|
4 THE NECESSITY FOR AN AGREED MINIMUM
ARCHIVE and promote equitable tendering.
5 DEFINITION OF A POTTERY ARCHIVE 5.1 Two levels of archive are identified: 5.2 The Site Archive (MAP2. Appendix 3). to consist of two phases: i Primary site data for pottery: prepared during or immediately after an excavation, specifically to aid the excavation process. ii Ceramic archive: basic ceramic data recorded post-excavation, usually covering all contexts irrespective of stratigraphic or ceramic value. 5.3 The Research Archive (MAP2, Appendix 6): the ceramic archive upgraded with further detail and quantification where necessary. 5.4 This document is concerned with the Site Archive only.
6 THE SITE ARCHIVE: PHASE 1: Primary site
data (MAP2, Appendix 3) 6.1 How the archive is compiled will
necessarily vary between different specialists and organizations
and is related to the circumstances prevailing, the nature of the site and availability
of trained personnel. Depending upon circumstances, work on the pottery during the
excavation may not be essential or feasible; in which case Phase I could be omitted.
It is recognized that the primary purpose of pottery work during an excavation is to
aid the excavator with dating and information potentially helpful to the progress of
the excavation, sampling policies and the interpretation and chronology of the site. 6.2 Thus the first phase of the Site Archive is limited to those aims and termed
PRIMARY SITE DATA FOR POTTERY, recording: i Rough quantity per context, in terms of small-medium-large. ii A terminus post quem iii Time span for the assemblage. iv A provisional date of deposition. v Optional assessment of condition, if exceptional. vi Optional comments on any notable level of residuality, possible intrusions, and broad emphasis. vii Date of the record. viii person responsible.
6.3 It is strongly emphasized that this work should be undertaken by the most experienced person available, since errors at this stage could have far reaching consequences. It is further emphasized that the sole purpose is to facilitate the excavation, and that in no way can this dating record be the basis for an assessment or the only work to be undertaken on the pottery. This record would normally only be used during or immediately after excavation; backlog sites are excluded.
6.4 The record should be a paper record, to be incorporated in either existing context or bulk finds records, and computerization is optional. It is essential that all records normally kept on site are photocopied or copied to disk to be kept safely elsewhere as insurance against loss.
6.5 While this is undertaken during excavation, it is emphasized that the CERAMIC ARCHIVE produced in the post-excavation period is equally part of the excavation, to be budgeted as an excavation cost.
7 THE SITE ARCHIVE: PHASE 2: The ceramic archive (MAP2, Appendix 3)
7.1 This should contain sufficient data to comply with the definition (above 2.1)
7.2 Depending upon the potential of the site and pottery, this record could be the only record or could form the basis for a RESEARCH ARCHIVE.
7.3 It is emphasized that the work must be undertaken by the most skilled specialist available. It must be a complete record of ALL pottery, including residual pre-Roman and intrusive post-Roman sherds, before extraction of specialist wares. and including objects made from sherds.
7.4 The post-excavation period is the optimum time to produce a CERAMIC ARCHIVE, when a provisional matrix or phasing is available. The pottery can be archived in stratigraphic sequence, with the component contexts of cut features viewed together, to gain the maximum information from the pottery, the PRIMARY SITE DATA for the pottery providing a useful objective double-check on both dating and condition.
8 THE CONTENTS OF THE CERAMIC ARCHIVE
8.1.1 The agreed definition is that imports, major traded wares and significant local wares should be individually recorded. It is not intended that detailed sub-divisions of grey wares fur instance should be involved, and it is stressed that the fabrics recorded should be identifiable macroscopically without undue use of microscopic examination.
8.1 .2 Fabric codes, whether numeric, alphabetic or a combination should be designed for the greatest ease, both in basic recording and for computerization and extraction. The use of prefixes for groups of fabrics is especially useful to aid their ordering in basic listing and in extracting and manipulating data. It is recommended that where possible, any nationally agreed codes are used, or that there is a concordance between those and the recording codes in use.
8.2 Record of vessel classes/forms and/or vessel types 8.2.1 The Students’ Guide to Roman Coarse Pottery defines form as ‘Classes of vessels sharing form and function’; the definition used here is that form means flagon, jar, bowl, while ‘type’ defines the type of form, as ring-necked flagon. reeded rim bowl etc.
8.2.2 A hierarchical code is recommended so that data can be extracted at different levels, i.e., all flagons or, in more detail, all flagons of ringed-neck type. It is recommended that vessel type should he recorded, as considered necessary by the individual specialist.
8.3.1 It is recommended that the record should be of quantities for fabric and vessel type, since the extra information over a presence/absence record involves minimal time and yields more information. More time is spent on the initial sorting into fabrics and vessel types than on the actual recording.
8.4 Measures 8.4.1 Since vessel types are to be recorded, sherd
count is the fastest measure (weight would involve endless weighing of single
sherds; it can be shown that 60-70% of records relate to single sherds). The
recommendation is sherd count, the use of’ weight being optional, particularly
where amphorae form a potentially useful group, and for some rural sites and
particularly where breakage may assist stratigraphic understanding. 8.4.2 It is emphasized that recording weight was
moving beyond the necessary level of basic archive into the research archive
although for some sites this may be advisable. The recording of’ EVEs does not
form part of the basic archive, and should be reserved during any further work
for only those groups where there is clear potential for useful information to
be gained from quantification. 8.4.3 If weight is added to the record, consideration should be given to confining this to
fabrics as groups, leaving the count as the basic record for vessel types;
weighing for vessel types would double the recording time. 8.5 Record of decoration, surface treatment, stamps 8.5.1 It is considered that stamps (potters’
names, trademarks or decorative) and graffiti are essential records, and that
decoration considered to be chronologically significant or potentially useful
should be recorded as an option. If the definition of fabric or ware covers
vessels which can be hand- or wheel-made, it is essential that the type of
manufacture is recorded where identifiable. 8.6 Record of obvious sherd links between contexts 8.6. 1 It is recommended that cross-links observed
between contexts during recording should be noted, without involving
time-consuming searches. 8.7 Record of
condition of group 8.7.1 This is essential when the condition is
exceptional (e.g., very abraded, fresh, burnt), and comments should be made on
unusual levels of residuality, possible intrusions and kiln products.
Assessments of the value of groups are useful to flag assemblages with
potential for further work. 4 8.8 Indications of probable illustration requirements 8 8.1 To facilitate estimates for future work,
it is recommended that vessels worth drawing for their intrinsic and
stratigraphic value are recorded separately. 8.9 Date 8.9.1 This should contain a terminus post quem. the
time span of the assemblage and a considered date of deposition. 8.10 Future potential 8.10.1 This section is only applicable to the SITE
ARCHIVE if the pottery is to he deposited in the appropriate museum without
further work: otherwise this belongs with the assessment document, prepared as
a preliminary to further work and a RESEARCH ARCHIVE. This should be a
summary of the complete pottery assemblage from the site. identifying contexts
of crucial importance for retention and those of dubious value, with clear
recommendations for their retention or dispersal. If dispersal is recommended,
it is essential that the archive record contains sufficient detail for future
research use. 9. 1 It is highly desirable that the archive is
computerised. and essential if the data relating to vessel types is to be
adequately used. Even small sites benefit from the ease of extraction of
information and its integration with site and other finds data. The financial
outlay on hard and software is small compared to not only the savings in labour
cost, but also the increased use arid availability of the data. 9.2 The ability to upgrade the ceramic
archive to both quantified and specialist ware records should be aimed for from
the inception of a scheme, and should present few difficulties, avoiding
duplication of effort. Any archive software must have the capability to
transfer data to a new datafile with extended fields to accommodate
quantification and specialist information. 9.3 Since the archive forms one of the most valuable assets of any
archaeological organization. it will require active curating including, as
changes occur in both hard and software, transfer to new formats. 5 ROMAN POTTERY ARCHIVE GUIDELINES : ACCOMPANYING NOTES An archive in the
wider sense accumulates over a period of time, each individual site adding its measure. The archive criteria of tomorrow
will not be those of today, but in the absence of crystal balls, this document
takes the view that the structure of today’s archive should be determined by
foreseeable demands upon it. The potential value and use of data is what
governs the extent of any record. An archive
contributes useful information at several different levels, some of which are
identified as under: 1 Most immediately to enable the crucial
assessment of the potential of the material to answer questions posed, whether
in the initial research design or arising during fieldwork. This is the MAP2,
Phase 3, and an essential prelude to any further work on the material. 2 For those sites with no obvious
potential for immediate further research, the archive is the factual statement
of what was found. This is also the case for contexts and groups within a site
that are deemed unsuitable for further work and quantification, but can still
contribute to the overall view of the site. 3 Low cost data for all sites for comparisons
either within an area or region, or within a city, as for instance, samian as a
proportion of the overall assemblage, or more finely as the proportions from
different samian factories. The spatial distribution with count as a measure
for any number of wares deemed to be useful to explore chronological
development and regression of settlement, status. trading patterns and economic
aspects. 4 If both fabric and vessel type is
recorded, the archive enables the identification of levels of residuality,
essential both for the assessment of future work on the pottery, hut equally
essential to aid the assessment of future work on artefactual and environmental
finds. Both also facilitate spatial analysis over an area to highlight
anomalous sites which may have a bearing on the type of occupation, status and
chronological changes. 5 Records even with low cost sherd count
as the measure enable examination of the proportional occurrence of both
fabrics and vessel types chronologically, and when related to the defined
deposition date and the size of the individual group, can aid definition of the
floruit of the fabric and/or type for future work. 6 The archive is a factual statement of
the contents of the physical archive and should enable its efficient curation
once deposited in a museum or depository, and aid both educational and research
aspects. The inclusion of a comments field in the archive for each record
facilitates the identification of, for instance, complete vessels useful for
display and educational purposes, while records defining both fabrics and
vessel types aid future research, i.e., all mortaria. All barbotine decorated
colour-coated ware, amphorae of a certain type etc. can be extracted for
assessment for a research project. 6 The following notes are attached to amplify various sections. The
archive provides useful information at several different levels, one more
immediately referring to the assessment of the potential information to be
gained from further work on the group, while
others contribute data to aid continuing and future research. Aspects of both are noted. 3 The extent of a
Ceramic Archive 3. I On the basis that all interventions have the potential
for producing useful evidence. no group of material should be excluded. 4 The Necessity for an agreed Minimum
Archive 4.1 This is viewed as
essential to future progress in archaeology as a whole. and in pottery studies
in particular. 5 Definition of a pottery archive 5.2 The Site Archive is split into a possible two phases dependent
upon the site and the circumstances of excavation primarily to ensure the most
effective and efficient data collection, the first being to facilitate the
excavation, and the second to achieve the maximum retrieval of’ ceramic
information. A flexible approach is essential; it may be necessary to complete the second phase during the
excavation and accept a certain loss of information see 7.4). 6 THE SITE ARCHIVE: Phase I : Primary Site
Data 6.2 This record should be an extremely rapid scanning of each
context. Every effort should be made to ensure that contexts are complete when
viewed, a common problem when viewing pottery during an excavation. Apart from
its primal), purpose of providing information to the excavator to aid the
progress of the excavation, it provides
the pottery specialist with an overview, but does NOT constitute the site archive for
pottery and does NOT provide sufficient data to enable a reasoned assessment of
potential. Some indication of quantity is essential, and weight or number of
trays or boxes (specifying the dimensions) would also be sufficient. 7 THE SITE ARCHIVE: 2: The Ceramic Archive It has been suggested that some sites of obvious importance should skip
the Ceramic Archive stage during which the sole minimum measure is sherd count
and proceed directly to the RESEARCH ARCHIVE, incorporating full
quantificationof estimated vessel equivalents (EVEs) and weight from the
beginning. This is an understandable response wishing to avoid getting the
pottery out and sorted twice, particularly bearing in mind that this work is
the most time consuming part of any record, whether a basic archive or a full
quantification. There are several problems
with such a procedure. The importance of the assessment phase, which integrates
the pottery with the stratigraphy and other finds, is to identify those
assemblages within a site worthy of future work. Such integration would be
impossible during an excavation, and the implication is that the entire pottery
assemblage, irrespective of ceramic or stratigraphic value, is to be fully
quantified. This goes against the accepted view of MAP2 and most pottery
specialists that resources should he targeted to achieve the maximum
information for the least cost, If the site was so exceptional in its character
or chronological span and full quantification was felt to be worthwhile, this
should be identified at the project design stage. For any site so treated, it
would be necessary to create an archive data file compatible with other
excavations (i.e., extracted from the quantified data file) to enable the site
to be used for any spatial analysis or survey. 7 It is feasible that
there should be a practical alternative open to negotiation in exceptional
circumstances, Dependent upon a rapid decision on the advisability of further
work, and bearing in mind the time consuming sorting and laying out work,
important contexts obviously requiring quantification could be sorted into
stackable trays, and stacked ready for quantification once the ceramic archive
has been completed. Once the assessment, based on the ceramic archive, has been
completed and decisions taken, the reserved material either proceeds to full
quantification or is boxed. If the former, clearly costs have been reduced; if
the latter, there has been merely the problem of temporary storage and a delay
in boxing. The upgrading of the Ceramic
Archive should be considered from the outset. It is essential that it is feasible to extract such basic data
as is required for contexts to be quantified to a separate database with extended
fields to accommodate the new information. Depending upon the coding system
adopted, this should encompass not only new fields for EVEs and weight (if not
already in the Ceramic Archive database), but also fields for any detailed
definition deemed necessary for fabrics and vessel types. An important feature of the
Ceramic Archive easily overlooked is that it covers ALL sites, irrespective of
value, at a consistent level. The original archive file should be retained with
any basic corrections necessary so that the data is still available for any
spatial analysis across areas, regions or urban areas. It is important that
additional information is added to separate databases. It may be considered
useful to add a field to the Ceramic Archive to cover samian dates, but to
retain the original consistency between sites which receive further work and
those that do not, such changes should be kept to a minimum that will not
interfere with the basic data and complicate its interrogation. Similarly to produce a
comprehensive archive of’ specialist wares. particularly samian, the basic
Ceramic archive data should be transferable to a specialist database with
extended fields to cover such information as potter’s name, die number,
specialist date, source (as Rheinzabern. Trier etc. expanding on East Gaul).
Similar databases enable area or city-wide surveys of mortaria, amphorae etc.
which may require the recording of the more detailed information received from
specialists. 8 The contents of the Ceramic Archive 8.1 Record of
wares or fabrics REASONS: Overall to
investigate trade, influences from neighbouring areas, status, the
inter-relationships between different potteries. For assessment purposes as a
dating indicator, to check for anomalies and residuality in the content of
groups, as for instance high percentages of amphorae in later Roman contexts,
or quantities of earlier fabrics occurring late, and as status indicators. 8.2 Record
of vessel classes/forms and/or vessel types. REASONS: Overall the possibility of
investigating differences between assemblages or sites based on vessel form.
arid use of the archive to define closer dates for types based on their dated
occurrence. For assessment purposes as a closer dating tool, vessel types being
a finer indicator than fabrics alone, and to search for any anomalies
indicative of high residuality. groups contain tug mixed date pottery of
debatable value for further work. If the codes are hierarchical. extracts may
be useful to examine any spatial differences across the site which may aid
Interpretation. 8 8.3 Record of
quantities. REASONS: To establish the content of
the assemblage and the relative occurrence of different fabrics and vessel
types. For assessment purposes, the relative occurrence of fabrics and types
can aid dating and the identification of any high residuality. 8.4 Measures Sherd count is not only the fastest measure but also
essential 1 fragmentation or brokenness is to be examined during further work,
If data on fragmentation beyond that recorded under 8.7 is considered desirable
to aid site Interpretation, every effort should be made to limit weighing to
those groups requiring such data. Overall context weights could equally be recorded
in the PRIMARY SITE DATA as a measure of quantity (6.2.i). 8.5 Record
of decoration, surface treatments, stamps. REASONS: Overall to facilitate further
work on stamps, graffiti, and for decoration or surface treatment, to aid
identification of datable decoration to extend use for dating, and any
relationships between industries. For assessment purposes, a further indicator
for dating in some instances, plus for specialist wares with name stamps, an
indicator of the quantity requiring specialist attention. 8.6 Record of obvious sherd links between
contexts. REASONS: Primary data for the
assessment phase to aid stratigraphic work, and the identification of obviously
residual sherds disturbed from earlier contexts. 8.7 Record of condition of group REASONS: To aid understanding of the
deposition of the context, plus assessments of the potential value of the
pottery. Particularly valuable to the assessment phase; high levels of fragmentation or
conversely freshness would be recorded here; also notably burnt or mortar
encrusted pottery etc. 8.8 Indication of likely illustration
requirements REASONS: To facilitate estimates for
future work. Particularly useful if graded between vessels needing illustration
for their own intrinsic value as new types or better examples, and those
necessary to substantiate the dating and interpretation 8.9 Date REASONS: To revise, if necessary, the
dating contained in the PRIMARY SITE DATA
FOR P0TTERY. 8.10 Future
potential REASONS: To provide guidance as to the
value of the pottery for future research, particularly in relation to decisions
on dispersal or retention. For assessment purposes, the future potential should
be part of the assessment document, based on information extracted during the
assessment process. 9. I It cannot be
stressed too strongly that computerisation of the archive enables it to be used
to the fullest extent. Computers provide the technology to enable the fullest
possible integration of pottery, finds and site data to the benefit of all. The
advantages to be derived from this, and developments in computer soft and
hardware likely to enhance the retrieval and use of evidence are part and
parcel of the modern archaeologists concern. Even a low power PC can extract
primary data more efficiently and cost-effectively, and enables its integration
with other data from the site or from comparative sites. Use of electronic data
in the dissemination of information will undoubtedly be a further development. 9 Computerised data
is particularly valuable in the early post-excavation stages when provisional
phasing is subject to change. Such changes can be easily made and their effect
assessed. The data can also be more easily interrogated to assess the potential
of the pottery for future work. A paper archive is
cumbersome and often unusable without considerable expense. A properly curated
computerised archive provides instant
access. 9.3 A paper copy of the complete archive and if felt necessary,
extracts of data, should be deposited with the pottery, and an additional copy
kept to safeguard against loss of electronic data. All primary paper records
should also be kept. Terminology is
problematical, but the following should amplify precisely what is meant in this
document. Ware A term used in two
senses: vessels from the same production centre (e.g. New Forest ware), or
vessels having the same basic characteristics in technique (e.g. colour-coated
ware). Fabric A term used to
describe the paste of a finished vessel, including texture, colour and nature
of both paste and surface. Paste The prepared clay from
which pots are thrown. In the finished vessel the term is often applied to the
body of a vessel, as distinct from the outer surface. Class Vessels are said
to belong to the same class if they are of the same general form and usable for
the same purpose. irrespective of date, fabric, and details of form. This term
appears to be interchangeable with “form” in common usage. Type Vessels are said
to be of the same type if they are identical in all significant features, and
not merely similar in general appearance. The
Romano-British coarse pottery: a student’s
guide specified type as vessels identical in all features of form and
fabric. Modern usage concentrates on the detail of the vessel form, which
facilitates comparison of different production centres. Group This is a vague
term but where used in this document is usually in the sense of a group of
pottery from a context. Assemblage This term is more
usually taken to mean larger groupings, as with Contexts sharing close
stratigraphic associations. or pottery from discrete phases of a site, or even
the entire stratified pottery from a site. 10 Fulford. M G and Huddleston, K 1991 The current state of Romano-British pottery
studies, English Heritage Occas. Pap.
I Frere. S S (ed). 1975 Principles of Publication in Rescue
Archaeology. DAMHB. MAP2 English Heritage. 1991 The Management of Archaeology Projects. Webster, U (ed). 1975
Romano-British coarse pottery: a student’s guide. CBA Res Rep. 6. Young C J (ed).
1980 Guidelines for the processing and publication of Roman pottery from
excavations. DAMHB. Occas Pap. 4. 11
|
Study Group for Roman Pottery (SGRP)