5. Moving Towards 3D - Challenges of Spatial Integration

5.1 Different ideals meeting 3D photogrammetric documentation

In dealing with the ideal of archaeological documentation, it is in part a question of how we perceive the archaeological record and in part the issues of absence and incompleteness and whether 'the total record' actually exists. This is illustrated by the classical controversy between Petrie, who argues for the idea of the selective record, and Pitt Rivers' idea of the total record with respect to the collection of archaeological evidence (Lucas 2012). On the other hand, spatial recording, naturally, lies at the centre of both ideals, and a common limitation is what we know as the paradox of excavation – that in order to record and document, we destroy the primary record. This does not apply to non-invasive archaeological prospection and remote sensing, of course. As we have seen, the recording strategy is deeply rooted in traditions and in a propagating documentation ideal - be it single context planning or schnitt/planum. In either instance, the notion of recording the past should perhaps be rephrased as 'the action of recording the process of investigating the past'. Arguably, the way we assure the value of archaeological spatial recording, is by accounting for the documentation process; meta- and paradata concerning the tools and the methodology applied, so that we may evaluate its validity and authenticity.

In comparing field archaeological traditions in the UK and Denmark, the different methodological approaches clearly relate to an ideal with consequences to our archaeological praxis – the role that documentation plays and the requirements it must meet. In traditional and single context archaeology, interpretation on site is not encouraged if not including post-processual reflexivity (Hodder 1997; Lucas 2001; 2012) but focus very profoundly on a final end-product of the excavation; a summary of the conclusions reached. Single context planning is as much an intellectual thought process of interpretations of what took place in the past in terms of sedimentary formations. This approach is, however, hampered by non-sedimentary events that do not relate to human activities, like biological processes – potentially undermining the premise for single context excavation.

The Danish excavation methodology over the course of the last decade has seen a wide acceptance of the more arbitrary schnitt, focusing less on the use of contexts. Instead, the excavation is considered an iterative process, by which the collection and evaluation of spatial data, primarily 'features', continuously improves and adds to the interpretation, and guides the excavation forwards. The excavation is often primed by a research agenda similar to the one proposed by Carver (2003; 2009) and the photogrammetric documentation is considered an observation-proximate control of hypotheses. Instead of interpreting the archaeological record as the product of archaeological contexts, the documentation itself becomes the framework of information, and organised by an agenda of visualisation. This is particularly clear from excavations by Aarhus University in the period 2002-2016 at sites such as Skelhøj, Jelling and Alken Enge (Holst et al. 2013; in press; Holst and Rasmussen 2013; Jessen et al. 2011). In targeting an illustration – in fact planning our documentation during the excavation on the basis of how we expect to be illustrating our results in the final report - the excavation is continuously being evaluated and the hypothesis is constantly questioned. As a consequence, a reconceptualisation of the basic documentation units is needed to dynamically support the iterative process and accommodate new spatial data both during the excavation and in post-processing. Basically, this means building the documentation from a series of data collections and documentation events, essentially recording the timeline of the documentation and interpretation process, rather than basing it on archaeological contexts (Jensen 2012).

As a comparison, the UK principle of single context recording actually fits well with the possibilities of photogrammetric recording, building up 3D models of individual units, interfaces or surfaces. It may even be the best way of documenting, considering that arbitrary sectioning, combined with photo-documentation, tends to basically record the excavation strategy and the location of cuts and sections. Individual contexts as 3D representations, on the contrary, directly reflect the archaeological record, or at least the surface of it, and much less the excavation layout. Single context archaeology is inevitably also being challenged by the new methods. Numerous individual contexts must be put together to create a 3D representation; an overall picture of 'what you can see'. In fact, the ideal of reaching a final interpretation within single context planning may be what is leading archaeology, particularly in the UK, to countless examples of visual reconstructions and applications of visualisation technologies. From a dissemination stand point and as an added bonus, this combines very well with the elaborate traditions of public outreach within British archaeology.

Interestingly, the two methodologies appear to approach each other and converge in dealing with the ideals of how we want to use the collected spatial data. Both focus on the ability to work with arbitrary surfaces; it is how both define their basic units of documentation. The main difference lies in whether the surfaces are actual physical entities or derived from visualisations or interpretations. Danish arbitrary surfaces will usually constitute physical sections or schnitt, which are available for scrutiny both during excavation and through observational, photographical documentation. The same section - or any other - could be constructed post-excavation in a single context excavation from the collected context plans, although based on the interpreted contexts as the primary evidence.

Arbitrary surfaces may be thought of as the middle ground of documentation principles, where single context planning and strict stratigraphical approaches meet the arbitrary pragmatic sectioning of features. In combination with 3D documentation, a less rigorous approach to single context recording is possible, as stratigraphical information is embedded within the model, through the absolute recording of elevation in a 3D-model. At the same time, arbitrary surfaces are very distracting to three-dimensional recording, as they tend to depict the layout of the excavation rather than the actual archaeological features or contexts. On the other hand, 3D-recording of physical, arbitrary surfaces inherently delivers some of the stratigraphical information that might otherwise be neglected in a large, open-area excavation conducted through arbitrary sectioning.


Internet Archaeology is an open access journal. Except where otherwise noted, content from this work may be used under the terms of the Creative Commons Attribution 3.0 (CC BY) Unported licence, which permits unrestricted use, distribution, and reproduction in any medium, provided that attribution to the author(s), the title of the work, the Internet Archaeology journal and the relevant URL/DOI are given.

University of York legal statements