Men of Bronze: Hoplite Warfare in Ancient Greece (54 page)

Pylos

The Pylos Survey was conceived in part as a “follow-on” from the Minnesota Messenia Expedition Survey (Alcock 2001: 190; Alcock et al. 2005). It was carried out by Jack Davis, Susan Alcock, and others in western Messenia, to the north and east of the Palace of Nestor at Pylos in 1991–95. Substantial reports have been published in
Hesperia
(Davis et al. 1997; Zangger et al. 1997; Alcock et al. 2005), and the project is also well published online at
http://classics.uc.edu/prap/
, including a full site gazetteer and pottery and small finds databases (though these are unedited and have not been updated since data entry in the field).

Very little material dating to the Submycenaean-Geometric range was discovered, and according to the published pottery report only three sites produced more than three certain sherds in the Geometric-Archaic range (Davis et al. 1997: 452–53). During the Archaic-Classical phase (seventh–fourth centuries BCE) there was a substantial increase in settlement and exploitation. But the investigators note the relative scarcity of Classical material, and the lack of evidence for small, dispersed rural farmsteads, suggesting that this might be a result of the Spartan domination on the region (Davis et al. 1997: 455–56; Alcock 2001: 195, fig. 13.2, 196–98; Alcock et al. 2005: 166–71). Indeed, the Archaic and Classical presence was substantially focused in one sector of the survey area (Area VI), which may well have been a sizable ancient village under the modern village of Romanou (Alcock 2001: 194–95 and fig. 13.3; Alcock et al. 2005: 163), and larger sites seem to have been the norm in this region (Alcock et al. 2005: 166–67). Not entirely surprisingly, the peak of occupation occurred in the Hellenistic period, after the liberation of Messenia (Davis et al. 1997: 456–57; Alcock 2001: 193). As in the case of the Boeotia Survey, it is possible to look directly at the raw ceramic data for activity/occupation in different periods (
table 10-3
, compiled using the unedited online pottery database, which may not be entirely accurate). It is clear from the sparse pottery finds that there is little evidence for substantial levels of settlement in the region before the seventh–sixth centuries BCE.

Kythera

The recent survey on the island of Kythera undertaken by Cyprian Broodbank and Evangelia Kyriatzi has not yet been fully published. The excellent project website
(
http://www.ucl.ac.uk/kip/survey.php
) clearly explains the sampling strategy and methodology, and presents preliminary results of data analysis. The project has discovered over two hundred sites. In the post–Bronze Age phases, Kythera shows a huge rise in site numbers during the Classical period, but only very small numbers of sites were occupied during the Archaic period. There is no evidence for Protogeometric or Geometric period occupation of rural sites (
http://www.ucl.ac.uk/kip/sitedates.jpg
).

TABLE 10-3
Finds from the sites in the Pylos Regional Archaeological Project identified as Dark Age through Archaic based on the unedited online pottery database at
http://classics.uc.edu/prap/

Survey Data and the Development of the Polis Countryside

It is clear from the discussion of the survey data above that there are some common trends in the rural settlement record across Greece. Strikingly, the Iron Age through the Late Geometric period is a low point in the settlement record virtually everywhere in Greece. In many areas there is almost no record of any activity after the Bronze Age until the Late Geometric (c. 750–700 BCE) period, and even at that point the evidence is very thin in most places. Small nucleated settlements appear to be usual, generally located in proximity to the best agricultural land. Excavation evidence confirms this conclusion (e.g., Lathouriza: Lauter 1985, Mazarakis Ainian 1994; Zagora: Cambitoglou et al. 1971, 1988; Oropos: Mazarakis Ainian 1998). Certainly there is no evidence of substantial levels of dispersed rural settlement anywhere in Greece during the eighth century; indeed the Greek countryside appears to have been quite empty at this time. Another common trend is that movement onto “marginal” sloping land, whether for habitation or for agricultural exploitation, is relatively late everywhere, and generally coincides with settlement peaks occurring well after the Archaic period.

Beyond these basic common trends there is a considerable amount of regional variation in the development of different Greek countrysides in their specific polis settings, and even within the territory of any particular polis. These variations between poleis are clearly linked to different trajectories of historical development. The cases of Laconia (where rural settlement peaked early in the late sixth–early fifth centuries BCE), and Pylos (where dispersed rural settlement was largely absent before the
Hellenistic peak) make this particularly clear, intertwined as they both are with the very particular development of Sparta as a polis, and its distinctive mode of domination over the southern and southwestern Peloponnese.

In most (but not all) Greek countrysides, dispersed rural settlement, consisting of a range of different kinds of sites (including graves and small sanctuaries), becomes characteristic at some point between the second half of the sixth and the third centuries BCE, at periods when the urban centers themselves are also densely populated. However, as the Boeotian case of Thespiai shows, it would be simplistic to interpret all of these as small isolated farmsteads belonging to the same sociopolitical group within the polis. It is plain that a range of “farmsteads” at different scales is likely to be represented, and distinguishing a single large “farmstead” from a cluster of smaller “farmsteads” using surface survey techniques alone is often not possible. In addition, the “sites” recovered reveal only the headquarters of a “farm”: it is likely in the majority of instances that the landholdings worked by the inhabitants would have included plots not directly adjacent to the structure itself (Osborne 198;, 2001; Foxhall 2001a, 2001b: 213). One of the things most difficult for archaeologists to judge is where the inhabitants of the sites revealed by surveys belong in the socioeconomic spectrum (Osborne 2004: 168–70). It may well be the case that even with the smallest sites we are not picking up the poorest farmers in the archaeological record at all, and that almost all of the “farmsteads” discovered belonged to relatively prosperous proprietors, with sufficient material goods to make an impact on disposal patterns. Slaves are certainly largely invisible, and even Morris and Pappadopoulos (2005) struggle to make a case for their presence on rural sites of the Classical period. On the other hand, wealth and poverty are relative, and some periods are clearly much wealthier overall in terms of the material goods, especially ceramics, that are available to a wide range of people, and which ultimately end up in the landscape.

That the peak of rural settlement varies so much in date from one area to another strongly suggests that dispersed habitation in the countryside may not have had much to do with widespread political changes or innovations in any one period, so establishing its link to a purported “hoplite revolution” across Greece is almost impossible. Rather, expansion (or decline) of dispersed rural settlement and changes in the exploitation of the countryside visible in the archaeological record are likely to be related to specific local factors. It is clear from the examples of Lakonia and Messenia that long-term political configurations may have had significant impact on the shaping of rural landscapes. However, it is probable that in most cases broadly socioeconomic factors and trends were most important (rather than overtly and narrowly political ones)—namely, those factors related most closely to the changing wealth and fortunes of the individual households that exploited the land. The patterns we see on the ground are likely to be the agglomeration of the activities and decisions of many different, individual households. Survey data generally provide no information about land tenure (Foxhall 2001b: 213–14), and a number of different land tenure arrangements could have produced the same archaeological patterns. Given the varied constitutions and political structures of the different poleis represented in the survey data, it may be remarkable that, beyond the differences in date, there is so much similarity in
these patterns of dispersed rural settlement when they occur. The underlying causes, of course, need not be the same in every case. However, what this broad similarity might suggest is that the fundamental relationship of households with land held as private property was very deep-seated in Greek culture, documented in texts from as far back as Hesiod’s
Works and Days
. In periods when investment in the countryside and pressure on land increased for whatever reasons (e.g., increased wealth, increasing population, additional sources of labor), individual households tried to make the most of the land to which they had access. Sometimes they might have carved out new cultivable lands where they had the means to invest in doing so. However, it is plain from the survey data discussed here that, except in periods when the best lands were densely settled, “marginal” hilly and mountainous lands and areas of poor soils were not intensively exploited or even occupied.

Conclusion

Where does this leave us with hoplites and the origin of the phalanx? For a start, the archaeological evidence strongly suggests that the Hanson hypothesis of an eighth-century move to dispersed agricultural settlement and the exploitation of “marginal” hillslopes and “
eschatiai
” is incorrect. The available archaeological data positively contradict this model. As van Wees (this volume) points out, practically all of the evidence that Hanson uses to support this view is derived from later periods, so that the picture he paints is anachronistic. Farmers in the eighth century appear to have lived in nucleated villages for the most part, using them as a base for cultivating the best agricultural land, generally located relatively close to these settlements. The evidence from archaeological surveys for this period is inadequate to produce even a very crude settlement hierarchy.

It is not until well into the sixth century in most parts of Greece that activity in the countryside becomes visible in any kind of meaningful way. However, even then, the extent of occupation and intensity of use is quite low in most places; Sparta seems to be somewhat exceptional in this regard. If we are looking for “the rise of the small independent farmer” (van Wees, this volume), we may not find him here either. There is no straightforward or easy way of mapping political structures such as the Solonian property classes, known from historical sources, onto the settlement hierarchies derived from archaeological data (although that has not stopped scholars from trying, e.g., Bintliff et al. 2007: 147–51). (And this is assuming that the production units associated with the Solonian property classes are correct—the “500 bushels” of the
pentakosiomed-imnoi
is, of course, the only one that is absolutely certain because it is embedded in the name: Foxhall 1997.) Even in periods of intensive exploitation of the countryside from dispersed residential sites, as appears in many regions from the later sixth through third centuries BCE, we do not know for certain that we are seeing “the small independent farmer” on the ground. And this is a period when we know from written and iconographic sources that the citizen-hoplite was unquestionably well established in many places. Archaeologically documented “farmsteads” certainly vary in size and scale of
production, and rural settlement is not uniform throughout the countryside. But could we tell whether these “farmsteads” were operated by small independent farmers? The example of the takeover of Thespian land by the Boeotians in the mid-fourth century, cited above, suggests that we cannot necessarily distinguish dependents from owner-occupiers in the archaeological record (Bintliff et al. 2007: 143).

I suggest that we are seeing different things in the archaeological and historical records. The written sources (and even the iconographic sources) primarily show us male individuals, often working together, but portrayed specifically as citizens, soldiers, and property owners. The archaeological data reveal households and their collective activities, not “homesteaders” or proprietors as individuals. The archaeological landscape that we see in the data is the aggregate pattern formed by the decisions and activities of many households. What it does show is a widespread engagement by the inhabitants of a polis with the agricultural landscape as property-
working
units (whether they were always property owners or not is another question) in, broadly speaking, later Archaic and Classical times. This creates an archaeological signature that is quite distinguishable from different ways of working the landscape visible in other periods (e.g., Roman and Late Roman phases; see Foxhall 2004). If this is correct, then it is hardly surprising that we cannot find individuals as “hoplite fighters,” or even the origins of the phalanx, in the landscape data.

Bibliography

Alcock, S. 2001. A simple case of exploitation? The helots of Messenia. In P. Cartledge, E. Cohen, and L. Foxhall (eds.),
Money, Labour and Land: Approaches to the Economies of Ancient Greece
, 85–99. London: Routledge.

Alcock, S., A. Berlin, A. Harrison, S. Heath, N. Spencer, and D. Stone. 2005. Pylos Regional Archaeological Project, part VII: Historical Messenia, Geometric through Late Roman.
Hesperia
74:147–209.

Alcock, S., and J. Cherry. 2004. Introduction. In S. Alcock and J. Cherry (eds.),
Side by Side Survey: Comparative Regional Studies in the Mediterranean World
, 1–9. Oxford: Oxbow.