10

HIDDEN LANDSCAPES: GREEK FIELD SURVEY
DATA AND HELLENISTIC HISTORY

Graham Shipley

The hellenistic period, in the Greek peninsula and homelands, is increasingly seen as a continuation of the classical, and field archaeology is playing a major part in this re-evaluation. Nevertheless, there were also changes in the social fabric of Old Greece after the reign of Alexander, partly as a result of new social and economic configurations and partly because of the expanded power of Macedonia. In this respect, too, archaeology promises to illuminate new developments. Southern mainland Greece has been the setting for a large number of field survey projects, and the past decade has seen the publication of many in preliminary or, less often, final form. Among those for which we have more or less detailed published data are the projects carried out in Boiotia, northern Keos, and at five locations in the Peloponnese: Nemea, the southern Argolid, Berbati–Limnes near Mycenae, Methana, and Laconia. This paper aims to explore the evidence for social change in rural landscapes, while not forgetting the impact of urban change. The period covered is chiefly before the early stages of Roman rule in the second and first centuries BC.¹

Systematic field survey, whether non-intensive (also called extensive) or more recently intensive, has been carried out in Greece for about forty years. As a result, there are enough data from published fieldwork to allow the methodological controversies of the 1980s, for example about the uses of statistical sampling and of written sources, to be laid to rest.² There is now a set of more or less standardized field methods, and for historical periods no one would do other than take full consideration of both written and archaeological evidence. We are now in a position to analyse the interplay between short-term événements of the kinds attested in literary or epigraphic texts, and the longer-term developments that surface survey is best suited to revealing. Earlier generalizations about social and economic change in Greek landscapes are being rewritten in the light of actual survey data, which reveal variations between and within different regions, as well as between and within particular periods.

Even before archaeological surveys were carried out, written sources and inscriptions pointed to key changes in society and in the economy that may have had an impact upon the countryside in the hellenistic period. On the one hand, democratic forms of government, in which decisions were made by council and popular assembly, were now widely diffused among Greek poleis. Despite the over-arching power of Alexander’s Successors, such constitutions were not necessarily a sham, as Peter Rhodes and David Lewis have shown with respect to civic decrees.³ Christian Habicht has also commented on the reality and effectiveness of democracy in hellenistic Athens.⁴ On the other hand, for many smaller poleis in southern Greece we have little detailed information about their constitutions before this time. Often the regime replaced by democratic institutions will have been a constitutional oligarchy of some kind, and in such a situation the ruling elite may not have been forced to give up its dominant influence. Even in places with stronger democratic traditions, such as Samos, inscribed civic decrees suggest that in the third century the propertied elite was strongly over-represented among those who were politically active.⁵ In many cities the elite increased their political and social power in the late fourth and third centuries.

All of this may have implications for what was going on outside the city walls. A few citizens of Greek poleis in this period became exceedingly rich, after the manner of Boulagoras of Samos,⁶ others moderately prosperous. Although long-distance trade and commercial investment were flourishing, many such men probably drew a large part of their incomes, directly or indirectly (an important qualification), from surpluses in primary produce, such as olive oil and wine, extracted from their own land. In general, it seems that high status remained closely tied to the ownership of land, and to the ability to dispose of large surpluses and of the labour of others. Polybios asserts that there was general depopulation, that childless Boiotians left their legacies to be squandered on feasts, and that wealthy Eleians preferred rural prosperity to civic participation. His claims have sometimes been dismissed as anecdotal or applicable only to the elite circles in which he moved.⁷ Yet the picture he paints is consistent, and it will be interesting to see whether the survey data point to real changes in rural settlement resulting from increasing disparities in land ownership. Important questions to pose of the data are to what extent they support the survival of citizen smallholder farmers, and whether they show the increasing separation of a landed elite through the growth of larger estates.

Since relations of economic dependency between rich and poor landowners were surely common in classical times, these, too, might be expected to be visible in survey data. The spectrum of dependency relations in classical Greece included, most notoriously, Sparta’s helots, who may be described as state serfs.⁸ Besides these, there were other groups of semi-free or serf-like dependants within Greek poleis. In the Peloponnese, for example, we hear of the Unclothed of Argos,⁹ the Club-bearers (or Sheepskin-tunic-wearers) of Sikyon,¹⁰ and the Dustyfoots of Epidauros.¹¹ A dozen or more such groups are attested in different parts of the Greek world at different times. In the fourth century, for example, Theopompos and Aristotle mention the Poor Men of Thessaly, who often rose in rebellion.¹² Other sources add the Dependants of Illyria and the Goodmasters of Syracuse (the latter, no doubt, an ironic or apotropaic name).¹³ Aristotle repeatedly notes that the perioikoi (‘about-dwellers’) of Cretan towns are tributary dependants. They may be the same as the people he calls Plot-men and those who elsewhere appear as Captives.¹⁴ There are others in the Aegean islands, Asia Minor, and the Black Sea, some datable to the hellenistic period. Kallistratos in the third century names the Mariandynian Gift-bearers of Herakleia Pontike,¹⁵ while in the same century Phylarchos describes similar relations between the people of Byzantion and the nearby Bithynians.¹⁶ The existence of other such groups can be inferred for the hellenistic period, such as the Plainsmen of Priene.¹⁷ Since we cannot be sure that other cities had followed Athens’ example in outlawing debt-bondage, there may well have been others, each with its own origins in local circumstances.

De Ste Croix is probably right to say that they are singled out for mention by our sources precisely because they were not typical, and that we should avoid over-generalizing from these examples. The regular form of non-free labour in the countryside was probably chattel slavery.¹⁸ However, the point I wish to make is that such intermediate, semi-free groups are symptomatic of a wider tendency on the part of Greek elites to foster relations of dependency between themselves and poorer folk. Even in Athens, where debt bondage had been outlawed, poor and landless citizens in the countryside were probably obligated to their richer neighbours in different ways. Along with independent citizens, dependent citizens, and the unfree, these marginal groups extend the range of possible interpretations for the small rural habitations frequently attested by survey data. Although we will not necessarily be able to identify such semi-free groups in archaeological surveys, we should be prepared to consider them as one possible explanation for the existence of small rural farmsteads, as Michael Jameson has pointed out.¹⁹ Such sites may have been home to free citizens, dependent free citizens, serf-like groups, or even unfree inhabitants. In short, while rural farms are sometimes evidence of flourishing citizen smallholders, they are not necessarily so.

The contribution of archaeological survey

Surface survey is designed to capture evidence of the silent majority, rarely mentioned in classical sources and then only seen through the eyes of the literate elite. Field-walkers transect the land surface systematically, regularly spaced in a line. (There is, of course, the constant temptation to deviate from one’s line in order to reach the shade of the nearest tree.) The principal aim is to observe and recover examples of all possible kinds of evidence of landscape use. In Greece, this chiefly means potsherds and fragments of roof-tile, with the occasional ‘small find’ such as a terracotta figurine or a coin. In some regions, teams may even find walls or standing remains of buildings, while roadways, olive-presses, sculptures, and other material may be recovered. These finds may be the remnants of settlements, graves, cemeteries, temples, small cult sites, and so on, or may indicate transient use of the landscape such as quarrying, storage, and grazing. Partly by design, and partly because of the near-impossibility of dating finds accurately when they are picked up, most surveys set out with the express aim of recovering material from all periods (or at least all pre-modern periods). It is therefore only at the stage of study and publication that dating and interpretation take place.²⁰

Particular difficulties inherent in the analysis of survey data result from the necessary use of arbitrary chronological divisions. Different projects use different conventions. The hellenistic period may begin at 350, 323, 300, or even later, and may end in 200, 150, 146, 100, or even in the reign of Augustus. It may be assumed without argument that actual changes in rural settlement rarely occurred at these precise dates. A related problem is that of accurately dating the finds. Until recently, hellenistic pottery was not so intensively studied as classical red- and black-figure, and we still rely for comparanda on published pottery from a handful of key excavated, urban sites such as Corinth or the Athenian Agora. Worse still, many rural surveys produce only small quantities of fine pottery; more often one finds unpainted, local varieties that can be dated only tentatively by their similarities to fine wares from the major excavated sites. At the best of times, it is rare that a sherd can be dated to within fifty or a hundred years. Since the number of sherds of a given period from a site does not often reach double figures, there is only so much we can deduce about the history of an individual site.

Though the meaning of the word ‘site’ has been controversial, most archaeologists now accept its use in a value-free sense, to denote a cluster of artefacts with no particular imputation of role or function. A cluster of finds may be evidence of a habitation, but in principle it could equally be a coincidental group of finds that is really part of the random background scatter; or it could be something in between. In the end, survey archaeologists have to make judgements about which sites represent habitations and which do not, and when possible they do so using rigorous criteria. These could include a minimum number of artefacts of a particular date, a minimum size of scatter, a minimum density of finds, or a combination of particular kinds of artefact. In many cases, however, certainty about the role of a particular site is impossible.

There are also problems with measuring the size of a site. This can be done quite rigorously, for example by a combination of statistically random sampling at specific points and total (or representative) collection at others. Yet a surface scatter, however accurately measured, need not be a true reflection of what lies under the ground or once stood on the surface. Different sites may spread and come to the surface in different ways. Quantities of finds may bear no direct relation with the number of people who lived here; we should beware what Nicholas Purcell has dubbed the ‘pots equals people’ fallacy. In the case of a multi-period site, the surface scatter will be an amalgam of material of different dates, and the site may never have been as extensive as the whole scatter. Even if we can link sherds of different dates to different parts of the site, the small numbers of sherds of each period and the imprecision of dating are a further constraint on certainty. A vessel made in the early third century may have been used for one, ten, twenty, or fifty years before being thrown away, and it may have been used alongside vessels manufactured both earlier and later. Unless finds are really plentiful, the implications of changes in the size of a scatter may be difficult to assess.

There is also the matter of a site’s territory. When we refer to the ‘area’ of a site, we normally mean the extent of the scatter. That, however, is only part of the story; we also need to know how much land each site controlled, by assessing the distribution of contemporary sites; yet all too often we cannot be sure we have found every site in a given locality. In principle, a small farmstead may have been located in a large parcel of land, a large one in a small parcel. Site territories may be contiguous, or separated by gaps. Cultivation need not have extended over the whole territory of a site.

A way of circumventing some of these problems is to recognize that small sets of data are much more informative when combined. We can improve the readability of the survey data by combining sherds of similar dates and forming an overall picture of the use of a landscape at a particular stage. If we find twenty sites from a particular period, we can assess the overall use of a landscape, and may be able to identify trends within periods as short as a third or a quarter of a century. The more we aggregate the data, the more closely they can be assumed to approximate to something real. This even allows us to get around the problems of multi-period sites changing their sizes through time, since broad-brush changes across a landscape will still be more or less accurately represented by the combined data. Although it will still be difficult to read characteristics of a society, such as population numbers, from survey data, the comparison of results from different surveys can point us towards regional differences. The work of Susan Alcock has highlighted the different trajectories followed by settlement patterns in different parts of Alexander’s empire.²¹ Here I wish to emphasize the variety and change that can be seen within Greece itself.

Hellenistic data from published surveys

In comparing different surveys, we must take into account differences in methodology, such as different period definitions. Fig. 1 shows total site numbers from raw data published from seven surveys, arranged by period according to the chronological divisions employed by each project. Even without adjusting for differences between different chronological schemes, it is clear that there is a pronounced dip in site numbers in, broadly, the late hellenistic period. (Perhaps Polybios had good reason to make his pronouncements after all.) In some regions, numbers pick up again in the Roman or late Roman period. Given that each project covers a different amount of territory, we may get further by calculating the density of sites in a survey area (Fig. 2). As before, there is a pronounced fall in site density after the early hellenistic period. Now, however, differences between regions become clear. Site density is consistently highest in Boiotia, lower in places like Methana, Berbati, and the southern Argolid. Finally, to bring out changes through time, we can choose a starting-point at which to index each survey as 100 (Fig. 3). Methana, Laconia, and Berbati see an increase in overall site density in the hellenistic period, while it falls in other regions. The same three places maintain this increased density in the Roman period, while others remain below their classical levels.

Fig. 1. Total site numbers from seven surveys, by period. Key to Figures. Cl. = classical, Hl. = hellenistic, R. = Roman; the prefixes E, M, and L signify early, middle, and late respectively.

Fig. 2. Density of sites from seven surveys, by period.

Fig. 3. Density of sites from seven surveys, by period; indexed so that classical = 100.

A closer look at changes through time in a range of surveys, including some non-intensive projects, reveals more detail. In the Laconia Survey, rural settlement numbers rise sharply in about 300 BC and remain high during the hellenistic period, though there may be a slight decline in the second and first centuries. A similar, long-lasting rise in site numbers is seen at Asea²² and at Pylos,²³ where an upturn took place slightly earlier, in about the fourth century, but persisted for two to three centuries. In other places the upturn is short-lived. In the southern Argolid, for example, the rise in small farms occurs between about 350 and 250, but is immediately followed by a fall in the second half of the third century.²⁴ Berbati–Limnes sees a similar rise around 300 and a fall by 200.²⁵ Data from Nemea,²⁶ and preliminary results from the territory of Megalopolis, ²⁷ suggest peak numbers in classical and early hellenistic times with a fall from the third century. Boiotia sees a fourth-century peak, with a fall during the third or second century.²⁸ In northern Keos, where classical sites were numerous, most were abandoned by 200.²⁹ Finally, in a few surveys there are different patterns. On Melos, for example, contraction of rural settlement began earlier, during the classical period, and continued in the hellenistic.³⁰ In Achaea, however, contraction began and ended later than elsewhere.³¹

There are, then, both common features and regional variation. Some places had a long-lasting late classical and hellenistic peak of rural settlement numbers. A rather larger number of regions saw a collapse of site numbers in the early or middle hellenistic period. In the Laconia Survey area there was a slow decline, or possibly none. In Achaea the contraction began and ended later. These differences will be the product of local circumstances. The surveyors of the southern Argolid, for example, have interpreted the changes in terms of fluctuations in the intensity of olive cultivation and the export of oil.³² Elsewhere one might invoke distance from a major conurbation, or from a Macedonian garrison. The north-western Peloponnese was affected by the rise of the Achaean league and later by economic intervention on the part of Romans and Italians. Other regionally specific accounts can be imagined.

We can also compare the sizes of sites in different regions. Different survey publications give more or less detail of site size, so for this purpose I have used a narrower selection. In order to compare like with like, we should not place too much weight on average size. The average (or ‘mean’) will be exaggerated by the inclusion of a few large sites (as in the southern Argolid, where four small poleis lie within the survey area), and does not necessarily give a representative indication. Fig. 4 gives an example of median site size combined with an indication of the interquartile range of sizes. (The median is the middle site in terms of size, the one with equal numbers larger and smaller than itself. The quartiles of a series are the points dividing it into four, rather than two, equal parts, the second quartile being identical to the median. The interquartile range represents the middle 50 per cent of sites, ignoring both the largest and the smallest 25 per cent; it is a statistical measure often used to indicate typicality.) While site numbers (and the total area occupied by all sites) fall sharply in the hellenistic period before rising in the late Roman, typical size increases steadily from about 0.2 ha in classical times to 0.4 ha in late Roman. When we consider the top-ranking sites, which are not taken into account in calculating the median and quartiles, it is evident that there is a relative increase in the numbers of medium–large sites in the hellenistic period. Average size also rises, partly because of an increase in the sizes of sites at the top of the scale, and provides supporting evidence of an overall increase in site size.

Fig. 4. Median site size (ha) and interquartile range of sizes: Southern Argolid.

The same calculations for Methana (Fig. 5) paint a different picture. Typical site size is smaller than in the Argolid. There is almost no change through time, only a very slight dip in the hellenistic and Roman periods. In Laconia (Fig. 6) sites are smaller at the start of the hellenistic period and there is even less change during the period, only a slight decline in average size.

These results suggest that in Methana and in the Laconia Survey area the way in which the rural landscape was being exploited (though not necessarily the intensity with which it was exploited) changed little. They suggest that there was no widespread rise of large elite estates. For the southern Argolid, however, they are clearly consistent with such a development.

Fig. 5. Median site size and interquartile range of sizes: Methana Survey.

Fig. 6. Median site size and interquartile range of sizes: Laconia Survey.

Case study: the Laconia Survey data

A case study of one region will illustrate how local trends can be interpreted with the help of historical evidence. The Laconia Survey data share features of other surveys, but there are contrasts too. (Fig. 7 shows the survey area.)

In most other areas of Greece, as we have seen, a sharp fall in site numbers occurred at some stage during the hellenistic period, most commonly in the third century or at its end. In the Laconia Survey area, this is not seen. Instead, the mid-sixth-century influx into previously unoccupied lands, itself typical of Greek surveys,³³ is suddenly thrown into reverse by a sharp fall in site numbers that comes much earlier than elsewhere, around 450 BC.³⁴ Around 300 BC, however, there is a rise in all three sectors of the survey area, most marked in the relatively less fertile uplands of the north (Fig. 8).³⁵ During the middle and late hellenistic period, in contrast to other regions, there are only faint signs of decline, or none. The nearest parallel for such trends, not a close one, is in Methana, where, after a fifth-century rise and a slight fall in the fourth century, settlement numbers rise in the early hellenistic period before dropping back again.³⁶ Elsewhere the classical or late classical period is usually the high point of site numbers.

Fig. 7. Schematic map of Laconia Survey area (A. Sackett and D. Miles-Williams).

	Late archaic– Early classical	Late classical	Hellenistic	Roman
North	10	6	26	26
West	20	15	24	13
South-east	54	23	25	12
Total	84	44	75	51

Fig. 8. Laconia Survey site numbers by period and sector

The strength of classical rural settlement is usually linked to the political primacy of citizen smallholders. As Anthony Snodgrass has commented, the chōra was endowed with

a political significance which it could scarcely possess under any other system … the period of maximum rural dispersal is also the period of maximum population, power and prosperity, for the community as a whole.³⁷

This is not true of Sparta, since the collapse in rural farmsteads in the second half of the fifth century coincides with Sparta’s maximum power. The losses of Messenian territory after the battle of Leuktra, and again after Chaironeia,³⁸ were still generations away.

Late classical and early hellenistic Sparta notoriously suffered from a shortage of manpower, which Aristotle called oliganthrōpia. Stephen Hodkinson has argued that this was due chiefly to the demotion of poor men from citizenship as they lost the ability to pay the requisite subscriptions. ³⁹ This, in turn, was a consequence of losing their land. At the same time, better-off families were getting richer. Economic polarization was the product of a combination of factors, including increasing social competition among the elite and Sparta’s peculiar rules about female inheritance. No doubt the frequency of casualties in battle in the late fifth and fourth centuries also increased the instability of land ownership, and presented some of the elite with opportunities.

We may assume that, because of these changes, some small farms within Sparta’s territory were being taken over, in some manner, by wealthier landowners and combined into large (though not necessarily contiguous) holdings. One might expect such a trend to be reflected in the survey data – small farmsteads, for example, being replaced by fewer, larger estate buildings – but this is not the case. Much of the survey territory is simply abandoned in the late classical period, especially in the parts further from Sparta. This suggests that the rise of large estates was not happening there. The elite may have had little interest in the survey area, which mostly comprises hill-land east and north-east of Sparta where the quality of the land is generally not high. As Richard Catling’s study of the archaic and classical data has suggested, they may have concentrated their efforts in the better farmland of the Eurotas valley, where estate enlargement was presumably taking place.⁴⁰ Polybios bears witness to the success of arboriculture in the Spartan plain at a later date.⁴¹ The abandonment of farms in the survey area in the late classical period may have been due to their economic failure. Situated on marginal land at a distance from the central market, many of them may have found it impossible to compete with larger estates in Sparta’s core territory, whose owners may already have made moves towards extensification and cash-cropping.

Who, then, were the settlers on the new farms in the early hellenistic period? Hardly successful citizen farmers; most of the sites are poor in finds in comparison with other surveys. In any case, we know from Plutarch that there was a crisis in Spartan citizen numbers by the mid-third century, when barely seven hundred citizens remained of whom about a hundred owned a significant amount of land. Presumably all seven hundred owned, or had the use of, enough land to qualify them as full Spartiates.⁴² A second possibility is that the new farmers were helots, sent to abandoned farms close to Sparta as a way of making up for the loss of agricultural surpluses from Messenia. Such a measure, however, would make better sense earlier, in the aftermath of 369. Third, it seems unlikely that excluded members of the citizen body or other marginal groups could unilaterally set up as squatters so close to Sparta in order to gain citizen status. This area had been settled, and presumably owned, by Spartan families as recently as the mid-fifth century. Titles to the land were surely preserved, orally or in writing. Moreover, since land was evidently still a precondition of citizenship, why the crisis in citizen numbers in the 240s?

A fourth possibility, that a class of dependent labour was involved, may now be considered. This need not have been an unfree or serf-like group. Two scenarios seem possible. (1) The early third-century settlers may have been demoted Spartans whom wealthier patrons, with the consent or encouragement of the state, were helping to reopen the land. Even if this help was given in return for a percentage of the crops, the settlers may still have qualified as citizens and formed part of Plutarch’s six hundred poorer landowners. On this scenario, Agis IV and Kleomenes III, in the 240s and 230s, may not have been the first Spartans to attempt a kind of land reform in order to strengthen the citizen body.⁴³ Equally, since site numbers in the Laconia Survey area remained stable or drifted downwards after the early hellenistic period, those kings’ reforms had no measurable effect upon this part of Laconia. (2) Alternatively, title in the land passed to the patron, and the early-third-century settlers (of whatever social origin) stood outside the citizen body. The poverty of the new sites, and the critical shortage of Spartan manpower by the 240s, both tend to favour the second scenario. The key point, however, is that, whether the new farms were occupied by poor citizens or by non-citizens, relations of dependency were involved.

Furthermore, the resettlement of part of the survey area was a special phenomenon arising from local circumstances. In this part of Laconia at least, a rise in small farms did not represent the continuation or revival of a ‘classical’ agricultural society dominated by citizen smallholders. Viewed in this light, the uniquely sharp rise in rural settlement in third-century Laconia looks less like a success story. Both the (almost unique) late classical collapse and the hellenistic revival can be explained as the direct or indirect result of elite land accumulation. Particularly favourable conditions for such a practice may have existed in Sparta. The probable rise of elite farms in other parts of its territory may explain the abandonment of some sites in the Laconia Survey area during the fifth century. It may also account for the unusually successful re-establishment of rural farms in the early hellenistic period. The longevity of the resulting settlement pattern may also reflect the relatively late and slow urban development of Sparta.⁴⁴

Conclusions

Archaeological field survey has revealed a collapse of rural settlement in many regions of southern Greece in the early or middle hellenistic period. The trend is clear; it is now the role of interpretation to ascertain what social changes lie behind this development. We must not only take account of local factors, but seek more general explanations.

Alcock has pointed to a combination of factors: population decline, urbanization, and the formation of elite estates.⁴⁵ Overall population levels are hard to establish, since migration into towns might account for a fall in rural population, with no overall decline. The data reviewed earlier seem to be consistent with a rise in larger farmsteads in some regions but not in others. Yet the fall in rural site numbers is widespread outside Laconia, and is not always accompanied by an increase in site size. The decline is unlikely to be due to Roman intervention, which can hardly have had a deep influence upon the landscape before the second century. We should look for a home-grown explanation.

In the third century, as noted earlier, the political and economic power of the upper classes was increasing. This was the case both in independent poleis and in organizations such as the Achaean league, which expanded to include many of Sparta’s Peloponnesian rivals. The alarm that reportedly spread through the Peloponnese in the 230s and 220s at the resurgence of Sparta under Kleomenes III may reflect the class interests of the civic elites of the Achaean league.⁴⁶ Since a sharp fall in rural population is consistent with migration into towns, was this the way in which intensified dependency relations were working through into the settlement pattern in most regions? Were the late third and early second centuries a time when elite land accumulation started to bite? In some places, this may have resulted in the rise of dependent small farmers, in others a rise in larger elite farms, elsewhere the abandonment of cultivated areas as small farmers were driven out of existence and into the towns. While the changes were not absolute or complete in any region, each scenario may mark, in one way or another, the decline of independent, free, normally citizen smallholders.

Paradoxical as it may seem, this development had its roots in classical times, the period when the citizen farmer was supposedly the archetypal free Greek. Yet, as we have seen, the citizen farmer was never the only kind of farmer, at least outside Attica.⁴⁷ To the extent that the ‘peasant-citizen’ culture of the polis was ever predominant, its decline was the culmination of a longer trend that received a boost whenever curbs on elite ambition were removed. Despite the dissemination of democratic forms of government, the propertied classes may have seized the successive opportunities presented by Macedonian domination and Roman rule to enhance their position at the expense of their fellow-citizens. At Sparta, the evidence of survey implies that dependency relations among members of the free population emerged in an acute form even earlier than elsewhere. For this, an explanation may be found in the peculiar history of Sparta’s social development.

Survey data must be handled sensitively. We must not expect to know too much about individual sites, and the data are most emphatic when grouped together. They raise questions of fundamental importance about the identity and status of those living in the countryside, about who owns the land, how much of it is exploited and in what ways, its economic relation to a central place, and so on. Written texts can only begin to suggest answers to these questions; it is for archaeology to put flesh on the bones. Sixty years ago, Mikhail Rostovtzeff could write of conditions in the Greek homeland that

The large majority of the working class … lived on what they earned by their manual labour as peasant landowners mostly overburdened with debts, as tenants of parcels of land owned by the cities, the temples, various corporations, and private persons, or as hired hands in agriculture and industry.⁴⁸

Survey data play their part in revealing regional differences and local variations that previously we could only infer in the most general terms. They tend to confirm the evidence of the sources and inscriptions that the elite were extending their power, and may even give some support to those who would rescue Polybios from opprobrium. The data also point to regional variation, illuminating the ways in which this common tendency was revealed differently in different landscapes. John Davies has recently called for new analyses of how economies worked at the levels of region and polis in the hellenistic world.⁴⁹ I suspect that as more survey data become available we shall be able to see further into regional changes, but will also detect more and more local variation in rural settlement patterns.

Acknowledgements

This investigation forms part of a project on change in hellenistic landscapes in the Peloponnese, which was supported in 1999 by a grant from the Research Leave Scheme of the UK Arts and Humanities Research Board (see also Shipley 2000a; Shipley 2002a; Shipley forthcoming). I have drawn upon the results of the Laconia Survey, for which I am grateful to Bill Cavanagh and Joost Crouwel. A version of the paper was read at the Triennial Meeting of the Hellenic and Roman Societies in Wadham College, Oxford, in July 2001; I thank Robert Parker for the invitation to take part, and members of the audience, particularly Peter Derow and Lene Rubinstein, for helpful comments. For comments on earlier versions of the text, I thank Anne Sackett and Sarah Scott. I am grateful to Daniel Ogden and Anton Powell for their invitation to contribute to this volume, as well as for acute comments on this paper.

Notes

¹ On the effects of the Roman conquest, see the excellent treatment in Alcock 1993.

² See the classic debates in Keller and Rupp 1983.

³ See Rhodes 1997.

⁴ Habicht 1997.

⁵ Shipley 1987, esp. 210–11, 214.

⁶ Austin 1981, no. 113 (SEG 1.366).

⁷ Polybios, 4.73.5–74.2 on Elis (Austin 1981, no. 85), 20.6.1–6 on Boiotia (Austin 1981, no. 84), and 36.17.5–10 on Greece generally (Austin 1981, no. 81).

⁸ de Ste Croix 1981, 149, followed by Cartledge 1988, 37–9. See also de Ste Croix 1988, 23–4.

⁹ Gymnēsioi of Argos: Stephanos of Byzantion, entry under Chioi (Cartledge 2002, 301 B. 9). Called Gymnētes: Pollux, 3.83. See further Jameson 1992, 138.

¹⁰ Korynēphoroi: Stephanos of Byzantion, entry under Chioi (Cartledge 2002, 301 B. 9). Katōnakophoroi: Theopompos (FGH 115), fragment 176. Cf. Whitehead 1981; Jameson 1992, 138–9.

¹¹ Konipodes: Plutarch, Greek Questions, 291 d–e. Hesychios, entry under konipodes, defines koniortopodes as agroikoi, ergatai (‘rustics, workmen’).

¹² Penestai: Theopompos (FGH 115), fragment 122; Aristotle, Politics, 1264a32–6, 1269a37–b5; Aristotle, fragment 586; also Kallistratos (FGH 348), fragment 4; Pollux, 3.83. All are translated at Cartledge 2002, 301 B. 5b, 302 C. 7a, 306 E. 2, 301 B. 7b, 303 C. 9, and 303 C. 13 respectively.

¹³ Prospelatai: Theopompos (FGH 115), fragment 40. Kallikyrioi: Aristotle, fragment 586. See Cartledge 2002, 302 C. 6–7.

¹⁴ Perioikoi of Crete: Aristotle, Politics, 1269b3, 1271b30–1272b18; cf. Shipley 1997, 217; part translated at Cartledge 2002, 306 F. 1. Klarōtai: Aristotle, fragment 586 (Cartledge 2002, 302 C. 7b); Kallistratos (FGH 348), fragment 4. Dmōïtai: Stephanos of Byzantion, entry under Chioi (n. 9 above).

¹⁵ Dōrophoroi: Kallistratos (FGH 348), fragment 4 (Cartledge 2002, 303 C. 9).

¹⁶ Phylarchos (FGH 81), fragment 8 (Cartledge 2002, 303 C. 10).

¹⁷ Pedieis: Hiller von Gaertringen 1906, nos. 60, 63, 139.

¹⁸ De Ste Croix 1981, 139–40 (intermediate statuses the exception to the free–slave dichotomy), 148–50, 160, 162 (Athens possibly unique in abolishing debt bondage), 171 (unfree labour may be widespread), 173 (serfdom in local forms, each apparently unique; but slavery the norm).

¹⁹ Jameson 1992, e.g. 135 (abstract), 145–6.

²⁰ For a general account of archaeological survey in Greece, see Snodgrass 1987, esp. ch. 4.

²¹ Alcock 1994, esp. 187–9.

²² Forsén, Forsén, and Lavento 1996, 91.

²³ Davis, Alcock, Bennet, Lolos, and Shelmerdine 1997, 455–7.

²⁴ Jameson, Runnels, and van Andel 1994, 383–4, 391, 393–4.

²⁵ Penttinen 1996, 229, 271–2, 279–81.

²⁶ Wright, Cherry, Davis, Mantzourani, Sutton, and Sutton 1990, 616–17.

²⁷ Roy, Lloyd, and Owens 1989, 149; Lloyd 1991, 189–90.

²⁸ Bintliff and Snodgrass 1985, 139, 145, 157.

²⁹ Cherry, Davis, and Mantzourani 1991, 331, 334, 343–4 (with figs. 17.6–7), 346. In Attica, where there appears to be a late classical peak of settlement, the rural demes may have declined in importance after 300 (Lohmann 1992).

³⁰ Wagstaff and Cherry 1982, 252–3 (though the reliability of the data has been questioned, see Catling 1984).

³¹ Lakakis and Rizakis 1992, 68–9; Petropoulos and Rizakis 1994, 190–2 (tables 1–2), 198.

³² See e.g. van Andel and Runnels 1987, ch. 6.

³³ Foxhall 1997, 123–7, identifies the late sixth century and/or classical period as the typical period of maximum settlement dispersal in southern Greece.

³⁴ See Catling 2002.

³⁵ Shipley 2002b.

³⁶ Gill, Foxhall, and Bowden 1997.

³⁷ Snodgrass 1987–9, 53 and 63; italics original.

³⁸ For the two-stage loss of Messenia, see Shipley 2000a; Roebuck 1948.

³⁹ Earlier discussions in Hodkinson 1986 and Hodkinson 1989; see now Hodkinson 2000.

⁴⁰ Catling 2002.

⁴¹ Polybios, 5.19.1: Amyklai is the finest place in Lakonike for trees and crops (kallidendrotatos kai kallikarpotatos). See Jameson 1992, 137 and n. 13.

⁴² Cartledge and Spawforth (2002, 42–3) suggest that the majority of these seven hundred may have had land that was mortgaged to the rich. Although the agōgē had apparently fallen into disuse by the mid-third century, it appears likely that land, if not mess contributions, remained a precondition for full citizenship.

⁴³ For these land reforms, see Plutarch, Agis 8.1–3, 13.2; Kleomenes 10(31).11–11(32).3.

⁴⁴ See Kourinou 2000.

⁴⁵ Alcock 1994, 188.

⁴⁶ Plutarch, Aratos 39. 8; Kleomenes 16–17. If, that is, Plutarch can be relied on in this respect; for, as Austin observes, ‘Aratus’ fear of the “contagion of revolution” … is not made clear in Polybius’ account’ (Austin 1981, 113). One would not expect Polybios to have missed an opportunity to highlight the threat to good order, as he saw it, posed by Kleomenes. On hellenistic elites, see also Shipley 2000b, 131–3, 191, etc.; also Shipley 1987, 202–28; on the Achaean league and Sparta, Shipley 2000b, 136–8, 145.

⁴⁷ On the difficulty of generalizing about changes in archaic and early classical Greek agriculture, see Morris 1998, esp. 74–9.

⁴⁸ Rostovtzeff 1941, iii. 1149.

⁴⁹ Davies 2001, esp. 34–6.

Bibliography

Alcock, S.E.

1993	Graecia Capta: The landscapes of Roman Greece. Cambridge, New York and Melbourne.
1994	‘Breaking up the hellenistic world: survey and society’, in I. Morris (ed.) Classical Greece: Ancient histories and modern ideologies, New Directions in Archaeology, Cambridge, 171–90.

Austin, M.M.

1981	The Hellenistic World from Alexander to the Roman Conquest: A selection of ancient sources in translation, Cambridge.

Bintliff, J.L. and Snodgrass, A.M.

1985	‘The Cambridge/Bradford Boeotian expedition: the first four years’, Journal of Field Archaeology 12, 123–61.

Cartledge, P.A.

1988	‘Serfdom in classical Greece’, in L.J. Archer (ed.) Slavery and Other Forms of Unfree Labour, History Workshop Series, London and New York, 33–41.
2002	Sparta and Lakonia: A regional history 1300–362 BC, 2nd edn, London and New York.

Cartledge, P. and Spawforth, A.

2002	Hellenistic and Roman Sparta: A tale of two cities, 2nd edn, States and Cities of Ancient Greece, London and New York.

Catling, R.W.V.

1984	Review of C. Renfrew and M. Wagstaff, An Island Polity, in Classical Review 98 [n.s. 34], 98–103.
2002	‘The survey area from the Early Iron Age to the classical period (c. 1050–c. 300 BC)’,in W. Cavanagh, J. Crouwel, R.W.V. Catling and G. Shipley, Continuity and Change in a Greek Rural Landscape: The Laconia Survey, i: Results and Interpretation, Annual of the British School at Athens, suppl. vol. 26, London.

Cherry, J.F., Davis, J.L. and Mantzourani, E.

1991	Landscape Archaeology as Long-term History: Northern Keos in the Cycladic Islands from earliest settlement until modern times, Monumenta Archaeologica 16, Los Angeles, Calif.

Davies, J.K.

2001	‘Hellenistic economies in the post-Finley era’, in Z.H. Archibald, J. Davies, V. Gabrielsen and G.J. Oliver (eds) Hellenistic Economies, London and New York, 11–62.

Davis, J.L., Alcock, S.E., Bennet, J., Lolos, Y.G. and Shelmerdine, C.W.

1997	‘The Pylos Regional Archaeological Project, part I: overview and the archaeological survey’, Hesperia 66, 391–494.

De Ste Croix, G.E.M.

1981	The Class Struggle in the Ancient Greek World, London.
1988	‘Slavery and other forms of unfree labour’, in L.J. Archer (ed.) Slavery and Other Forms of Unfree Labour, History Workshop Series, London and New York, 19–32.

Forsén, J., Forsén, B. and Lavento, M.

1996	‘The Asea valley survey: a preliminary report of the 1994 season’, Opuscula Atheniensia 21, 73–97.

Foxhall, L.

1997	‘A view from the top: evaluating the Solonian property classes’, in L.G. Mitchell and P.J. Rhodes (eds) The Development of the Polis in Archaic Greece, London and New York, 113–36.

Gill, D., Foxhall, L. and Bowden, H.

1997	‘Classical and hellenistic Methana’, in C. Mee and H. Forbes (eds.) A Rough and Rocky Place: The landscape and settlement history of the Methana Peninsula, Greece, Liverpool Monographs in Archaeology and Oriental Studies, Liverpool, 62–76.

Habicht, C.

1997	Athens from Alexander to Antony (trans. D. Lucas Schneider), Cambridge, Mass., and London.

Hiller von Gaertringen, F.

1906	Inschriften von Priene, Berlin.

Hodkinson, S.

1986	‘Land tenure and inheritance in classical Sparta’, Classical Quarterly 80 [n.s. 36], 378–406.
1989	‘Inheritance, marriage and demography: perspectives upon the success and decline of classical Sparta’, in A. Powell (ed.) Classical Sparta: Techniques behind her success, London, 79–121.
2000	Property and Wealth in Classical Sparta, London.

Jameson, M.H.

1992	‘Agricultural labor in ancient Greece’, in B. Wells (ed.) Agriculture in Ancient Greece, Skrifter utgivna av Svenska Institutet i Athen, 4o, 42, Stockholm, 135–46.

Jameson, M.H., Runnels, C.N. and van Andel, T.H.

1994	A Greek Countryside: The Southern Argolid from prehistory to the present day, Stanford, Calif.

Keller, D.R. and Rupp, D.W. (eds.)

1983	Archaeological Survey in the Mediterranean Area, British Archaeological Reports, International Series 155, Oxford.

Kourinou, E.

2000	Σπάρτη: συμβολὴ στὴ μνημειακὴ τοπογραφία της (διδακτορικὴ διατριβὴ), Ἡ μεγάλη βιβλιοθὴκη, Athens.

Lakakis, M. and Rizakis, A.D.

1992	‘Les grandes étapes de l’occupation du sol de l’âge de pierre à l’époque contemporaine’, in A.D. Rizakis (ed.) Paysages d’Achaïe, i: Le Bassin du Péiros et la plaine occidentale. Meletemata 15, Athens, 59–75.

Lloyd, J.

1991	‘Farming the highlands: Samnium and Arcadia in the hellenistic and early Roman periods’, in G. Barker and J. Lloyd (eds.) Roman Landscapes: Archaeological survey in the Mediterranean region, Archaeological Monographs of the British School at Rome 2, London, 180–93.

Lohmann, H.

1992	‘Agriculture and country life in classical Attica’, in B. Wells (ed.) Agriculture in Ancient Greece, Skrifter utgivna av Svenska Institutet i Athen, 4o, 42, Stockholm, 29–57.

Lotze, D.

1959	Mɛταξὺ ἐλɛυθέρων καὶ δούλων: Studien zur Rechtsstellung unfreier Landbevölkerungen in Griechenland bis zum 4. Jahrhundert v. Chr. Deutsche Akademie der Wissenschaften, Schriften der Sektion für Altertumswissenschaft 17, Berlin.

Morris, I.

1998	‘Archaeology and archaic Greek history’, in N. Fisher and H. van Wees (eds.) Archaic Greece: New approaches and new evidence, London, 1–91.

Penttinen, A.

1996	‘The Berbati–Limnes Archaeological Survey: the classical and hellenistic periods’, in B. Wells and C. Runnels (eds.) The Berbati–Limnes Archaeological Survey 1988–1990, Skrifter utgivna av Svenska Institutet i Athen, 4o, 44, Stockholm, 229–83.

Petropoulos, M. and Rizakis, A.D.

1994	‘Settlement patterns and landscape in the coastal area of Patras: preliminary report’, Journal of Roman Archaeology 7, 183–207.

Rhodes, P.J.

1997	The Decrees of the Greek States (with D.M. Lewis), Oxford.

Roebuck, C.

1948	‘The settlements of Philip II with the Greek States in 388 BC’,Classical Philology 43, 73–92. Reprinted in C. Roebuck, Economy and Society in Early Greece: Collected Essays, Chicago, 1984, 129–50.

Rostovtzeff, M.

1941	The Social and Economic History of the Hellenistic World, 3 vols., Oxford.

Roy, J., Lloyd, J.A. and Owens, E.J.

1989	‘Megalopolis under the Roman empire’, in S. Walker and A. Cameron (eds.) The Greek Renaissance in the Roman Empire, Bulletin of the Institute of Classical Studies, suppl. 55, 146–50.

Shipley, G.

1987	A History of Samos 800–188 BC, Oxford.
1997	‘ “The other Lakedaimonians”: the dependent perioikic poleis of Laconia and Messenia’, in M.H. Hansen (ed.) The Polis as an Urban Centre and as a Political Community, Acts of the Copenhagen Polis Centre 4; Det Kongelige Danske Videnskabernes Selskab, Historisk-filosofiske Meddelelser 75, Copenhagen, 189–281.
2000a	‘The extent of Spartan territory in the late classical and hellenistic periods’, Annual of the British School at Athens 95, 367–90.
2000b	The Greek World after Alexander: 323–30 BC, Routledge History of the Ancient World, London and New York.
2002a	‘Rural landscape change in hellenistic Greece’, in K. Ascani, V. Gabrielsen, K. Kvist and A.H. Rasmussen (eds.) Ancient History Matters: Studies presented to Jens Erik Skydsgaard on his Seventieth Birthday, Rome, 39–45.
2002b	‘The survey area in the hellenistic and Roman periods’, in W. Cavanagh, J. Crouwel, R.W.V. Catling and G. Shipley, Continuity and Change in a Greek Rural Landscape: The Laconia Survey i: Methodology and Interpretation, Annual of the British School at Athens, suppl. vol. 26, London, 257–337.
Forthcoming	‘ Κοινωνικὲς μεταβολὲς στὴ Σπάρτη καὶ στὴ Λακωνία κατὰ τηὴ ἑλληνιστικὴ περίοδο ἡ συμβολὴ δεδομένων ἐπιφανειακῆς ἔρευνας ’, in Πρακτικὰ τοῦ ΣΤ' Διεθνοῦς Συνεδρίου Πελοποννησιακῶν Σπουδῶν, Τρίπολις, 2000, Athens.

Snodgrass, A.M.

1987–9	‘The rural landscape and its political significance’, Opus 6–8, 53–70.
1987	An Archaeology of Greece: The present state and future scope of a discipline, Sather Classical Lectures 53, Berkeley, Los Angeles and Oxford.

Van Andel, T.H. and Runnels, C.

1987	Beyond the Acropolis: A rural Greek past, Stanford, Calif.

Wagstaff, M. and Cherry, J.F.

1982	‘Settlement and resources’, in C. Renfrew and M. Wagstaff (eds.) An Island Polity: The archaeology of exploitation in Melos, Cambridge, 246–63.

Whitehead, D.

1981	‘The serfs of Sicyon’, Liverpool Classical Monthly 6, 37–41.

Wright, J.C., Cherry, J.F., Davis, J.L., Mantzourani, E., Sutton, S.B. and Sutton, R.F., jun.

1990	‘The Nemea Valley Archaeological Project: a preliminary report’, Hesperia 59, 579–659.