William Cavanagh, Christopher Mee, and Peter James (henceforth: C, M, & J) have given us an archaeological survey volume of more than ordinary interest — a volume as innovative in its (re)iterative research and methodology as it is novel in its documentation of select rural sites. The primary aims of the Laconia Rural Sites Project (hereafter: LRSP) were to produce a more nuanced understanding of small rural sites and to posit more accurately their function within the rural economy of ancient Laconia. It was toward these aims that the authors revisited a subset of 20 sites of various periods located by the Laconia Survey (1983-1988). At each site, the authors mobilized a number of non-intrusive techniques — total material collection, geophysical prospection, trace element and phosphorus analyses of sampled soils, and other modes of determining soil morphology. Although a bit too technical in some respects for a non-specialist audience (i.e. in the presentation of the geomorphology and pedology), the work justly deserves the consideration of economic historians, landscape/regional specialists, and archaeologists at large.
The book is divided into four chapters: introduction, methodology, sites, and conclusions. In the introduction the authors present us with the key matter of concern regarding the ‘functionality’ of the rural site. As a continuation of the site-based approach of the original survey, the authors deal specifically with on-site material distributions.1 They begin by questioning the appropriateness of the category of ‘farmstead’ and assumptions about the nature of occupation in the classification of the rural site.2 C, M, & J then present us with various forms of evidence — literary, epigraphic, ethnographic, and material — for a much wider array of rural sites than can be encompassed by the category of farmstead and the assumption of permanent residence. In closing this chapter, the authors discuss the character of rural residence and land tenure in Laconia from the Prehistoric through the Ottoman periods.
The methodology chapter focuses on issues of site transformation, survey, and sampling. As a way into the LSRP methodology, a broader theoretical/practical background is provided to surface collection, the impact of post-depositional processes, geophysical prospection, and soil analyses in the Mediterranean. These considerations build toward the key question of the relationship between materials on the surface and buried features. C, M, & J are positive in their answer. Despite displacement processes, from the geomorphology of the plot (soil creep, fluvial erosion, gravity, etc.) to the agricultural action of the plow, the authors hold that gross correlative patterns between what’s on the surface and what’s underground can survive over the long term. Such confidence is critical to the sampling procedure. It is toward the latter that the remainder of the methodology chapter is oriented.
Here, the LRSP sampling procedure warrants attention if only in brief outline. The sites were overlaid by basic sampling grids which were divided into units made up of 5 X 5 meter squares. These units provided common referents for the superposition of artifact, geophysical, and soil data. While all ceramics and small finds were collected, all tile fragments were weighed and counted on site. Geophysical prospection then took the form of magnetometry, resistivity, and a variety of soil analyses. In the LRSP approach all these modes of sampling were then compared to gain an understanding of spatial, subsurface, and surface relations. It should be noted that the LRSP had varying degrees of success with the geophysics and soil analyses. Even so, ‘some degree of correlation’ was found between artifact distribution, geophysical anomalies, and concentrated pockets of elements in the soil on some sites.
The original Laconia Survey covered a contiguous area of approximately 70 sq km east of the River Eurotas.3 The sites dealt with in the LRSP were concentrated in the central area of the original survey due east of Spárti. Chapter 3 begins with a discussion of the criteria for selecting the sites, which included: small size, the potential for good preservation, the representation of the main chronological periods, and the representation of a variety of sites with material concentrations from heavy to light. Through this sieve fell larger settlement sites, complex, multi-period sites, special purpose locales such as quarries, bridges, or sanctuaries, sites located in areas of steep relief, and sites located on schist soils. As a result, large tracts of the original survey were not revisited. This should not take away from their accomplishment. While the LRSP personnel have substantially reduced the number of sites — a total of 420 — dealt with on the original survey, they have substantially amplified our understanding of the 20 focused upon (though the degree of difference in the detail between these sites will no doubt be a factor in comparative regional studies).4
After a brief discussion of the physical setting and environmental change, Chapter 3 presents the 20 sites in chronological order. The presentation of each site is divided into three sections: 1) an introduction, which includes summary of the data from the Laconia Survey; 2) a results section that contains the artifact data, geophysical prospection, and soil analyses; and 3) an interpretation section. The novelty of the LRSP’s documentation rests upon consistent use of mixed media for the sites — a combination of text, map, photograph, material illustrations, and other plans are given for each. These extremely important publication standards are, to be sure, more difficult to emulate with larger survey volumes dealing with hundreds of sites where images and plans cannot be shown for all. Also of note is the holistic treatment of the site evidence. Rather than split materials up into individual chapters focused upon general typologies, ceramics, lithics, and small finds are presented together with the rest of the geophysical, soil, and feature data. This contextual approach to each site is a welcome complement to the standard, specialized typological arrangement of materials as published in Volume 2 of the Laconia Survey.
In Volume 2 of the Laconia Survey, categories of site and material were presented in more neutral terms (i.e. they did not label sites as “farmsteads” or even as “occupied structures” in the catalog, rather these designations were reserved for the interpretative syntheses in Volume 1).5 LRSP provided the occasion to revisit the interpretations of sites and accumulate more evidence toward a particular classification. So S437, “a rather dense scatter” with a wide variety of domestic materials6 becomes LP14, an occupied site as suggested by the presence of “tableware, cooking vessels, and loom weights” with two concentrations of tile (p. 148). Though they did not detect any features of interest, the latter lead the authors to wonder whether there were two structures and, on the basis of the combined evidence, speculate that these may have been a “farmhouse and an annexe used as a storehouse” (p. 148). Likewise, M350, a “thinnish but well-defined scatter” with tile, a variety of ceramics, and a worked block7 is transformed into LP13 a potentially residential site, “at least on a temporary or seasonal basis” as suggested by the variety of ceramics. In addition, a magnetic anomaly along with charcoal in the topsoil may be indicative of a kiln or some other mode of “industrial activity” (p. 230). These processes of revisiting, reworking, and transforming archaeological knowledge, along with a more neutral and more unassuming approach to categories of site, are representative of good practice. They are demonstrative of good science. And, to be sure, a number of previous interpretations were not only refined, but were radically altered. U498, a “very small, thin scatter” which became LP17, turned out to be probably not a site at all. Indeed, a number of high temperature features such as kilns, furnaces, or burnt areas would have not been suspected without the subsurface prospection (p. 319). The recognition of such features adds to the range of activities present on rural sites.
Chapter 4, the conclusions, presents a series of reflections on the choice of sites, along with discussions of the effectiveness of the various methods enrolled in the survey, the material collection and analyses (tile, pottery, chipped stone, and soil), the correlation of artifact, geophysical and soils data, as well as the post-depositional effects. The chapter is brought to a close with a chronologically arranged discussion of site function followed by a short section on retrospect and prospect. Here the interpretive discussion of site function will be of particular interest to many as it seeks to demonstrate how the LRSP approach can help refine site categorization a little further. In this regard a little more emphasis is warranted as it gives us an indication of the range of site specificity and detail.
The authors identify the Early Bronze Age sites of LP7 and LP8 as small farmsteads or hamlets, while they regard Middle to Late Bronze Age sites of LP10 and LP20 as settlements with perhaps as many as a few dozen inhabitants. Their discussion of these sites is couched in terms of settlement hierarchy. The image of the Archaic-Classical sites is crisper (the fate of LP17 has been discussed). While LP11, due to the presence of pithoi and mortars, was regarded as a storehouse, LP18, LP19, and LP14 contained a wide variety of artifacts which are suggestive of residence. At LP18 high phosphate values along with a magnetic anomaly, which C, M, & J cautiously posit as a potential cistern, lend some support to the presence of animal stock. And while, the character and duration of the occupation could not be assessed, the authors speculate that Spartiates owned these sites. Eight sites are Hellenistic and/or Roman. Two of the Hellenistic sites, LP3 and LP15, were short-lived. One, LP4, was located in a prime agricultural area near the Eurotas. Here geophysical prospection revealed a structure. The initial and seeming similarity of Roman period sites due to their lack of domestic pottery was misleading as these sites turned out to be quite diverse. Both sites, LP 9 and LP16, of the Byzantine-Ottoman periods were elusive and as a result little could be said.
The LRSP methodology adds to our ability to posit a diversity of site categories through non-intrusive means. Yet while the authors speak of variability through time, they admit this to be ‘difficult to track’ (p. 315). We still do not have a solid handle on how such rural sites are transformed in use over time.8 Structures occupied by people can be made over into keeps for companion species, storage buildings, or refuse dumps. Indeed, to add not a wholly insignificant cavil, important debates concerning categories of rural site and their transformation in the Journal of Mediterranean Archaeology were neither cited nor discussed by the authors.9 This is unfortunate, for the LRSP contributes a great deal to the discussion.
There are, moreover, a few regrettable caveats. First, as the authors lament, it is most unfortunate that the ultimate goal of excavation could not be realized. This is a missed opportunity for further testing and verifying the correlation of what’s on the surface with what’s below the ground. Had this been accomplished it would have perhaps helped to solidify our confidence in the wide variety of techniques brought to bear upon each site. Likewise, excavation can potentially lend something more to the issue of site transformation over time.10 Second it is regrettable that C, M, & J did not cast their net wider in relation to the diversity of potential sites for survey. Here the ultimate goal of excavation partly determined the course and goals of the project and thus partially limited the scope to focusing only on sites with the potential for good preservation. The LRSP approach could have proved fruitful across a much larger area of the original survey. Still none of this should take away from the fact that the LRSP volume represents an important contribution which challenges others to invest in similar research.
Making good on the hopes that the interpretive volume of the Laconia survey was a “beginning, not an end,” C, J, & M have demonstrated how the sites, materials, and features of intensive surveys may be further modified and refined through (re)iterative (multistage and multiscalar) research. (Re)iterative research involves reworking materials in the context of new questions. While this is a mainstay of archaeological practice, rarely do survey practitioners return to the landscapes and sites themselves to test presuppositions and produce new knowledge.11 In so doing the LRSP has strong claims upon being treated as a model of good (re)iterative survey practice (especially given the extreme difficulties posed by conducting multistage and multiscalar research in Greece) by archaeologists at large. Here we may place our hope that all survey projects be considered beginnings and not ends.
Notes
1. For further comments on the site-based approach of the Laconia Survey refer to Pettegrew, BMCR 2003.09.42.
2. The authors in the original Laconia survey volumes were careful about the designation of a particular site without supporting evidence to corroborate its function within the catalogue which was published prior to the interpretive volume (refer to n. 4). Other surveys of the time offered interpretations of sites within the site register. For example, of the 328 sites located by the Argolid Exploration Project, 166 either attained or were candidates for being a ‘farmstead’ — by far the most of any other category of site on that survey. Refer to M.H. Jameson, C.N Runnels, and T.H. van Andel, A Greek Countryside: The Southern Argolid from Prehistory to the Present Day (Stanford 1994) 415-538.
3. For a concise description and rational for the selection of the original Laconia Survey area refer to W. Cavanagh, J. Crouwel, R.W.V. Catling, and G. Shipley, Continuity and Change in a Greek Rural Landscape: The Laconia Survey. Volume I, Methodology and Interpretation (London 2002) 1-15.
4. For more on the issue of comparative survey data refer to S.E. Alcock and J.F. Cherry, Side-by-Side Survey: Comparative Regional Studies in the Mediterranean (Oxford 2004).
5. Though this is not the place for a detailed discussion and critique, this separation of data and interpretation is not without its major problems. Refer to I. Hodder, The Archaeological Process: An Introduction (Oxford 1999).
6. W. Cavanagh, J. Crouwel, R.W.V. Catling, and G. Shipley, Continuity and Change in a Greek Rural Landscape: The Laconia Survey. Volume II, Archaeological Data (London 1996).
7. Refer to Cavanagh et. al. (supra n. 5) p. 387.
8. Pettegrew has emphasized ‘reuse’ and taphonomy as major issues to be considered in discussions of rural sites. Such ephemeral processes are best addressed through excavation. Refer to D.K. Pettegrew, ‘Chasing the Classical farmstead: Assessing the Formation and Signature of Rural Settlement in Greek Landscape Archaeology’, Journal of Mediterranean Archaeology 14(2) (2001) 189-209.
9. As citation goes through 2003 there seems to have been plenty of time to incorporate the debate. Refer to Pettegrew (supra n. 7), R. Osborne, ‘Counting the Costs: Response to David K. Pettegrew’ and L. Foxhall, ‘Colouring the Countryside. Response to David K. Pettegrew’, Journal of Mediterranean Archaeology 14(2) (2001) 212-22. Also J. Bintliff, E. Farinetti, P. Howard, K. Sarri, and K. Sbonias, ‘Classical Farms, Hidden Prehistoric Landscapes and Greek Rural Survey: A Response and an Update’, Journal of Mediterranean Archaeology 15(2) (2002) 259-265 and D.K. Pettegrew, ‘Counting and Coloring Classical Farms: A Response to Osborne, Foxhall and Bintliff et al’, Journal of Mediterranean Archaeology 15(2) (2002) 267-273.
10. Though excavated structures are not without their own problems in this regard. For a succinct discussion of depositional problems and transformations in excavated houses refer to I. Morris ‘Archaeology, Standards of Living, and Greek Economic History’, in J.G. Manning and I. Morris (ed.), The Ancient Economy: Evidence and Models (Stanford 2005) 117-122.
11. Interestingly the potential for (re)iterative research was one of the major legitimating factors of survey in contrast to excavation. Refer to J. Cherry ‘Frogs Round the Pond: Perspectives on Current Archaeological Survey Projects in the Mediterranean Region’, in D.R. Keller and D.W. Rupp, Archaeological Survey in the Mediterranean Area, (Oxford 1983) p. 384. Here also I should underline the Pylos Regional Archaeological Project’s efforts in revisiting the survey area of the pioneering University of Minnesota Messenia Expedition.