This post, which briefly outlines processes of archaeological investigation that take place after excavation, follows on from the previous post on the excavation process, in order to answer a few of the questions that archaeologists frequently receive from those with no knowledge or experiences of the subject. Subsequent (but not necessarily following) posts will continue by providing more information on the ideas that often influence the ways in which archaeologists study the past, and on the use of archaeological evidence alongside other sources.
Post-excavation analysis – ‘post-ex’
After excavations have concluded, all the material must be examined, (in most cases of pre-construction archaeology) primarily in order to create a record of what’s been found, but also to attempt to understand this material, particularly what is may tell us about the ‘history’ of the site. Whilst ‘diggers’ sometimes continue their investigations in the office by proceeding with post-excavation analysis after they have finished on site, depending on the size and composition of excavation ‘units’, those who excavate sites may otherwise move on to new sites, with other staff continuing with the post-ex. Information on post-ex undertaken by the author can be found here.
Previous posts mention the importance of accurately recording the discoveries made on site. This is not only to enable interpretation during post-ex analysis, but is also to provide a record that will be available for use in future research. Contexts and finds are catalogued from the data records that were created on site, such as context sheets (often using familiar database programmes, such as MS Access, but in a museum environment, artefacts may be recorded by using more specialised software, such as MODES); and diagnostic artefacts are photographed and drawn. The site, contexts, and features plans are hand-drawn and digitised (using software such as ArcGis and AutoCAD to draft plans in relation to their geographical location), and along with any digital survey data (see previous post) developed in a format that is suitable for publication. This information, the context database, and site notes, are used to interpret stratigraphic relationships (which are discussed in a previous post) – commonly diagrammatically recorded using a system known as the Harris Matrix. The Sometimes unit archaeologists will have the necessary knowledge and skills to investigate and interpret features, artefacts and other material (such as environmental data: see previous post); at other times, the relevant experts will be commissioned.
As already noted, the spatial and temporal relationships between the traces of human activity that have been found are very important to the archaeologist, as they allow more to be said about historical development, by making comparison with similar material and sites elsewhere – both to help date material, and to place particular contexts into their wider historic setting. Such comparisons are fundamental to archaeology, as most of the material derived from excavations is dated by using ‘relative dating’ techniques.
Relative dating depends upon the location of material in relation to other material (the ‘stratigraphy’: see the previous post), which may be determined during post-excavation analysis by consulting the excavation records, and by undertaking research on similar sites and material to make comparisons. In some cases, an artefact (such as a coin) is known to have been made at a particular time. This is often thought to readily provide a close date for the deposit with which such material is associated (e.g. if a coin or newspaper dating to 1910 is found on a floor, the other things beside the coin on the floor might be presumed to have been put on the floor in 1910). However, the situation is not quite as straightforward as might at first be thought. Despite such finds being attributed definite dates, they can rarely be used to determine the exact date at which the object, or the associated material, was deposited: instead, they can usually only provide certainty of the date after which (in archaeology known as a terminus post quem – sometimes shortened to t.p.q, and more accurately translating as ‘limit after which’) the deposit was formed. This is because many objects are kept and used for some time before they are discarded and replaced – as can be seen by examining money contained in purses and wallets today: (unless just having received newly minted coins, perhaps after a trip to the bank), the coins are likely to have various dates, and may sometimes be several decades old. But comparisons with other contexts, and the investigation of other forms of evidence in conjunction with the material evidence, can sometimes reveal patterns in behaviour, that may enable dates to be narrowed down. For example, if a newspaper is found in the bottom of a birdcage, there is likely to be more time between the manufacture and deposition of this object, than a newspaper found on a chair, alongside a pair of reading glasses. The condition of an object may also be of use – for example, a worn coin is likely to have been in circulation for a while. The notion of terminus ante quem – effectively but not literally ‘time before which’ – is useful, but is less regularly adopted, as requires certain knowledge of when the context above the context in question was deposited. For example, there may be certain evidence to indicate that a car park was laid in 1975, so anything sealed beneath this deposit should date before 1975. However, as with the t.a.q., this is where the limits of the evidence can be seen, as the exact time and date of deposition is rarely known. The clay pipe beneath the 1901 house wall was not necessarily dropped before 1901, just as the rubbish bin placed on the car park surface need have been placed there after 1975: these separate events may have only occurred within a short time before and after the construction of the wall and car park, respectively. As only broad dates can usually be obtained, the possibility within some situations of temporal proximity requires caution in providing t.p.q.s as ‘at/to or after’, or t.a.q.s as ‘at/to or before’.
Similar caution must also be applied when considering technological and stylistic change as an indicator of date, but the context of a deposit is again very important. A common misunderstanding encountered when discussing the history and archaeology of standing buildings is the assumption ‘old’ fittings and fixtures necessarily date to the original phase of a house: an example might be a cooking range found within the room that originally functioned as the kitchen of a Victorian terraced house. Although we perhaps modernise and redecorate properties more often that occupants in the past (enabled by lower costs for materials, and reflecting an increased interest in ‘DIY’ during and after the second half of the 20th century), phases of modernisation during and after the early 20th century are commonly evident. With regard to the previous example, technological developments during the early decades of the century encouraged many tenants (and in some cases landlords, as well as the relatively small group of owner-occupiers) to replaced the cast iron Victorian range with a more efficient (and consequently cheaper to run), often enamelled (and therefore demanding less time for cleaning), newer model.
Conversely, whilst a wealthy household may be fitted with more up-to-date objects, it might be many years before a low-income household comes to afford even the least expensive technological or stylistic developments found in other homes. With regard to archaeology of the recent past, this is further complicated by the development of mass manufacturing techniques (particularly during the early 20th century) that brought down costs, so enabling often enable poorer households to purchase objects reflecting quite new designs; in these circumstances, the quality of the materials used was usually poor. It can be very difficult to determine the quality of objects, as such a criterion is open to a good deal of subjectivity, although the systematic and objective analysis of quality has been attempted by some archaeologists of the recent past, and attempts to develop new approaches are underway (this topic will be returned to at a later date). But problems remain: there were (and are) other mechanisms in place by which some poorer houses might obtain quality consumables. This includes trade in second-hand goods, and distribution of quality material as gifts – for example from ‘mistresses’ to those employed in domestic service, and of course; in these instances, analysis of condition might be one way to investigate the socio-economic conditions of households in the past. There may be additional, non-economic, reasons why ‘old fashioned’ objects are retained and continue to be used – the ‘heirloom’ is a case in point. Another is the continued use of objects from previous phases of activity, in which case the earlier material might be described as ‘residual’.
However, dating a context from the material within it, and in relationship to other layers, is problematic is the deposit is not ‘sealed’ (i.e. if there are signs of later disturbance). A number of circumstances may result in later material might finding its way into earlier deposits: e.g. through the action of burrowing animals. In these situations, it may be difficult to differentiate between the material that belongs to the original deposit, and that which is intrusive.
One of the most important ‘rules’ in dating undisturbed contexts is that they – and the deposits that lie above them – must date to or after the latest (i.e. ‘newest’) object. Going back to the example of coins, if a purse containing coins dating to 1898, 1911, and 1923, is found within a particular context, then it cannot have been deposited before 1923, because coins bearing this date were not manufactured before this date: 1923 provides a t.p.q for the context. If the coin is very fresh, it might not have been deposited too long after this date (though certainty is still not possible – the coin could have been stored for sometime before being placed within the purse). If very worn, it was probably deposited sometime after (unless the wear does not correspond to what might be expected – by comparisons to other coins – from the coin being in circulation for many years).
In some cases, more accurate – ‘absolute’ – dating techniques may be used: such as Archaeomagnetic dating, radiocarbon dating, and dendrochronology; these techniques can now obtain quite close dates for the deposition of certain materials. But absolute dating techniques are not applicable to all circumstances, are usually expensive, and can sometimes take a while, so they are primarily used to date contexts that are seen as being of particular special interest.
When all of the information has been analysed, a report is created to describe the historic background of the site, the approaches that were adopted for investigation, what was found, and interpretations of the material. Sometimes the report is published as a monograph; often it is published within an academic journal, or a local archaeological society journal. But due to limited budgets, most pre-construction excavations are recorded in a format often referred to as ‘grey literature’ (an example from the Hungate excavations can be found here). These documents are accessible through the excavation unit, sometimes local museums and local studies libraries, and increasingly through internet storage systems, such as Oasis.
The intention is to create a future post to consider how archaeological evidence might be studied in conjunction with other historical sources; but the next few posts will return to the case studies that are being investigated through LIP (as part of the Dec20 project). This will begin by an appraisal of the census evidence for No. 8…