Longenecker B.W. - Exposing The Economic Middle. A Revised Economy Scale For The Study of Early Urban Christianity (JSNT 2009)

Journal for the Study of the New Testament
http://jnt.sagepub.com Exposing the Economic Middle: A Revised Economy Scale for the Study of Early Urban Christianity
Bruce W. Longenecker Journal for the Study of the New Testament 2009; 31; 243 DOI: 10.1177/0142064X08101524 The online version of this article can be found at: http://jnt.sagepub.com/cgi/content/abstract/31/3/243
Published by:
http://www.sagepublications.com
Additional services and information for Journal for the Study of the New Testament can be found at: Email Alerts: http://jnt.sagepub.com/cgi/alerts Subscriptions: http://jnt.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.co.uk/journalsPermissions.nav Citations http://jnt.sagepub.com/cgi/content/refs/31/3/243
Downloaded from http://jnt.sagepub.com by Eduardo de la Serna on October 26, 2009
JSNT 31.3 (2009) 243-278 The Author(s) 2009 Reprints and Permissions: http://www.sagepub.co.uk/JournalsPermissions/nav http://JSNT.sagepub.com DOI: 10.1177/0142064X08101524
Exposing the Economic Middle: A Revised Economy Scale for the Study of Early Urban Christianity Bruce W. Longenecker
St Marys College, University of St Andrews, Fife, KY16 9JU, Scotland bruce.longenecker@st-andrews.ac.uk
Abstract
In 2004 Steve Friesen proposed a poverty scale for Graeco-Roman urbanism as a backdrop against which to assess features of the earliest urban Christian communities. This article offers an assessment of the strengths and weaknesses of Friesens scale, not least in relation to binary taxonomies of Graeco-Roman economic stratication, rhetorical conventions of the ancient world, and the middling groups of Graeco-Roman urbanism. It proposes adjustments to the scale (renamed as the economic scale) and gives consideration to the signicance of those adjustments for the reconstruction of early Christianity relative to ancient poverty.
Key Words
Friesen, middling groups, poverty, economic scale, urbanism, early Christianity
1. On Modelling Graeco-Roman Poverty The economic status of the rst urban Christians has been a hotly contested issue in recent years. While economic status is merely one indicator to be considered alongside others when proling early Christian communities,1 an economic prcis needs nonetheless to be included as a critical part of that overall prole.2 This is especially the case in relation to our
1. Other factors inuencing social status include such things as family heritage, ethnic origins, ordo, citizenship, personal liberty, wealth, occupation, age, sex, and public ofces or honours (Meeks 1983: 55). 2. Cf. Woolf 2006: 92: The institution of the census made wealth one of the most explicit and formal measures of an individuals social standing and a key component of his public identity. One consequence was that impoverishment carried with it the threat of a form of social death. Cf. Patterson 2006: 188; Friesen 2004: 338. Of
244
Journal for the Study of the New Testament 31.3 (2009)
understanding of (1) the economic location of the rst urban Christians, (2) factors that might have attracted adherents to the early Christian movement, and (3) how adherents to the Christian movement might have been variously related and tasked within their communities. Consequently, positive contributions to the economic proling of the Graeco-Roman world in general and of early Christian communities in particular are to be welcomed. One such contribution is Steven Friesens so-called poverty scale for the general population in Roman urban contexts.3 In his proposed poverty scale, Friesen set out seven economic categories of rst-century GraecoRoman urbanism, and supplied estimated percentages for each of the categories (Friesen, 2004; 2005; see Table 1). As Friesen himself made clear, the outlining of an economic taxonomy for Graeco-Roman urbanism is an enterprise that is fraught with difculties, due to the many complexities that append themselves to any such effort. The extent to which a taxonomy of this kind has constructive utility within the study of Christian origins has yet to be tested by a cohort of scholars. But in my view, Friesens poverty scale offers merit enough to be taken seriously in analyses of Graeco-Roman urbanism in general and the early Christian movement in particular, offering signicant heuristic potential. Ignoring Friesens efforts is not a live option. There are two main reasons for this appraisal. The rst pertains to the nature of the extant evidence. That a heuristic device is especially necessary in analysis of ancient poverty is clear from the fact that the literary remains of the Graeco-Roman world show no standardization in their terminological usage. It is often postulated, for instance, that in ancient literature the term pe/nhj refers to one who has resources of some kind but struggles to survive through the harsh conditions of toil (i.e., owners of small farms, and the like), whereas the term ptwxo&j was used to signal the beggar without any resources whatsoever (cf. Aristophanes, Plutus 552-54).4 But neither the terminology nor the conceptualization was
course, other features of ones status could at times register more forcefully than economic prole. 3. Although Friesen never states what part of the urban population his scale is meant to measure, it seems clear that he means to include the whole urban population within his percentages. So when laying out his reasoning with regard to the scales percentages, he speaks of the inhabitants, men, women and children, families, the population under Roman imperialism and city populations (2004: 340-43). 4. So Louw and Nida (1988/89: 57.50) dene pe/nhj as a person who is poor and must live sparingly, but probably not as destitute as a person spoken of as ptwxo&j.
LONGENECKER Exposing the Economic Middle

Table 1. Steven Friesens poverty scale Scale PS1 Description Imperial elites Includes imperial dynasty, Roman senatorial families, a few retainers, local royalty, a few freedpersons equestrian families, provincial ofcials, some retainers, some decurial families, some freedpersons, some retired military ofcers most decurial families, wealthy men and women who do not hold ofce, some freedpersons, some retainers, some veterans, some merchants some merchants, some traders, some freedpersons, some artisans (especially those who employ others), and military veterans many merchants and traders, regular wage earners, artisans, large shop owners, freedpersons, some farm families % 0.04
245
PS2
Regional or provincial elites
PS3
Municipal elites
1.76
PS4
Moderate surplus
PS5
Stable near subsistence level (with reasonable hope of remaining above the minimum level to sustain life) At subsistence level (and often below minimum level to sustain life) Below subsistence level
22
PS6
small farm families, laborers (skilled and unskilled), artisans (esp. those employed by others), wage earners, most merchants and traders, small shop/tavern owners some farm families, unattached widows, orphans, beggars, disabled, unskilled day laborers, prisoners
40
PS7
28
246
wholly stable. In his play Dyskolos, for instance, Menander applies the term ptwxo&j to the land-owning and slave-owning character Gorgias (ll. 284-86). Conversely, Philo uses the term pe/nhj of destitute people without land (e.g., Virt. 90; 97; Laws 2.85, 105; Gaius 123). The terminology is not stable within the ancient record.5 Relatedly, some ancient authors show a preference for using only one term to connote a spectrum of poverties, rather than dissecting poverty into various terminological compartments. Whereas the authors collected in the New Testament make far greater use of ptwxo&j than pe/nhj, Josephus and Philo much prefer pe/nhj and its cognates over ptwxo&j. So when Philo wants to speak of a poor man becoming destitute, he uses not ptwxo&j but the comparative form of penixro&j to depict destitution (penixro&teron, Laws 2.75; cf. Virt. 84). Moreover, while it is the case that ancient rhetoricians frequently collapsed non-elite status into one overarching economic category of poverty (see below), at times they also relegated less-afuent elite to that same category. In the rst century CE the poet Martial described Gaurus as poor (Epigram 4.67), although Gaurus is said to have possessed 300,000 sesterceshardly a triing amount. In Marcians Novel 4 (midfth century CE) a woman is depicted as a pauper who is also said to be worthy of marrying an elite senator. As Caroline Humfress suggests (2006: 188), this is probably a woman of (what we might term) middling means, perhaps even a woman of high rank who had lost her family patrimony and had thereby become poor, or rather poorer. She was certainly not one of the destitute permanent poor. Furthermore, in some rare instances, the ancient elite found it to their own advantage to depict themselves as poor.6 This rhetorical strategy is used by Philo of Alexandria in his discussion of the laws of Moses, when he lists himself as among the poor (we poor men, oi9 pe/nhtej h(mei=j, Laws 2.20)although clearly we should be wary of accepting this as an accurate description of his economic prole in absolute terms. Accordingly, extant texts from antiquity demonstrate how relative and imprecise both the terminology and the conceptualization of poverty were, being open for persuasive denition and ascription according to
5. Osborne (2006: 11) notes that, even if the general scheme has general validity, neither of the terms [pe/nhj and ptwxo&j] was sufciently laden with associations with a particular level of need to prevent its use in quite other circumstances. 6. On the rhetorical advantages of presenting oneself as poor, see Woolf 2006: 9799.
247
context, and even more open to remaining vague and ambiguous.7 When dealing with the literary remains of the ancient world, the rule of thumb must be that each reference to poverty and the poormust be read in its particular case-specic context (Humfress 2006: 203). Since ancient authors demonstrate signicant slippage in their use of terms pertaining to poverty, scholarly discourse needs to be extra vigilant in arriving at conceptual precision. Friesens model contributes signicantly towards that goal, offering the potential of grounding scholarly discourse in a workable taxonomy from which to progress our discussion and debate. The second reason for thinking that Friesens poverty scale has heuristic merit is that it outlines economic stratication in a more sensitive and nuanced manner than any form of binary modelling, in which an extremely small group of elite is contrasted with a huge and undifferentiated majority struggling at subsistence level (as depicted, for instance, by Alfldy 1986: 52-53, 55; Toner 2002: 50-51). Of course, the binary model reects the rhetoric common in GraecoRoman literature, with its differentiation of the respectable elite on the one hand and the ordinary poor on the other. Differentiations between the few honestiores and the huge majority of humiliores are numerous in elite Graeco-Roman literature.8 In his Histories, for instance, Tacitus differentiates between those who are virtuous and associated with the great houses and the dirty plebs (1.4), just as in his Annals he contrasts the rabble and the citizens of repute (3.36). Generally speaking, among the Graeco-Roman elite the term the poor normally referred to anyone who was not of the ruling orders, and this became institutionalizedby the division between honestioresand humiliores (Whittaker 1993: 8). But binary rhetorical constructs of this sort are not to be taken as economic descriptors of the Graeco-Roman world. The point will be returned to below. For now, it is enough to note that binary modelling has a strong foothold in some sectors of New Testament studies. The most signicant recent advocate of a binary model is Justin Meggitt. In his important book Paul, Poverty and Survival, Meggitt argues that a bleak material existencewas the lot of more than 99% of the inhabitants of
7. Osborne 2006: 15. With regard to legal denitions of poverty in the Roman world, Humfress suggests that there is no general or constant denition of poverty, rather the tacit recognition that the poor would include those lacking all property (2006: 202), even if non-propertied persons would not exhaust the category of the poor. 8. Some examples can be found in Whittaker 1993: 1-3; see also Hellerman 2005: 6-8. Other examples could be added to his survey, such as Cicero, Mur. 1; Rep. 2.40.
248
the Empire (1998: 99). This is a bold claim, perhaps the starkest of claims by those who advocate more binary forms of modelling the Graeco-Roman economy. Occasionally scholars who advocate a binary model attempt to nuance their model somewhat, in an effort to take account of data that would otherwise compromise their model. But their attempts often reveal how tenuous their binary models can be. In his 2005 SNTS monograph, for instance, Joseph Hellerman recognizes that a two-strata perspective blurs somewhat in real life, while at the same time awkwardly postulating that it possesses the advantage of clarity (2005: 168 n. 14). This represents an odd historiography. Nor does Hellerman improve his case when noting that for the purposes of macrosociological analysis of ancient society, it is fair to retain a twofold division, since this is precisely the way in which the ancients themselves viewed their social world (2005: 168-69 n. 14). In this, Hellermans interests in macrosociology have simply entangled him in the rhetorical constructs of the ancient elite, rather than coming to better grips with the realia of (what Hellerman calls) real life. The distinction between the elite and the plebs that is so embedded within the rhetoric of Graeco-Roman literature must be seen for what it is: a rhetorical construction that served to reinforce the binary values and worldview of the elite, legitimating the accumulation of power by the elite over and above the masses.9 As Neville Morley notes, elite sources often treat the rest of the population as an undifferentiated mass, and apply the vocabulary of poverty indiscriminately, with such rhetorical constructs being deliberate attempts at constructing and promotingan image of society for particular purposesi.e., to reinforce the social structure that kept the masses in their place.10 The point is made explicitly by Aelius Aristides in the second century CE: the existence of inferiors is an advantage to superiors since they will be able to point out those over whom they are superior (Or. 24.34). In short, it served elite purposes to relegate all who are not among the elite to the category of the poor
9. The fact that the same unrened differentiation could be adopted by non-elite authors only testies to the way in which the differentiation had entrenched itself in the rhetorical constructs of the ancient world. 10. Morley 2006a: 29 and 26 respectively; see also 39. An exception to this rhetorical distinction is representing oneself as poor, despite ones elite status. This is evident in Martial, for instance, who describes himself as one who will always be a poor man (Epigrams 5.13), although in the same context he also crows about his equestrian rank and imperial patronage (5.17, 19). See footnote 6 above.
249
indiscriminately.11 Consequently, rhetorical constructs of this kind are illegitimate economic indicators in the reconstruction of ancient economic situations. The ancient elite usually collapse the categories [of non-elite economic status] into one and thereby offer little sense of the gradations of poverty that characterized their world.12 Although a binary differentiation between elite and the poor had become entrenched as a standard rhetorical construct in the Graeco-Roman world, economic analysis of the Roman world should not be constrained by that differentiation. With its seven-point scale and apportioned percentages, Friesens model goes much further than a strictly binary model towards offering a nuanced tool for the analysis of economic stratication. (The point will be returned to below, when assessing Friesens own percentage estimates.) For at least two reasons, then, Friesens proposed model has merit: (1) the instability of ancient discourse on poverty necessitates the construction of a heuristic tool in order to stabilize academic discourse; and (2) Friesens model is not only articulated (as opposed to implicit), but is sufciently nuanced (as opposed to a sharply dichotomous binary model). Of course, it is the case that Friesens model has weaknesses. I note two.13 First, Friesens seven-point scale is often informed by status indicators that outstrip mere economic indicators, especially at the top end of the scale. Social status was comprised of any number of factors, of which the economic indicator was merely one component. The multifaceted
11. The tone of this comment by Wallace-Hadrill (1994: 144) comes to the same conclusion: the vast riches squandered by the elite of the late Republic and early Empire, and the contrast with the undoubted squalor experienced by the poor, tempt us into polarizing the culture of the elite and that of the masses. It is easy (and perhaps for us morally satisfying) to dramatize this contrast. 12. Woolf 2006: 94. He goes on to note: Neither Juvenal nor Seneca have much to gainfrom portraying a more gradated social hierarchy; elsewhere the condemnation that both show to social climbers implies forcefully that no unbreachable gulf existed between rich and poor. 13. One feature that I do not include among its potential weaknesses is its tendency, as a snapshot, to mask economic mobility of both the upward and downward sorts. The loss of economic resources through misfortune was probably common enough to keep even the afuent elite mindful of their vulnerabilities; and the promotion of some humbler men to positions of relative comfort (see below in connection with the Augustales and apparitores) was probably common enough to keep some poorer but enterprising men watchful for any opportunity to advance themselves within the res publica. It was not Friesens goal to document movement from one economic stratum to another, but his model can at least x our discourse about what such movement might have involved in any given instance.
250
character of status in the ancient world makes it problematic simply to equate, for instance, the extremely high social status of a senator with a corresponding status at the top of the economic taxonomy.14 But while this is a weakness, it is one that we can live with. While we might wish to be economic purists, the goal is itself beyond our grasp, due to the fact that the evidentiary data required for the task has simply been lost to us.15 As such, we will be forced to explore the general interlock between some forms of social status and economic status. While at times inconsistencies within ones status registry could be pronounced, in general terms the full register of ones social status was often in general agreement with that persons economic status, or at least was usually not notably out of alignment with it. Second, as is generally recognized, models can skew data when their practitioners are resolutely committed to their ubiquitous applicability. A model of urban economic stratication in the Roman world is inevitably going to be heuristic, characterized by both the strengths and weaknesses that accompany heuristic tools. On the one hand, a model of economic stratication will be dogged by the fact that the Roman economy was not a homogenous entity, but rather a series of interlocking regional ones (Mattingley 2006: 285-86). Moreover, many parts of it were driven by agrarianism while others were distanced from agricultural extraction to one degree or another. As such, the Graeco-Roman economy was multifaceted and non-uniform. On the other hand, study of economic stratication requires a starting point, and modelling of one kind or another will always be operative in the handling of data. Models that are articulated have the advantage of being recognized as models, whereas models that
14. For this criticism of Friesen, see Oakes forthcoming. Nonetheless, in general, wealth was central to the denition of those suitable for membership of the senatorial and equestrian orders (Patterson 2006: 201; so too 202). See also Friesens discussion in relation to decurions, 2004: 360 n. 113. 15. At the Society of Biblical Literature in 2007 (San Diego), Friesen unveiled a wholly revised poverty scale in which he admirably attempted to strip the scale of all indicators except economic ones. The result was a scale comprised not of 7 but of 120 different levels. I am not sure that this laudable effort to introduce an economically pure model has improved the situation, however. Granted, the 7-point scale may well be vulnerable in coupling economic and other social indicators, and it may well require the eventual inclusion of further categories within it (not least within ES4). But a 120-point scale, while laudable in itself, is much more vulnerable than the 7point scale to the inaccessibility of data from the ancient world. Consequently, I much prefer Friesens early and less rened model to his later and precise model, whose very precision compromises its utility, owing to the poverty of extant data.
251
operate implicitly, without articulation, can too easily skew data without their effects being recognized. If erring is to be done on one side or the other, preference is to be given to a model that is articulated, reasoned and accessible, as long as it is not allowed to ride roughshod over vast swathes of data.16 At best, then, a heuristic model like Friesens should only be used as a helpful starting point that sensitizes analysts to general patterns, but which is also open to inevitable revision in the light of specic differentials that come into play in any given instance.17 The point will be returned to again below, but for now it is enough to note that Friesens model makes a constructive contribution in the articulation of a model that is at one and the same time both helpfully heuristic and unavoidably preliminary. This is the inevitable character for all constructed and constructive models pertaining to the ancient world. As Keith Hopwood writes (1989: 171):
One of the major problems facing the ancient historian is the lack of evidence at his [sic] disposal for many of the basic questions of the dayto-day organization of life in the ancient world. Consequently he [sic] is continually forced to construct models to describe and explain ancient social structures and modes of behaviour, models which are in need of constant renement.
Friesens own proposals are of precisely this order. In the discussion that follows, I will abandon my indebtedness to Friesens term poverty scale, preferring instead the term economic scale, since the latter is less value-laden. As a consequence, discussion of the various points on the scale will shift from PS4 to ES4, from PS5 to ES5, and so on. 2. On the Middling Groups Although I nd Friesens proposals to be generally helpful and constructive, his data are nonetheless somewhat skewed at places. This has nothing to do with the fact that Friesens numbers total 99.8% instead of the expected 100%. Instead, it has to do with the allocation of percentages to
16. The case is well articulated in Carney 1975: 1-46, a section entitled Models, What Are they and What Do they Do? 17. Cf. Holmberg 2004: 269-70: Models should not be seen as ideas that a scholar starts from and lets his research be guided by, but rather as the results of empirical research, which serve to simplify, abstract, or generalise the ndings obtained, so that they can be further tested elsewhere [H]istorically based models can turn a guess into an educated guess, but no more.
252
certain groups. There would be little point, of course, in arguing over slight adjustments to percentages, especially since those percentages can never be anything more than rough guides. But Friesens percentages are subject to qualication signicant enough to suggest that a reworked model is called for. This is especially the case with regard to the middling groups of ES4. Demonstration of the point follows in four sections below, with respect to three interrelated issues: (1) Friesens manner of calculating percentages (2.1); (2) Friesens indebtedness to Finleyan orthodoxy (2.2); and (3) further evidentiary analysis (2.3). After considering these issues, a revision of economy-scale percentages will be presented (2.4), followed by a prole of two important groups in the middle and upper bands of ES4. 2.1 Friesens Method of Calculating Percentages It will be argued below that Friesens percentages are subject to signicant qualication. Initially, however, it should be noted that there is no need to call into question the percentages that Friesen awards to the upper end of the scale. Friesens percentage for the elite (ES1 through ES3) is secure. At roughly 3%, this gure is the most reliable on the scale, being standard within studies of economic proling in the Roman world, which usually place the urban elite somewhere between 1% and 5% of the population.18 Less reliable, however, are the percentages applied to the lower end of the spectrum, due to the way in which Friesen arrived at his percentages for ES6 and ES7. When determining percentages for those economic levels, Friesen is heavily reliant on an important study by C.R. Whittaker (1993), who supplied estimates of how much of the population was engulfed in poverty in ancient Rome. Whittaker estimated that those in the equivalent of ES7 equated to 24-28%.19 Taking Whittakers upper percentage estimates, Friesen identies 28% of the population as belonging to ES7. Whittaker also estimated that some 30-40% of the population fell
18. Friesen includes an appendix in which he outlines his reasoning in support of these percentages. According to Scheidel, land registry evidence suggest that a truly tiny membership [within the elite orders] would hardly have been possible (2006: 51), so he imagines the elite landholders as comprising the top 3 to 5 per cent of the population (2006: 54). Ekkehard and Wolfgang Stegemann place the gure at between 1 and 5 percent (1999: 77). 19. Whittaker 1993: 4. In Whittakers analysis, this group is comprised of two subgroups: those unable to earn a living, who comprised 4-8% of the population, and those permanently in crisis who comprised 20% of the population.
253
into the equivalent of ES6. Taking Whittakers upper percentage estimate, Friesen lls ES6 with 40% of the population. In this way, when compiling his percentages for ES7 and ES6, Friesen has simply applied Whittakers upper estimates in each case. But this is precisely what makes Friesens percentages vulnerable. Whittakers gures allow for a variable of 14 percentage points for those two levelsa potential slippage that Friesens model simply erases. In this way, ES6 and ES7 are pegged to percentage points at the upper-most limit suggested by Whittaker, whose scale would also have permitted pegging at 14% lower than Friesens model.20 Questioning Friesens percentages for ES6 and ES7 has a consequent knock-on effect for his percentages for ES4 and ES5. These two levels on Friesens scale are derived by adding the percentages of the population already allocated to ES1 to ES3 and ES6 to ES7, and then making a rough guess as to how much of the remaining percentage should be allocated to ES4 and ES5. With 71% of the population already assigned to the other ve points on the poverty scale (3%, 28%, 40%), Friesen estimates that most of the remaining 29% should be allocated to ES5, making ES4 much smaller than ES5 because of the endemic character of poverty in the Roman empire, because of structural impediments in the economy, and because of the large amounts of wealth required to move up the poverty scale (2004: 346). So Friesen awards percentages of 7% to ES4 and 22% to ES5. In theory, however, if we choose to work with the percentages suggested by Whittaker, any or all of the indenite 14% of Whittakers ES6 and ES7 groups could have been entered within ES4 or ES5, in any proportion, taking the combined percentage of those two categories up from 29% to a maximum of 43%. As will be shown below, there is good reason for thinking that the percentage for ES4 requires augmentation.
20. Whittakers gures are themselves extrapolations from gures for pre-industrial cities. Two observations arise from this. First, this shelters Friesen from the potential criticism that he has taken gures applicable to Rome (as in Whittakers discussion) and applied them to Graeco-Roman urbanism in general; in fact, the gures that Friesen adopts are not tied to Romes own proleitself a city with quite a distinctive economic prole in relation to the urban centres of the ancient world. Second, the fact that Whittakers gures are based on gures for pre-industrial cities rather than Roman cities is simply a product of the phenomenon highlighted by the Hopwood quotation cited above. To criticize either Whittaker or Friesen for this extrapolation is simply to show a lack of cognizance as to the frailty of the data record of the ancient world.
254
2.2 Friesens Indebtedness to Finleyan Orthodoxy Friesen speaks early in his article of wanting to get beyond binary constructions of poverty. So he writes: The problem with binary terminology of rich and poor is that the term the poor has to cover at least 90% of the population with little or no differentiation among them (2004: 340). Clearly Friesens seven-point scale is well suited to get beyond a binary form of construction. But when Friesen applies his percentages to his seven-point model, his scale becomes little more than a binary construct heavily differentiated at the upper end (ES1 through ES3) and the lower end (ES5 through ES7), with only a small 7% group in the middle. Friesens seven-point scale is weighted in such a way and with such a glaring gap between the upper end and the bottom that it amounts to little more than a nuanced binary system.21 Friesens somewhat binary instincts (i.e., attributing only 7% to ES4) are explicitly indebted to the inuential work of Cambridge classicist Moses Finley. Finley argued that ancient economies, being almost exclusively agrarian in character, were without notable stimulation from either integrated commercial trade or signicant technological advance (1965; 1975, esp. chs. 5 and 6). Within that economic environment, there is little opportunity for middling groups to become established. Friesen cites his indebtedness to Finley in this way:
The crucial scholar in thisis Moses Finley [who] succeeded in demonstrating that the ancient economy was qualitatively different from modern economies. The ancient economy was agrarian and relatively static. It allowed for little growth or entrepreneurial initiative because the organization and use of resources was acquisitive rather than productive. Hence, there was no united market for the commercial exploitation of the empire nor any middle class to undertake such activities. So most inhabitants of the empire were born into poverty, and their only chance to escape it was the tomb (Friesen 2004: 338-39).
Finleys position has served as a discursive watershed in the study of the Graeco-Roman economy for the past four decades or so. One of its most notable inuences has been its feeding of binary economic models that virtually obliterate economic levels between the elite and the poor. In an economic situation of the kind imagined by those adopting Finleyan
21. The same is true of the Stegemanns proposed stratication (1999: 53-95). For them, three economic levels differentiate the tiny number of Graeco-Roman elite, while three levels also differentiate the massive number of poor. This six-point taxonomy is little other than a nuanced binary schema.
255
orthodoxy, there is little scope for an innovative and emergent middling group.22 Acknowledging his indebtedness to Finley, Friesen awards a gure of only 7% to the middle group of ES4. What Friesens account fails to make clear to its readers, however, is the extent to which Finleys position has been repeatedly and signicantly challenged. For some, Finleys reconstruction requires only some tweaking, while for others it is largely problematic. But either way, there arises more scope than Finley had allowed for a range of different levels of wealth to emerge between the two binary extremes. Within ten years of the publication of Finleys monograph, for instance, Keith Hopkins (one of Finleys students) had argued repeatedly that Finleys model required revision in order to take account of economic growth in the rst two centuries of the common eraa growth largely facilitated by the spread of technical and social innovations (1983: xiv; cf. Hopkins 1978; 1980 [reworked in 1995/6]). Hopkins did not consider the data he tabled to completely overturn Finleys model of Roman economy, since the Finley model of the ancient economy is sufciently exible to incorporate this modest dynamic (1983: xxi). More recently, however, scholars have not been quite so generous towards Finley. This is especially the case, for instance, in the work of archaeologists Kevin Greene and Andrew Wilson. Taking note of Finleys minimalist interpretations and dismissive attitude towards archaeological evidence (Greene 2000b: 29-30), both have tabled signicant archaeological evidence that suggests that technological progress within the Roman world was not at all inconsequential (Greene 1986; 2000a; 2000b; Wilson 2002; 2004; 2006).23 In contrast to Finley, for instance, Wilson depicts a picture of an
22. For discussion of the extent to which Finley has been misrepresented, see Saller 2002 (for a critique of Sallers view, see Silver 2003). For the purposes of this article, it is notable that Friesen taps into Finleys (alleged) position when advocating a minimalist view of middling economic groups. 23. For others involved in correcting the position of Finley, see Fredriksen 1975; DArms 1977; Christ 1980; Vittinghoff 1980; Carandini 1989; Pleket 1990; Descat 1995; Paterson 1997; the preponderance of articles in Mattingly and Salmon 2001a; Andreau 2002. For discussion of the place of Finleys views in their historical context, see Wilson 2002: 2. For discussion of the ideological function of Finleys views, see Morley 2006b: 41-44. Whereas Finley and others question the extent to which archaeological evidence can negotiate between economic models, Woolf rightly contends that archaeology offers the best means of choosing between these alternatives (1992: 283); see also Andreau 2002: 37-40.
256
economic boom of the rst and second centuries A.D., which can be attributable to a variety of factors. One factor is
the boost to state nances given by the use of advanced mining technologies, on top of a very healthy agrarian base which grew in the provinces under the stimulus of the opening up of new markets as vast swathes of territory came under Roman control (Wilson 2002: 30).
For Wilson, certain forms of economic growth stimulated further forms of economic growth. This is testied to, for instance, in the material remains of the construction industry, which underwent notable technological improvements during this period (Wilson 2006). For example, through the modularization of materials and the standardization of material sizes, building materials could be produced for mass distribution, rather than simply for local or regional distribution. This helps to explain why bricks forged in one region have ended up in buildings far from where they were originally forged.24 The benets that followed in the wake of expansive trade opportunities would have been accrued at various points along the economic spectrum. It would not be surprising if the elite would have had their ngers in the economic pie of any developing industry and would have beneted richly from it, but benets of technological advances would not have been restricted simply to them. According to Wilson, private brick makers and artisans throughout a broad expanse of the empire would have beneted economicallythat is, some of those listed in Friesens ES4 and others at the top end of ES5 (i.e., traders).25 So too, according to John Patterson, urban growth and prosperity led to increasing prots for artisans and traders of the cityand [to] social groupings just below the traditional elitethat is, those within Friesens ES4. Similarly, Dennis Kehoe notes that during the early imperial period,26
24. In contrast to the view that economic growth predominated in the imperial west, Alcock (2007: 696) argues that [a]ggregate growth in productionto some as yet unquantiable, but arguably considerable degreemust be accepted for the early imperial east. 25. In Wilsons reconstruction, it is not hard to imagine brick makers of ES5 nding ways of capitalizing on the increasing economic expansion that technological advances in the construction industry provided, with some of the more fortunate moving into ES4 as a consequence. Cf. Mattingly and Salmon 2001b: 10: local production may in some instances have been signicant in the accumulation of local prosperity, and probably also of individual fortunes. 26. Kehoe 2007: 550; cf. 569: it is likely that there were manypeople who could

[f]ortunes could be made in commerce and manufacturing, which certainly occupied a signicant place in the Roman economy [T]he chief beneciaries of the opportunites that the Roman empire offered for generating wealth in these sectors were generally people of more modest social and economic status [than the elite], from wealthy freedmen to independent artisans.
257
Such examples of technological and trade improvements are causing many to question the adequacy of radical binary modelling of the GraecoRoman world. According to Wilson, having failed to properly consult the archaeological record, historians have massively underestimated the degree of technological development in antiquity and its impact on the economy. Wilson and others advise that we should abandon the assumption that the ancient economy was either stable or stagnant, favouring instead a model of intensive per capita economic growth in the rst two centuries AD (2004: 144).27 Or as Mattingly and Salmon contest (2001b: 9), much work since (and, of course, before) Finley, especially that which depends essentially on archaeological evidence, demonstrates that both individuals and groups successfully established and exploited markets in particular cases, often on a large scale. If Finley erred by underestimating growth within Roman trade and technology, his many detractors at times have a tendency to err by overemphasizing the importance of that growth. Two issues intertwine here, but should not be confused: (1) the effect of technological advance and improved trade on the Roman economy overall, and (2) the implications of technological advance and improved trade on the modelling of Roman (socio-) economic stratication, of the kind that Friesen has pioneered. With regard to the rst issue, if it is right to imagine 75-80% of the Roman economy being agriculturally based, developments in trade and industry may not have much overall effect on economic growth in the Graeco-Roman world, at least on a macro scale. The point has been made by Saller (2002), for instance, and by Scheidel and von Reden (2002: 71). Whereas the debate has primarily been framed as an either-or among scholars (i.e., Finley or anti-Finley), a middle way seems preferable. The material record testies to the existence of various kinds of economic systems within the Graeco-Roman world. The predominant economic system was agrarian and acquisitive, beneting the elite landholders
take advantage of the business opportunities that Roman rule created. See the case study by Drinkwater 2001. 27. See also Hitchner 1993; 2005; Jongman 2007: 612-15.
258
primarily; to a notable degree, then, the basic Finleyan model remains intact. But other economic inuences were also in evidence, economies beyond agriculture (so Mattingly and Salmon 2001a), many of which were disengaged from agrarianism to one extent or another (e.g., ceramics, textiles, mining and quarrying). The Roman world comprised various economic motors, only some of which Finleys model does justice to (as in point two of 3 below). With regard to the issue of modelling economic stratication, correcting Finleys underestimation of trade and technology in the rst two centuries CE has the potential to destabilize severely binary schematizations that are indebted to Finley. The Graeco-Roman world testies instead to a spectrum of statuses between the wealthiest senators at the top and the most oppressed slaves at the bottom (Patterson 2006: 261). The consequence is that the size of the Graeco-Roman middling groups may well have been notably underestimated by Finley and others. Peter Garnsey and Richard Saller note that groups distinguished from both the elite orders and the humble masses were sizeable in the Graeco-Roman world (1987: 116).28 Willem Jongman notes that [e]ven many ordinary citizens were moderately prosperous, and there were also many moderately wealthy people in between the masses of modest means and the rich but small political elite.29 The view has been strenuously articulated by Scheidel (2006), as will be noted below (2.3). Consequently, Friesens percentage of 7% for ES4, which is ultimately attributable to a Finleyan perspective of the ancient economy, is quite vulnerable. Since Finleys day, there have been considerable changes in our factual knowledge and interpretation of ancient technology and economics (Greene 2000b: 29). In his point by point rebuttal to Finley, Greene charges that sufcient evidence has been adducedto show that a change in emphasis has taken place in the interpretation of the classical economy since Finley placed his stamp upon it in 1965, and repeated the same view in 1973 (2000: 52).30 Or, as Wilson notes, since Finleys day
28. Garnsey and Saller do not allow their perception of this sizeable group to do much to offset a binary distinction between the honestiores and the humiliores, even while recognizing that a rough, binary classicationrun[s] the risk of oversimplifying and thus distorting the social reality (1987: 116). 29. Jongman 2007: 597. 30. It is important to note, of course, that Greene recognizes that Finley was no doubt right to stressthat traders and manufacturers were in a minority without social status or political or economic inuence (2000a: 45), and that gross disparities in wealth, the importance of political power and social status, and the limitations of nancial systems, are not in dispute (2000a: 52). What is in dispute is
259
thinking on the ancient economy has moved on considerably (2002: 2). Or, as Jean Andreau puts it (2002: 34-35), Finleys model has not triumphed; instead,
after years in which his conclusions continued to be advancedthis is no longer the case. The questions are Finleyan, the methods and ways of thinking bear the stamp of his inuence, but the answers are moving farther and farther away from his own.
Friesen, however, gives his readers no hint that this change in emphasis has taken place, that thinkinghas moved on considerably, or that scholarly opinion is moving farther and farther away from Finleys own. While Finleys views have merit in relation to some important economic dimensions of the Roman world, they exhibit deciencies in relation to economies beyond agrarianism. The consequence of this pertains especially to attempts to model the middling groups. The increasingly prominent role of trade and technology within the Graeco-Roman economy undermines the legitimacy of stark binary schemes. Hanging in the balance is the extent to which middling groups might have been evident within the economic spectrum of the Graeco-Roman world. For many historians, there would have been a far greater presence of such groups than the Finley model suggests. But Friesen registers nothing of this sort within his article. By tying itself so closely to Finley, Friesens analysis disguises the scope and importance of artisanal and commercial wealth, indicating an apparent gap in his imagining of the ancient economic landscape. Although Friesens Finleyan proclivities are the primary grounds for disputing his percentages for ES4, other data lend conrmation to the view that Friesens analysis disguises the true extent of middling groups within the Graeco-Roman economy, as noted in 2.3 below. 2.3 Further Evidence and Scheidels Estimate If there is reason to suspect a notable presence of middling groups in the Graeco-Roman world, awarding percentages for such groups within an economic scale is not an uncomplicated exercise. According to Walter
the extent of growth in production in the Roman world. In 2006, Greene claimed that in the 20 years since his important book Archaeology of the Roman Economy (1986), the optimistic views expressed in 1986 have been amplied by the quality and diversity of subsequent archaeological discoveries and publicationsreinforc[ing] my conclusion that the level of economic activity and complexity of production found in the Roman provinces was not matched again until the postmedieval period (2006: 132).
260
Scheidel (2006), however, approximate percentages for this economic level can be estimated through a series of calculations that gravitate towards a much higher percentage than the 7% that Friesen awards to ES4. As a case in point, Scheidel tables the ve-tiered census taxonomy for property owners of the late Republic.31 With the rst to fth tiers corresponding with decreasing levels of annual income, Scheidels descriptors of the third and fourth tiers correlate relatively well with Friesens descriptors for ES4. In Scheidels estimate, these two census categories clearly exclude the afuent elite (i.e., Friesens ES1 to ES3). Instead, the fourth census tier (i.e., second to the bottom) would have applied to a reasonably secure commoner household[that] would have been reasonably well cushioned against chronic want, with the same being even more true of the [higher] third class. Notably, Friesen usually describes ES5 in relation to subsistence-level groups ES6 and ES7, describing them as those who lived near the subsistence level (p. 345). So Scheidels descriptors for census tiers three and four are out of alignment with Friesens general descriptor of ES5. Clearly, there is little reason to think that an exact match is possible between the census tiers and Friesens scale, and it is possible that some at the lower end of the fourth census tier might arguably fall within the upper levels of ES5. But in general, Scheidels descriptors of these two census tiers coincide with Friesens descriptors of ES4, not least in Scheidels claim that they would have constituted a middling group (2006: 49). Listing these census tiers as middling, Scheidel also estimates that the middling property-owners would have comprised a notable part of the population: I nd it hard to imagine, he writes, how in any reasonable scenario the number of middling property-owners could be (even) smaller than20 per cent (2006: 51). He nds conrmation of a gure of 20plus percent in various land registers of the ancient world. The land register of Ligures Baebiani of southern Italy in the second century CE, for instance, refers to a notable spectrum of estates, with most of them falling within categories two to four of the census, and a signicant number of estates falling into the fourth tier of the census taxonomy (Duncan-Jones 1976: 26; 1990: 129-31). Similarly, land registry data from imperial Egypt gives indication of similar stratication. In his study of Egyptian land registers, Roger Bagnall notes the existence of a broad band of middlerange men capable of bearing public obligations in civic contexts (1992:
31. For the view that the census taxonomies of the late Republic continued to have some inuence in imperial times, see Rathbone 1993: 144.
261
142). In Scheidels guring, this broad band of middle-range men would have comprised between 20% and 25% of the population in imperial Egypt (2006: 53-54). Similar impressions emerge with regard to housing gradation in the ancient world. Archaeological discoveries in various locations throughout the Roman world have noted the existence of a continuum of housing standards in urban contexts of various sorts. Excavations of large farming centres (large towns or small cities)32 in Roman Syria have registered the existence of a graduated spectrum in the quality of housing, ranging from the comfortable residence, through villas of steadily increasing simplicity and small farms, to humble shanties.33 The same case can be made about gradation in housing quality in relation to other urban centres of the Roman worldnot least Roman Corinth, Ostia and Pompei. With regard to Corinth, Dirk Jongkind has noted the following (perhaps unwisely articulating his case with the questionable word class):
A consideration of living spaces in ancient Corinth suggests that it is not possible to characterize its society as one made up merely of a very small number of lite alongside vast numbers of non-lite who were extremely poor. The variety of housing suggests the existence of another class.34
32. Many of what we would call villages today, with populations of 3,000 or so inhabitants, would have been known as small urban cities in the ancient world, whose inhabitants earned their living by farming. So it would be anachronistic to suggest that the practice of farming alone is an indication of a rural environment; such a view does not necessarily coincide with ancient perceptions. So Theissen 2001: 72. Friesens urbanism is focused on cities of 10,000 or more (roughly 5% of the GraecoRoman population), but the judgment to restrict urbanism to that number is somewhat articial. As he notes, a decurion presence was probably characteristic of most urban centres, excluding only the smallest urban centers, which he identies as small towns, villages, way stations etc.), all of which were too small or had insufcient resources to be organized in relation to a decurionate. The case of Philippi might be a case in point. Meeks identies it as one of the rather small towns of the empire, primarily an agricultural centre (1983: 49), and being comprised of not more than six hundred to eight hundred meters from wall to wall along its east-west axis (1983: 46). 33. De Ligt 1990: 51 (translating the ndings of Tchalenko 1953: 358). These excavations also reveal the existence of well-off farmers in those centresprobably a group that would fall within ES4, although they are not included in Friesens descriptor of that group. 34. Jongkind 2001: 139. He concludes in this way: The variety of housing in Corinth shows us a society which cannot be simply divided into an elite and a very poor non-lite. The non-lite were not all extremely poor nor did all lead a life just
262
Similarly, Andrew Wallace-Hadrills study of the archaeological remains of Pompeii demonstrates that a graduated housing spectrum existed there. He speaks of the strikingly mixed distribution of houses in Pompeii, of the interlocking jigsaw of large, medium, and small houses that repeats itself constantly throughout the city (1994: 67, 99). Although he is rightly cautious about making the leap from a graduated spectrum of house quality to a graduated spectrum of economic stratication, WallaceHadrill notes nonetheless that in broad terms we are dealing with a [graduated] progression of wealth and status of inhabitants (WallaceHadrill 1994: 90). For reasons of this sort,35 Scheidel seeks to undermine a stark binary construct and to augment notably the percentage of those who fall within the middling groups of the Graeco-Roman world. He expresses the view in this way (2006: 54, 51-52):
[T]here is sufcient evidence in support of the notion of an economic continuum from a narrow elite to a steadily broadening middling group as we move down the resource ladder. Sources ranging from Republican Italy to imperial Egypt and Syria all point in the same direction. It is perfectly possible to reconcile the dominance of a disproportionately afuent elite with the presence of a substantial middle Not necessarily formally beholden to landed elites as Mediterraneanist clients, they [the middling groups] would have provided the backbone sorely lacking from a hypothetical, more extremely dichotomous society that pitted a few wealthy toffs with their entourage of slaves and freedmen against countless marginalized subsistence peasants and day-labourers.
2.4 A Revised Economy Scale In view of these factors, a revised economic scale is called for. As we have seen, Scheidels percentage for Friesens ES4 is something in the area of 20-25%.36 While we should probably be guided by Scheidels
above abject poverty. Such generalisations are not supported by the archaeological evidence and cannot be used in describing Pauls relation to the wealthy and the poor in Roman society (2001: 148). 35. Other data can also be entertained. For instance, Jongman (2007) analysed wages, annuities and slave prices and concludes about them in this way (602) : it is remarkable that these three estimations of the extent to which the incomes of ordinary citizens exceeded bare subsistencereturn such similar results. They all conrm that for many (though not slaves) standards of living were well above subsistence, at least for a while. 36. In this way, Scheidels contribution addresses one of the concerns that Barclay
263
gures, my own impression is that his upper percentage of 25% for the middling group is too high, and that his percentage range of 20-25% should drop by approximately 5% to 15-20%.37 This proposed adjustment to Friesens percentages for ES4 has a signicant effect on the model of economic stratication in the GraecoRoman world. We have seen reason to believe that Friesens model might benet from adjustments not only in relation to ES4, but also to ES6 and ES7, and in equal measure. It has been noted that Friesens percentages for ES6 and ES7 are potentially too high, perhaps by as much as 14%.38 And it has been noted that Friesens percentage for the middling group of ES4 is probably too low, perhaps by as much as 13% (i.e., up to 20%, instead of Friesens 7%). In light of this, I propose three simple adjustments to Friesens otherwise helpful model: 1. the percentage for the ES4 middling groups should be registered at 17%. This requires taking 10% from other ES points in the Friesen model, and consequently, 2. the percentages for ES6 and ES7 should be comprised of gures from the lower end of Whittakers estimates: for ES7, 25% in Whittakers range of 24-28%,39 and for ES6, 30% in Whittakers range of 30-40%, and 3. the percentage for ES5 should be raised slightly from 22% to 25%.
(2004: 365) appropriately expressed about Friesens model: It is particularly debilitating that we cannot determine at all the likely percentages in Friesens categories PS 4 and 5i.e., ES4 and ES5. 37. Scheidels 20-25% seems somewhat high. For instance, Scheidels proling of Ligures Baebiani, with its Gini coefcient of 0.435, needs to be set in relation to other samples of property-size differentiation in the ancient world. Duncan-Jones (1990: 129-38) gathers Gini coefcients of 0.394 for Volcei in southern Italy, 0.447 for Lamasba in Numidia, 0.526 for Velecia in Northern Italy, 0.679 for Magnesia on the Maeander in Asia and 0.856 for Hermopolis in Egypt. While the rst two are on a par with that of Ligures Baebiani, the latter two are signicantly higher. 38. Although it is neither my intention nor within my expertise to do so, it is of course possible that these categories themselves might be rened further. Oakes (2004: 368) has offered suggestions about how we might move towards a more rened model, but his own proposals would seem to outstrip the bounds of the extant data. 39. It is notable that the proposed 25% for ES7 falls roughly in line with Oakess estimate that 20% of the Philippian population were living below subsistence level (2001: 47-49).
264
In the table below, to the right of Friesens seven-point enumeration is Whittakers estimate range for (the equivalents of) ES6 and ES7, and to the right of that appears Scheidels estimate range for (the equivalent of) ES4. At the far right of the table is the proposed economy scale after adjustments.
Friesens Whittakers Percentages Percentages ES1 ES3 ES4 ES5 ES6 ES7 3 7 22 40 28 Scheidels Percentages 20-25 30-40 24-28 Proposed Percentages 3 17 25 30 25
2.5 Proling Some Important Middling Groups The revised economy scale permits what is arguably a more realistic estimate of the presence of economically middling groups within Roman urbanism. But the gure of 17% for ES4 disguises what is likely to be a much-elongated economic category. Friesens own descriptors for this stratum are helpful up to a point: some merchants, some traders, some freedpersons, some artisans (especially those who employ others), and military veterans. But such descriptors tend to conceal the civic importance of this economic stratum at its middle and upper end. For that reason, it is important to give some indication of the signicance of these points within ES4 to the organizational structures of Roman urbanism. Whereas Friesen places retainers only within the 3% elite of ES1 to ES3, the adjective has at least as much relevance to certain groups within the middling level. One group that would have fallen towards the middle of ES4 is the apparitores. These men worked for civic magistrates as scribes, messengers, lectors and heralds. They were usually appointed to their apparitorial positions by patronage but, once appointed, could serve any magistrate. They varied in status, performing tasks from humbler to more responsible functions, with honour being accrued to them by virtue of their role in the proper running of the civic order. They consisted of men originally of relatively humble means who, being deemed able and trustworthy to civic goals, were moved into positions of relative security. As such, the apparitores play two key roles in any attempt to prole the economic stratication of the Roman world. First, they represent a sector
265
of society that was generally upwardly mobile.40 They may have been relatively exceptional in this regard, but their upward potential is not without parallel elsewhere in the ancient record (see also the Augustales, below).41 Second, they represent a liminal group, occupying a position on the social boundary between the elite and non-elite (Edmondson 2006: 273). In his classic study of apparitores, Nicholas Purcell describes this group as numerous, powerful, self-aware and self-esteeming (1983: 127). Purcell advised in 1983 that their intermediate status must be taken into account if we are to understand the [economic] stratication of Roman society (1983: 171). Another group of notable civic importance is the Augustales, who should be registered towards the top of ES4, sometimes spilling over into ES3.42 This group consisted predominately of freedmen43 who, due to their servile past, were generally excluded from securing decurial posts but who, upon gaining their freedom and in view of the viability of their nancial resources, were willingly conscripted and appointed by decurial patrons
40. Purcell (1983) makes this point repeatedly: These posts helped the lowly to aspire to the privileges of the higher echelons of the society (162); it is surely time to abandon the old contention that there were no bridges across the gulf between men of servile and of free origin, as well as the view that, with a few exceptions, men of less than equestrian rank could never dream of approaching such a distant and lofty status (171); the apparitorial position in the res publica allowed it to provide an excellent entre into the world of patronage which characterised Roman public life. It constituteda licensed mechanism for social mobility (171). 41. Cf. Carney 1975: 94: It would be wrongto think that there was no chance of upward social mobility in these societies [of the past]. For in fact there were two major avenues for such ascent: warfare and service under an autocrat. 42. Some have suggested that there were notable differences between the Augustales, the magistri Augustales, the seviri Augustales and the seviri, but most likely these terms were somewhat interchangeable, without signicant differences between them, at least for our purposes here. It is important to note that the Augustales were primarily a western phenomenon, with a lesser prole east of Macedonia. See especially Duthoy 1974: 148-91. But as Oliver has shown (1958), the same functions that were performed by the Augustales in the west were performed by the Gerusiae in the east. The Augustales, then, are unlikely to be simply a western phenomenon within the empire; they appear to have been absent at Rome, precisely because other groups perform the same functions (i.e., the vici magistri), just as they are lesser in the east due to the presence of the Gerusiae there. 43. Duthoys estimate (1974: 135-36) that 80-90% were freedmen is commonly accepted.
266
to enhance local civic life.44 Although they usually did not enjoy the enormous resources of the elite, they were nonetheless upwardly mobile (Edmondson 2006: 273) and were in a markedly different economic location to those who struggled in relation to subsistence living (Patterson 2006: 245; cf. 190-91).45 On occasions when food and money were distributed (i.e., formal banquets), the Augustales received less than decurions but more than the plebs (Patterson 2006: 246-49). Similar testimonies to the public perception of the middling status of Augustales come in the form of gifts awarded by leading citizens to sectors within the local populace. Such gifts frequently differentiated between amounts awarded per head for decurions, the Augustales and the general populace (i.e., sporting events, care for children), with decreasing amounts in each case (Duncan-Jones 1964).46 According to Patterson, the Augustales comprised
44. Debate exists over whether they served religious roles in relation to the imperial cult, or simply held civic duties apart from the imperial cult. The point does not require attention here, but helpful views are expressed by DArms 2000: 129 and Patterson 2006: 243. There is little dispute, however, with regard to the view that the Augustales served the double purpose of simultaneously afrming the superiority of wealth to poverty and the inferiority of servile birth to free birth; see Garnsey and Saller 1987: 121; Patterson 2006: 251. 45. The classic study of the Augustales as a middling group is that of Abramenko 1993. Scheidel (2006: 42) argues (unconvincingly in my view) that they do not form part of the middling group but fell wholly within the economic elite. I place them generally at the top of the middling group, perhaps with some overlap at the bottom of the elite stratum. A slightly confusing picture of the Augustales appears in Crossan and Reed 2004. There the Augustales are initially described as a lower-class group who could only make their inuence felt by banding together and putting their resources together collectively in order to win them civic visibility (2004: 300). On Friesens scale, the terminology tends to correspond with ES5. But what the Corinthian Augustales erected in their city was a huge blue marble step with an enormous statue of Augustussomething that a band of ES5 freedmen would be unlikely to have been able to resource. Later in Crossan and Reed 2004, discussion of the Augustales is couched in terms that apply most appropriately to ES4, sometimes even ES3 (32325). Even accounting for the fact that Augustales often ended their lives much wealthier than they had been upon gaining their freedom, the description of the Augustales as a lower-class group (2004: 300) is far too skewed to be generally accurate. 46. It may even be possible to estimate the general percentage of the urban population represented by the Augustales alone, although noting that local variations would obviously apply. In his analysis of private gifts to cities, Duncan-Jones nds two occasions in which evidence can allow us to reconstruct the size of the decurial and Augustales populations. In the case of Italian Petelia, Duncan-Jones gives an
267
one of the three central groups in civic life (Patterson 2006: 260), sandwiched between the decurions of ES3 and the members of collegia below that economic level. In light of this, ES4 looks to be a signicantly elongated category of wealth, with Friesens descriptor of it as involving moderate surplus pertaining to its lower levels, but being an inadequate descriptor of its upper levels. For this reason, it might be appropriate to move in due course from a seven-point scale to an eight-point scale. Or perhaps a distinction between ES4a (its upper end) and ES4b (its lower end) is called for. But regardless of how the taxonomy is articulated, it is important to recognize that (1) just below the decurial level wealth could be amassed at relatively signicant levels (rather than simply being moderate), and (2) those whose wealth placed them at the upper band of the middling groups were often committed to the enhancement of the civic environment that the Roman imperial order cultivated. 3. Further Observations The adjusted gures proposed above make visible the excluded middle that has largely been missing from more binary schemas of GraecoRoman economic stratication. Seven further observations need to be registered in relation to the adjusted economic scale. First, as we have seen, despite the potential of his model to move beyond binary schemas, Friesen tends towards a binary model nonetheless. As such, his discourse frequently reverts to the language of binary opposition: for nearly everyone in the Roman empire, poverty was a way of life (2004: 358). But an estimate of that kind does not apply to one fth of the Graeco-Roman population (ES1 to ES4; 20%) in the adjusted economy scale proposed here. Nor is it wholly clear that the phrase poverty was a way of life is a justied description of the 25% of the population represented by ES5. Those within ES5 would clearly have been conscious
estimate of 80-100 Augustales and 100 decurions (1964: 199-200). Although this suggests a ratio of almost 1:1, Duncan-Jones notes that this ratio is probably unrepresentative, since the evidence of most gifts would suggest a higher proportion of Augustales. More representative is the case of a generous gift given to Spanish Barcino, where the ratio is closer to 2.5 Augustales for every 1 decurion (1964: 205). If we accept Duncan-Joness reasoning on this, and if we accept Friesens estimate that 1.1% of the urban population was comprised of decurions, then we can estimate that, in general, the Augustales would have comprised somewhere in the order of 2.5% to 3% of the urban population.
268
of their economic vulnerabilities and their proximity to poverty, but to lump them together with those in ES6 and ES7 is to over-generalize. That poverty affected all three groups is clear, but it affected them in different ways. As John Barclay aptly notes, the difference between ES5 and ES7 is simple and stark: life versus death (2004: 366 n. 2). It might be vulnerable life versus impending death, but the difference is notable nonetheless. Second, the proposed adjustments to the Friesen model need to be modied by the same qualication that Friesen applied to his percentages (see Friesen 2004: 338, 347). Just as any model is, by necessity, an abstraction, so too the adjusted economy scale is only a proposed starting point for considerations of economic stratication in the Graeco-Roman world. Indigenous variations are bound to have occurred, and any application of this adjusted economy scale will need to be attentive and responsive to those variations.47 But as an articulated construct, and especially in the adjusted form proposed here, the economy scale has the advantages outlined earlier in this article. Third, proposing a gure of 17% for ES4, as I have done here, should not be seen as re-establishing the notion of a middle class into our economic taxonomy of the ancient world. The claim is frequently made that there was no middle class in the ancient world, and this assertion is rightly made, since the notion of class itself is a relatively modern construct, grounded in Marxist analysis, at least in the sense of a class consisting of an identiable stratum of society with a relatively stable shared prole and common socio-economic interests in relation to wealth production. As Thomas Francis Carney notes well, the Roman world was generally founded on the principles of patronage, not class stratication (1975: 90, 94, 121 and 171; cf. Smith 1966):
[S]ociety resembled a mass of little pyramids of inuence, each headed by a major familyor one giant pyramid headed by an autocratnot the three-decker sandwich of upper, middle, and lower classes familiar to us from industrial society Patronage virtually precludesthe emergence of class consciousness Under the divisive local loyalties of patronage the latter [i.e., the sub-elite] were unaware of its class interests The client of a power wielderbecomes a powerful man and himself in turn attracts clients. Even those marginal hangers-on to power attract others,
47. For instance, poverty might be differently apportioned in urban and rural contexts. And uctuations in population percentages will occur even within urban contexts. For an interesting case for the virtual absence of poverty in Roman Egypt, see Rathbone 2006.

more disadvantageously placed, as their clients It is quite different from the three-layer sandwich of a class society.
269
Although this portrait might require some slight qualication, it nonetheless serves rightly to warn against imagining the existence of a middle class in the ancient world. But this exclusion of the middle class from the ancient world has sometimes been misconstrued to mean that, in the ancient world, the enormous gap between the elite and the poor was lled only by a signicant vacuum. When P.A. Brunt wrote in 1970 that There is no evidence for a middle class in the city intervening between them [the upper class] and the poor (1971: 383), he was as right as he was wrong (at least if by the poor he is envisaging those at the bottom end of the economy scale). There may have been no middle class, but there was a middling groupor better, middling groups. In this, two things are to be recognized simultaneously: (1) a substantial number of people comprised middling groups in GraecoRoman urbanism, and (2) the strategies for survival and security available to those middling groups were much different to the strategies open to the so-called middle class in industrial and post-industrial societies. Whenever the balance between these two is skewed, things tend to go wrong in one direction or another: either a modern economic theory is applied to the ancient world (i.e., class) or a broad band disappears from the economic spectrum of the Graceo-Roman world (i.e., middling).48 Fourth, the adjusted economy scale should not be seen as an attempt to peripheralize or trivialize the poignant issues pertaining to structural poverty in the ancient world. With 55% of the Graeco-Roman world skimming the surface of subsistence and occasionally dropping down below it (ES6 and ES7), and with another 25% living in an extremely fragile suspension above subsistence level (ES5), studies of the early Christian movement cannot be immune to the pressing realities of poverty that affected the majority of the imperial world. The relative absence of nuanced discourse on poverty for its own sake within the literary remains of the ancient elite should not blind us to the realities of life that pressed down on vast swathes of the Graeco-Roman population. The important contribution of Meggitt and others should continue to serve as poignant and sobering registers of the squalid conditions affecting the lives of
48. For instance, Gill uses Garnsey and Sallers point about the absence of a middle class in the ancient world to criticize Wayne Meekss (alleged) view that somehow Christians fall into a middle social group (1994: 118 n. 79).
270
multitudes of people at the time when the early Christian movement was nding its feet.49 Fifth, and relatedly, adjusting the population percentages for the economy scale should do nothing to erode a necessary focus on economic exploitation, oppression and injustice in the agrarianism of the GraecoRoman world. This is where the economy scale for population percentages needs to be correlated with consideration of the socio-economic machinery that drove advanced agrarian systems and the matrices of socio-economic, religio-political status codes that animated most cultures of the ancient world, not least the Graeco-Roman world. Sixth, increasing the presence of middling groups within the GraecoRoman world in general does not necessarily lead to the conclusion that a similar percentage must be applied to middling groups within early Christian urban communities. In my view, it would be very difcult to argue for a double-gure percentage for members above ES5 in early Christian communities. Leaving ES1 to ES3 out of consideration, Friesens depiction of economic stratication in the Graeco-Roman world is virtually replicated in his depiction of economic stratication in Pauline communities, with a tiny percentage in ES4 and a huge percentage in ES5 to ES7. But this results in a rather monochrome comparison that dulls us to the potential differences between economic stratication in Graeco-Roman urban contexts generally and in early Christian communities particularly.50 This is especially true of ES4 on the economy scale, where an estimate of 17% for the Graeco-Roman population should probably drop to single gures for estimates of the constituency of early Christian communities. At stake, then, is the way we conceptualize the relationship of the early Christian movement to the society/societies around it. And seventh, it would be instructive to test the heuristic viability of the economy scale in relation to the interpretation of early Christian texts and our conceptualization of the rise of early Christian movement. For instance, how are we to conceptualize the economic realities at play within Pauls urban churches? To what extent can an economic taxonomy help to ground our discussion of the economic level of particular individuals known to us from Pauls letters? Is there a general economic level that Paul presumes of his communities when he advises his addressees to
49. For a helpful catalogue of life at or near subsistence in the Graeco-Roman world, see Meggitt 1998: 53-73. 50. A monochrome analysis is even more evident in Meggitts work, with 99% of the Graeco-Roman worlds population being comprised of those in poverty, as compared, evidently, with 100% of Pauline communities.
271
work with your own handsand be dependent on no one (1 Thess. 4.1112; cf. 2 Thess. 3.12; Eph. 4.28; Acts 20.34-35; see also Pauls own selfdescription in 1 Cor. 9.12-18; cf. 2 Thess. 3.6-10), or when he directs his communities, Each of you is to put aside and save whatever extra you earn (1 Cor. 16.2)? Might it be that many of the individuals named in Pauls texts belonged to ES4, even though Pauls advice about how to work and save for others is generally applicable to the unnamed of ES5? To what extent were Pauls communities supported by ES4 members acting as benefactors? To what extent would those ES4 members have been civic-minded in their aspirations (as in 2.5 above) and thereby complicit with a pro-Roman civic ethos? To what extent, then, might the presence and function of ES4 members in Christian communities have affected the contours in which urban communities appropriated the Christian message of the crucied lord and saviour? Are there indicators that Pauls communities were comprised largely of people oundering around the subsistence levels of ES6 and ES7, or is Meeks right to say that there is no specic evidence of people who are destitute within Pauline communities (1983: 73)?51 Or how are we to prole the situation of Heb. 10.34, with its mention of the addressees having cheerfully accepted the plundering of your possessions, knowing that you yourselves possessed something better and more lasting? Does the mention of plundering of possessions refer to elite conscation of land formerly held by those in, say, ES4 or ES5?52 And if so, have some of the addressees of the letter slipped from ES4 or ES5 to an extremely vulnerable position in ES6 perhaps?53 Or when comparing benevolent practices in the ancient world, might it be appropriate to locate (1) the benefactors of civic euergetism generally at ES4 and above, with their targeted beneciaries at ES5 and above, and (2) the benefactors of Christian charity in the rst century generally at ES4 and ES5, with their targeted beneciaries at ES6 and ES7?54
51. For discussion of these issues, see Longenecker forthcoming. Friesen discusses the economic location of Pauline Christians in 2004: 348-58. 52. This example illustrates that Friesens model is nuanced enough to register the stages of conjunctural poverty (i.e., slippage from ES4 to ES6, for example), in a manner that a starkly binary model is much poorer at doing. 53. Josephus offers us an example of this within rst-century Palestine. In his Antiquities 18.36-38, Josephus describes the enforced habitation of Tiberias, the controversial city established by Herod Antipas in 1819 CE. Those bribed to populate the city were often rewarded with new gifts of landi.e., land most likely acquired through conscation. 54. Whittaker 1993: 24: Widows, orphans, migrants and the sick, whom the
272
We might even ask whether such a scale is applicable in more rural contexts. For instance, when Jesus speaks of the poor as an economic group, we might well debate what point on the economy scale he is speaking from and referring to. Is he speaking as one from ES6, perhaps, referring to those in ES7? Alternatively, is he speaking as one from, say, ES5, referring to those in ES6 and ES7, or simply ES7?55 An economy scale, albeit merely a heuristic device, will not answer such questions for us, but it will enable our debates to be more focused than would otherwise be the case.56 In my view, it would be interesting to see scholars making use of the economy scale over the next decade or so in order to determine the extent of its utility in both (1) exegetical precision with regard to early Christian texts and (2) historical conceptualization with regard to the rise of the early Christian movement in the larger context of Graeco-Roman urbanism.
Church tended, were those whom Roman euergetism almost ignored. They were not the deserving poor. Cf. Parkin 2006: 60: Christian charity did not develop out of pagan municence. The two were concerned with fundamentally different sectors of ancient society. 55. In the lower levels of the economy scale, categories ES5 to ES7 might be relatively transferable from urban to rural contexts, such as rst-century Galilee. 56. Similarly, a poverty scale will enable us better to register cases such as that of the fth-century Christian Salvian, whose writings reveal a disconnect between a theological construction of radical poverty inherited from Christian discourse and his own experience of relative poverty. According to Grey (2006), the kind of poverty that Salvian seems cognizant of in his vicinity is the poverty of the small landowner, or even less wealthy members of local aristocracies (2006: 180), but not the poverty of utter destitution. Nonetheless, even if his own conceptualization of poverty gravitates towards ES5 or even ES4, the rhetoric that he applies in his theological discourse gravitates towards ES7.
273
References
Abramenko, A. 1993 Die munizipale Mittelschichte im kaiserzeitlichen Italien: Zu einem neuen Verstndnis von Sevirat und Augustalitt (Frankfurt: Peter Lang). Adams, E. and D.G. Horrell (eds.) 2004 Christianity in Corinth: The Quest for the Pauline Church (Louisville, KY: Westminster/John Knox Press). Alcock, S.E. 2007 The Eastern Mediterrannean, in Scheidel, Morris and Saller 2007: 671-97. Alfldy, G. 1986 Die rmische Sozialgeschichte (3rd edn; Wiesbaden: Steiner Franz Verlag). Andreau, J. 2002 Twenty Years after Moses I. Finleys The Ancient Economy, in Scheidel and von Reden 2002: 33-49. (Originally: Vingt ans aprs Lconomie antique de Moses I. Finley, in Annales ESC 50 [1995]: 947-60.) Atkins, M. and R. Osborne (eds.) 2006 Poverty in the Roman World (Cambridge: Cambridge University Press). Bagnall, R.S. 1992 Landholding in Late Roman Egypt: The Distribution of Wealth, JRS 82: 128-49. Bang, P.F., M. Ikeguchi and H.G. Ziche (eds.) 2006 Ancient Economies, Modern Methodologies: Archaeology, Comparative History, Models and Institutions (Bari: Edipuglia). Barclay, J. 2004 Poverty in Pauline Studies: A Response to Steven Friesen, JSNT 26: 36366. Brunt, P.A. 1971 Italian Manpower, 225 BCAD 14 (London: Oxford University Press). Carandini, A. 1989 Italian Wine and African Oil: Commerce in a World Empire, in K. Randsborg (ed.), The Birth of Europe: Archaeology and Social Development in the First Millenium AD (Rome: LErma di Bretschneider): 16-24. Carney, T.F. 1975 The Shape of the Past: Models and Antiquity (Lawrence, KS: Coronado Press). Christ, K. 1980 Grundfragen der rmischen Sozialstruktur, in W. Eck, H. Galsterer and H. Wolff (eds.), Studien zur antiken Sozialgeschichte (Cologne: Bhlau). Crossan, J.D. and J. Reed 2004 In Search of Paul (San Francisco: HarperSanFrancisco). DArms, J.H. 1977 M.I. Rostovtzeff and M.I. Finley: The Status of Traders in the Roman World, in J.H. DArms and J.W. Eadie (eds.), Ancient and Modern: Essays in Honor of Gerald F. Else (Ann Arbor: University of Michigan Press): 15779.
274
2000

Memory, Money, and Status at Misenum: Three New Inscriptions and the Collegium of the Augustales, JRS 90: 126-44.
Descat, R. 1995 Lconomie antique et la cit grecque, Annales 50 (1995): 961-89. Drinkwater, J.F. 2001 The Gallo-Roman Woollen Industry and the Great Debate: The Igel Column Revisited, in Mattingly and Salmon 2001a: 297-308. Duncan-Jones, R. 1964 Human Numbers in Towns and Town-Organisations of the Roman Empire: The Evidence of Gifts, Historia 13: 199-208. 1976 Some Congurations of Landholding in the Roman Empire, in M.I. Finley (ed.), Studies in Roman Property (Cambridge: Cambridge University Press): 7-33. 1990 Structure and Scale in the Roman Economy (Cambridge: Cambridge University Press). Duthoy, R. 1974 La fonction sociale de laugustalit, Epigraphica 36: 134-54. Edmondson, J. 2006 Cities and Urban Life in the Western Provinces of the Roman Empire, 30 BCE250 CE, in D.S. Potter (ed.), A Companion to the Roman Empire (Oxford: Blackwell Publishing): 250-80. Finley, M.I. 1965 Technical Innovation and Economic Progress in the Ancient World, Economic History Review 18: 29-45. 1975 The Ancient Economy (London: Chatto & Windus). Fredriksen, M.W. 1975 Theory, Evidence and the Ancient Economy, JRS 65, 164-71. Friesen, S. 2004 Poverty in Pauline Studies: Beyond the So-called New Consensus, JSNT 26: 323-61. 2005 Injustice or Gods Will: Explanations of Poverty in Proto-Christian Communities, in Richard A. Horsley (ed.), Christian Origins (A Peoples History of Christianity, 1; Minneapolis: Fortress Press): 240-60. Garnsey, P.D.A, K. Hopkins and C.R. Whittaker, 1983 Trade in the Ancient Economy (London: Chatto & Windus). Garnsey, P.D.A. and R.P. Saller 1987 The Roman Empire: Economy, Society and Culture (London: Duckworth). Gill, D.W.J. 1994 Acts and the Urban lites, in David W.J. Gill and Conrad Gempf (eds.), The Book of Acts in its First Century Setting. II. The Book of Acts in its Graeco-Roman Setting (Grand Rapids, MI: Eerdmans): 105-118. Greene, K. 1986 The Archaeology of the Roman Economy (London: Batsford; Berkeley: University of California Press). 2000a Industry and Technology, in A.K. Bowman, P. Garnsey and D. Rathbone (eds.), The Cambridge Ancient History. XI. The High Empire, A.D. 70192 (Cambridge: Cambridge University Press): 741-68.

2000b 2006 Grey, C. 2006
275
Technological Innovation and Economic Progress in the Ancient World, Economic History Review 53: 29-59. Archaeological Data and Economic Interpretation, in Bang, Ikeguchi and Ziche 2006: 109-36. Salvian and the Poor in Fifth-Century Gaul, in Atkins and Osborne 2006: 162-82.
Hellerman, J.H. 2005 Reconstructing Honor in Roman Philippi: Carmen Christi as cursus pudorum (Cambridge: Cambridge University Press). Hitchner, R.B. 1993 Olive Production and the Roman Economy: The Case for Intensive Growth in the Roman Empire, in M.-C. Amouretti and J.-P. Brun (eds.), La production du vin et de lhuile en Mditerrane (Athens: cole francaise dAthnes): 499-508. 2005 The advantages of wealth and luxury: The Case for Economic Growth in the Roman Empire, in J. Manning and I. Morris (eds.), The Ancient Economy: Evidence and Models (Stanford: Stanford University Press): 20722. Holmberg, B. 2004 Methods of Historical Reconstruction, in Adams and Horrell 2004: 255-72. Hopkins, K. 1978 Economic Growth and Towns in Classical Antiquity, in P. Abrams and E.A. Wrigley (eds.), Towns in Societies: Essays in Economic History and Historical Sociology (Cambridge: Cambridge University Press): 35-79. 1980 Taxes and Trade in the Roman Empire, 200 BCAD 400, JRS 70: 101-25. 1983 Introduction, in Garnsey, Hopkins, and Whittaker 1983: ix-xxv. 1995/6 Rome, Taxes, Rents and Trade, Kodai: Journal of Ancient History 6/7: 4175. Hopwood, K. 1989 Bandits, Elites and Rural Order, in Andrew Wallace-Hadrill (ed.), Patronage in Ancient Society (London: Routledge): 171-88. Humfress, C. 2006 Poverty and Roman Law, in Atkins and Osborne 2006: 183-203. Jongkind, D. 2001 Corinth in the First Century AD: The Search for Another Class, TynBul 52: 139-48. Jongman, W.M. 2007 The Early Roman Empire: Consumption, in Scheidel, Morris and Saller 2007: 592-618. Kehoe, D.P. 2007 The Early Roman Empire: Production, in Scheidel, Morris and Saller 2007: 543-69. de Ligt, L. 1990 Demand, Supply, Distribution: The Roman Peasantry between Town and Countryside: Rural Monetization and Peasant Demand, Mnstersche Beitrge zur antiken Handelsgeschichte 9.2: 24-56.
276
Longenecker, B.W. forthcoming Socio-Economic Proling of the First Urban Christians, in Still and Horrell forthcoming. Longenecker, B.W. and K. Liebengood (eds.) forthcoming Engaging Economics: New Testament Scenarios and Early Christian Reception (Grand Rapids, MI: Eerdmans). Louw, J.P. and E.A. Nida 1988/89 GreekEnglish Lexicon of the New Testament Based on Semantic Domains (New York: United Bible Societies). Mattingly, D.J. and J. Salmon 2001a Economies beyond Agriculture in the Classical World (London: Routledge). 2001b The Productive Past: Economies beyond Agriculture, in Mattingly and Salmon 2001a: 3-14. Mattingley, E. 2006 The Imperial Economy, in David S. Potter (ed.), A Companion to the Roman Empire (Oxford: Blackwell Publishing); 283-96. Meeks, W.A. 1983 The First Urban Christians: The Social World of the Apostle Paul (New Haven, CT: Yale University Press). Meggitt, J.J. 1998 Paul, Poverty and Survival (Edinburgh: T&T Clark). 2004 Sources: Use, Abuse, Neglect: The Importance of Ancient Popular Culture, in Adams and Horrell 2004: 241-54. Morley, N. 2006a The Poor in the City of Rome, in Atkins and Osborne 2006: 21-39. 2006b Narrative Economy, in Bang, Ikeguchi and Ziche 2006: 27-47. Oakes, P. 2001 Philippians: From People to Letter (Cambridge: Cambridge University Press). 2004 Constructing Poverty Scales for Graeco-Roman Society: A Response to Steven Friesens Poverty in Pauline Studies, JSNT 26: 367-71. forthcoming Methodological Issues in Using Economic Evidence in Interpretation of Early Christian Texts, in Longenecker and Liebengood forthcoming. Oliver, J.H. 1958 Gerusiae and Augustales, Historia 7: 472-96. Osborne, R. 2006 Roman Poverty in Context, in Atkins and Osborne 2006: 1-20. Parkin, A. 2006 You do him no service: An Exploration of Pagan Almsgiving, in Atkins and Osborne 2006: 60-82. Paterson, J.J. 1997 Production and Consumption, in P. Jones and K. Sidwell (eds.), The World of Rome: An Introduction to Roman Culture (Cambridge: Cambridge University Press): 181-207. Patterson, J.R. 2006 Landscapes and Cities: Rural Settlement and Civic Transformation in Early Imperial Italy (Oxford: Oxford University Press).

Pleket, H.W. 1990
277
Wirtschaft, in F. Vittinghoff (ed.), Europische Wirtschafts- und Sozialgeschichte in der rmischen Kaiserzeit (Stuttgart: Ernst Klett Verlag): 25160.
Purcell, N. 1983 The apparitores: A Study in Social Mobility, PBSR 51: 125-73. Rathbone, D.W. 1993 The Census Qualications of the assidui and the prima classis, in H. Sancisi-Weerdenburg and H.C. Teitler (eds.), De Agricultura: In Memoriam Pieter Willem de Neeve (19451990) (Amsterdam: J.C. Gieben): 121-52. 2006 Poverty and Population in Roman Egypt, in Atkins and Osborne 2006: 10015. Saller, R. 2002 Framing the Debate over Growth in the Ancient Economy, in Scheidel and von Reden 2002: 251-69. Scheidel, W. 2006 Stratication, Deprivation and Quality of Life, in Atkins and Osborne 2006: 40-59. Scheidel, W. and S. von Reden (eds.) 2002 The Ancient Economy (Edinburgh: Edinburgh University Press). Scheidel, W., I. Morris and R. Saller (eds.) 2007 The Cambridge Economic History of the Graeco-Roman World (Cambridge: Cambridge University Press). Silver, M. 2003 Review of Walter Scheidel and Sitta von Reden (eds.), The Ancient Economy. EH.Net Economic History Services, 3 January 2003. URL: http.//eh.net/bookreviews/library/0570; accessed 11 March 2008. Smith, M.G. 1966 Pre-Industrian Social Stratication Systems, in N.J. Smelser and S.M. Lipset (eds.), Social Structure and Mobility in Economic Development (Chicago: Aldine Press, 1966): 141-76. Stegemann, E.W. and W. Stegemann 1999 The Jesus Movement: A Social History of its First Century (ET: Minneapolis: Fortress Press, 1999). Still, T. and D.G. Horrell (eds.) forthcoming After the First Urban Christians: The Socio-Historical Study of Pauline Christianity Twenty-Five Years Later (Grand Rapids, MI: Baker Academic). Tchalenko, G. 1953 Villages antiques de la Syrie du nord, I (Paris: P. Geuthner, 1953). Theissen, G. 2001 The Social Structure of Pauline Communities: Some Critical Comments on J.J. Meggitt, Paul, Poverty and Survival, JSNT 84: 65-84. Toner, J.P. 2002 Rethinking Roman History (Cambridge, MA: Polity Press). Vittinghoff, F. 1980 Soziale Struktur und politisches System in der hohen rmischen Kaiserzeit, Historische Zeitschrift 230: 31-55.
278
Wallace-Hadrill, A. 1994 Houses and Society in Pompeii and Herculaneum (Princeton: Princeton University Press). Whittaker, C.R. 1993 The Poor in the City of Rome, in Land, City and Trade in the Roman Empire (Aldershot: Variorum Ashgate, 1993), article 7. Wilson, A. 2002 Machines, Power and the Ancient Economy, JRS 92: 1-32. 2004 Cyrenaica and the Late Antique Economy, in Ancient West and East 3: 143-54. 2006 The Economic Impact of Technological Advances in the Roman Construction Industry, in Elio Lo Cascio (ed.), Innovazione tecnica e progresso economico (Bari: Edipuglia): 225-36. Woolf, G. 1992 Imperialism, Empire and the Integration of the Roman Economy, World Archaeology 23: 283-93. 2006 Writing Poverty in Rome, in Atkins and Osborne 2006: 83-99.

Longenecker B.W. - Exposing The Economic Middle. A Revised Economy Scale For The Study of Early Urban Christianity (JSNT 2009)

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Longenecker B.W. - Exposing The Economic Middle. A Revised Economy Scale For The Study of Early Urban Christianity (JSNT 2009)

Uploaded by

Copyright:

Available Formats

Journal for the Study of the New Testament

Downloaded from http://jnt.sagepub.com by Eduardo de la Serna on October 26, 2009

Downloaded from http://jnt.sagepub.com by Eduardo de la Serna on October 26, 2009

Journal for the Study of the New Testament 31.3 (2009)

Downloaded from http://jnt.sagepub.com by Eduardo de la Serna on October 26, 2009

LONGENECKER Exposing the Economic Middle

Regional or provincial elites

Downloaded from http://jnt.sagepub.com by Eduardo de la Serna on October 26, 2009

Journal for the Study of the New Testament 31.3 (2009)

Downloaded from http://jnt.sagepub.com by Eduardo de la Serna on October 26, 2009

LONGENECKER Exposing the Economic Middle

Downloaded from http://jnt.sagepub.com by Eduardo de la Serna on October 26, 2009

Journal for the Study of the New Testament 31.3 (2009)

Downloaded from http://jnt.sagepub.com by Eduardo de la Serna on October 26, 2009

LONGENECKER Exposing the Economic Middle

Downloaded from http://jnt.sagepub.com by Eduardo de la Serna on October 26, 2009

Journal for the Study of the New Testament 31.3 (2009)

Downloaded from http://jnt.sagepub.com by Eduardo de la Serna on October 26, 2009

LONGENECKER Exposing the Economic Middle

Downloaded from http://jnt.sagepub.com by Eduardo de la Serna on October 26, 2009

Journal for the Study of the New Testament 31.3 (2009)

Downloaded from http://jnt.sagepub.com by Eduardo de la Serna on October 26, 2009

LONGENECKER Exposing the Economic Middle

Downloaded from http://jnt.sagepub.com by Eduardo de la Serna on October 26, 2009

Journal for the Study of the New Testament 31.3 (2009)

Downloaded from http://jnt.sagepub.com by Eduardo de la Serna on October 26, 2009

LONGENECKER Exposing the Economic Middle

Downloaded from http://jnt.sagepub.com by Eduardo de la Serna on October 26, 2009

Journal for the Study of the New Testament 31.3 (2009)

Downloaded from http://jnt.sagepub.com by Eduardo de la Serna on October 26, 2009

LONGENECKER Exposing the Economic Middle

Downloaded from http://jnt.sagepub.com by Eduardo de la Serna on October 26, 2009

Journal for the Study of the New Testament 31.3 (2009)

Downloaded from http://jnt.sagepub.com by Eduardo de la Serna on October 26, 2009

LONGENECKER Exposing the Economic Middle

Downloaded from http://jnt.sagepub.com by Eduardo de la Serna on October 26, 2009

Journal for the Study of the New Testament 31.3 (2009)

Downloaded from http://jnt.sagepub.com by Eduardo de la Serna on October 26, 2009

LONGENECKER Exposing the Economic Middle

Downloaded from http://jnt.sagepub.com by Eduardo de la Serna on October 26, 2009

Journal for the Study of the New Testament 31.3 (2009)

Downloaded from http://jnt.sagepub.com by Eduardo de la Serna on October 26, 2009

LONGENECKER Exposing the Economic Middle

Downloaded from http://jnt.sagepub.com by Eduardo de la Serna on October 26, 2009

Journal for the Study of the New Testament 31.3 (2009)

Downloaded from http://jnt.sagepub.com by Eduardo de la Serna on October 26, 2009

LONGENECKER Exposing the Economic Middle

Downloaded from http://jnt.sagepub.com by Eduardo de la Serna on October 26, 2009

Journal for the Study of the New Testament 31.3 (2009)

Downloaded from http://jnt.sagepub.com by Eduardo de la Serna on October 26, 2009

LONGENECKER Exposing the Economic Middle

Downloaded from http://jnt.sagepub.com by Eduardo de la Serna on October 26, 2009

Journal for the Study of the New Testament 31.3 (2009)

Downloaded from http://jnt.sagepub.com by Eduardo de la Serna on October 26, 2009

LONGENECKER Exposing the Economic Middle

Downloaded from http://jnt.sagepub.com by Eduardo de la Serna on October 26, 2009

Journal for the Study of the New Testament 31.3 (2009)

Downloaded from http://jnt.sagepub.com by Eduardo de la Serna on October 26, 2009

LONGENECKER Exposing the Economic Middle

Downloaded from http://jnt.sagepub.com by Eduardo de la Serna on October 26, 2009

Journal for the Study of the New Testament 31.3 (2009)

Downloaded from http://jnt.sagepub.com by Eduardo de la Serna on October 26, 2009