The size of the Helot population under Spartan control in the classical period has not often been the focus of separate studies, although it has been an issue very often addressed in general appreciations of Spartan society and political history. Until recently, the estimates of Helot numbers have had an impressionistic coloration, serving to shed light on their authors’ appreciation of the basic qualities of Spartan life and the structures of Laconian society rather than investigating this subject in its own terms. I shall attempt to show, however, that it is feasible to situate within useful limits the carrying capacity of the Laconian agrarian system for various social classes.
The nature of ancient demographic studies and Sparta
Demographic research about Greek poleis labors under a heavy load of evidentiary disability. [1] We lack the detailed data about births, baptisms, marriages, and deaths provided by administrative records for later European history. [2] There are indeed isolated exceptions. Demographic studies of fourth-century Athens have profited from examination of lists of members of the Council of the Five Hundred, rosters of young men receiving ephebic training, and other enumerations of quantifiable civic groups. [3] Inscribed casualty lists from the fifth century comprise another, albeit largely unexploited, resource. [4] In contrast, our evidence for most other times and places is primarily literary, and of two types. They may be called the “summary” and the “military” categories. {193|194}
In instances from the former category, an ancient authority will state the size of a specific demographic segment or a total population. As an apposite example vis-à-vis servile groups like the Helots, I would note the notorious passage in Athenaeus that provides us with slave numbers in excess of 400,000 for each of the cities of Athens, Corinth, and Aigina (6.103.272C–D; cf. Westermann 1941; Sallares 1991:50–55). Athenaeus’ sources help to illustrate the quality of the information contributing to such figures. For Corinth, the important west Greek historian Timaeus is cited (FGrHist 566 F 5), and for Aigina the constitutional treatise on that city attributed to Aristotle (fr. 475 Gigon; cf. Schol. Pind. Ol. 8.30i). For Athens, a census conducted by Demetrios of Phaleron (probably in 317/16) is the original source cited by the intermediary Ktesikles (FGrHist 245 F 1). This composite citation nicely combines three modes of demographic testimonia: descriptive historiography, constitutional historiography, and documentary historiography. And all three figures strain credulity beyond its breaking point. Through the process of transmission and homogenization, whatever was originally valid has been distorted or prejudiced, and is only to be restored through conjecture (e.g. Figueira 1981:22–52; Canfora 1982:36–45). We happen to lack any similar ancient summary for the number of Laconian Helots, in large part because their place in Spartan society did not activate even these limited modes of commentary. Nonetheless, if the daimones of evidence survival had so graced us, it is most unclear not only to what degree such a figure could be credited, but also whether its point of reference could even now be determined.
Figures for troop and ship mobilizations comprise the second class of evidence. Naturally they need considerable massaging in order to render population totals. The relevant classes of individuals serving have to be determined relative to the sum of adult male citizens, who in turn must be defined in relation to the total for adult males. A model life table, ideally, or some other reckoning of the likely proportion of adult males to the entire population, must then be applied (note, e.g., Coale and Demeny 1966:782–783). Examples of this type of data are the figures for the mobilization of Spartan troops that provided one basis for my article in 1986. For Sparta, this line of analysis has a venerable history, extending back into the nineteenth century. I reproduce here my table from 1986, with some additions and corrections, as Table 8.1. Its basic conclusions have not, to the best of my knowledge, been systematically challenged. [5] {194|195}
Context
Spartan Army
Morai
Spartiates (20–49)
Perioeci (20–49)*
Spartiate Decline
1.
Plataia (479)
10,000 (20–49)
9400 (20–49)
5000
5000
N/A
2.
First Army Reorganization (455–446)
8004–8196 (20–49)
7104–7296 (20–49)
3055 (?)
4541 (+ 600 Skiritai)
39% (1. to 2.)
(300 Hippeis, 600 Skiritai)
3a.
Pylos (425)
7320 (20–49)
6720 (20–49)
2755
3965 (+ 600 Skiritai)
45% (2. to 3a.)
3b.
Pylos Adjusted for Perioeci
7704–7896 (20–49)
7104–7296 (20–49)
2755
4349–4541 (+ 600 Skiritai)
45% (2. to 3b.)
4a.
Mantineia (418)
7744 (20–54)
6144 (20–54)
2251
3239 (+ 600 Skiritai)
18% (3b. to 4a.)
4b.
Mantineia Adjusted
7744 (20–54)
6144 (20–54)
2086–2141
3349–3404 (+ 600 Skiritai)
25% (3b. to 4b.)
5.
Nemea River (394)
6600 (20–54)
5400 (20–54)
< 1833
2991 (+ 600 Skiritai?)
12% (4b. to 5)
6.
Leuktra (371)
4000–5000
3456 (20–54)
938
2150 (+600 Skiritai?)
51% (5. to 6.)
7.
Post-Leuktra
4000–5000
876+
7%> (6. to 7.)
* This column estimates only the number of Perioecic hoplites.
TABLE 8.1 The decline of Spartan manpower
The number of Helots under Spartan control has long been a matter of scholarly interest. It was the essence of the Spartan social order that the Helots supported the class of full citizens or Spartiates, so that their numerical strength conditioned the size and prosperity of the civic body. Moreover, the measures taken to control the Helots strongly influenced Spartan life. The Spartans themselves negotiated terms for Attic assistance in the event of a Helot revolt in their alliance with Athens of 421 (Thuc. 5.23.3), and these terms may recapitulate provisions in place as early as the Hellenic League of 481 (Figueira 1993:107, 294). Thucydides said that the design of Spartan institutions was predicated on security vis-à-vis the Helots, and outlined a Spartan atrocity to underline his point (4.80.2–4). Plato in the referred, perhaps inaccurately, to Spartan anxieties over Helot unrest circumscribing their response at a critical moment, specifically the Persian incursion of 490 (698E; note Dusanic 1997). Aristotle {195|196} vividly described the Helots as lying in ambush for the Spartans (Pol. 1269a37–b5). And Critias discussed a series of precautions taken by the Spartiates in daily life because of perceptions of danger from the Helots (fT. 37 DK [Lib. Or. 25.63]).
To gauge the magnitude and impact of Spartan fear of Helot unrest, scholars have attempted for many years to establish their numbers. Yet great care must be exercised in this line of speculation, because identification of the precise triggers for heightened levels of Spartan anxiety is a more complex issue than usually recognized, for such perceptions were profoundly ideologized. [6] Furthermore, perceived change in demographic proportions rather than absolute numbers may cause tension, as seen recently in Kosovo and Macedonia, where popular perceptions of the birthrate of Albanian speakers have profoundly and negatively influenced the politically dominant Slavs. For Sparta, it is noteworthy that our evidence for Spartan anxiety over the Helot threat derives from the period after the earthquake of c. 465, an event that altered the proportion between the exploited and exploiting social components (see below, pp. 224–7).
Unfortunately, our two categories of ancient demographic evidence are bound to serve us poorly in gauging the population of Spartan Helots. As a dependent group, the Helots were unlikely to generate the conventional type of literary reference. Nonetheless, several passages have been customarily cited for our question of Helot numbers. First, Thucydides observes that the people of the island of Chios had the most slaves except for the Spartans (Thuc. 8.40.2; cf. Theopompus FGrHist 115 F122a–b; Athen. 6.88–89, 265B–266F). That surely implies that the Spartan Helot population exceeded slave numbers not only at late fifth-century Athens, but also at other maritime centers such as Corinth, Aigina, and Samos. [7] Hence, Thucydides could intuitively place Helot numbers in a sequence of populations, the amounts of which can unfortunately no longer be determined. Second, Herodotus gives a number that might (marginally) count in either of our categories. Each Spartiate serving as a heavy infantryman mobilized for the campaign that led to the battle of Plataia was accompanied by seven Helots (9.28.2). That amounted to 35,000 Helots with 5,000 Spartiate hoplites. Third, in his discussion of the Spartan manpower crisis, Aristotle remarks that Spartan territory could support 1,500 cavalry and 30,000 infantry (Pol. 1270a29–31). That observation {196|197} naturally reflects on the carrying capacity of whatever segment of Spartan territory Aristotle was considering. Suffice it to emphasize again that even such valuable comments as these stand removed by several interpretative stages from actual estimates of a population for the inhabitants of Lakônikê and, specifically, for the Helots.
In the earliest scholarship, informed speculation mainly prevailed concerning the number of the Helots. This quality has persisted in more recent, cursory considerations. I say this not to belittle—since thoughtful and learned historians have been at work—but chiefly to rein in expectations. I list some earlier estimates in Table 8.2. Their wide range is noteworthy. [8] An investigative approach, emerging somewhat later, bypassed the absence of ancient evidence by focusing on the agrarian assets of the Spartans available to support the Helots and on modern economic and demographic parallels. This methodology was pioneered by Karl Julius Beloch in his groundbreaking monograph of 1886 (note Gallo 1990). After {197|198} refinement, this line of analysis could exploit data from modern Greek censuses, surveys on land usage, and research on the Greek and Mediterranean agrarian economy (see Gallo 1984:7–22).
Authority
Estimate
Relevant date
Spartiates:Helots
Manso 1800:127–140
312,000–800,000
479
>1:14
Müller 1839(2):44–45
224,000
479
1:7
Beloch 1886:146–148*
175,000
c. 400
Guiraud 1893:412
220,000
1:7
Grundy 1908:81
375,000
c. 480
1:15
Cavaignac 1912:272–274
140,000
5th century
1:7
Kahrstedt 1919:291
180,000
Jardé 1925:109–112
240,000–320,000
5th century
1:10
Busolt-Swoboda
1:20
GSK 1926:641–642
Coleman-Norton 1941:63
250,000
c. 600
1:10
Ehrenberg 1960:31
140,000–200,000
371
1:16–29
Cartledge 1987:174
175,000–200,000
Talbert 1989:23
170,000–224,000
* Beloch offers alternative minimum figures both of 100,000 based on Helot numbers exceeding Chian, Athenian, and Corinthian slave totals and of 60,000 based on an emended deployment of the Spartiates at Plateia and Herodotus’ ratio of 7 helots to 1 Spartiate.
TABLE 8.2 Estimates of Helot numbers
Moreover, unusually, in the case of Sparta, this methodology could be enriched by a quantitative appreciation of the actual processes of the Laconian economy, because the sort of speculative investigation of alimentation common for other poleis could in the case of Sparta be supplemented by data on the movement of products. Spartan subsistence was dominated by fixed or stereotypical transfers of output. The nature and scale of these movements of goods can be read backward to illuminate their implications for the nature of the underlying classes that sustained or were sustained by them. Moreover, our purpose in positing values for variables such as grain production per hectare and food consumption per capita is not predominantly to offer suppositions about lost data, although we may strive for the most accurate quantitative estimates. Rather we utilize the best hypotheses for the economic data to enable well designed “thought experiments” that test theories about the nature of the relevant social processes (such as a klêros system involving either sharecropping or fixed rents).
Our procedure is complicated by debates over the nature of Spartan property holding. Investigation is more straightforward for those who, like myself, accept the historicity of an archaic redistribution of a critical mass of real properties (the politike khora of Polyb. 6.45.3; cf. Plut. Lyc. 8.3; Plato Laws 684D) and the existence of a period during which klêroi were not exclusively transmitted by conventional, partible inheritance. [9] Analysis is more problematical, however, for those who see the tradition of the equality of the klêroi as solely a late ideological construct (Ducat 1983; cf. Papazoglou 1993). Even then, we would not be authorized to ignore the traditional figures on the number of klêroi, the level of rents, and the scale of mess dues. Their full ramifications would still deserve investigation in deference to the heuristic function of these figures in indigenous Spartan quantitative thinking about social relations. That conceptual status would truly be manifest, if it were the intention of the Hellenistic reformers to recreate a system of klêroi predicated on these amounts (cf., e.g., Plut. Agis 8.2). If these aspects of the system of klêroi, however, are to be condemned as merely Hellenistic constructs, the same details must consequently also be credited as Hellenistic appreciations, albeit intuitive, of the design parameters of various agrarian orders that may have been possible in Lakônikê. {198|199}
Some recent treatments
There have been three significant recent discussions with a bearing on the linked issues of the Spartan agricultural economy and the number of the Helots: my own article in 1984, a study of Michael Jameson in 1992, and sections of a book of 2000 by Stephen Hodkinson. [10]
My earlier synthesis and reconsiderations
In my article in 1984, I took my departure from the rents in kind paid by the Helots to their Spartiate masters (Plut. Lyc. 8.7; cf. Myron FGrHist 106 F 2). These included 82 of barley each year, which I estimated to equal c. 2,493 kilograms (kg) of barley by weight, to which I added a sixth to account for grain to be reserved for seed for the next harvest. That gave me an amount of 2,992 kg of barley for the annual rent and associated seed grain from each klêros or allotment. These figures for the weight of barley per medimnos and for the weight relationship of barley to wheat have been largely confirmed by the recently published Attic grain law of 374/73 (Figueira 1984:92–94 [esp. n. 14]; cf. SEG 36.146.21–25; Stroud 1998:54–55). For the productivity of klêros-land in barley, I adopted a range of yield of 750 kg to 900 kg for a hectare (ha = 10,000 square meters), having based myself on ancient and modern analogies (Figueira 1984:99–100). Thus I reasoned that the rent-producing land of each klêros was c. 3.3–4.0 ha, which I doubled to allow for alternate year fallowing. To my resultant units of 6.6–8.0 ha (for producing the klêros rents in grain), I next added 0.6 hectares for grape production to allow for payment of the monthly mess dues in wine of each klêros-holder (Plut. Lyc. 12.3; Dicaearchus fr. 72 Wehrli). So my rent-producing klêros land was c. 7.2–8.6 hectares.
One Spartan tradition going back to Tyrtaeus puts the rents of the Helots at 50% of their production (fr. 6 W; Paus. 4.14.4–5; Ael. VH 6.1). Such an even split of output or assets between an invader and his victim is a feature of Dark Age raiding (Il. 18.509–512; 22.114–121). Furthermore, a 50% maximum rent is common in share-cropping arrangements (Figueira 1984:103–104; Hodkinson 1992). Naturally, the question arises about the currency of this rule during the classical period, when fixed rents were in place (according to the Peripatetic authors of constitutional treatises on Sparta). Some have chosen to take the fixed rent and the levy of 50% as alternative traditions or as separate arrangements that were perhaps employed successively (see, most recently, Hodkinson 2000:125–131). In that case, {199|200} however, it would be passing strange if the simpler 50% rule superseded the more specific amounts.
I would instead suggest that the fixed rents were a specification of the 50% levy and the fixed rents and the 50% levy were necessarily simultaneous provisions in the late archaic and classical periods (note Paus. 4.14.4–5). [11] The Helots of each klêros paid fixed rents that were envisaged as 50% of the “normal” production of the allotments. Using calculations comparable to those for splitting agricultural holdings in active cultivation and fallow, the klêroi were configured to yield the requisite rents from a 50/50 division. Moreover, in exigent conditions, the klêros-holder had to be content with 50% of actual production. This very arrangement is attested from classical Attic land leases in order to compensate for enemy incursions (IG II2 2492.12–14; SEG 21.644.11–16). Plutarch Moralia 239D–E, which also appears to derive from a Spartan politeia, envisages just this arrangement, a fixed rent resting on an ancient tradition. [12] Here a prohibition against exceeding the apophora was enforced by religious sanction (ἐπάρατον).
There are a number of further considerations that support this reconstruction. First of all, the combination of the two stipulations was practically helpful whenever land had to be assigned to a klêros-holder or a klêros had to be reconfigured. The klêros could not in fact comprise a fixed amount of land—let alone, a discrete, unified block—for that would require periodic wider redistribution to accommodate the creation of any new allotments. [13] Instead, it was merely any set of nearby parcels for which 50% of the likely output met the criteria for rent payment, such as 82 medimnoi of barley. [14] Second, a fixed rent was administratively easier, because the klêros-holder need not concern himself with concealment of production in normal years (unlike someone supervising a sharecropping arrangement). Furthermore, since the stipulations added two ceilings to the income from the klêros for its holder, both a fixed amount and 50% of the crop, the incentives for a Spartiate to manage his klêros in the conventional manner of Greek landholders were greatly lessened. That inhibition promoted the Spartiate’s concentration on {200|201} political and military communal activities (cf. Plut. Ages. 26.6–9; Mor. 213F–214). [15] Spartiates were non-khrêmatistic and non-oikonomic klêros-holders in their own terms and non-opportunistic or non-optimizing property holders in ours. Third, high fixed rents and relative stability of the configuration of a klêros offered strong incentives to the Helots to meet or exceed the expected output, because they appropriated that addition, while protecting themselves against their masters’ displeasure (itself a punitive incentive). Finally, we shall discover below that a system of simple 50/50 sharecropping is hard to reconcile with the actual resource base of Lakônikê.
Consequently, if 50% is a default rent for the klêros, my calculation of the rent-producing part of the allotment had to be doubled to produce an estimate for a minimum size of the klêros that is consistent with any amount of rent in kind. Total minimum size of each klêros would be 14.4–17.2 ha. Unsurprisingly, the klêroi were larger than the size of ordinary Greek family farms (3.6–5.4 ha), even those supporting the families of hoplites (Burford Cooper 1977–78:168–172; Gallant 1991:82–87; Jameson 1992:137; Isager and Skydsgaard 1992:77–80).
There are five aspects of this reconstruction that I might now reconsider. First, I am now more tempted by the possibility that the measures of volume in this Plutarch passage are Aiginetic measures and not Attic, comparable to the citation of the mess dues in the same work (Lyc. 12.3). I would still note that such a stipulation does make it difficult to accommodate the system on the ground, because each klêros would require significantly more land. That revision would entail reworking all my calculations to account for rents about 50% higher in barley. Secondly, I would shift the seed grain into the portion of the output from the klêros that was retained by the Helots. Thirdly, I would emphasize more forcefully that there were rents levied in olives or in olive oil from each allotment, and that mess dues, calculated in terms of oil, were contributed by each Spartiate. These amounts, however, happened not to have been transmitted by the Peripatetic authors of constitutional works on Sparta.
Next, I would clarify the likely role of the production of wine grapes for the klêros and its rents. As I have noted above, our only figure involving wine is the monthly contribution to the mess. This contribution is likely to have been a larger share of the rent in wine than were the contributions in grain, because more wine than grain was probably consumed in the company of messmates than without them. In addition, it is less likely that the Spartiates split wine production equally with their Helots despite {201|202} the rule of 50% rent. Thus, it is unclear whether we ought to double the size of the acreage for vines of the klêros. In my actual calculations in 1984, this discrepancy was offset by using only the mess dues for estimating the size of the klêros and not a doubled or higher amount (to allow for Helot retention of one half the wine production). Wine is a high value agricultural product for its volume. Thus, wherever the value of farm output is to be transported over distances, stored, and perhaps redistributed, the prominence of wine, [16] just as in the mess dues, may indicate that the social processes at work transcend the simple conveyance of calories from producers to consumers.
To sum up, the cumulative force of these reconsiderations (of which the first is the most significant) might be to displace upward the range of the size of the klêros toward 18 to 21 hectares.
Finally, I would bring out more explicitly the implication of attributing a single particular yield in cereals to many thousands of hectares of agricultural land in Lakônikê, when so very broad a range of productivity may be justified. The crudity of that mode of estimation vitiates our incorporating an allowance for the attrition through mishandling, waste, spoilage, damage by vermin, and misappropriation, i.e., the effects that occur at each stage between harvesting and consumption. Rather than cloak our speculations about the Spartan agrarian system with a spurious exactitude, we are better advised to choose a “realizable output” per hectare that is sufficiently below the highest conceivable yields to include the wastage that ensued in subsequent processing. [17]
Later work on grain production in Greece and specifically Messenia has been divided on outputs for production per hectare, but has tended to stress the practice of alternate year fallowing. A study by Sanders (1984) has argued persuasively for higher yields for Melos (while correcting Wagstaff et. al. 1982), but there has been a tendency to downplay the relevance of Melos because of the fertility of its volcanic soils. The predominant trend, however, also based on comparative evidence, opts for lower yields, so that the force of my conclusions would be strengthened rather than diminished (Ruschenbusch 1988b; Sallares 1991:372–389; Garnsey 1988:10–14, 95–96, 1992, 1999:26–28). Many high figures, however, do exist from the early twentieth century. [18] If we adopt a lower yield of barley for each hectare of, for {202|203} example, 650 kg per hectare, our range for the size of the klêros might find its upper limit at 24 hectares. That seems an outer possibility. Messenia was an especially fertile region and, in general, the klêroi occupied the best land in Lakônikê. [19]
At this point, I could have crudely estimated the number of Helots for any number of equal klêroi, so long as I observed the rule that the rents were fixed and could not exceed 50%, and that seed grain (at least a sixth) was subject to provision. The non-rent-producing land committed to grain would support about 8 people at bare total subsistence, which I take to be around 250 kg of wheat or its equivalent a year (Figueira 1984:91–92, 107). Therefore, 6,000 such allotments would support 48,000 persons, and 9,000, 72,000 people. Adopting Aiginetic measures in place of Attic measures brings the number of people supported for each klêros to around 12 and for 9,000 klêroi 108,000. We ought to bear in mind this range of magnitude when we later consider the number of Helots more intently.
In point of fact, in my article in 1984, I did not take this approach. In the first place, it was clear that the Spartiate population had fallen from its high point by the period (450–370) for which I was investigating the operation of the messes (see pp. 224–7 below). That meant not only an aggregation of property, but also the likelihood that the procedure for the transmission of that property was evolving and land unit sizes were differentiating. More important, however, was my intention to illustrate the way in which the highly articulated material transfers in the Spartan social order tended to intertwine the economic and demographic destinies of citizen and dependent to an unusual degree.
Nevertheless, in 1984, I was compelled to focus on the allotments in Messenia and their Helot inhabitants. The Laconia Survey (Cavanagh et al. 1996, 2002) and the Pylos Regional Archaeological Project (PRAP) lay well in the future (Davis et al. 1997; Zangger et al. 1997; Davis 1998; Alcock 2002). Yet the Minnesota Messenia Expedition (MME) provided some valuable material on ecology and settlement patterns (McDonald and Rapp 1972; Rapp and Aschenbrenner 1978). Accordingly, I joined my estimates for the output and size of klêroi with the figure of 6,000 that I took to be not only a traditional number for a sub-unit of the klêroi, but one that also roughly fit the relative areas available for allotment in Messenia and Laconia (1984:101–102). The total area of Messenia was 2,872 square kilometers (km2), of which 1,276 km2 was cultivated in 1960 (or 44%, well above the average for Greece). My figures for labor inputs for cereal farming were taken from modern Greek surveys, where the effects of modern mechanization could be discounted (cf. Gallant 1991:75–76). {203|204}
Out of the modern total for cultivated land, I had to subtract the land exploited by Perioecic settlements (Figueira 1984:102–103). My results for the extent of arable land are outlined in Table 8.3. [20] Without the invaluable recent work of Graham Shipley (1992, 1996a, 1996b, 1997, 2000) and Jacqueline Christien (1982–83, 1989a, 1989b, 1992a, 1992b, 1998) on the topography of Lakônikê and in particular Perioecic sites, I was quite dependent on Niese 1906. In my review in 1984 Perioecic settlements comprised about 28% of Messenia.
Any such procedure of estimation can forward no great claim to precision. In the appendix, I assign each commune in modern Messenia either to klêros-land or non-klêros-land. Shipley’s synthesis (1996) has reinforced the Perioecic status of {204|205} most of the candidates from the earlier literature. [21] Of the more doubtful examples, changing our mind on the status of Leuktron and Pephnos will not throw off our estimate. [22] Their position on the western littoral of the Tainaron peninsula between the probable Perioecic communities of Kardamyle and Thalamai makes the existence of klêroi in their vicinity unlikely.
0.74 mt/ha
0.9 mt/ha
1. Perioecic Land
351 km2
351 km2
a. Known cultivation: 226 km2
b. Est. cultivation: 125 km2
2. Rent Producing klêros-land:
516 km2
432 km2
3. Helot Supporting klêros-land:
409 km2
492 km2
4. Barley product of line 3:
15,375 mt
22,230 mt
wheat equivalent of line 3:
9,949 mt
14,450 mt
5. Helots sustained by line 4:
39,976
57,800
6. Man-days to farm lines 2 & 3:
2,513,950
2,513,950
7. Male man-days for lines 2 & 3:
1,759,550
1,759,550
8. Man-units avail. from line 5:
22,786
32,946
9. Male man-units from line 5:
11,860
17,149
10. Man-days/166 day year (1438 hrs):
3,782,476
5,469,036
11. Male man-days/166 day year:
1,968,760
2,846,739
12. Man-days/120 day year:
2,737,320
3,953,520
13. Male man-days/120 day year:
1,423,200
2,057,880
TABLE 8.3 The agricultural economy of Messenia (mt = metric ton)
The status of the next group of sites, including Abia, Alagonia and Gerenia to the north is, however, more questionable. [23] Some classical remains and pottery notwithstanding, refoundation or discontinuity cannot be definitely excluded. A similar reaction stands for the next group to the north, the two communities that lay to the south of the Nedon River, Pherai and Kalamai. Pherai was probably a Perioecic town. [24] Kalamai is attested by Pausanias as a Dark Age/archaic Messenian community (4.31.3) and was a Hellenistic Messenian town (Polyb. 5.92.4). [25] Doubtful location confuses the issue: it was either at modern Sola, MME #140 (Perivolakia), primarily a Mycenaean site, or at Yiannitsa, MME #537, where there are significant traces of classical and Hellenistic settlement. This area probably lay in one of the five traditional regions of Messenia, the Mesola (Strab. 8.4.5, 7). Unfortunately, we cannot ascertain whether Pherai and its environs approximated Thouria and its setting, that is, a Perioecic town amid klêros-land, or whether the Nedon River marked a more decisive boundary, below which there lay only Perioecic communities between Pherai and Kardamyle. One indication of the {205|206} presence of klêroi might be the loss of the region north of Gerenia to newly founded Messene (cf. Shipley 2000:385–386). [26]
To be factored into our judgment is the tradition that the mid-eighth-century Spartan king Teleklos founded three towns, Ekheiai, Poiaessa, and Tragion, from the temple of Athena Nedousia (Strab. 8.4.4). These could be mythological names for some of the archaeologically or textually attested and possibly Perioecic sites in this area. Yet the tradition could also represent a later archaic “charter” myth, in which Messenian rebellion, hostility, or sacrileges against Athena Nedousia were intended to justify the creation of klêroi here. Hence it cannot be ruled out that southeastern Messenia provided some additional land for klêroi.
Nonetheless, in 1984, following Niese (1906:124), I had not addressed the possibility that Kyparissos was a Perioecic site. Shipley has now marshaled the evidence, placing Kyparissos/Kyparissiai in his grouping of attested archaic/classical settlements whose polis-identity is established in the Hellenistic or Roman period. [27] Kyparissia may be equated with modern Kástro Kyparissías (MME #70), a placement with which most of the citations are consistent, although the strong association with Nestor and Pylos may indicate translocation. [28] Yet, there is also a strong association of Kyparissia with Triphylia. [29] Accordingly, the Kyparissioi may have inhabited a district stretching north to the Neda river and Triphylia (Polyb. 5.92.5), possibly encompassing Aulon, and south toward Pylos. Such a determination might have far-reaching ramifications for habitation patterns in northwest Messenia. Not only does Kástro Kyparissías have considerable agricultural land (c. 1,600 ha within a 2.8 km. radius), but a number of other known sites in its district (MME #69, #200, #409) might also be conceivably considered Perioecic. Thus, our belief in a Perioecic Kyparissia raises the possibility that a considerable amount of the land in this region should be subtracted from that available for accommodating klêroi.
When this analysis, however, is combined with my closer review of modern Greek communities in the appendix, only slight changes to our view of the classification of {206|207} Messenian land results. Non-klêros-land appears to have constituted at least 28% of Messenia (with 33% standing as an upper limit).
Returning to my earlier calculations, there were two important conclusions that emerged. First, it was clear that so much grain was extracted from the allotments in rents that very large numbers of Helots could not be supported at subsistence from the land committed to grain cultivation. This conclusion not only arises out of the assumption that around 50% of production went to pay rents. It was also implied by my investigation of the land left for the subsistence of the Helots after the land needed to pay the rents and the land held by Perioecic communities was subtracted. My analysis implied a global Helot population between 60,000 and 87,000, when a possible c. 3,000 Laconian klêroi and their Helots were factored into the estimate. That ceiling for the number of Helots did, however, appear low intuitively. Thus I posited recirculation of food from the Spartiates back to the Helots, in part mediated by the messes, as one mechanism for supporting a higher Helot population. [30] Such circulatory regimes using incentives among servile agriculturalists can be paralleled. [31] Any inclination to adjust the size of the klêros upward or to lower the output in barley of klêros-land has the impact of intensifying these effects. I have been disappointed in some scholarly reaction to this hypothesis.
Let me note first the discussion of J. Ducat (1990:61–62). He compares the daily siege rations for the Spartans on Sphakteria of 2 khoinikes of alphita and 2 kotylai of wine that were agreed upon with the Athenians in 425 (Thuc. 4.16.1), and objects that the mess ration of alphita only exceeds the siege ration by 20%. I had been less impressed with this disparity in discussing the same comparison (1984:88), but it is important here to make my position more explicit.
First, the regime for food provision at Pylos was to provide the total intake of food, while only a single, albeit the main, meal of the day was consumed in the mess. Second, the meal in the mess was supplemented by cheese, figs, opsônion, meat from hunting, olives and olive oil, and other edible contributions, while the siege ration included only a portion of meat (probably small, as the amount is unspecified by Thucydides). Third, Ducat did not offer an explanation for the mess dues in wine, which are quite remarkably 220–240% of the Pylos siege ration in wine. As noted above, wine was a particularly convenient means for conveying or redistributing value in a barter economy, centering on natural goods (see Figueira 2002). {207|208}
Fourth, I emphasized the extraordinary total value of the mess dues, when expressed in terms of wheat, equaling at least c. 1280 kg, that is, enough economic output to support 5 people at bare subsistence. Furthermore, a mess of 15 Spartiates contributed annually an immense store of foodstuffs, equivalent in economic value to at least 20 metric tons of wheat. Truly it mattered not to a Helot whether he carried away a kotylê of alphita or the wine, figs, cheese, olives, vegetables, and meat from the mess contributions that could be traded for barley and wheat (cf. 1984:91–96). Finally, the mess dues were indeed not the only source for grain and other products for redistribution. After each Spartiate had tendered his mess dues, he still possessed 44 medimnoi of barley from each intact klêros, not to mention other agricultural output that cannot now be quantified (Figueira 1984:98–99). Any portions of this rent not consumed in a Spartiate oikos constituted a large supply from which to pay wages and offer emoluments in kind during recirculation to the Helots.
Even were we, however, to accept the validity of Ducat’s comparison of unlike alimentary regimes and downplay the assets for redistribution other than mess contributions in grain, it would still not entirely vitiate my earlier hypothesis of recirculation. Even on Ducat’s reckoning, 0.25 medimnos individually or 3.75 medimnoi for a whole mess of 15 Spartiates would be available monthly from surplus grain, enough to support 6 additional Helots at their Pylos rations, and (in Ducat’s view) a maximum of 2,000 Helots in all. That last estimate is erroneous even on its own terms. The actual adult Spartan manpower of 8,000 c. 480 is roughly compatible with the traditional and conventional number of klêroi of 9,000 (Hdt. 7.234.2; Figueira 1986:167–170). Eight thousand mess members gives us 3,200 more Helots supportable under Ducat’s conditions, a number that comprises a substantial increment of additional adult males to the entire servile population and to the dependent labor force.
Nevertheless, arguing over the amount of additional Helots sustainable from recirculation can be misleading, if we suppose that increasing that total population was the chief goal of such a process instead of its beneficial byproduct. A servile redistributive system operates in the realm of marginal decision-making. No Helot oikos was ever faced with the option of supporting one or more additional family members for the next year in return for any one action. Rather, some Helots were daily confronted with the chance for some extra food in return for an acquiescent demeanor, adherence to specific orders, undertaking or executing some task more efficiently, or some act of pro-Spartiate initiative. From the totality of Helot decisions, contingent on the pleasure of their masters, more resources were available to sustain their lives. Their quality of daily life was radically conditioned by such choices because of the magnitude of the transfers of their production to their {208|209} Spartiate masters. Does this incessant grind of agonizing over compliance or passive resistance not truly lie at the heart of “le mepris des Hilotes?” [32]
Misreading the social texture, scaling, and complexity of food recirculation as a technique for controlling a dependent labor force also affects the comments of W. Singor (1993:45–46). Recirculation through the messes influenced precisely those male Helots who interacted most intensely with the Spartiates, those who came into the “field of effect” of the syssitia. Hence, Singor’s caricature of my view, in which there is conjured up a fantasy of mule trains back and forth over Taygetos, is an unnecessary complication. [33] Nor were recirculation through the Spartiate oikos or recirculation through the messes ever exclusive mechanisms, either in practice or in my piece of 1984. Nonetheless, it was significant that some considerable recirculation was probably enacted in the public space of the messes rather than in the private space of an oikos at Sparta or on the klêros. In the messes, such material exchange was pervaded by class-to-class psychological affects rather than shaped by a behavioral dynamic transacted by individual masters and their individual dependents. That context for performance not only restricted any managerial tendencies for the Spartiate, but also inhibited his development of a mode of personal paternalism over “his” Helots.
Moreover, the hypothesis of redistribution not only helped account for a larger Helot population and illuminated one mode by which the Helots were conditioned to their role of dependency, but it also dramatized the fragile equilibrium of the whole system. As my elaboration of the labor inputs necessary to farm the klêroi suggested, the productivity of the klêroi was probably vulnerable to any significant subtraction of manpower (Figueira 1984:104–106). [34] For lower levels of productivity such as might prevail under wartime conditions, the flight or rebellion of male Helots brought the Spartan rural economy below the level of necessary inputs (cf. Gallant 1991:75–76). This factor is probably implicated in the sharp decline of Spartiate numbers in the years after the Attic occupation of Pylos (Figueira 1986:192–197). Many Spartiates then lost their ability to pay their mess contributions. The same cause accounts for an echo of this initial decline in the fourth century, when possible family limitation in the late fifth century seems to have had a marked delayed effect (Figueira 1986:202–206).
One approach that has not been suggested previously but which would affect my earlier estimate of Helot numbers would be to hypothesize that some Helots {209|210} were supported by non-klêros land. This option can be admitted on our theory that the core territory of the Spartan villages was not subjected to redistribution during the seventh or early sixth centuries. That hypothesis may be invoked to explain the persistence of a Spartan aristocracy that had carried over intact, into the regime of homoioi, an early archaic steep hierarchy of landholdings (see Figueira forthcoming). Larger estates in this core area may have always been tilled by dependent workers who originated in the weaker communities of the upper Eurotas valley and its watershed that had been absorbed by the Spartans or who derived from more vulnerable groups among the Spartans themselves. Smaller estates in these core holdings had originally been cultivated by their owners. After the consolidation of the system of klêroi, syssitia, and agôgê, however, even these former smallholders needed to induce either poorer Perioeci or (more to our point here) Helots from the klêroi to work in their place. [35] How much the acceptance of this surmise might shift upward the range for Helot numbers is uncertain, but an additional 20% of the Laconian Helots would number 4,000–5,780 people in my earlier reconstruction.
Comments of M. H. Jameson
In a paper delivered in 1990 and published in 1992, Michael Jameson drew attention to several features of the rural economy of Sparta that have a bearing on the issue of Helot numbers.
1. He compared the output of the klêros to the 200 medimnoi production of the zeugite census class at Athens, the class from which the bulk of the Athenian phalanx was raised (Jameson 1992:137). This comparison, however, would perhaps be better made with the next census class, that of the Attic hippeis, whose estates produced 300 measures of wet and dry agricultural product annually. If the Helots’ rent was in Aiginetic measures and constituted 50% of production, the normative klêros produced at least 246 dry measures and 12 wet measures of wine (specifically for paying the mess dues). That gives a total of 258 measures, without considering all the production of olive oil, whatever wine production was not paid over in dues, and other products tendered in rent. Certainly, this analogy reminds us that one of the outstanding features of the “Lycurgan” order was its blending of aristocratic warfare patterns with a hoplite polity. The more elite quality of Spartan full citizenship demanded a notably broader base in agrarian and human assets than comparable institutional orders elsewhere. {210|211}
2. Jameson also revisited the issue of the land available for Spartan klêroi, focusing on the Eurotas valley and the Helos plain in Laconia. He concluded that no more than 21,000 hectares were available, which allowed for 2,000 10.8 hectare klêroi, his lower limit (cf. Wagstaff 1982:48–64). Above, we have found that lower limit unrealistically low. At his upper limit for klêros size, which was 18 hectares, Jameson notes that only 1,200 klêroi could be accommodated in Laconia. Jameson graciously sent me a draft of a part of his research some years before this paper’s delivery, so that I commented on some of his calculations in 1984 (n. 48, p. 102). At that time, I estimated that his appraisal of the arable land in the Eurotas valley and Helos plains would support 1,221 to 1,456 klêroi. Jameson also criticized Paul Cartledge for offering figures between 50,000 and 75,000 hectares for the land available for klêroi in Laconia (Cartledge 1987:174–174). Although Jameson was not offering exhaustive discussion here, my fairness to both parties compels me to observe that Jameson is actually criticizing a long tradition of discussion on the topic, one that included such prominent representatives as Jarde (1925:112–113), Kahrstedt (1919:280–281), and Bolte (1929:1339–1340; cf. Figueira 1984:102; Hodkinson 2000:132, table 1). With Cartledge, I believe that there were probably more than 1,500 klêroi in Laconia at the height of Spartan population between 480 and 465. Mobilization figures seem to support this conjecture.
Nevertheless, Jameson’s investigation of the topography does suggest several observations about the Laconian klêroi. The existence of klêros-land outside the Helos plain should be envisaged. Shipley has drawn attention specifically to the plain around Kyparissia (1992:219–220). There may well also have been encroachments of klêros-land in the vicinity of the Perioecic communities of the Eurotas valley like Pellana, Geronthrai, and Khrysapha, a doubtful case as Perioecic. [36] At Khrysapha, the Laconia Survey confirms a wave of sixth-century occupations. Moreover, some land near Sparta itself may have been distributed in klêroi because it belonged to Spartans excluded from the creation of the class of homoioi, just like the Partheniai dispatched as colonists to Taras.
Moreover, Laconian klêroi had probably undergone some evolution in their composition. It may be that subdivision, along with the creation of new klêroi, had occurred as Spartiate and Helot numbers had increased over the sixth century. More marginal farming land may have been incorporated into existing klêroi, with teams of Helot families conducting their operations in several proximate locales. In fact, it may make sense to envisage the Helots themselves as the motor of these changes, attempting to increase their production so that more was left after their payment of the fixed rents. Concealment of production doubtless played some role in this {211|212} process. If we suppose that the 50% calculation was not constraining in non-exigent periods, and that klêroi were only examined at intervals for their output (as I have hypothesized above), the resultant periods of stable occupation were probably long enough to justify more intensive cultivation practices. The Spartans could be seen as periodically identifying and appropriating the advances of longer duration made by Helot farmers.
Therefore, we cannot speak dogmatically about the pattern of Helot settlement in Lakônikê (Laconia and Messenia). We can easily envisage that the Spartans would have acted to break up existing Dark Age settlements, like Nichoria in lower Messenia, for security reasons (cf. Lukermann and Moody 1978:92–95; McDonald and Coulson 1983:326; Spencer 1998). Scattered farmsteads constituting klêroi would then become more prevalent in these locales. And, in the Helos plain, there does seem to be evidence for more nucleation in the Hellenistic period, after the breakdown of the earlier socio-economic order (cf. Shipley 2000:382–383). Elsewhere, however, the process of creating new klêroi as the Spartans exploited the expansion of cultivated area by increasing numbers of Helots would be consistent with more nucleated settlements. Such settlements would be autonomously shaped by the Helots consistent with the forces militating in favor of such conglomeration in other areas of rural Greece. [37] This hypothesis would be consonant with the good evidence for nucleation along with low density of occupation in the Pylos region found by PRAP. [38] Archaic Spartan agrarian “policy” was limited to simple interventions to carve out new klêroi as necessary through inspections (made at harvest times?) in a manner appropriate to the primitive administrative and managerial techniques that prevailed (cf. Hdt. 2.109.1–2).
3. Jameson recognized that Aristotle’s observation about the carrying capacity of Spartan territory is problematical for his understanding of the limitations of the arable land available within Laconia (117,536 ha in modern Lakonia and Kynouria {212|213} in 1961). He countered by applying Aristotle’s estimate of 1,500 cavalry and 30,000 heavy infantry to Lakônikê prior to 371, when the Spartans held all Messenia as well (Jameson 1992:137–138n14).
While not an impossible inference, since Aristotle does refer implicitly to the battle of Leuktra in this passage, this conjecture does seem a little incongruous when the refounding of Messene would have preceded the earliest forerunner of the Politics by a considerable lapse of time. At the time of the Persian wars, Sparta had at least 16,000 Spartiate and Perioecic hoplites (18 or older). Adult male Perioeci of non-hoplite rank may have numbered as many as 30,000, a good number of whom would have been hoplites in almost any other city-state than Sparta. After all, a figure of 30,000 Perioecic allotments is found in the “constitutional” tradition (Plut. Lyc. 8.3; cf. Agis 8.1; note Lotze 1993/94). Messenia alone could perhaps support, solely from agriculture, 60,000 male and female Perioeci at bare subsistence (Figueira 1986:182–183). Thus it hardly makes for a striking point if Aristotle observed that all of Lakônikê, as constituted before Leuktra, could have produced 30,000 men of hoplitic rank under a different, more egalitarian social order and with a conventional Greek economy (cf. Polyb. 2.38.3).
Accordingly, our question is better formulated in these terms: could Aristotle, with some allowance for exaggeration, have believed that Laconia itself, along with the Spartan-held parts of the Arcadian borderlands, Messenia, and Kynouria (for which, see Shipley 2000; also Magnetto 1994), still had this demographic potential (31,500 of hoplite census) at the time of the composition of the Politics? If he did, he perhaps placed a high evaluation on the fertility of Lakônikê and on the aggregate income from non-agricultural sources. Yet, a legislator starting completely afresh could probably carve out at least 19,000 minimum hoplitic farms producing 150 medimnoi of barley out of Laconia and Kynouria alone, without considering non-cereal and non-agricultural income and without adding the production of the Spartan-controlled areas of Arcadia and Messenia. [39]
Comments of Stephen Hodkinson
Finally, let us consider Stephen Hodkinson’s recent contributions, which build on his piece published in 1992. I call attention to several of his conclusions that are particularly relevant to the subject of Helot numbers.
1. Hodkinson discusses the issue of the arable land available for the klêroi both in Laconia and Messenia (2000:131–145). Using the 1971 census, he has carefully explored the amount of cultivated land in modern Greek communities that might {213|214} have been available for ancient klêroi. Noteworthy is his emphasis on the need to include the arable land in the Helos plain that now belongs to the Eparchia Epidaurou Limiras (2000:141; cf. NSS 1978:1.218). The significant result is to restore the land available in Laconia for klêroi up from Jameson’s c. 21,000 ha toward Bolte’s level of 50,000 ha. [40] His result for Messenia is in line with earlier work that opted for around 90,000 ha, but is welcome confirmation indeed. Although Hodkinson is not comfortable with ancient figures about the number of klêroi, I would recall that one traditional split of the received total of 9,000 klêroi into groups of 6,000 and 3,000 is nicely accommodated by his understanding of the geography. Hodkinson offers various figures for the size of citizen estates (2000:382–385). These are a mean landholding size of 20.77 ha for 6,500 households, or, alternatively, the combination of an elite holding of 44.62 ha for 585 households and for 5915 households an ordinary Spartiate estate of 18.41 ha (in the range that I posited in 1984).
2. Let us turn to his estimate of Helot numbers (2000:385–395). Hodkinson begins his estimate of the population of the Helots from the 35,000 Helots taken on the Plataia campaign, which he reduces by 5,000 for the Helot batmen regularly accompanying their masters on campaign, so leaving 30,000 Helot farmers. That distinction looks artificial to me, and I would probably collapse this category into his reduction that is made to account for young males. For him, that implies 45,000 adult male Helot farmers between 20 and 60, by applying the two-thirds levy to the Plataian mobilization, and 52,000 of all ages. With 20% unmarried males, the remaining 41,600 Helot family units give a total Helot population of 187,000 by applying a multiplier of 4.5. He believes that this reconstruction is consistent with a Helot land holding of 3.25 ha for each family. Thus, these Helot land holdings must occupy 135,200 ha, or virtually all the arable land available for klêroi in Lakônikê (in Hodkinson’s view). Alternatively, Hodkinson offers the 7 to 1 ratio of the Plataia campaign as the Spartan perception of Helot numbers vis-à-vis themselves. Hence 8,000 Spartiates give ca. 56,000, and, subtracting batmen and youths, a sum of 36,000 Helot families and 162,000 Helots. This yields an average farm of 3.75 ha. {214|215}
This reconstruction is useful in my view because it illuminates the upper limits of a feasible Helot population. Its 3.25 ha average holdings would produce in barley between 1056 and 1463 kg, if we apply respectively a low output of 0.65 metric ton (mt) and a high output of 0.9 mt per ha and crop in alternate years. A 50/50 sharecropping arrangement will leave the Helots between 528 and 732 kg of barley each year. These incomes in barley are equivalent to between 323 and 476 kg of wheat. These outputs will not have supported Helot families, even if we adopt high incomes from non-agricultural sources. A population of 187,000 needs at least 46,750 mt of wheat equivalent to survive, or the output of a little below 100,000 ha of barley (at 0.75 mt per ha; 79,915 ha at 0.9 mt per ha); 200,000, with alternate-year fallow; 400,000 with 50/50 sharecropping. I shall not repeat my calculations to discuss Stephen Hodkinson’s lower alternative (36,000 Helot families), but note that it differs from his higher estimate by a reduction of 13% in any of our calculations. This reconstruction demands either much more fertile land than Lakônikê could offer or a very high and unrealistic output from sources other than annual crops.
This Helot population seems too high in absolute terms for the agricultural base of Lakônikê, but that is not the only difficulty with this interpretation. A large contributing factor to its unfeasibility lies in the conjecture of 50/50 sharecropping without recirculation of food. That arrangement draws off far more output from the Helots than does the hypothesis of a system of fixed rents and an exigent exaction of 50% of production. That is because the achievement of higher levels of productivity for individual units of grain-producing land does not yield benefits in the short or middle term to the Helot farmers. Moreover, the hypothesis of fixed-rent klêroi suggests a much more workable system when the number of Spartiates had fallen below the level of 8,000 or 9,000 (the scenario imposed by conventional reconstructions of Spartiate demography). On the assumption that each klêros-fragment was to pay its set share of the fixed rent, handed down from an earlier intact klêros, the demographic decline of the Spartiates removed the pressure of marking off new klêroi so that any increases in production were retained by the Helots for their own subsistence. In contrast, 50/50 sharecropping is immune to such amelioration, as it does not correlate with changes in Spartiate numbers.
Simple sharecropping on a 50/50 basis shares the drawbacks of all reconstructions that do not incorporate a cycling of foodstuffs back to the Helots, whether through the messes or by individual Spartiate dispensation. The Spartiate class was never large enough to consume all the food that would have been extracted from the klêroi through 50/50 sharecropping. Unless redistribution is posited, the entire surplus of rent left over after Spartiate consumption must be imagined to have exited a closed klêros system either to other groups within Lakônikê (e.g. the Perioeci) or to consumers outside Lakônikê. While there is no impediment to visualizing such flows {215|216} in principle, their scale—on the basis of estimates offered below for recirculation— would be surprising prima facie and because of the silence of our sources. In the absence of recirculation too few Helots are left behind to satisfy traditional estimates of their numbers and, more significantly, to account for the 35,000 males mentioned by Herodotus. Admitting the necessity for redistribution not only changes the reasoning by which the population of Helots is estimated, but it calls into doubt the supposition of a 50/50 sharecropping system. Recirculation mediated by individual Spartiates is in practice variable rate sharecropping.
Summation
After this welter of figures and estimates, one may well ask what firm basis is left for estimating the number of Spartan Helots. Our answer is that the work of the last several decades has marked out considerable common ground. Five important premises are these.
1. Estimates for the size of the ordinary Spartiate klêros seem to be settling around 18 hectares. [41] Accordingly, the klêroi were larger than average ordinary Greek farms or even than the estates for most non-Spartan hoplites elsewhere, but much smaller than the large establishments once envisaged by authorities such as Jarde and Kahrstedt.
2. The amount of arable land available for klêroi probably lay between 115,000 and 145,000 hectares, a range that is once again much lower than that prevailing in an important portion of the earlier scholarship (see the appendix for discussion of the agricultural land of Messenia).
3. Further investigation of our topic seems to generate new constraints for the Spartan agrarian system to operate within, such as limitations on available land or lower cereal yields. While it is possible to mitigate the impact of such parameters in our modeling of the problem (as the arguments above suggest), there has been no breakthrough in recent scholarship in the opposite direction, that is, on behalf of hypothesizing significant new sources for output for Spartan agriculture. One might concede that some latitude to our calculations is provided by hypothesizing Helots working on non-klêros land, but that is not only conjectural but also relatively modest in its impact. Therefore, any emerging impression envisages Spartan landholding as characterized by both a fragile equilibrium and a prodigality in its use of resources. {216|217}
4. High multiples of Helots to Spartiates seem increasingly improbable. While earlier scholars were comfortable with ratios of 10 to 1 or higher, the proportion of 7 to 1 implied by the mobilization for Plataia now seems a high upper limit. There remain, however, significant differences about the aggregates to which any such ratio may be applied.
5. Consequently, the very high numbers sometimes offered for the Spartan Helots that once appeared everywhere in accounts of Sparta appear more and more unlikely.
A new synthesis
What then is the likely range for the population of Spartan Helots? Let us consider a paradigmatic klêros c. 480–479 on the basis of our previous discussion. I emphasize that recirculation of food paid in rents back again to the Helots must be a factor in this economic context of low productivity. We shall use the amount of 0.9 mt of barley as the yield for each hectare. This is high, especially as it will be viewed as a “realizable output,” but it is a ceiling for Helot numbers that is being sought. Moreover, the scale of the rents might have motivated more intensive cultivation that would raise the disposable resources (Halstead 1987:85–86). If the rent in grain and wine were paid in Aiginetic measures, it would total 3,740 kg of barley (2880 of wheat equivalent). This rent and yield would fit a klêros of 17.4 hectares. [42] An amount equivalent to the rent is left for the Helots, but seed grain for rent and the residuum must be taken from it, let us say at 1:6, leaving 2506 kg of barley. This sum of barley will be considered equivalent to 1930 kg of wheat. [43] That would support 7.7 people at bare subsistence (250 kg of wheat equivalent per annum).
If we raise the total food consumption of these persons to 310 kg of wheat equivalent per year (to align with the common daily ration of one khoinix of wheat), we shall have to posit 462 kg of wheat equivalent as redistributed from the rent (or 16% of the rent in grain). Further, we might assign the Helots a total annual consumption of all goods of 360 kg of wheat equivalent in order to bring our Helots in line with nineteenth- and twentieth-century subsistence agriculturalists. We could do that by positing an additional 50 kg per capita of non-cereal production. That would give the Helots only a tiny output on the klêros outside of production of grain and grapes of 7%. {217|218}
Yet, there would have been in point of fact significant non-cereal production, including olive cultivation and animal husbandry (Hodkinson 2000:133–134, 151–152). [44] Moreover, in this reckoning the 7.7 Helots on each klêros would be generating 801 kg of wheat equivalent per capita per annum (although only consuming 360 kg of wheat equivalent). That level of productivity appears too high (cf. Clark and Haswell 1970:77–78). Accordingly, 7.7 Helots on each klêros or 69,300 for 9,000 klêroi is too few. By setting our parameters for production high, we have paradoxically given our cohort of Helots too high a bar of productivity to achieve. If we raise the amount of recirculated grain to 37% of the rents (1066 kg of wheat equivalent) our Helots for each klêros could be 9.7. And if we posited that 20% of Helot economic output was in goods other than grain and grapes, we might reach c. 13 Helots for each klêros. Our assumption would have to be that this additional output had only a slight effect on the size of our paradigmatic klêros through inter-cropping, domestic production, and exploitation of marginal land. These c. 13 Helots would be consuming 350 kg of wheat equivalent and producing c. 693 kg of wheat equivalent per capita. [45]
Nine thousand klêroi would give us 117,000 Helots. [46] The upper range of my crude estimation of 7–11 Helots per klêros provided 99,000 Helots. My earlier reconstruction of Messenian agriculture suggests 86,700 Helots in Lakônikê, without factoring in redistribution and non-cereal/grape production. Utilizing more pessimistic parameters for productivity in barley per hectare, or for the area available for cultivation, in Lakônikê will create much lower populations. For example, a cogent case can be constructed for a Helot population as low as 75,000. [47] Admittedly, it is hard to imagine that 30–35,000 members of such a population could have been marched off toward central Greece in 479 for the Plataia campaign. Yet it should be stressed that 75,000 Helots in toto is not at all an unreasonable number in isolation of the literary texts, considering ancient agricultural techniques, the available arable land in Laconia and Messenia, and the omnipotent fact of the whole issue, namely that a large Spartiate class had to be supported by rent without any input of its labor. {218|219}
Nonetheless, 9000 klêroi of 17.4 hectares would need 156,600 hectares, an amount that seems too large an area of cultivated territory compared to the best estimates. To bring down the amount of land required, we may well be forced to admit that income in kind flowed to the Helots from non-klêros land, as has been hypothesized above. That device might help to reduce the total needed below the hectares that appear a reasonable upper-limit for the klêros-land of Lakônikê. Alternatively, we could surmise some mixture of higher yields, higher work incentives outside the klêroi, and higher non-agricultural income to generate smaller average klêroi, enough to offset the 11,000 or more hectares that cannot be found on the ground.
An attraction of breaching the 100,000 barrier for Helot numbers is that it enables us to utilize the 35,000 Helots marching toward Plataia to postulate something significant about Spartan demography (cf. Hdt. 9.10.1, 28.2, 29.1). First, we must reject that this was a “military” muster in any sense, unlike the two-thirds levy that appears in Spartan mobilization contexts. [48] There were unlikely to have been rolls of the Helots preserved in any administrative archive. Secondly, we must reject that there were seven Helot families, or seven adult males on each klêros. Such numbers could not be fed in a period in which the adult male Spartiates numbered at least 8,000. Since Sparta had only the most primitive governmental apparatus, the best way to envisage the levy of Helots for Plataia is as a security measure. With the possibility of Persian raids on Lakônikê not entirely excluded, the mass of Helots could not be left behind as potential dissidents.
Herodotus did indeed repeatedly emphasize this proportion of seven Helots to each Spartiate. [49] Yet the force of his initial phrasing has not received proper recognition: {219|220} καὶ ἑπτὰ περὶ ἕκαστον [Spartiate] τάξαντες τῶν εἱλώτων (9.10.1). [50] This usage might well connote that each Spartiate was ordered to bring seven Helots with him, much as a financial levy might be apportioned, because other explanations of the participle τάξαντες here seem less satisfactory. [51] Herodotus believed the Spartiates were largely successful in fulfilling this requirement. We have no reason to doubt him, but also no warrant to assume the total of 35,000 is as “hard” a figure as the number of 5,000 for mobilized Spartiates, or that the total of 35,000 rests on anything independent of the tradition about the very terms of the muster.
To continue, let us posit exempli gratia that our 35,000 Helots were the males between 15 and 65 years of age, minus 10% or 3,500 left behind to maintain the klêroi, who were in turn assisted by boys, very old men, and women. The entire Helot population might then number 118,000, using a model life table (Coale and Demeny 1966, stable population, mortality level 4). Thus, our range for the Helot population in Lakônikê c. 480–479, when the Spartans were themselves close to their apogee in numbers, is 75,000 to 118,000, perhaps 3–5 times the number of the Spartiates.
Helot population in perspective
Other Peloponnesian poleis, such as Argos, Sikyon, and Epidauros, possessed groups with secondary civic statuses and dependent populations that approximated the Spartan pattern. [52] An important difference, however, was that the other exploitative systems benefited a sociopolitical elite, not the entire civic body. These dispositions gave way to more consolidated socio-political orders under various influences: hoplite military tactics, reactions to prevailing aristocratic leadership, the emergence of tyrannies, and the monetization of the local economy. In contrast, Sparta developed a unique type of civic or hoplitic dependent relations. An important factor in Sparta’s divergent development was certainly the scale of its institution of servile labor, both on the basis of Helot numbers and because of the large territory necessarily occupied by klêroi. The scale of Helotage moved agrarian dependency to the central point in the social structure. The conquest of Messenia shifted Laconian {220|221} dependency above a crucial threshold not otherwise met in the Peloponnesus. This conclusion would be reinforced, if we were indeed compelled to envisage (as suggested above) the Laconian klêroi as incrementally encompassing, during the late archaic period and early fifth century, less fertile or less contiguous land at the margins of the Helos plain, in the Eurotas valley, and on Cape Malea.
Therefore the status of Messenia and its population was at all times critically important for the status of Helotage. Some features of Messenian ecology were constants throughout the Spartan hegemony. First, Messenia was comparatively fertile. Second, the high density of Bronze Age settlement transformed the environment so dramatically that the very topography of later Messenia probably offered a ready-made template for human exploitation (even absent the possible reception of oral traditions about land resource utilization). In this light, the legacy of the massive cultivation of olive trees by the Mycenaeans is particularly relevant (McDonald and Coulson 1983:424–425; Zangger et al. 1997:592–593). Thirdly, Iron Age Messenia was relatively secure from extra-regional intervention behind mountainous terrain in its sea-girt corner of the Peloponnesus, until the Athenians mastered and put to use large-scale amphibious, flotilla naval warfare in the fifth century.
The conquest period
Sparta, as the chief political power in Laconia, benefited from the salient differences existing between that region and Messenia. Early archaic Laconia was relatively more densely populated. The polis structure was established early at Sparta, and hoplite warfare became a subject of precocious experimentation. While Laconian dependence may have had a complex aetiology, any eighth-century Messenian conquests were achieved through raiding that stimulated flight by some indigenous inhabitants and harried those remaining into surrendering a part of their output. Depopulation was always a powerful tool for archaic Spartan “imperialism,” since it removed the most resistant and engendered compliance in those remaining by allowing the acquiescent to divide a proportionately larger, accessible resource base. [53] The evidence from Nichoria seems to demonstrate Dark Age II (975–850) interaction with Laconia. [54] By Dark Age III (800–750), raiding from Laconia may already have reduced the size of {221|222} the community at Nichoria, while the boundary between the Dark Age III and Late Geometric (early archaic in a historical perspective) saw the termination of the village (McDonald and Coulson 1983:326–328; Coulson 1983b:332).
Down to the middle or late seventh century, Messenian communities were probably dependent on the Spartans globally, without individual klêroi, although the advantages of this dependency perhaps accrued disproportionately to the Spartan elite (as elsewhere in the early archaic period). I would date the principle of the one-half tribute burden to these earliest conquests. The annual declaration of war on the Helots by the Spartan ephors also derived from this period (Aris. fr. 543 Gigon with Plut. Lyc. 28.4), when the arrival of raiding parties offered a choice to their victims of tendering “gifts” or resistance. [55] The krypteia began its reinstitutionalization in this period. Given the primitive character of the governmental apparatus, the krypteia was an important administrative mechanism over Messenia. Clearly, this whole dispensation was fragile, not only because of the existence of non-compliant Messenians, but also because growth in Messenian numbers endangered the status quo. The poems of Tyrtaeus portray this dispensation under challenge.
The significantly greater task of subduing all Messenia could not, however, be consummated without the thorough implementation of a hoplite polity, a painful evolution also reflected in Tyrtaean poetry. This challenge brought to acute significance the issue of division of spoils. I would hypothesize that the distribution of shares in a colony became a model for the creation of the system of klêroi and syssitia involving the entire citizen body. That reorganization affected conquered, Helot-tilled land, but left the core of civic property in Laconia with its hierarchy of holdings intact. Under prevailing administrative and military conditions, incremental expansion was possible with or without reduction to servile status of the conquered. So extensive an incorporation, however, as all Messenia with all its population in situ, however, was probably infeasible, so that its pacification in the second half of the seventh century involved another depopulation. Once again, the relict population, whether it became Perioecic or was Helotized, was a partial beneficiary along with the Spartiates of any reduction in regional inhabitants. Non-compliant Messenians had an escape route in colonial Greece, where settlers were needed, a phenomenon clearly attested at Rhegion and much dramatized by the later Messeniaka (local histories written from the standpoint of liberated and [re]founded Messene). That alternative may have simplified Spartan policy, as it provided a viable alternative to death or subjugation for some Messenians. {222|223}
The consolidation period
The scale of the Spartan conquest decoupled a usage of hoplite tactics from the pressure for political inclusivity that was the rule elsewhere. Yet the system of klêroi, rents, and messes labored under massive disabilities in the form of disincentives to productivity and a lack of coordination between asset holding and managerial behavior. Its heyday was the later archaic period. At first, a low population put little pressure on agricultural resources. Isocrates reports a tradition that the Spartans originally numbered 2,000 when they occupied Laconia (12.255). Such a tradition cannot derive from a mythic Dorian invasion, but might rather transmit an archaic Spartan contention that an appropriate social dispensation, one true to mythological precedent, ought to include a civic body of this size.
If the early archaic Spartiates (prior to 600) did indeed at some point number 2,000, that tradition would help substantiate the conjecture, reasonable in its own terms, that Spartiate and Helot numbers grew in conjunction during the sixth century (contrast Figueira 1986:181–182). Any rigidities within procedures for assigning klêroi or tensions over their transmission were mitigated by availability of unoccupied land for distribution. The powerful inducement of allowing young Helot males to marry and to form households early promoted the formation of new klêroi and lowered restiveness among the dependent population. Internal colonization mixed the Helots, creating the single identity that is attested in the fifth century (Figueira 1999:223–225). During this period, the demographic morale of the Spartans is clear in their willingness to suffer casualties in the duel of champions with the Argives, to undertake a distant expedition to Naxos and Samos, and to send out colonists under Dorieus (see Figueira 1986:170–175).
The early fifth century
By the period of 479 and thereafter, the number of klêroi and thereby Spartiates was reaching the carrying capacity of Lakônikê. The received figure of 9,000 for the number of klêroi, rather than being a design feature from an omniscient legislator, is merely a reflection in tradition of a detail from the system at its height. If my discussion of the land and agricultural output is broadly correct, the 5,000 Spartiates who marched off to Plataia would imply a total male population of a little over 10,000, the number offered by Aristotle as a traditional maximum (Pol. 1270a36–38). Spartan numbers may indeed have risen further in the 14 years between Plataia and the Great Earthquake. It might have been becoming more difficult simply to assign klêroi to every Spartiate at will, although there could have been land held in reserve in the Thyreatis and other regions of ongoing occupation for this purpose (Figueira 1993:299–308). Toward the end of this period, Pausanias {223|224} began his intrigues with the Helots, which I interpret not only as a reaction to the emergent military challenge from the Athenians, but also as an attempt at a redefinition of social mobility. [56] Assuredly, any confrontation with Athens required many triremes manned by Perioecic and Helot rowers.
The later fifth and early fourth centuries
The Great Earthquake of 465 changed the socio-economic order of Lakônikê dramatically, because it lowered the number of Spartiates significantly (Table 8.1). [57] It struck Sparta and Laconia more directly than Messenia, so that the proportionate reduction in the number of Spartiates was probably much greater than any immediate fall in numbers of the Helots. Secondarily, the earthquake led to a great Helot revolt that killed many Spartiates and subtracted from the Helot workforce. The difficult in subduing the Helot rebels short of reaching terms for their withdrawal indicates that thousands of Helot fighters remained under arms to the end. The emigration of these and their dependents marked still another of the depopulations that characterized the history of Messenia during the Spartan hegemony. In principle, fewer Spartiates meant more numerous and more affluent Helots. Not only was the pool of resources for recirculation increased, but the marginal costs of policing Helot adherence to the norms of dependency (including scrupulous rent payment) were also heightened, so that concealment of production became more viable. The earthquake and revolt also meant more economic differentiation among the Spartiates. The ensuing period appears to have opened an era of more intense social competition among non-aristocratic Spartiates. [58] It was also a period when the Messenian civic identity was constructed, both within and without Messenia, although a substratum of earlier rural populism doubtless underlay the later ideological structures. [59] {224|225}
Thus Lakônikê emerged from the late 450s with a lower population owing to the earthquake, the rebellion, and emigration. There was, however, a differential effect on various social components: the Spartiates were most depleted through catastrophic and combatant mortality; the Helots were next affected through hostilities, flight, and defection, and the Perioeci perhaps experienced the least demographic impact. [60] Yet even they were reduced through the earthquake affecting the nearest to Sparta of their towns, through the defection of two of their towns, Aithaia and Thouria, and through military casualties. The relative change in the balance of demographic components seems to have elicited from the Spartans bipolar responses. The Helot population was recognized as a potential political asset. I have suggested that the Helots rewarded by the Spartans, as reported in Thuc. 4.80.3–4, had earned their recognition in service during the hostilities of 465–446 before the Thirty Years Peace. These manumissions were followed later in the century by the Brasideioi, by the Neodamodeis (Willetts 1954 for references), and by the Helots who were mobilized to serve as rowers on Spartan warships (cf. Myron FGrHist 106 F 1; Xen. Hell. 7.1.12; cf. 5.1.11, 16, 24). Military service, however, also had the effect for Sparta of relieving it of many members of a repressed and, accordingly, distrusted social stratum.
Therefore, regarding the Helots as a demographic resource for warfare was counterbalanced by the fear of the Helots, so eloquently discussed by Thucydides. In the aftermath of the Athenian and Messenian raids from Pylos, he comments upon Spartan counteractions with these remarks (4.80.2–4):
Here a key phrase, τὴν νεότητα, which ought to mean ‘males of an age for military service’, is not textually certain, but likely in interpretation (cf. Hdt. 9.12.2 with {225|226} Figueira 1984:101–102 n. 47). [61] After the losses of Helots to warfare, flight, and Spartan assassination, a new generation of Helots had reached manhood during the Thirty Years Peace and the first years of the Archidamian War. Not only was the global plethos ‘multitude’ of the Helots disturbing, but a “baby boom” of potential rebels was also especially threatening.
Moreover, to believe Thucydides, the here-mentioned liberated Helot benefactors, who had been singled out for honors for their earlier services, had been subjected to an effective covert attrition at the hands of the Spartiates. In the context of this period of trial after the disaster at Pylos, however, Sparta’s need for its Brasideioi, freed Helots enlisted to accompany Brasidas to Thrace, and later the Neodamodeis was too acute to permit any systematic mistreatment, however secret. Nonetheless, the Spartans did exhibit a willingness to keep soldiers of Helot extraction occupied on distant campaigns (in accordance with the sentiments attributed to them here) and, at least initially, were uncertain where they, with their families, were to be accommodated.
Not only was the proportionately larger Helot population of the period 455–371 paradoxically both an asset and a liability simultaneously, it was also both superfluous and insufficient at the same time. We have just noted the ability of the Spartan government in these years to redress the imbalance between free and servile Laconians by freeing and elevating favored Helots to the status of Perioeci. Yet, it is also clear that there was a significant reduction in the number of homoioi in the 420s and 410s, owing not only to losses on campaign, but also to their incapacity to tender their mess dues. While some of this shortfall in income from the klêroi is to be attributed to direct damage from raids, whether by Attic flotillas or from forays mounted from Pylos and Cape Malea, the flight of Helot workers was a more decisive factor. That is why at several junctures Thucydides emphasizes Helot flight in his descriptions of the Spartan predicament after the debacle on Sphakteria (4.41.3; 5.14.3, 35.7; cf. 4.56.1–2; 56.2–3).
Just as the earthquake triggered intensified social competition among Spartiates, the vicissitudes of the war seem to have created a competition over Helot labor, in which Spartiates lost workers to flight and conscription differentially. The winners moved upward socially; the losers became hypomeiones, ‘Inferiors’. Let me note one scenario to aid in accounting for this phenomenon. We have supposed that some Messenian klêroi had been gradually established in the sixth century through internal colonization. These new klêroi benefited persons with less access {226|227} to other assets like private property, such as younger sons. These lineages were the least able to endure economic vicissitudes in the late fifth century, when Messenia was more vulnerable to Attic attack and Helot flight (Figueira 1986:203)
The result for Lakônikê of the Spartan demographic crisis of the late fifth and early fourth centuries was still another depopulation. Xenophon describes Lakônikê in general as typified by erêmia ‘depopulation’ (Hell. 6.5.23, 25). Messenia had a low population even after the refoundation of Messene as a polis and the repatriation of thousands who claimed Messenian extraction. At the summons of Epameinondas, these claimants had gathered from Italy, Sicily, and Libya (Paus. 4.26.3, cf. 27.2; Plut. Ages. 34.1, Pelop. 24.9; cf. Dipersia 1974). The Sicilian contingent clearly drew on the refugee community at Tyndaris that included many naturalized “Messenians” (Diod. 14.78.5). Volunteer settlers were also enlisted to constitute πολλοὺς … οἰκήτορας ‘many … settlers’ (Diod. 15.66.1). The impressive fortifications of refounded Messene surrounded ample open spaces, intended to be built upon in a comprehensive urban program (cf. Themelis 1993:36–37, 1994; Morizot 1994), and also protected agricultural land (Alcock 1998:180–182). Furthermore, the palynological evidence is instructive. Cultivation of olive trees in the area of Pylos only reached its post-Helladic apogee after Messene had been refounded (Zangger et al. 1997:594; Yazvenko 1998). The new community was militarily weak, dependent on Arcadian and Boiotian aid against even an enfeebled Sparta. In conjunction, these circumstances indicate that there were no longer all that many Helots living in pre-liberation Messenia (see also Luraghi, this volume, pp. 121–4). The demographic state of the region mirrored the fate of the Spartiate class, although the Perioeci and Helots in Messenia experienced a less dramatic decline.
The interdependence of Helot and Spartiate conjoined the demographic fates of the two groups in a way different from that of the classes within other classical poleis. The more we probe Sparta’s version of the “peculiar institution” of douleia, the more that we seem to come to appreciate these departures. To note just two of those which have emerged from the foregoing sketch, I would first choose the way in which successive, reconfigured Spartan social and political orders were punctuated by episodes of depopulation (at least, in Messenia), and, second, how the formulaic character of material interaction between Spartiates and Helots begins in sharp definition, but ends in an indeterminacy in which the same population can be at once too many and too few, an asset and a liability. [62] {227|228}
Appendix
Using the material collected in MME 1, NSS 1978, and various other topographical aids, I attempted to assign all the communities of the modern Nomos of Messenia to the categories of klêros-land and non-klêros-land. The vast majority of the latter category is composed of the territory of Perioecic communities, although some territory in the valley of the Neda or lying to its north or northeast probably lay completely outside Lakônikê. Such a classification labors under that customary disability of investigation of the ancient rural economy. Even though we possess extensive survey work on Messenia, attested or hypothetical ancient sites fall far short of demonstrating a cultivation of the territory of many modern communes. I shall not claim that this classification represents the final word on the subject, hoping that scholars with active field experience with Messenian topography will offer corrections. The census of 1971 states the extent of all the enumerated agricultural holdings in Messenia to have been c. 119,977 ha. It is hardly surprising, granted the movement from agricultural labor to other modes of production, that this total is reduced from 1951. My interest was less with the total of arable land in 1971, but in assigning it to, or withholding it from, the klêroi. This distribution is 72% klêros-land and 28% non-klêros-land. The strongest reservations in our analysis concerned communities in southeast Messenia (marked with an asterisk below) that were only doubtfully perioecic (cf. pp. 134–6 with ns. 21–6). This adjustment might add c. 2% to the amount of klêros-land.
Inasmuch as modern Messenia is characterized by a commercial agricultural economy that includes significant viticulture and olive cultivation, a determination of the extent of land producing annual crops is not likely to be indicative of ancient conditions (c. 31,634 ha). Because non-klêros-land is for the most part more elevated, less fertile, and more marginal, proportionately more of it is devoted to grain cultivation (for example) than vineyards and olive groves in contemporary Messenia. If one considers the proportions of the klêros-land and non-klêros-land as suggested by acreage of annual crops, the division becomes 67% and 33%. This 67% may represent a lower limit for the proportion of klêros-land in ancient Messenia. {228|229}
Alcock, S. E. 1998. “Liberation and Conquest: Hellenistic and Roman Messenia.” In Davis 1998:179–191.
———. 1999. “The Pseudo-History of Messenia Unplugged.” Transactions of the American Philological Association 129:333–341.
———. 2002. “A Simple Case of Exploitation? The Helots of Messenia.” In Money, Labour and Land: Approaches to the Economies of Ancient Greece, ed. P. Cartledge, E. E. Cohen, and L. Foxhall (eds.), 185–199. London.
Baltrusch, E. 2001. “Mythos oder Wirklichkeit? Die Helotengefahr und der Peloponnesische Bund.” Historische Zeitschrift 272:1–24.
Beloch, J. 1886. Die Bevölkerung der Griechisch-Römischen Welt. Leipzig.
Bockisch, G. 1985. “Die Helotisierung der Messenier: Ein Interpretationsversuch zu Pausanias IV 14, 4f.” In Antike Abhängigkeitsformen in den griechischen Gebieten ohne Polisstruktur und den rämischen Provinzen, ed. H. Kreissig and F. Kuhnert, 29–48. Berlin.
Bolte, F. 1929. “Sparta: C. Geographie.” Realenzyklopädie der classischen Altertumswissenschaft. 2.3:1294–1373.
Bradeen, D. W. 1964. “Athenian Casualty Lists.” Hesperia 33:16–62.
———. 1969. “The Athenian Casualty Lists.” Classical Quarterly 19:145–169.
———. 1974. The Athenian Agora. Vol. XVII, Inscriptions: The Funerary Monuments. Princeton.
Buckler, J. 1977. “Land and Money in the Spartan Economy—A Hypothesis.” Research in Economic History 2:249–279.
Burford Cooper, A. 1977–78. “The Family Farm in Ancient Greece.” Classical Journal 73:163–175.
Canfora, L. 1982. “Il soggetto passivo della polis classica.” Opus 1:33–48.
Carothers, J., and W. A. McDonald. 1979. “Size and Distribution of the Population in Late Bronze Age Messenia: Some Statistical Approaches.” Journal of Field Archaeology 6:433–454.
Cartledge, P. 1987. Agesilaos and the Crisis of Sparta. London.
———. 1991. “Richard Talbert’s Revision of the Spartan-Helot Struggle: A Reply.” Historia 40:379–381.
Catling, R. W. V. 1996. “The Archaic and Classical Pottery.” In Cavanagh et al. 1996:33–89.
———. 2002. “The Survey Area from the Early Iron Age to the Classical Period (c. 1050–c. 300 BC).” In Cavanagh et al. 2002:151–256.
Cavaignac, E. 1912. “La population du Peloponnese aux Ve et IVe siecles.” Klio 12:261–280.
Cavanagh, W., J. Crouwel, R. W. V. Catling and G. Shipley. 1996. The Laconia Survey: Continuity and Change in a Greek Rural Landscape. Vol. 2, Archaeological Data. Annual of the British School at Athens, Supplementary Volume 27. London.
———. 2002. The Laconia Survey: Continuity and Change in a Greek Rural Landscape. Vol. 1, Methodology and Interpretation. London.
Christien, J. 1982–83. “La Laconie orientale: topographie antique et problèmes historiques.” Peloponnesiaka 9:58–76.
———. 1989a. “Promenades en Laconie.” Dialogues d’histoire ancienne 15:75–105.
———. 1989b. “Les liaisons entre Sparte et son territoire malgré l’encadrement montagneux.” In Montagnes, Fleuves, Forets dans l’histoire: Barrieres ou lignes de convergence? ed. J.-F. Bergier, 17—44. St. Katharinen.
———. 1992a. “L’étranger à Sparte.” In L’etranger dans le mondegrec, ed. R. Lonis, 147–167. Nancy.
———. 1992b. “De Sparte à la côte orientale du Péloponnèse.” In Polydipsion Argos: Argos de la fin des palais myceniens à la constitution de l’État classique. Bulletin de correspondance hellénique, Supplément 22, ed. M. Pierart, 157–171. Athens.
———. 1998. “Sparte et le Péloponnèse après 369 B.C.” In Πρακτικὰ τοῦ Ε᾽ διέθνους Πελοποννησιακῶν σπουδῶν (῎Αργοσ-Ναύπλιον 1995). 433–467. Athens.
Coale, A. J., and P. Demeny. 1966. Regional Model Life Tables and Stable Populations. Princeton.
Coleman-Norton, P. R., ed. 1941. The Greek Political Experience: Studies in Honour of William Kelly Prentice. Princeton.
Cornelius, F. 1973. “Pausanias.” Historia 22:502-504.
Coulson, W. D. E. 1983a. “The Pottery.” In McDonald, Coulson, and Rosser 1983:61–259.
———. 1983b. “The Site and Environs.” In McDonald, Coulson, and Rosser 1983:332–339.
Davis, J. L., ed. 1998. Sandy Pylos: An Archaeological History from Nestor to Navarino. Austin.
Davis, J. L., S. E. Alcock, J. Bennet, Y. G. Lolos, and C. W. Shelmerdine. 1997. “The Pylos Regional Archaeological Project, Part I: Overview and the Archaeological Survey.” Hesperia 66:391–494.
Dipersia, G. 1974. “La nuova popolazione di Messene al tempo di Epaminonda.” In Propaganda epersuasione occulta nell’antichità (Contributi dell’istituto di storia antica 2), ed. M. Sordi, 54–61. Milan.
Ducat, J. 1983. “Le citoyen et le sol à Sparte à l’époque classique.” In Hommages à Maurice Bordes.Annales de la Faculté des lettres et sciences humaines de Nice 45:143–166.
———. 1990. Les Hilotes. Bulletin de correspondance hellénique, Supplément 20. Athens.
Dusanic, S. 1997. “Platon, la question messénienne et les guerres contre les Barbares.” In Esclavage, guerre, économie en Grèce ancienne: Hommages à Yvon Garlan, ed. P. Brule and J. Oulhen, 75–86. Rennes.
Ehrenberg, V. 1960. The Greek State. Oxford.
Figueira, T. J. 1981. Aegina. New York.
———. 1984. “Mess Contributions and Subsistence at Sparta.” Transactions of the American Philological Association 114:84–109.
———. 1986. “Population Patterns in Late Archaic and Classical Sparta.” Transactions of the American Philological Association 116:165–213.
———. 1991. Athens and Aigina in the Age of Imperial Colonization. Baltimore.
———. 1993. Excursions in Epichoric History: Aiginetan Essays. Lanham.
———. 1999. “The Evolution of the Messenian Identity.” In Sparta: New Perspectives, ed. S. Hodkinson and A. Powell, 211–244. London.
———. 2002. “Iron Money and the Ideology of Consumption in Laconia.” In Sparta: Beyond the Mirage, ed. S. Hodkinson and A. Powell, 137–170. London.
———. Forthcoming. “The Nature of the Spartan Klêros.” In Sparta: Structures, Ideologies, and Personae. London.
Finley, M. I. 1959. “Was Greek Civilization Based on Slave Labour?” Historia 8:145–164.
Fogel, R. W., and S. L. Engerman. 1974. Time on the Cross: The Economics of American Negro Slavery. Boston.
Foxhall, L., and H. A. Forbes. 1982. “Sitometreia: The Role of Grain as a Staple Food in Classical Antiquity.” Chiron 12:41–90.
Gallant, T. W. 1991. Risk and Survival in Ancient Greece. Stanford.
Gallo, L. 1984. Alimentazione e demografia della Grecia antica. Salerno.
———. 1990. “Beloch e la demografia antica.” In Aspetti della storiografia di Giulio Beloch, ed. L. Polverini, 113–168. Naples.
Garnsey, P. 1988. Famine and Food Supply in the Graeco-Roman World. Cambridge.
———. 1992. “The Yield of the Land in Ancient Greece.” In Agriculture in Ancient Greece, ed. B. Wells, 147–153. Stockholm. Repr. 1998 in Cities, Peasants and Food in Classical Antiquity. 201–213.Cambridge.
———. 1999. Food and Society in Classical Antiquity. Cambridge.
Golden, M. 1987. Review of Hansen 1986. Phoenix 41:443–46.
———. 2000. “A Decade of Demography. Recent Trends in the Study of Greek and Roman Populations.” In Polis & Politics: Studies in Ancient Greek History Presented to M. H. Hansen, ed. P. Flensted-Jensen, T. H. Nielsen, and L. Rubinstein, 23–40. Copenhagen.
Grundy, G. B. 1908. “The Population and Policy of Sparta in the Fifth Century.” Journal of Hellenic Studies 28:77–96.
Guiraud, P. 1893. La propriété foncière en Grèce. Paris.
Halstead, P. 1987. “Traditional and Ancient Rural Economy in Mediterranean Europe: Plus Ça Change?” Journal of Hellenic Studies 107:77–87.
Halstead, P., and G. Jones. 1989. “Agrarian Ecology in the Greek Islands: Time Stress, Scale and Risk.” Journal of Hellenic Studies 109:41–55.
Hansen, M. H. 1986. Demography and Democracy: The Number of Athenian Citizens in the Fourth Century B.C. Herning.
———. 1988. “Demography and Democracy Once Again.” Zeitschrft für Papyrologie und Epigraphik 75:189–193.
———. 1989. “Demography and Democracy—A Reply to Eberhard Ruschenbusch.” Ancient History Bulletin 3:40–44.
———. 1994. “The Number of Athenian Citizens secundum Sekunda.” Echos du monde classique 13:299–310.
Harrison, A. B., and N. Spencer. 1998. “After the Palace: The Early ‘History’ of Messenia.” In Davis 1998:147–162.
HCT = A. W. Gomme, A. Andrewes, and K. J. Dover. 1945–1981. A Historical Commentary on Thucydides. Oxford.
Hodkinson, S. 1992. “Sharecropping and Sparta’s Exploitation of the Helots.” In Φιλολά: Lakonian Studies in Honor of Hector Catling, ed. J. M. Sanders, 123–134. London.
———. 2000. Property and Wealth in Classical Sparta. London.
Hornblower, S. 1996. A Commentary on Thucydides. Vol. 2, Books IV–V.24. Oxford.
Hunt, P. 1997. “Helots at the Battle of Plataea.” Historia 46:129–144.
Isager, S., and J. E. Skydsgaard. 1992. Ancient Greek Agriculture: An Introduction. London and New York.
Jameson, M. H. 1992. “Agricultural Labor in Ancient Greece.” In Agriculture in Ancient Greece, ed. B. Wells, 135–146. Stockholm.
Jarde, A. 1925. Les céréales dans l’antiquité grecque. Vol. 1, La production. Paris.
Kahrstedt, U. 1919 “Die spartanische Agrarwirtschaft.” Hermes 54:279–294.
Legrand, P.-E. 1954. Hérodote. Histoires: Livre IX. Paris.
Link, S. Forthcoming. “Snatching and Keeping: The Motif of Taking in Spartan Culture.” In Sparta: Structures, Ideologies, and Personae, ed. T. J. Figueira. London.
Lotze, D. 1959. Μεταξὺ ἐλευθέρων καὶ δούλων. Studien zur Rechtsstellung unfreier Landbevölkerungen in Griechenland bis zum 4. Jahrhundert v. Chr. Berlin.
———. 1971. “Zu einigen Aspekten der spartanischen Agrarsystem.” Jahrbuch für Wirtschaftsgeschichte 2:63–76.
———. 1993–94. “Burger zweiter Klasse: Spartas Periöken: Ihre Stellung und Funktion im Staat der Lakedaimonier.” Akademie gemeinnütziger Wissenschaften zu Erfurt: Sitzungsberichte der Geisteswissenschaftlichen Klasse 2:37–51.
Lukermann, F. E., and J. Moody. “Nichoria and Vicinity: Settlements and Circulation.” In Rapp and Aschenbrenner 1978:78–107.
Luraghi, N. 2001. “Der Erdbebenaufstand und die Entstehung der messenischen Identität.” In Gab es das Griechische Wunder? Griechenland zwischen dem Ende des 6. und der Mitte des 5. Jahrhunderts v. Chr., ed. Papenfufi and V.M. Strocka, 279–301. Mainz.
———. 2002a. “Helotic Slavery Reconsidered.” In Sparta: Beyond the Mirage, ed. A. Powell and S. Hodkinson, 227–248. London.
———. 2002b. “Becoming Messenian.” Journal of Hellenic Studies 122:45–69.
Magnetto, A. 1994. “L’intervento di Filippo II nel Peloponneso e l’iscrizione Syll.3 665.” In Ἱστορίη: Studi offerti dagli allievi a Giuseppe Nenci in occasione delsuo settantesimo compleanno, ed. S. Alessandrì, 283–308. Lecce.
Manso, J. C. T. 1800–1805. Sparta. 3 vols. Leipzig.
McDonald, R. A. 1993. “Independent Economic Production by Slaves on Antebellum Louisiana Sugar Plantations.” In Cultivation and Culture: Labor and the Shaping of Slave Life in the Americas, ed. I. Berlin and P. D. Morgan, 275–299. Charlottesville.
McDonald, W. A., and W. D. E. Coulson. 1983. “The Dark Age at Nichoria.” In McDonald, Coulson, and Rosser 1983:316–329.
McDonald, W. A., and G. R. Rapp, eds. 1972. The Minnesota Messenia Expedition: Recontructing a Bronze Age Environment. Minneapolis.
McDonald, W. A., W. D. E. Coulson, and J. Rosser. 1983. Excavations at Nichoria. Vol. 3, Dark Age and Byzantine Occupation. Minneapolis.
Morizot, Y. 1994. “Le hiéron de Messéné.” Bulletin de correspondance hellénique 118:399–405.
Müller, C. O. 1839. The History and Antiquities of the Doric Race. 2nd ed. London.
Niese, B. 1906. “Neue Beiträge zur Geschichte und Landeskunde Lakedämons.” Nachrichten von der königlichen Gesellschaft der Wissenschaften zu Göttingen, Phil.-Hist. Kl.:101–142.
NSS = National Statistical Service of Greece (= Ἐθνικὴ στατιστικὴ ὑπερηρεσία τῆς ῾Ελλάδος) Results of the Agricultural and Livestock Census of March 14, 1971 (= Ἀποτελέσματα ἀπογραφῆς γεωργίας καὶ κτηνοτροφίας τῆς Ι4ης Μαρτίου 1971). Volume 1. Athens 1978.
Oliva, P. 1971. Sparta and Her Social Problems. Amsterdam.
Papazoglou, F. 1993. “Sur le charactère communitaire de la propriété du sol et de l’hilotie à Sparte: À propos d’une thèse de J. Ducat.” Ziva Antika 43:31–46.
Powell, J. E. 1977. A Lexicon to Herodotus. 2nd ed. Hildesheim.
Rapp, G., and S. E. Aschenbrenner. 1978. Excavations at Nichoria. Vol. 1, Site, Environs, and Techniques. Minneapolis.
Rhodes, P. J. 1980. “Ephebi, Bouleutae and the Population of Athens.” Zeitschrift für Papyrologie und Epigraphik 38:191–201.
———. 1981. “More Members Serving Twice in the Athenian Boule.” Zeitschrift für Papyrologie und Epigraphik 41:101–102.
———. 1984. “Members Serving Twice in the Athenian Boule and the Population of Athens Again.” Zeitschrift für Papyrologie und Epigraphik 57:200–202.
Roebuck, C. A. 1945. “A Note on Messenian Economy and Population.” Classical Philology 40:149–165.
Roobaert, A. 1977. “Le danger hilote?” Ktema 2:141–155.
Ruschenbusch, E. 1979. “Der soziale Herkunft der Epheben um 330.” Zeitschrift für Papyrologie und Epigraphik 35:173–176.
———. 1981a. “Epheben, Buleuten und die Bürgerzahl von Athen um 330 v.Chr.” Zeitschrift für Papyrologie und Epigraphik 41:103–105.
———. 1981b. “Noch einmal die Bürgerzahl Athens um 330 v.Chr.” Zeitschrift für Papyrologie und Epigraphik 44:110–112.
———. 1982. “Die Diaitetenliste IG II/III2 1927. Zugleich ein Beitrag zur sozialen Herkunft der Schiedsrichter und zur Demographie Athens.” Zeitschrift für Papyrologie und Epigraphik 49:267–281.
———. 1983. “Ein weiteres unbeachtetes Zeugnis zur Bürgerzahl Athens im 4. Jh.v.Chr.”Zeitschrift für Papyrologie und Epigraphik 50:202.
———. 1984a. “Die Diaiteteninschrift vom Jahre 371 v.Chr. IG II2 143 frg. a, b, c, d und Hesperia 7 (1938) 278ff. Nr.13 (= H).” Zeitschrift für Papyrologie und Epigraphik 54:247–252.
———. 1984b. “Zum letzten Mal: Die Bürgerzahl Athens im 4. Jh.v.Chr.” Zeitschrift für Papyrologie und Epigraphik 54:253–270.
———. 1985. “Die Sozialstruktur der Bürgerschaft Athens im 4. Jh.v.Chr.” Zeitschrift für Papyrologie und Epigraphik 59:249–251.
———. 1988a. “Demography and Democracy. Doch noch einmal die Bürgerzahl Athens im 4. Jh.v.Chr.” Zeitschrift für Papyrologie und Epigraphik 72:139–140.
———. 1988b. “Getreideerträge in Griechenland in der Zeit von 1921 bis 1938 n. Chr. Als Maßstab für die Antike.” Zeitschrift für Papyrologie und Epigraphik 72:141–153.
———. 1988c. “Stellungnahme.” Zeitschrift für Papyrologie und Epigraphik 75:194–198.
Sallares, R. 1991. The Ecology of the Ancient Greek World. London.
Sanders, G. D. R. 1984. “Reassessing Ancient Populations.” Annual of the British School at Athens 79:251–262.
Scheidel, W. 2001. “Progress and Problems in Ancient Demography.” In Debating Roman Demography, ed. W. Scheidel, 1–81. Leiden.
Sekunda, N. 1992. “Athenian Demographic and Military Strength 338–322 B.C.” Annual of the British School at Athens 87:311–355.
Shipley, G. 1992. “Perioikos: The Discovery of Classical Lakonia.” In Φιλολάκων: Lakonian Studies in Honour of Hector Catling., ed. J.M. Sanders, 211–226. London.
———. 1996a. “Archaeological Sites in Laconia and Thyreatis.” In Cavanagh et al. 1996:235–313.
———. 1996b. “Site Catalogue of the Survey.” In Cavanagh et al. 1996:315–438.
———. 1997. “‘The Other Lacedaimonians’: The Dependent Perioikic Poleis of Laconia and Messenia.” In The Polis as an Urban Centre and as a Political Community. (Acts of the Copenhagen Polis Centre 4), ed. M. H. Hansen, 189–281. Copenhagen.
———. 2000. “The Extent of Spartan Territory in the Late Classical and Hellenistic Periods.” Annual of the British School at Athens 95:367–390.
Singor, H. W. 1993. “Spartan Land Lots and Helot Rents.” In De agricultura: In memoriam Pieter Willem de Neeve, ed. H. Sancisi-Weerdenburg et al., 31–60. Amsterdam.
Sloan, R. E., and M. A. Duncan. 1978. “Zooarchaeology of Nichoria.” In Rapp and Aschenbrenner 1978:60–77.
Spencer, N. 1998. “Nichoria: An Early Iron Age Village in Messenia.” In Davis 1998:167–170.
Stein, J., and G. Rapp. 1978. “Archaeological Geology of the Site.” In Rapp and Aschenbrenner 1978:234–257.
Stroud, R. S. 1998. The Athenian Grain-Tax Law of 374/3 B.C. (Hesperia, Supplement 29). Princeton.
Talbert, R. J. 1989. “The Role of the Helots in the Class Struggle at Sparta.” Historia 38:22–40.
Themelis, P. G. 1993. “Damophon of Messene: New Evidence.” In Archaeology in the Peloponnese, ed. K. A. Sheedy, 1–37. Oxford.
———. 1994. “Artemis Ortheia at Messene: The Epigraphical and Archaeological Evidence.” In Ancient Cult Practice from Epigraphical Evidence, ed. R. Hagg, 101–122. Stockholm.
Thompson, W. E. 1987. “The Corruption of Thucydides 4.80.3.” Ancient History Bulletin 1:61–64.
Valzania, S. 1996. “L’esercito Spartano nel periodo dell’egemonia: dimensioni e compiti strategici” Quaderni di storia 43:19–69.
van Wees, H. 2000. “The Myth of the Middle-class Army: Military and Social Status in Ancient Athens.” In War as a Cultural and Social Force: Essays on Warfare in Antiquity, ed. T. Bekker-Nielsen and L. Hannestad, Copenhagen:45–71.
Wagstaff, J. M. 1982. The Development of Rural Settlements. A Study of the Helos Plain in Southern Greece. Amersham.
Wagstaff, M., S. Augustson, and C. Gamble. 1982. “Alternative Subsistence Strategies.” In An Island Polity: An Archaeology of Exploitation in Melos, ed. C. Renfrew and M. Wagstaff, 172–180. Cambridge.
Westermann, W.L. 1941. “Athenaeus and the Slaves of Athens.” Athenian Studies Presented to W.S. Ferguson. (Harvard Studies in Classical Philology, Supplement 1). 451–70. Cambridge, MA.
Willetts, R. F. 1954. “The Neodamodeis.” Classical Philology 49:27–32
Yazvenko, S. 1998. “From Pollen to Plants.” In Davis 1998:14–20.
Zangger, E., M. E. Timpson, S. B. Yazvenko, F. Kuhnke, and J. Knauss. 1997. “The Pylos Regional Archaeological Project. Part II: Landscape Evolution and Site Preservation.” Hesperia 66:549–641.
Footnotes
[ back ] 1. Note Hansen 1988, but with the observations of Golden 1987.
[ back ] 2. For recent overviews on ancient population studies, see Golden 2000; Scheidel 2001.
[ back ] 5. Valzania 1996 does offer a type of contestation, accompanied by a limited citation of scholarship and an unusual selection of the ancient evidence in order to dismiss some attested figures. He specifically denies the sharp declines after the earthquake and between Mantineia and Leuktra, opting for a constant strength of 2,000 Spartiates! Cf. Valzania 1996:34, 45, 47.
[ back ] 6. Talbert 1989 is particularly wrong-headed, especially in its dismissal of Helot resistance during the Peloponnesian War (cf. Figueira 1986:196–198, 1999; Cartledge 1991 and this volume, pp. 21–3). Cf. also Roobaert 1977; Baltrusch 2001.
[ back ] 7. Note Jameson 1992:138–140. There is no reason to believe that Thucydides was referring to the density of slave population measured against the free population (cf. Finley 1959:151, 163; HCT 5.86–87). That interpretation is not only unsupported by the text, but would also be an unusual way for a fifth-century observer to view the issue. It does remain uncertain whether the Chians had more slaves than the Athenians when Attic slave numbers peaked during the Pentakontaeteia. By winter 412/11, however, Athenian slave numbers were reduced through impoverishment, defection, military action, and manumission (cf. Thuc. 7.27.5).
[ back ] 8. Oliva 1971:53n3 has a helpful conspectus of authorities.
[ back ] 9. Figueira (forthcoming) will speak to this complex of problems.
[ back ] 10. Earlier examples of this type of analysis are more useful on other features of the Laconian economy. Note Kahrstedt 1919; Lotze 1971; Buckler 1977.
[ back ] 11. See the commentary on Paus. 4.14.4–15 of van Wees (this volume, pp. 36–7) and Luraghi 2002a:235–236.
[ back ] 12. Moralia 239E (Instituta Laconica 41): οἱ δὲ εἵλωτες αὐτοῖς εἰργάξοντο τὴν γὴν ἀποφορὰν τὴν ἄνωθεν ἱσταμένην τελοῦντες. ἐπάρατον δ᾽ ἦν πλείονός τινα μισθῶσαι, ἵν᾽ ἐκεῖνοι μὲν κερδαίνοντες ἡδέως ὑπηρετῶσιν, αὐτοὶ δὲ μὴ πλέον ἐπιξητῶσιν. Note also here, however, the defensive and propagandistic tendency.
[ back ] 13. The incremental settlement of the area of the Laconia Survey which started after 600 (especially regarding the south-eastern sector) would be consistent with this practice. See Catling 2002:157–161, 224–225; cf. pp. 233–237.
[ back ] 14. As many kleroi experienced later fission and consolidation in the fifth and fourth centuries, these underlying assignments of various sub-rents to portions of the kleroi may have demarcated Helot responsibilities to klêros-holders. Cf. Hodkinson 2000:126.
[ back ] 16. Appalachian farmers often arrogated the right to convert grain into more easily transported whiskey in defiance of the U.S. tax law, partly in order to overcome the challenges set by geography and a primitive transportation system. Concomitantly, the Spartans may have imposed high rents in wine on the Helots to simplify their problems of manipulating agricultural output.
[ back ] 17. Occasional years of high yield should not distort our estimates: there was not only considerable interannual variation because of climatic conditions, but also significant local variability from year to year (as G. Shipley reminds me per litt. [5/5/02]).
[ back ] 18. Note Gallant (1991:75–78) who observes: “… I suspect ancient yields might well have been higher” (p. 78).
[ back ] 20. Although it was approached by a different methodology, I should note in comparison Roebuck’s estimate of 789 km2 available for all cereal cultivation in liberated Messenia and 875 km2 for total cultivated area (1945:156–157).
[ back ] 21. Of the 13 possible perioecic communities with which I dealt, Shipley 1997 classifies as contemporaneously-attested poleis: Aithaia (Shipley #18 = MME #136), Asine (19 = 512), Aulon (20 = 601), Thouria (21 = 137), and possibly Methone (36 = 412). Among Hellenistic/Roman poleis that are attested as archaic/classical settlements, he places Thalamai (33 = 150) and Kardamyle (34 = 147). Archaeological remains generally support their status as perioecic communities. Christien (1992:154–163) prefers a different site for Aithaia, equating it with Aipeia/Korone, but a location at Rizomilo/Nichoria (MME #100) is problematical. It is hard to reconstruct the effect for our estimate of this placement, which would exchange one good agricultural site for another. Shipley has provided me with a tentative, reconsidered classification (per litt. 5/5/02). Definite: Asine, Mothone (Methone); probable: Aithaia, Thouria; possible: Aulon, Kardamyle, Korone, Kyparissos, Pharai/Pherai, Thalamai.
[ back ] 22. Shipley classifies Leuktron = Leuktra (48 = 148) as a retrospectively attested polis (1996:190, 239), and Pephnos as an unconfirmed polis (122 = 549); cf. Shipley 2000:385–386, 387.
[ back ] 23. Shipley classifies them as poleis attested in post-classical sources: Alagonia (38 = 548), Abia (55 = 144), and Gerenia (57 = 146). More recently, Shipley classes these sites as places that “cannot be shown to have existed as a settlement in Ar(chaic)/Cl(assical) times” (per litt. 5/5/02).
[ back ] 24. Pharai = Pherai (79 = 142) is classed by Shipley as an early or legendary polis, but this status is probably a trick of the survival of our evidence, if Pharai is modern Kalamata (with remains dating back to the archaic period). Note Xen. Hell. 4.8.7, cf. Nepos Conon 1.1 (coloniam Lacedaemoniorum); Steph. Byz. Eth. 658; note also the heroic oecist in Paus. 4.30.2.
[ back ] 25. Shipley classifies under “legendary or early poleis (and possible) poleis” (1997:263 for #102). His more recent classification (per litt. 5/5/02) denies the presence of an “Archaic/Classical settlement.”
[ back ] 26. Christien (1998:437–439) puts the new border further south. Yet the Boiotian character of Leuktron, Kharadra (= Pephnos?), and Thalamai (Strab. 8.4.4), and the Argive oecist of Oitylos (Paus. 3.25.10), are better interpreted as indicative of the heroic pedigrees fabricated as precedents that preceded an archaic reception by Sparta of immigrants from central Greece and the Argolid as Perioeci.
[ back ] 27. IG V.1 1421 = SIG3 952 = SEG 11.1026, 22.314; SEG 11.1025; Shipley 1997:242–243 on #35. Shipley’s more recent classification (per litt. 5/5/02) lists Kyparissos as a possible perioecic polis.
[ back ] 30. Compare Scheidel (this volume, pp. 241–5) where a “simplified model” for Helot numbers posits recirculation along with higher yields and higher arable totals as means to account for higher estimates for the Helot population.
[ back ] 31. For positive and negative incentives in the antebellum south, see Fogel and Engerman 1974:144–157, 239–243. On Louisiana sugar plantations, slaves grew corn for their own advantage, selling it to their masters, who used it for the slave rations of the plantation (McDonald 1993:279–282).
[ back ] 32. Note Figueira 1984:108 for its discussion of Athen. 14.74.2 657C with Theopompos FGrHist 105 F 22. Here a Thasian offer of delicacies to Agesilaos and the Spartiates is refused as illicit, while such an offer was properly made to the Helots, who might more appropriately be corrupted by such fare.
[ back ] 33. Note my earlier specific rejection of this approach: Figueira 1984:95–96n25.
[ back ] 34. On such constraints, see now the discussion in Halstead and Jones 1989:47–50.
[ back ] 35. Yet the preserved economic hierarchy of the core territory was not static. The western zone of the Laconia Survey on the margins of the Eurotas plain experienced sixth century settlement on land to which earlier claim was probably laid by elite Spartiates and so presumably outside the system of klêroi. Helots and poorer Perioeci may have provided the workforce. See Shipley 1996b:332–335, 355–377, 380–389; Catling 2002:166–167, 180.
[ back ] 37. There may be some unappreciated literary evidence for such nucleation in a fragment of the Philolakon of Stephanos, a poet of New Comedy (fr. 1, PCG 7.614–615 [Athen. 11.469b]; cf. T 13): {Σ.} τούτῳ προέπιεν ὀ βασιλεὺς κώμην τινά. / {Β.} καινόν τι τοῦτο γέγονε νῦν ποτήριον; / {Σ.} κώμη μὲν οὖν ὲστι περὶ τὴν θυορίαν. / {Β.} εἰς τὰς Ροδιακὰς ὅλος ἀπηνέχθην ἐγὼ / καὶ τοὺς ἐφήβους, Σωσία, τοὺς δυσχερεῖς. While various candidates, including Philip II, Pyrrhos, and Alexander the Molossian, have been suggested for the king in line 1, surely there is a more economical hypothesis. A play about a Laconizer mentioning a grant of a kômê ‘village’ near Thouria in Messenia suggests a Spartan king. The context may be that an early Hellenistic Spartan king intends to reconquer southeastern Messenia so that he can reduce to dependency the formerly Helot communities in its vicinity, enabling him to make land grants (not traditional kleroi) to his supporters. Compare the foundation by an Athenian of Kolonides (Paus. 4.34.8). The kômê in question could derive from a nucleated Helot settlement.
[ back ] 38. Note especially the sites of Romanou Romanou and Romanou Glyfadaki in survey area VI. See Davis et al. 1997:454–469; Harrison and Spencer 1998:158–162; Alcock 2002:192–196.
[ back ] 39. Note van Wees 2000 for a spirited argument on behalf of the thesis that hoplites at Athens were recruited from more prosperous thetes, in which he uses 150 medimnoi as an important threshold level.
[ back ] 40. One reservation, however, concerns the argument that the Malea fortification of the Athenians during the Peloponnesian War (Thuc. 7.26.2) necessitates the existence of kleroi in the vicinity of Neapolis on the southern end of the peninsula (Shipley 1997:203; Hodkinson 2000:141). The modest population of potential Helot defectors from an out of the way, small district could not have justified such a base on their own account. A strongpoint there was well placed, however, to receive runaways by boat from throughout the Laconian Gulf, while it accommodated raiders in a spot conveniently distant from Spartan bases.
[ back ] 41. Cf. Catling 2002:161–163, 193–195, where some sites within the study area of the Laconia Survey may imply farmsteads controlling estates of 10–20 ha. I hasten to add, however, that kleroi need not have been contiguous estates of any certain size, but merely obligations of groups of Helots exploiting assets that yielded rents of an appropriate scale.
[ back ] 42. C. 3740 kg of barley @ 900 kg per ha requires 4.2 ha. At alternate year fallowing, 8.4 ha is needed for rent-producing land. On the supposition of a 50% klêros rent, that amount is doubled again to make 16.8 ha. We shall again add 0.6 ha as a minimum for grape production.
[ back ] 43. Using a ratio of 1 kg of alphita = 0.77 kg of wheat, which is halfway between my earlier ratio of 0.72 kg and the 0.82 kg of Foxhall and Forbes 1982:75–81.
[ back ] 44. Nichoria, likely to have been an eighth-century site affected by Spartan domination over Lower Messenia, exhibits a pattern of intensive raising of livestock, among which cattle are notably prominent; see Sloan and Duncan 1978:76–77; Stein and Rapp 1978:256.
[ back ] 45. That is a considerable level of output for subsistence agriculturalists. Clearly, however, we could manipulate redistribution, lower consumption, and reduce non-cereal output to lower per capita output to a certain degree.
[ back ] 46. Roebuck (1945:162–165) estimates a population of 90,000 (upper limit of 112,500) for free Messenia.
[ back ] 47. Note the low numbers posited for the Bronze Age kingdom: McDonald and Rapp 1972:128, 141, 254–256; Carothers and McDonald 1979.
[ back ] 48. Hunt 1997 is ill-judged in its revival of the hypothesis (Cornelius 1973) that the Helots provided the bulk of the Spartan phalanx, which was fronted by the 5,000 Spartiates. It would take a lengthy presentation to untangle all its improbabilities. Let me outline some counter-arguments instead: 1) the presence of Helot dead does not entail activity in the phalanx; service as skirmishers and flank guards would be sufficient (ἐφύλασσον in Hdt. 9.28.2 has this technical sense; cf. 9.85.1); 2) the Helots are not functionally differentiated from the other psiloi makhimoi in 9.29.1–2; 3) one rank is too shallow a Spartiate frontage, especially when the area of lethality of the phalanx is remembered; 4) the engagement of the second and third ranks in active fighting made light-armed troops this far forward very risky (Cornelius suggested a deeper phalanx); 5) this mass of Helots to the Spartiate rear without a rearguard posed a safety risk to their masters; 6) the Helots ought not be thought to have usurped the Perioecic role as comprising the rear ranks of the Spartan phalanx; 7) Helot training and social conditioning ill suited them for this role; 8) Helots later fought with the promise of emancipation or after emancipation and they served in separate units; 9) there remain doubts over the feasibility of a 5 km. front for any hoplite army and its practicality on the ground at Plataia.
[ back ] 50. Legrand (1954:15) recognizes a problem that he addresses by emending καὶ to κατ᾽ or excising on the grounds of omission from some manuscript traditions.
[ back ] 51. Powell (1977:351) opts for a rare non-military connotation of τάσσω in this passage (9.10.1). Yet Powell’s other comparable examples can be seen as metaphorical or extended usage based on the military (75 examples) and administrative (33) senses of τασσω (cf. 3.68.5; 8.98.1; 7.36.3). Another passage (3.91.4:… ες τὠυτὸ τεταγμένοι …), speaking of a classification for tribute payment, points us toward a better solution. The deployments of τάσσω relating to the Helots belong to the significant financial/taxational connotation (13 instances) where levies or fines are imposed (e.g., 2.65.5; 2.109.2; 3.89.1; 3.90.1; 6.42.2 ter, 6.79.1 [Kleomenes at Sepeia]).
[ back ] 52. E.g. Pollux Onom. 3.83. See Lotze 1959:53–56; van Wees, this volume, pp. 38–54, 64.
[ back ] 53. Compare Luraghi this volume, pp. 110–5 and van Wees this volume, pp. 34–7; cf. Bockisch 1985. Our view differs in that Spartan expansionism conditioned or even modulated Messenian depopulation. The hypothesis that the assemblage of the demographic components of classical Messenia was a complex evolution may reconcile the insights won from both approaches, seeing them as valid about distinct elements of the classical population. See Raaflaub this volume, pp. 171–2, who notes the importance of “homogenization of various forms of dependent labor.”
[ back ] 54. Witness the shared motifs between Dark Age II pottery at Nichoria and sherds from Sparta and Amyklai (Coulson 1983a:78–9, 111).
[ back ] 55. For an independent, parallel exploration of similar insights, see Link (forthcoming). Contrast Luraghi this volume, pp. 132–3. Note also Bockisch 1985:33–34.
[ back ] 56. Thuc. 1.132.4; Nepos Paus. 3.6; cf. Aris. Pol. 1307a1–5. Note Figueira 1999:224–225.
[ back ] 57. The considerable reduction of the number of sites in the study area of the Laconia Survey during the mid-fifth century indicates the scale of the upheaval, and suggests that valuable workers were relocating to more productive locations. See Catling 2002:175–178, 248–252.
[ back ] 58. The large building at Kopanaki in the Soulima valley of northern Messenia, initially mistaken for a Roman villa, may instead be an establishment designed for a mode of industrial or plantation agriculture c. 450/25 (for which Catling 1996:34n13) and be indicative as well that a threshold in property concentration and entrepreneurial/managerial engagement had been passed. See Kaltsas 1983; Harrison and Spencer 1998:161–162; Alcock 2002:195–196; Hodkinson this volume, p.271. A date earlier in the century might lead us to hypothesize a royal estate.
[ back ] 59. See Figueira 1999; also Christien 1998:454–459; Luraghi 2001:293–301; Luraghi 2002b. See also Alcock 1999 for the role of tomb cult in this process, and Coulson 1983b:334–336 for the classical tomb cult at Nichoria.
[ back ] 60. Yet many of the 20,000 direct casualties of the earthquake were Perioeci (cf. Diod. 11.63.1). See Figueira 1986:177–178.
[ back ] 61. Both νεότητα and οκαιότητα have manuscript authority, and a paleographical judgment cannot be made between the two alternatives. Gomme seems prudent to note that τὸ πλῆθος makes better sense with τὴν νεότητα (HCT 3.547). Hornblower (1996:264–265) agrees, but complicates the issue with the translation ‘youthful vigour’. Thompson 1987 is helpful for its discussion of the textual issues. His attempt to defend σκαιότης as Spartan conceptualization (inspired by the suppression of the great revolt) of Helot persistence as stupidity misreads the past and present contexts.
[ back ] 62. I should like to thank the other participants in the conference Helots and their Masters, and especially the organizers, Professors Nino Luraghi and Susan E. Alcock, for their invitation, hospitality, and intellectual encouragement and assistance with this paper. I also express my gratitude to Professor Graham Shipley of the University of Leicester, who was kind enough to read this piece in draft and offer me criticisms and suggestions.
