2.2 Why model it at all?

Assisting interpretation

Explanations of archaeological sites tend to express themselves in terms of broad phasing and the overall character of a site, focusing on especially important features that the archaeologist deems worthy of more detailed investigation. This approach allows plausible hypotheses to be constructed round often scant archaeological evidence but can lack detail. Attempting to model a site in detail can highlight the areas of the interpretation that can be easily ignored in largely textual methods. Questions regarding the actual position and form of the roof cannot be ignored in a three-dimensional visual representation, neither can inconvenient aspects of the building's morphology and in this respect, 3-D modelling forces an archaeologist to consider them. The modelling process can also demonstrate whether interpretations that seem plausible in the archaeologist's mind actually fit in 3-D space.

One such problem area in the House of the Surgeon is the roof over the service area, which does not readily fit into a neat pattern owing to its irregular shape in plan. In any textual account it would be enough to say that the area was roofed; however, the issue cannot be ignored in a 3-D model without producing awkward-looking images.

Assisting communication

The old saying that a picture paints a thousand words applies very well to archaeology, where there is a long history of visual speculative reconstruction. Once 3-D models are created, images can be relatively easily produced, meaning the archaeologist now has an unparalleled means of communicating interpretations to other specialists and the public at large.

The massive difference between the house as an ancient domestic property and as a modern archaeological site ensures that the ability to visualise the house as it was will have an appeal, particularly to non-specialists. It can be difficult enough for experienced archaeologists to 'imagine in' details and create a mental image of the house as it might have been. This applies even more to members of the general public, so the ability to provide visual cues to stimulate interest is a vital part of any attempt to convey the results of the project to a wider audience. The human angle – what the house was like to live in, how it would have been decorated, what the difference was between master and servant, etc. – cannot be effectively conveyed to a largely non-specialist public without some kind of attempted reconstruction, 3-D or otherwise.

The tourist site's emphasis on preserving the ruins as they are rather than attempting to re-create the past environment means that there is a gap in the visualisation market. This gap is currently partially filled with books where images on acetate overlay photos of the ruins themselves, allowing the viewer to see an ancient and modern view of the same scene. A more sophisticated version of this is possible with 3-D modelling and its ability to create any view of a model.

Previous analyses of elite Pompeian housing have focused mainly on architecture (Richardson 1989) or decoration (Clarke 1991). Artefactual evidence seems to have been largely ignored until Allison's work on the significance of house contents (1992 and 2004). One reason for this is the ubiquity of upstanding remains and the paucity of reliably recorded artefactual data (Berry 1997, 186). Yet Allison convincingly argues (1992, 15) that previous interpretations based on a mixture of textual sources and analysis of upstanding remains are often insufficient to create a plausible account of room function, particularly in the less architecturally distinct rooms. Allison's analysis of 30 Pompeian houses (2004) provides much needed complexity to the oversimplified assumptions of architectural interpretations as well as indicating just how cluttered certain parts of a Pompeian house could be. This aspect of ancient domestic life is almost wholly absent from the visitor experience due to problems of artefact security and preservation. The incorporation of this evidence into visualisations of Pompeian houses would present a radically different image of domestic space than that presented by the site today.

Visualisation is not just restricted to the transmission of information to the general public. 3-D models can be used as a format for idea exchange between specialists (Cantone 2002), especially if both parties are familiar with the same software. When dealing with a 3-D monument it is helpful to work with 3-D data. Traditional plans and elevations present the data in a format that is well understood among specialists but several of these are required to re-create a 3-D site. One unified model with an accurate method of measuring distances would enable other interested specialists to interrogate the 3-D data and possibly create their own version, complete with modifications, as a basis for discussion.

Why model it in 3-D?

When attempting to recreate the experience of inhabiting a 3-D space, it follows that a 3-D approach will be much more effective. Even though we travel on the floor in a mainly two-dimensional fashion, we experience space in 3-D. Traditional 2-D representations such as plans and elevations take an orthogonal view of space that corresponds to no actual viewpoint and as such cannot replicate the experience of being in a place and seeing it in real life. Plans and elevations are scientific abstractions of data, not re-created experiences. Thus, any phenomenological approach to space benefits from a 3-D approach, being more accurate in re-creating the built environment. This is especially useful in Pompeii where one of its biggest selling points, both as an archaeological site and as a tourist attraction, is the height of the surviving ruins. Visitors come to walk around an actual 3-D environment and any attempt to expand on this experience must also deal in three dimensions. Visitors currently see the monument as largely denuded of furnishing, mainly lacking wall plaster and populated only by tourists, archaeologists and the custodi who look after the site, three groups of people largely if not completely absent in antiquity. A populated Pompeii is possible by 3-D computer visualisation with a flexibility and reality possible only with more narrative forms such as poetry and film.

With specific reference to the House of the Surgeon I hope to demonstrate that 3-D modelling is a much more effective method of displaying the hypothesised upper floor configurations. The upper floors are in different areas of the house and also occur at different levels. The ability to display both location and height information simultaneously greatly increases the effectiveness of the model.

In publications such as the 'Häuser in Pompeji' series detailing individual Pompeian domestic properties (e.g. Strocka 1984), the authors take a mainly orthogonal approach to recording wall decoration. Elevations and photos effectively record the standing remains but the limitations of 2-D representation ensure that it is impossible to see wall decoration in context. It is highly likely that wall decoration was chosen in antiquity as part of an overall scheme rather that as a set of discrete entities. The ability to display the decoration as a coherent whole and also to view the walls from 'eye level' gives a more experiential impression than the necessary but contextually divorced elevations.

Why use computer-based technology?

Computer-based technology offers a flexibility that no other form of visualisation can match. Once a model is built it can be modified to reflect many varying conditions. The tourist's ability to see the site in a variety of different conditions is limited by practical concerns. The site is closed after dark and although there is a 'Pompeii at night' event where visitors can wander the ruins with an accompanying audio-visual presentation, this is restricted to the forum and parts of the Via dell'Abbondanza and includes no individual houses. As a result, images of Pompeian houses at night are very rare. Guidobaldi (2002, 309) provides one of the few examples but it is clearly artificially lit, creating quite a pleasing image but not necessarily accurately reflecting the experience of being in a Pompeian house at night.

Tourists also only experience the site with the weather conditions of the particular day they visit. Far more people have impressions of Pompeii that are formed in the mid-morning of a hot summer's day than on a cool November evening. The southern Italian climate has quite a range of diversity about it. It is hard to imagine needing sources of light and heat in a Pompeian house when viewed at 11am in mid-July so the ability to at least replicate the dullness and darkness of dusk in winter is adding significantly to the communication of the site to non-specialists.

As use of modelling software becomes more prevalent in the archaeological community, there will be a growing body of models in the public domain. These models, if made freely available with appropriate metadata, can act as a useful library of objects to future researchers. When attempting to re-create staircases for which insufficient evidence survives, it is possible to draw on the work of Adam (1994, 200-5) who illustrates many examples from Pompeii and Herculaneum with dimensions and angles. It is not too optimistic to think that in the near future enough models will have been created to enable such elements as staircases to be reused, saving time and making the modelling process easier. There are significant dangers with this approach along with significant benefits. The benefits include placing modelling capability within the reach of researchers who currently have neither the time nor the expertise to model buildings from scratch. This opens up whole new avenues of enquiry to current researchers. One of the disadvantages lies in the tendency of this cut and paste approach to result in pastiches of archaeological sites rather than well-considered models. It also presents problems of inaccuracies going unnoticed and being replicated in further models. Accurate metadata attached to the objects, detailing source information of both the archaeological data and the 3-D model would be an essential element in preserving some kind of academic basis for these models.


© Internet Archaeology/Author(s) URL:
Last updated: Tues Feb 5 2008