This wiki service has now been shut down and archived
Topic Leader: Professor Vince Gaffney, University of Birmingham
Rapporteur: Professor Gary Lock, University of Oxford
Scale: a meagre product of reality
Professor Vincent Gaffney
Visual and Spatial Technology Centre
Institute for Archaeology and Antiquity
University of Birmingham
The title of this paper originates in a quote by Friedrich von Schiller who exhorted us to “measure not by the scale of perfection the meagre product of reality”. It is actually a quote that I have used before in a published paper on scale but, perhaps, the sentiment of the quote deserves further consideration within the context of a wider discussion on the significance of scale within the humanities (Gaffney and Gaffney 2006). In attempting to do so now I would acknowledge that Schiller would have been appropriately reassured by the certainty that archaeology and the Humanities more generally are only rarely troubled by the problems of perfection. Moreover, within the context of archaeological analysis, at least, it is also a moot point whether many of our digital products are, or should be considered, a product of reality. Such observations may, perhaps, be dismissed as trite but the underlying point is a serious one and impinges upon any discussion associated with the issues of scale. Within the historical studies, the substance of our study, past action and meaning, can never be engaged with directly. This is in contrast to any metric description of an existing material residue, its associated physical context or the larger natural environment. Consequently, any aspect of representation carries significant implications in relation to how we understand or interpret our data. In spatial terms, an understanding of scale and, by association, the allied concept of resolution should therefore be central to our research.
Archaeology, in common with most humanistic studies, is inherently a spatial discipline. In empirical terms its practitioners are usually comfortable with discussion relating to the two factors discussed here, scale and resolution, which are encountered almost continuously during the process of data collection, manipulation and display. The definitions provided in the ADS GIS Guide to Good Practice are as useful as any in establishing a primary point of reference. Here we learn that scale is the ratio of the distance measured on a map to that measured on the ground between the same two points whilst resolution is the smallest distance that can be usefully distinguished on a map with a given scale. Whilst acknowledging the significant related issues of precision and accuracy, a prime quality of both scale and resolution is that, together, they represent significant abstractions of a potentially experienced reality. In cartographic terms the process of abstraction is often associated with map generalisation: a procedure associated with a considerable technical literature but which, in empirical terms, is almost invariably concerned with problems of direct representation. Whilst this is undoubtedly of value when ordering our data this also presents us with significant problems. If we regard scale as being a derived, simplified and therefore manufactured value, there must be some concern not with what scaled data represents but what it signifies, adds or subtracts from our interpretative schema. In other words, what is excluded from a data set as a consequence of scale may not be as important as the significance of the derived interpretation.
We can explore some of these matters through a consideration of the issues associated with the relatively innocuous interpretation of routes and communication – a mainstay of landscape analysis in many disciplines. As we move between larger and smaller scales the nature and interpretative significance of routes varies accordingly. Here, tracks, streets or roads may have an explicit or formal existence but routes, for instance, may be almost conceptual in their scale of representation or significance. The nature of the Silk Route, for instance, is a highly contentious issue. Another characteristic of these phenomena is that the behavioural significance of routes may also vary, not only according to the subject of study, but also according to the scale at which any analysis may be undertaken. Different processes may operate at different scales and, therefore, scale or resolution is not an objective optic on the past or indeed the present (Gaffney and Gaffney forthcoming).
Figure 1. The significance of scale (modified from Roberts 1996, figure 2.2, and reproduced in Gaffney and Gaffney (forthcoming).
These issues can become more problematic if one considers the associated qualities of routes. Accessibility, for instance, is a more general characteristic measuring the relative ease or difficulty with which one may cross or communicate across or within an area of land. Communication may also be a product of a route or the general accessibility of a landscape and it may, equally, take static or mobile forms. Issues affecting these qualities, and one might readily acknowledge discussion related to the use of viewsheds in archaeology, are equally affected by scale and resolution. Behavioural issues associated with the choice of scale are therefore substantial and these may become increasingly complex when temporal factors are incorporated into our analyses.
Having acknowledged the significance of scale it is important to consider, at least, whether the choices we make in relation to scale, and the undoubted limitations that arise as a consequence of such decisions, are actually required. There must be an argument that in some instances the increasing resolution of our spatial description must approach, if not perfection, at least adequacy on occasion. Laser metrical survey, whether as air-based LiDAR or ground-based 3D surface scanning, might be considered in this light, as may some aspects of remote sensing (Gaffney and Gaffney 2006). The increasing capacity of parallel or GRID networks to represent or analyse such data might also prompt such an observation. Unfortunately, whilst these achievements are significant in their own terms, the adequacy of measurement does not presume interpretative significance. We cannot necessarily finesse a path to understanding through an increasing resolution of measurement or enhanced access to computational power (Gaffney and Fletcher 2007).
The concepts underlying scale and the nature of scale effects are therefore likely to remain an issue of considerable academic concern. In disciplines, including the humanities, where behavioural, cognitive or phenomenological issues remain central to our interpretative position, the current awareness of the multivocality of scale will probably increase and become increasingly contentious. Paradoxically, this debate may deepen with increased access to massive computational power. Whilst likely to facilitate more substantive analysis of scale-related data, we may have to be more critical of our own analytical procedures as a consequence. In an imperfect world scale is, indeed, a meagre product of reality. This positional paper will discuss the issues raised by our current reliance on scaled data and consider options for future development.
- Roberts, B. K. 1996. Landscapes of Settlement. Routledge London.
- Gaffney, V. and Fletcher, R. 2007 Always the Bridesmaid and never the Bride! Arts, Archaeology and the E-Science Agenda. In Clarke P., Davenhall C., Greenwood C. and Strong M. (Eds.) Lighting the Blue Touchpaper for UK e-Science - Closing Conference of the ESLEA Project. PoS(ESLEA)031 (http://pos.sissa.it//archive/conferences/041/031/ESLEA_031.pdf)
- Gaffney, V. and Gaffney, C. 2006 No further territorial demands: on the importance of scale and visualization within archaeological remote sensing. In “From Artefacts to Anomalies: Papers inspired by the contribution of Arnold Aspinall. University of Bradford 1-2 December 2006. (http://www.brad.ac.uk/archsci/conferences/aspinall/ and http://www.brad.ac.uk/archsci/conferences/aspinall/presentations/Gaffney&Gaffney.pdf)
- Gaffney, V. and Gaffney, H. Forthcoming. Modelling routes and communications. In Külzer A. (Ed.) Handelsgüter und Verkehrswege. Commodities and Traffic Routes. Aspekte derWarenversorgung im östlichen Mittelmeerraum (4. bis 15. Jahrhundert)
- Thyveetil, M. A., Manos, S., Suter J. L. and Coveney, P. V. Use of UKLight as a Fast Network for Grid Infrastructures In Clarke P., Davenhall C., Greenwood C. and Strong M. (Eds.)Lighting the Blue Touchpaper for UK e-Science - Closing Conference of the ESLEA Project. PoS(ESLEA)013 (http://pos.sissa.it//archive/conferences/041/013/ESLEA_013.pdf)
Case Study: The Mass Project
Professor Tony Wilkinson, University of Durham
The MASS Project (Chicago-Durham) is employing agent based modelling to simulate the development of Bronze Age societies on the Near East, and show how they developed (or succumbed ) to stresses. Simulations, undertaken mainly at the Argonne National Labs in Chicago, are based on data derived from landscape surveys in northern Syria and Iraq, as well as satellite remote sensing, data derived from ancient texts, ethnographic reports and other sources.
The simulations produce output analogous to social processes, because agents (households and individual household members) can develop, grow and decline through time, but equally we can see how land use zones grow as population grows. A key part of the MASS project is to interactively link processes at the scale of the household, the settlement and its agricultural catchment and ultimately the settlement region. The model is also incorporating mobile pastoralists which adds complexity, increased spatial scale, as well as richness to the simulations. Although we use GIS and remote sensing to analyse the basic landscape data, the simulations run within Argonne’s own dedicated software.
Initial results of the simulations have recently appeared in American Anthropologist vol. 109 (1): 52-68: Urbanization with a dynamic environment: modelling Bronze Age communities in Upper Mesopotamia.
Case Study: Title TBA
Dr Mark Lake, University College London
* Scale * Heterogeneity * Standards and Metadata