on-line access to BAB (although not necessarily as a substitute to hard copy). In terms of publication form and content, a few consumer studies relating to individual publications have been undertaken, for example, the use of British Archaeology (Heyworth 1996) and Internet Archaeology (Vince 1997). As far as we know, the present survey contains the first thoroughgoing investigation of the archaeological community's views upon electronic media for the publication of archaeological fieldwork.

A survey of user needs relating to digital data in archaeology, undertaken by the Archaeology Data Service (Condron et al 1999), found that access to computers and the Internet varied for different constituencies, but that, for example, over 80% of field archaeologists were using computers, and the majority of archaeologists have Internet access (Condron et al 1999, 23-24) although sizeable minorities (over 30%) of field archaeologists, consultants and local government archaeologists did not have Internet access. More surprisingly perhaps, the survey found that over 50% of society members have access to the Internet, either at home or through their place of work. Extrapolating from the survey results, the authors suggested that all archaeologists will be working with computers by 2004, although they acknowledged that there will always be those who do not feel the need to work with digital data, or who cannot afford to do so (ibid, 26).

The only further initiative that focused more generally on the publication of fieldwork projects was that undertaken by the Society of Antiquaries of London in the early 1990s. For reasons we have seen (Section 2.3.5), the diverse responses (which are difficult to assess and compare) to that Report were negative in flavour because they tended to focus on problems arising from proposals put forward in the report, rather than the needs of users in general.

2.8 Summary

Next section

PUNS report index page