chap t e r 1

Introduction: Roman poverty in context

Robin Osborne

What are we studying when we study poverty? Are we studying the social
and economic structure that means that a proportion of the population has
barely adequate access to the resources required for life? Or are we studying
those in a society who at any moment happen to have less than some
particular, and more or less arbitrary, threshold of resources? Or again, are
we studying how the society in question analyses its own structure, how
it classifies those with least resources, what it does about them and how it
justifies to itself what it does or does not do?
Studying poverty in contemporary societies is closely linked to the ques-
tion of what to do about it; ‘make poverty history’ is the political slogan
of 2005. Doing something about it depends on understanding the nature
of the problem to begin with. Are the poor a random collection of people
who for different reasons have fallen on hard times but can be expected
to improve their lot in better times (‘conjunctural poverty’ as it is some-
times called)? Or are the poor trapped by the structure of economic system,
whether that be feudalism, capitalism, or whatever, so that in good times
as well as hard times they will remain impoverished (‘structural poverty’)?
Is poverty an economic problem (because a given society does not pro-
duce enough resources to go round), or is it a social problem (because the
resources are there but for social reasons are maldistributed)?
Understanding poverty in the contemporary world is inevitably a politi-
cal matter, and the politics do not always assist the understanding. For this
reason, it can help us to see the issues involved if we study poverty in a
historic society, particularly in one well removed from the roots of twenty-
first century social and economic problems. Studying poverty in the Roman
world – and in this volume we are primarily concerned with the Roman
world in the first four centuries ad – has a peculiar interest. The size of the
city of Rome – the first western city to reach a million inhabitants – created
issues of food supply quite unlike those faced by Greek city-states or even
the great Hellenistic cities, and the equally unprecedented size of Rome’s
1
2 robin osborne
empire meant that Roman government could both call upon an extraor-
dinarily diverse productive base and had responsibility for ensuring the
well-being of the isolated as well as of those at the centre. Rome thus gives
a case study in the sustenance of a population that is extremely unequally
distributed in a world where communications were slow and uncertain.
But Rome is also of particular interest because the arrival of Christianity
gives an opportunity to examine the impact of changing systems of belief
upon the classification of and attitudes towards the poor.

past work on roman povert y

‘There are no studies specifically on poverty in ancient Rome’. So C. R.
Whittaker, in his chapter on ‘The poor’ in Giardina’s collection originally
published as L’uomo Romano.1 Since these words were published in 1989,
poverty at Rome has begun to attract more attention. Peter Brown’s Poverty
and Leadership in the Later Roman Empire has brought to the forefront
of discussion issues of the changing position of and attitudes towards the
poor in late antiquity, which were flagged up long ago by Bolkestein and
reinforced by Patlagean. In addition, Marcus Prell has given us a socio-
economic study of poverty in Rome between the Gracchi and Diocletian.2
What is more, detailed work has been done on poverty in specific areas of
the Roman empire.3
Two related issues dominate discussions of the poor in the Roman world:
the emergence of the poor as a distinct social group, and the changing ways
in which poverty is represented and the poor are thought about. Although
throughout Greek and Roman history it was acknowledged that some men
were poor, only in the late Roman Republic and the imperial period did
poverty begin to be seen as a social and political problem which required
some sort of consistent and systematic treatment, and even then the poor
never came to constitute a distinct class.4 It was not until the early empire, as
Bolkestein stressed, that people began less to think of the poor as necessarily
morally corrupt and more to see giving monetary relief to the poor as a
virtue. Once this alteration in the view of the poor had occurred, the
beneficence which had earlier been bestowed upon communities generally,
and to which the work of Veyne has done so much to attract attention,
came to be seen as properly directed at the poor.5
1 Whittaker (1993) 299.
2 Bolkestein (1939); Patlagean (1977); Brown (2002); Prell (1997).
3 Hamel (1989); Holman (2001). 4 Prell (1997) ch. 3.
5 Veyne (1990); on which see Garnsey (1991b)
 Roman poverty in context 3
Much recent scholarship has repeated the idea that there was a move
from a civic notion of virtue, in which it was the general well-being of
the whole community which was promoted by the well-doing of the rich,
to a more narrowly economic definition of benefaction, in which largesse
consisting in money or consumable goods was bestowed specifically upon
the impoverished. There is no general agreement, however, about the date
of the change of attitude, and the reason for it. For Bolkestein, whose study
of pre-Christian antiquity embraced Egypt and Israel as well as Greece
and Rome, the change was visible as early as the first century ad, and was
consequent upon oriental influence which caused priority to be given to
poor relief in the Graeco-Roman world just as poor relief had been given
priority in Israel. Bolkestein thought it significant that Seneca, in Letters
to Lucilius 95.51, included giving a coin to the beggar and a crust to the
starving in an otherwise tralatician list of minimum moral demands on
any man (on which see Parkin, below p. 66). He noted parallels with Philo
and Josephus, and saw the mark of eastern influence.6 By contrast, for
Patlagean and for Brown this same change is a feature of late antiquity,
emerging ‘slowly in centuries that followed the conversion of Constantine
in 312’.7 But whereas for Patlagean the crucial factor was a massive change
in the structure of late antique society in general, partly consequent on
significant demographic change, for Brown, as his choice of 312 as a key
date indicates, the crucial factor was the influence of Christianity.8
One major weakness to date of work on poverty in the Roman world has
been the absence of any study which spans the whole period from Republic
to late antiquity: Bolkestein and Prell stop with the rise of Christianity,
Brown and Patlagean show no great interest in the Roman world in the
pre-Christian period. A second is that those who, like Bolkestein, Hands,
and Brown, interest themselves in attitudes to the poor tend to look only
superficially at what it was actually to be poor, while those who, like Prell and
Patlagean, interest themselves in the actual conditions of the poor pay little
attention to ideas about the poor. Peter Garnsey’s scholarship is marked by
a unique interest in the ways in which ideas played themselves out in prac-
tice – in the relationship between legal privilege and social status, in ideas
of slavery and the conditions of the slave, in how the nutritional value of
foods and conventional attitudes to foodstuffs relate to their consumption

6 Philo ap. Eusebius, Praeparatio Evangelica 358d; Josephus, Ap. 2.29.1; for some questioning of the
truth of this see Hands (1968) 84.
7 Brown (2002) 111.
8 Brown (2002) 75–6 for arguments against Patlagean. With Brown’s own position compare also the
conclusion of Prell (1997) 296.
4 robin osborne
and its consequences.9 In this collection of essays by his pupils, brought
together to honour Peter and to demonstrate something of what we have
learned from him, we attempt to bridge both the divide between poverty
as image and poverty as reality and the divide between earlier and later
Roman empire in a set of papers which discuss both the realities and the
representation of poverty in the Roman world both before and after the
conversion of Constantine. In this introduction I outline the big issues
involved by asking whether there was anything distinctive about poverty
in the Roman world, by asking how the representation of poverty at Rome
compares with the representation of poverty in the Greek world, and by
offering a synopsis of the chapters which follow.

was roman povert y d ist inct ive?

The Roman world was pre-industrial. Its economy was fundamentally based
in agriculture, and its population was largely rural. In modern terminology
‘the Roman economy was underdeveloped’.10 Life expectancy was low (life
expectancy at birth was somewhere between twenty and thirty and probably
closer to twenty).11 Nutritional deficiencies were widespread.12 But in none
of these features was the Roman world clearly distinct from the Hellenistic
world or from the world of the archaic and classical Greek city-state.
Poverty in this pre-industrial world was largely determined by access to
land.13 Those who owned, or were able to secure the rental of, land could
secure their subsistence provided that the area of land at their disposal was
large enough, and the climatic conditions favourable enough. How large
the plot of land needed to be has been much debated: it is clear that the
productivity of land is directly related to the labour put into it – gardening
is more productive per unit area than farming – but also that the law of
diminishing returns applies – repeatedly doubling the number of gardeners
does not repeatedly double the output of the garden.14 What counts as
favourable climatic conditions depends upon the nature of the land (‘the
grimness of the terrain’15 ) and the crops grown (barley can withstand drier
conditions than wheat). What it is possible or reasonable to grow, however,
will often, in turn, depend upon the relationship of the farmer to the
market: farming régimes that optimise the yield of the land in calorific

9 Garnsey (1970), (1988), (1996), (1999). 10 Garnsey and Saller (1987) 43.
11 See Scheidel (2001b) and chapter 3 below; cf. Brown (2002) 52, citing Simon Keay’s work on Tarraco.
12 Garnsey (1999). 13 Garnsey (1998) 201–13, esp. 213.
14 See most recently the papers in Van der Veen (2005), and especially Jones (2005).
15 Garnsey (1999) 1 (specifically of Palestine).
 Roman poverty in context 5
terms may not produce the kind of food a family needs to consume. In
general large landowners do better than small out of drought conditions, but
how badly the small farmer fares will depend upon access to the market.16
Many people, therefore, had reason to be anxious about food, but for those
who had access to land the threat of hunger was episodic, not endemic.17
Not all who were without land or access to land were impoverished. From
the eighth century bc onwards in both Greece and Italy there was signifi-
cant urbanisation.18 Although the proportion of the population employed
in craft activity or service industries of one sort or another never approached
the proportion employed in agriculture, nevertheless a significant number
of people was securely fed, and in some cases significantly enriched, by
non-agricultural activities. Towns were an important focus of such activ-
ities, though not the only one: those activities which depended upon the
exploitation of natural resources – above all mining – were necessarily
located in the countryside. Political developments further diversified the
possible sources of livelihood: at any one time a large number of mercenary
troops, infantry or rowers, were to be found in active service in the classical
or Hellenistic Greek worlds.
Since land was the main acceptable security for loans, it was hard for
those without land to achieve wealth, but in times of plenty all who were
able-bodied could expect to subsist. In the country even those who did not
own land could gather food from the land beyond cultivation.19 What was
gathered could be consumed directly or marketed in towns and villages. In
the town there were possibilities of casual employment that might involve
working alongside slaves but which would give an irregular income.20 For
the able-bodied, poverty was conjunctural.
Times of dearth divided communities between those who had and those
who had not managed to fill their storehouses. Those compelled to pay the
soaring prices of foodstuffs in the market quickly found their conditions of
life deteriorating as the need to secure food caused other economic activity
to contract. It was in such times that individuals were no doubt tempted to
sell themselves or their children into slavery – a practice legislated against
by Solon in Athens but still encountered by Augustine.21
For those who were not able-bodied, all times were times of dearth. The
disabled relied on the charity of their families, their friends, and ultimately
16 Garnsey (1998) 212. 17 Garnsey (1988), (1999) 2.
18 For a survey of early urbanisation in the Mediterranean see Osborne and Cunliffe (2005).
19 See, for twentieth-century Greece, Clark (1976), (1997), Forbes (1997).
20 Brown (2002) 50–51 on cities constructing a safety net for the destitute; but I am sceptical about his
claim that the real poor were in the countryside.
21 Brown (2002) 63 on Augustine.
6 robin osborne
of strangers. If they exhausted local charity and moved away to seek alms
from larger pools of beneficence they risked finding themselves isolated
from all with whom they had affective bonds. For such people, poverty was
structural.
Both in Greek city-states and during much of the Roman Republic polit-
ical status was of greater significance than levels of wealth. As a result, the
poor were not thought of as a distinct social group. It is true that Greek city-
states, including democratic Athens, and Republican Rome both restricted
certain economic opportunities (above all landownership) to citizens and
made certain political rights depend upon wealth. In this way rights of par-
ticipation might be curtailed, both theoretically and practically, by poverty.
However, citizenship and the legal privileges which went with it were for-
feited only by seriously unbecoming conduct. Citizens, however indigent,
remained distinct in their political rights from both free non-citizens and
slaves, and the possession of citizenship and freedom, in that order, were
ideologically, if not always practically, privileged over considerations of
wealth. The importance of political status that Finley saw as rendering
Marxist class analysis unsuitable for the ancient world ruled out the per-
ception, or self-perception, of ‘the poor’ as a particular group just as it ruled
out the development of a ‘working class’.22
What scholars call ‘civic’ models of poor relief are based on the privi-
leging of political status over economic need. The sharing out among all
citizens of the profits that had accrued to a polis is attested for the archaic
period, when the Siphnians shared the profits of the silver mines there,
and later, in the early fifth century, when Themistocles intervened to boost
the Athenian navy at the expense of such a hand-out in Athens.23 Acts
of beneficence (euergesia) by rich individuals towards their communities
are attested in Greek cities from the classical period onwards and become
increasingly prominent in later Greek epigraphy from the Hellenistic and
Graeco-Roman worlds. But ‘very few euergetists would have described what
they were doing as poor relief’.24
The principles of sharing out city resources were applied also to the
sharing out of grain. At times of crisis city magistrates might be charged
with buying grain, and might distribute it at a fixed price, but the prin-
ciple of distribution was that it was to citizens.25 However, it is with the
question of grain distribution and its recipients that we encounter Roman
22 Finley (1973) 49.
23 Herodotus 3.57.2 for Siphnos, 8.144.1–2 and [Aristotle] Ath. Pol. 22.7 for Themistocles; Humphreys
(1976) 145 for the general principle.
24 Garnsey and Saller (1987) 101. 25 All this definitively documented in Garnsey (1988).
 Roman poverty in context 7
distinctiveness. Finley observed that at Rome the decision in 58 bc to dis-
tribute free grain again restricted recipients to citizens, but he stressed that
in this instance ‘the ancient sources are unanimous in their view of the dole
as a form of poor relief won by the plebs after considerable struggle’.26 Why
was grain distribution regarded like this at Rome when it had not been so
regarded in other cities of the Greek or Roman world? Despite emphasising
the exceptional nature of this Finley offers no discussion of the reasons for
the exceptional conception of grain distribution at Rome.27 Two factors
can, however, surely be isolated. One is the sheer size of the population of
Rome in the late Republic, the other is the potential political power of the
Roman poor. Each of these demands some further discussion.
The economic impact of Rome’s unprecedented size was first drawn to
ancient historians’ attention by Keith Hopkins, in an unpublished paper,
and it has been set out in detail by Neville Morley.28 The concentration
of people in Rome created demands for both foodstuffs and other basic
necessities of life, such as clothing and housing, and also for the goods
required to secure and display status in a place where all ranks of society
gathered. A city of a million inhabitants that was the centre of an empire
extending all round the Mediterranean and beyond was quite unlike any
other town or city. Along with Rome’s peculiar demands for goods went
also demands for labour, not least to sustain a supply system that had to
draw on the surplus of a much wider area than any other city and to ensure
that the goods required reached those who needed them.
As far as the way in which the poor were perceived and perceived them-
selves is concerned, however, what was important about Rome was not
that its economy was differently configured but that the sheer number of
citizens present in Rome meant that the fiction of the citizen state could
no longer be maintained. As recent work has made ever more clear, only
a tiny proportion of citizens resident in Rome could ever physically cast
their vote in a Roman voting assembly, let alone have their votes make any
difference to the result.29 As Aristotle had pointed out, if a population grew
to beyond a few thousand citizens the organisation of the city-state would
be threatened, since no herald would be physically able to address them all
(Politics 1326b). The citizen population of Rome could no longer envisage

26 Finley (1973) 170–71 with second edition (1985) 201; and cf. 40.
27 Finley (1973) 201–2 devotes rather more space to the question of the reasons for Trajan’s alimenta
schemes, withdrawing his initial support for Veyne’s view that the motivation was demographic and
preferring to see the projection of the emperor’s power as the crucial factor.
28 Morley (1996). Hopkins was inspired by Wrigley (1967) on London.
29 Mouritsen (2001), engaging with Millar (1998).
8 robin osborne
itself as a distinct community when it could neither gather together in one
place nor engage together in even the most minor of political activities.
Sheer weight of numbers crushed both the distinction between citizens
and other urban residents and the political machinery invented for a small
town. The breakdown of the political machinery manifested itself in the
politics of violence, the destruction of the distinction between citizens and
other urban residents manifested itself in the birth of the poor. It is no
accident that the Clodius who introduced the free grain dole was also the
prime exponent of political violence.30
But if the sheer size of Rome made it inevitable that the meaning of
citizenship would be transformed, it was Roman imperialism that spread
awareness of, and self-awareness among, the poor, and in two ways. First,
the incoming wealth of empire encouraged everyone to have higher hopes of
material riches. ‘Debates over poverty . . . tend to flourish in the context of
rising expectations.’31 Second, in order to ensure that Rome could raise the
size of army required to maintain and expand its empire, Rome abolished
the traditional requirement that to serve as a soldier one had to possess a
certain (gradually reduced) level of property.
Rome’s need for military manpower on a scale, both in terms of numbers
and in terms of length of service, quite different from that of any Greek city,
impacted directly upon the economic and political ambitions of the citizen
body. The lowering and eventual abolition of the property qualification
for legionary service during the second century bc fundamentally altered
the relationship between the army and the land.32 It also meant that at the
end of every military campaign poor Roman citizens were in a position,
with the minimum of organisation, to make their presence felt in such
numbers that traditional means of expressing political views, such as the
ballot box, became irrelevant. Although Catiline’s conspiracy seems in the
end actually not to have mobilised the poor in significant numbers, and
although many of Clodius’ activities themselves relied not upon the poor
but upon slaves, the potential that had been feared in 63 bc and was then
enabled by the tribune’s legislation of 58 bc was real enough. Other cities
needed to provide a cushion for their whole population only in times of
crisis in the grain supply; at Rome, by contrast, the abolition of property
qualifications for military service led to an identification between legionary
and landless such that there was a permanent need to provide subsidised
food for the landless citizen poor.

30 On violence in Rome see Nippel (1995). 31 Shaw (2002) 43.

32 So, famously, Brunt (1962/1988).
 Roman poverty in context 9
Augustus famously acknowledged the political importance of grain dis-
tributions when he refrained from abolishing them on the grounds that they
were bound to be reintroduced at some point per ambitionem (Suet. Aug.
42.3). The senatorial aristocracy in the late Republic and the new régime
in the early principate cashed in grain for power. In the late empire, as the
city of Rome itself lost its overwhelming dominance, it was a more gen-
eral concern for the poor that emperors cashed in for power. Peter Brown
insists on the continued importance of political interests when he argues
that the emergence of a discourse on the poor in the fourth century was
directly related to the need of the new (Christian) imperial state to assert
its presence.
It is hard to separate the transformation that the size of Rome brought
about in the effective civic status of its poor inhabitants from that which
it wrought in their material conditions. For all the importance of urban
neighbourhoods, clichés about the deracination and anonymisation of the
individual in the metropolis retain their force: Rome remained notable
into late antiquity for the presence of a population living ‘informally in the
crevices of the towering buildings, sleeping rough in tabernae or huddled
in the vaults beneath the seating of theatres, circuses and amphitheatres’.33
On the one hand, the system necessary to provision the huge urban popu-
lation inevitably involved a level of wastage sufficient to support significant
numbers; on the other, those who wanted to be regarded as the greatest
men in the city had not only to cream off the wealth of empire to build
houses and gardens of extraordinary luxury but also to be seen to have
throngs of men dependent upon them. The princeps sustained his position
as primus inter pares by ensuring that the calendar of the poor as well as the
well-to-do was structured around festivals and events that were linked to
himself and that brought material as well as immaterial pleasures to all.34
If the growth of Rome and Roman imperialism had already destroyed
the civic ideal in Rome itself by the late Republic, that ideal continued
to thrive outside Rome. Some of the clearest manifestations of civic bene-
faction come from the cities of Italy and the Greek east in the first and
second centuries ad. But as the Roman world gradually transformed itself
from a collection of semi-autonomous cities subordinated to the power
of an alien Rome, and to a single political and economic unit, the civic
ideal came under pressure outside Rome also. The provincial elite were
incorporated into central government through recruitment to the senate

33 Purcell (1996) 784, citing Ammianus Marcellinus 14.6.25.

34 Purcell (1996) 799–806.
10 robin osborne
or to the imperial service: ‘provincial wealth flowed to Rome as they pur-
chased houses on the Esquiline and in other fashionable areas and set up
their considerable establishments’.35 Initially such men continued to wish
to display themselves to their native communities through benefactions,
but the more local citizenship came to be a matter of obligations rather
than of opportunities, the more the old civic idealism became irrelevant
to the way in which people’s lives were organised and envisaged. Division
between those who were in a position actively to participate in imperial
rule (essentially the wealthy and those who served in the legions), and those
who were not, became formalised already under Hadrian in the distinction
between honestiores and humiliores.36 When Caracalla extended citizenship
to all free-born inhabitants of the empire in the Constitutio Antoniniana
of ad 212, the civic model was doomed. Where there was no distinction
of political status to back them up, distinctions of social status could not
survive unless they were also distinctions of economic status. The death of
the city-state inevitably brought about the birth of the poor.
Looked at from the bottom up, the Roman world was recognisably the
same under-developed world as the world of classical Greek city-states or
Hellenistic kings. Political unification had an economic impact, reducing
the risks and therefore the costs of long-distance transport. This speeded
up the ‘brownian motion’ which had, as Horden and Purcell have taught
us, long been a feature of the Mediterranean’s corrupting sea, and so
enabled both the primate city and the leading men to become far wealth-
ier. But those changes occurred within an unchanged economic structure
within which even the achievement of per capita economic growth is
debated.37
The revolution which was effected by the Roman empire was not eco-
nomic (or socio-economic) but political (or socio-political). Roman con-
quest and Rome’s own revolution from city-state to imperial power brought
about the slow decline in the domination of the civic ideal over the self-
perception of the free inhabitants of the empire. The habit of defining
oneself in contrast to various Others, which has been seen as so central
to classical Greeks,38 could no longer be sustained when some of the fun-
damental divisions upon which it rested were first effectively and then
formally dissolved in a world empire. As the myth collapsed according to
which the citizens of each city-state were peculiar and particular, a myth
which had successfully prevented material circumstances from bringing

35 Edwards and Woolf (2003b) 11. 36 Garnsey (1970) ch. 11.

37 Garnsey and Saller (1987) 51–63. 38 Cartledge (1993).
 Roman poverty in context 11
about divisions among those citizens, so material divisions in the condition
of life imposed themselves. Imperial, philosophical (Stoic) and religious
(Christian) visions of world citizenship had to find new ways of coming to
terms with economic and social variety within that world vision.

representing povert y in the greek and roman world

What impact did the distinct manifestation of poverty in the Roman world
have on how its richer inhabitants saw and related to the poor? Rome
not only came to deal with the poor and with poverty in ways that were
distinctly different from those prevailing in the Greek city but also to think
about poverty and the poor differently.
It is a striking feature of discussions of poverty in Greek texts that poverty
is always relative.39 Of the various terms available to describe the poor, none
is attached to any absolute level of destitution. In Aristophanes’ Plutus the
personification of Penia, the term normally translated ‘poverty’, reacts to the
suggestion that she is the sister of Ptocheia, the term normally translated
‘beggary’, by saying that the life of a ptochos ‘is to live having nothing’,
whereas ‘the life of a penes is to live a sparing life, working hard, with
nothing to spare but not falling short’ (lines 552–4). But Menander, in his
play Dyskolos will have Gorgias describe himself as a ptochos even though
he owns land (lines 284–6). Demosthenes can even describe as ‘without
means’ (aporoi) men who belonged to the liturgical class in Athens, the
richest 10 per cent of the citizens (18.108). Although it is generally true to
say, as Finley does, that ‘Penia, in short, meant the harsh compulsion to
toil, whereas the pauper, the man who was altogether without resources,
was normally called a ptochos, a beggar, not a penes’, neither of the terms
was sufficiently laden with associations with a particular level of need to
prevent its use in quite other circumstances.40
In as far as ‘the poor’ constitute a political group for classical Greek
writers, it is as the majority who are not rich. Aristotle insists that it is
rule by the wealthy that constitutes oligarchy, rule by those without means,
the aporoi, that constitutes democracy, and that this would be true even if
the wealthy were in fact the majority and the aporoi the minority (Politics
1279b26–1280a6). In the case of democracy, the important point is not
that those who are resourceless in fact dominate it, but that resources
are, in theory, irrelevant to political power in a democracy. Similarly, the
dominance of oligarchy by the wealthy derives from the fact that wealth

39 See more generally Hands (1968) 62–76, 77–88. 40 Finley (1973) 41.
12 robin osborne
defines eligibility for full citizenship in such a régime, whereas there is no
property qualification for citizenship in a democracy. Plato had already
remarked that the use of property qualifications led not only to power
not necessarily being given to those who could make best use of it, but
also to those below the property qualification having no reason not to sell
real estate. As a result, he observes, they may cease to have any stake in
the city and end up destitute, so that the city becomes divided into two
groups, those with and those without resources (Republic 551b–552b). For
the ideal community of the Laws Plato legislates to make land inalienable
and to prevent the wealthiest becoming more than four times as wealthy
as the poorest (Laws 744d–745a). It is the political effects of differences of
wealth, not the problem of absolute poverty, which exercises both Plato
and Aristotle.
The issue of the really destitute arises when Aristotle considers what
makes a democracy durable. He is critical of redistributing state surplus
to the destitute, on the ground that they will spend and remain destitute
(‘such assistance for the aporoi is a large jar with a hole in the bottom’). He
recommends instead that redistribution should be undertaken sufficient to
enable those without means (the aporoi) to acquire land or set themselves up
in business (Politics 1320a17–1320b3). Aristotle’s concern here is not with the
welfare of the destitute as such, but with the political behaviour consequent
on there being some within a community who are heavily burdened with
taxes, and others who have no resources of their own and rely on state pay
and other handouts. Plato had suggested that indebtedness was the primary
cause of revolution from oligarchy to democracy, provoking Aristotle to
point out that this is not the only source of such a change (Politics 1316b6–
27).
The redistribution mechanisms to which Aristotle refers themselves con-
firm the absence of concern with a distinct group of really resourceless
people. Cities of democratic persuasion might offer pay for taking public
office, pay for attending public meetings, pay for military service, and var-
ious free handouts at public festivals. All of these distributions were made
to the citizen body in general, without any redirection of them specifically
to the needy. Only in the case of those who were disabled, did democratic
Athens recognise a case for meeting a manifest need with targeted help.
At one point in the Laws Plato argues that ‘if the state and society he lives
in is run with only average skill’ no virtuous person will ever be reduced
to ‘final ptocheia’ – and in consequence he makes a law that beggars shall
be expelled from the ideal city (936b3–c7). That denial that there were
any virtuous poor was made easier by the long-standing Greek habit of
 Roman poverty in context 13
describing the wealthy as ‘good’ and ‘best’ (chrestoi, beltistoi), the poor as
‘bad’ and ‘worse’ (poneroi, kheirous).41 Such terminology, equating virtue
and worldly success, only began to be challenged in the fifth century, and
some remnants remain in fourth-century authors. The idea, fundamental
to Hesiod’s Works and Days, that if a man is prepared to work honestly
and hard he will be able to provide for himself adequately, lies consistently
behind classical Greek texts. In consequence, not only are those who really
are poor necessarily not themselves good, but poverty itself cannot be or
breed virtue.42
Despite the concerns of Plato and Aristotle, and despite the existence
of ‘abolition of debts’ and ‘redistribution of land’ as revolutionary rallying
cries, it remains unclear how important a role poverty and the poor played
in the practical politics of classical Greece.43 By contrast, in Rome, from
at least the middle Republic, poverty plays an important part in political
discourse and the poor have a significant role in practical politics.
The place of the poor in Roman political discourse seems to be initially
linked to the invention of the virtuous poor man. Greek writers sometimes
criticise a life of truphe, ‘luxury’, or express nostalgia for the simple country
life (e.g. in plays of Aristophanes), but they never extol the life of the poor
man as in any way exemplary. The message of Hesiod’s Works and Days, that
the hard life of labour is imposed upon men by the gods, and it is for men
to knuckle down and make the best of it, is the message that runs through
classical Greek texts. Luxury is associated not simply with wealth, but with
non-Greeks, particularly with the east, and the opposition is not between
luxury and poverty but between ‘barbarian’ and Greek behaviour.44 By
contrast, the Romans had already in the middle Republic developed the
image of the virtuous hard-working citizen, who had no time for anything
except earning his living on his farm and doing his civic duty. The truth
of the exemplary stories (e.g. of Cincinnatus or of the Elder Cato) does
not matter; the importance is that those stories were told, and both are
clearly part of the stock of exempla doing the rounds in the late Republic
(Cic. Sen. 56 for Cincinnatus; Plut. Cat. Mai. 3.1–3 for the Elder Cato).
Although the exemplary Cincinnatus and Cato were hardly destitute, the
honour that they bestow upon the labouring life which enjoys no luxury
offers the foundation upon which a positive evaluation of poverty can be
built. And that positive evaluation we find in such fictions as the speech
41 See further Osborne (2004) 11–12. 42 cf. De Ste Croix (1981) 425–6, 431–2.
43 Social factors in political unrest have been stressed by Fuks (1984) and minimised by Gehrke (1985).
See the balanced review by Austin (1994) 528–35.
44 Hall (1989) 81–3, 126–9, 209–10.
14 robin osborne
which the Elder Seneca puts into the mouth of Arellius Fuscus, discussed
below by Greg Woolf, in which poverty is extolled as the best defence
against the corruption of riches.45
The discourse of the corruption of riches is well represented in Sallust’s
Catiline. In the introduction to that work, Sallust presents riches as the
root of all evil. The state whose success is built upon justice and hard
work is undermined by leisure and wealth, which become an impossible
burden (Cat. 10.1–2). Desire for money is followed by desire for power
and all manner of evils follow as greed undermines honesty and loyalty
(Cat. 10.3–4). As riches are themselves honoured and become the root of
power and glory, poverty becomes seen as criminal and established values
are overturned (Cat. 12.1–2). This then attracts all who are resourceless, hate
the status quo, and desire change, since change can bring them no loss –
the group in question, Sallust says, includes ‘practically the whole plebs’
(Cat. 37.1–3). Men who have seen others around them becoming wealthy
are moved by the desire to reverse their own misfortune and looking to
their own material interest prefer the handouts they can get in the city to
honest labour (Cat. 37.4–8).
Analyses of what goes wrong with a constitution in terms of corruption
and greed are familiar in Greek sources from Thucydides on civil strife
(stasis) (3.82.8) through to Polybius (6.57), and greed plays a catalysing role
in Aristotle’s analysis of stasis (Politics 1302b; cf. Plato Republic 555b).46 But
wealth and poverty play a much more prominent part in Sallust’s analysis
than in any Greek text.47 Although Sallust is not himself consistent in his
description of the reactions of the city residents to Catiline,48 there is little
doubt that the picture he paints in Cat. 37 puts so much stress on the role
of the economic position of Catiline’s urban followers precisely because
poverty had become a political issue.
Finley observed that ‘Not even the state showed much concern for the
poor. The famous exception is the intensely political one of the city of
Rome.’49 But the fact of the exception is crucial: in Rome the poor had
become a political force as they had never been in any other city. As we
have already seen, the plebs frumentaria created by the grain dole recognised
and gave an identity to a large body of more or less impoverished citizens

45 Arellius’ formulation suggests that Finley’s claim that ‘Fundamentally . . . “Blessed are the poor” was
not within the Graeco-Roman world of ideas’ (1973: 38) is wrong: it is not within the Greek view,
but it is within the Roman.
46 Balot (2001) 46–8.
47 And are more prominent in the Catiline than in the corresponding passage (ch. 41) of the Jugurtha.
48 Compare Cat. 31 and Cat. 48. 49 Finley (1973) 40, cf. 171.
 Roman poverty in context 15
with a political voice.50 Emperors’ toleration of popular protest about food
shortages stemmed in part from the opportunity that solving perceived
crises gave to display imperial power; but at the same time imperial officials
often listened to what the crowd said, and responded positively.51
There is no sign that the poor of late Republican Rome came to be
considered to have any greater moral claim to support from the more well
off than had been possessed by the poor in any Greek city. Their political
power did not make them virtuous, and writers and politicians continued
to treat them as the dregs of society, responsible for their own destitution by
their own moral failings. But what the political power of the poor did was
to draw attention to the contrasts between rich and poor, between those
whose unusual political power gave them wealth and those whose common
destitution gave them political power. It remains as true for Rome as it was
for Athens that ‘poor’ was a relative term, open for persuasive definition and
ascription according to context, and even more open to remaining vague
and ambiguous. It remains true for Rome that the poor were more often
a topic for thinking with than a practical problem to be solved. It remains
true in Rome, as it was in Athens, that there was only a discourse of wealth,
not a discourse of poverty. But for all that, the invention of the poor as a
political problem had a profound effect on the ways in which life was lived
and theorised. In classical Athens, the moment when buying expensive
fish led to suspicions of aiming for tyranny passed, and even the eastern
connotations of the luxury lifestyle came to be positively appreciated.52
At Rome by the late Republic excessive luxuria had come to create an
expectation of both moral and political depravity.53

the shape of this volume

It is this new world of the poor and distinctive Roman attitudes towards
poverty and representation of the poor that the chapters in this collection
proceed to explore. They take up the story of poverty and its representation
in the Roman world from the beginning of the principate and trace issues of
what it was to be poor through until the great late Roman law codes. They
revisit and review from various angles the question of the impact upon the
poor of the peculiar life of the city of Rome and the unprecedented size and
coherence of its empire, on the one hand, and of the spread of Christianity
on the other.

50 Garnsey (1998) 237–9; (1988) 211–14, 236–43. 51 Garnsey (1988) 244 for both points.
52 Davidson (1993); Miller (1997) chs. 8–10. 53 Edwards (1993).
16 robin osborne
We begin with Neville Morley’s discussion of poverty in the city of Rome.
Morley takes up the discussion of the politics of Roman poverty from that
offered above, expanding the discussion not only of the politics of poverty in
the Roman world but also of the politics of poverty in scholarly writing since
David Hume. He looks closely at the various ways in which we might define
poverty and at the problems of finding in Rome the people so defined. He
lays emphasis on the poor as a social and cultural, rather than an economic
group, picking out in particular their characteristic vulnerability, exclusion
and the shame that attaches to poverty. In the final section of his chapter
he explores what a history of the Roman poor might look like, asking
how the various changes in civic organisation and the economy during the
principate affected the poor. His final words bring the discussion full circle
by noting how the politics of poverty in the late Republic and early empire
provided a resource on which nineteenth-century discussions of the poor
were able to draw.
Walter Scheidel takes up in Chapter 3 Morley’s challenge of seeing the
poor in Rome in relation to more recent discourse of poverty by situating
the study of poverty and the poor in the Roman world in the context of
studies of the poor more generally. He sets the poor within the sociology
of the Roman empire and within the debate over the ways in which the
formal ranking of Roman society did or did not make for stratification
into separate classes. He argues against the dichotomisation of Roman
society and against the notion that all Mediterranean societies have always
been characterised by extreme inequality of land ownership and large-scale
patronage. In the face of a prevailing view that there was no significant
‘middle class’ in the Roman world, he attempts to show, by quantifying the
different census classes across the empire, that there was in fact a substantial
‘middling’ group. Our whole conceptualisation of the society of the Roman
empire generally and our understanding of the world presupposed by our
literary texts are at issue here. Scheidel then goes on in the final section
to consider the difficulties of assessing living standards; he suggests that
Roman Italy was ‘developed’ in rather different ways from classical Athens,
and that Roman Egypt was significantly different again.
Anneliese Parkin (Chapter 4) turns attention away from what made the
poor poor, to the ways in which the condition of the poor was relieved.
In discussing pagan almsgiving she insists that although ‘the generosity
of Veyne’, in Peter Garnsey’s own phrase, was not aimed at the destitute,
the destitute were not in fact merely ignored before Christian charity was
directed towards them. Parkin examines the philosophical discussions about
giving to the poor, noting that Stoic resistance to pity for the poor went
 Roman poverty in context 17
together with a willingness practically to help them. Similarly the insistence
that help should be given only to those who are able in some way to
reciprocate did not mean nothing was given to beggars, whose continued
existence indicates otherwise. But she suggests that much of the giving to
the destitute may well have come from people who were not themselves
among the elite and who may have been little affected by philosophical
arguments, and whose giving may well have gone along with a certain
disgust at the beggars themselves. Almsgiving should not be seen as purely
a moral matter: fear of the beggar may itself have played a part. However,
late legislation to outlaw begging by the able-bodied and to divide the poor
according to their labour capacity suggests that beggars needed to be in
some sense pitiable.
Greg Woolf’s discussion ‘Writing Poverty in Rome’ in Chapter 5 turns
to the question of the literary image of the poor in Rome – above all in
the early principate. Woolf faces up to the question of how the ‘realism’
of literary fictions can be deployed for historical purposes, and insists that
understanding the relationship between literature and life is as essential to
understanding literature as it is to understanding life. Woolf argues that in
the early principate there was no single discourse of poverty, but that poverty
was a topic thought about in the context of the dominant discourses, such
as those on wealth and on luxury. Woolf explores the ways in which poverty
was treated as ‘unwealth’ in particular in the poetry of Martial, and argues
that the persona of poverty was attractive to Martial in part because the
negative condition of not being wealthy covered so great a social range
and left the reader having to decide the degree of honesty or irony to be
read into any particular claim. But he also argues that this poetic play with
poverty had an effect on the destitute, who were depersonalised and treated
as eminently ignorable.
Dominic Rathbone’s chapter (6) moves the discussion from Rome to
Egypt, but keeps the issue of image and reality in the centre of the dis-
cussion. Is the invisibility of the poor and problems of poverty in Egypt
during the early principate and their visibility in late antique Egypt a prod-
uct of increasing poverty, of increasing visibility for the poor who were
always present, or of a Christian invention of poverty? Rathbone looks at
the evidence for widows and their condition, arguing for the possibility
that widows did not remarry because they had sufficient means to remain
independent, and at the evidence for standards of living, arguing for relative
prosperity. Although he finds some reason to suppose that conditions did
worsen in late antiquity, he suggests that the prominence of poverty in the
Christian source material gives undue emphasis to poverty as a problem.
18 robin osborne
With Sophie Lunn-Rockliffe’s discussion in Chapter 7 of Ambrosiaster’s
treatment of almsgiving we turn firmly to the Christian sources. Lunn-
Rockliffe’s analysis further illuminates the concern of the late antique
church to offer space – and salvation – to the rich. She shows how Ambrosi-
aster notably avoids tussling with those scriptural passages which con-
demn riches. But he is prepared, as writers such as Clement had not
been, to acknowledge that wealth needs to be taken into account when
assessing other actions – an acknowledgement which builds, at least in
part, on the allowance for status made in Roman law. Lunn-Rockliffe
argues that Ambrosiaster’s position cannot be understood on the assump-
tion that there were monolithic ideologies of poverty and wealth. She
suggests that the poor could be both disdained, because of their phys-
ical condition, and admired, because of their spiritual wealth, and that
the question of how means affected virtue was explored in a sophisticated
way.
Richard Finn in Chapter 8 takes further the issue of what can and should
be concluded about the relationship of texts to the realities of impoverish-
ment and destitution, now in relation to Christian texts. Finn argues that
the visibility of the poor in late antique Christian texts should not be exag-
gerated, and that attention needs to be paid to the instances in which the
poor are unreasonably absent as well as instances in which they are the
focus of attention. In an analysis of Augustine’s Enarrationes in Psalmos he
suggests that the contrast between the low incidence of encouragement to
almsgiving in the brief exegetical notes on Psalms 1–32 and the high inci-
dence in the expositions of Psalms 33–98 is a consequence of promotion of
almsgiving being one of the prime duties of a bishop. Finn analyses the way
in which Augustine draws the materially poor man into relationship with
the spiritual needs of all men, so breaking down the distance between rich
and poor. There are some reflections of an insecure and uncertain age in
these expositions, but avoiding detailed descriptions of the poor and using
the term beggar only infrequently to refer to the recipient of alms helped
keep small the perceived distance between rich and poor. Finn suggests that
the attitudes of the well-off can be read between the lines of Augustine’s
sermons and the strategies of argument that Augustine chooses to employ.
The much greater visibility of the poor in saints’ lives, in terms of descrip-
tions of their circumstances, must also be read in the light of the purpose
and readership of these lives: they achieve their effects by replaying the
episodes of Christ’s encounters with the poor in the Gospels, and hence
emphasising the Christlikeness of the saint, as much as, or more than, by
shocking the reader with recognition of daily reality.
 Roman poverty in context 19
It is with a saint’s life that Lucy Grig begins her discussion in Chapter 9 of
the parties thrown for the poor that figure prominently in some late Roman
sources and which have been seen by some as central to Christian charity.
Grig analyses the literary construction of these party stories in an extended
treatment of Paulinus of Nola’s thirteenth letter. That letter spends much
time on the sumptuousness of the setting of the ‘poverty party’, the basilica
of St Peter’s in Rome, and Grig contrasts Paulinus’ belief in the glorification
of God through material splendour as well as through charity with Jerome’s
belief in the absolute priority of the poor and with Ambrose’s use of the
story of St Laurence, who presented the poor as the riches of the church,
to justify giving away the church’s material wealth. Grig notes that there
was a sense in which the church relied upon desire for both material and
spiritual riches and in which the poor played only an instrumental role in
the church’s courtship of the elite.
The world of Paulinus of Nola contrasts strongly with the graphic pic-
ture of the perils of life in fifth-century Gaul painted in the diatribe De
Gubernatione Dei written by Salvian, which is the subject of Cam Grey’s
chapter (10). Salvian’s work is an argument, rather than a description, but
his themes of the responsibility of those in power, the importance of reci-
procity in vertical social relations, and the need for communities to have a
unity of purpose, reflect the issues of the day. For all that his generalisations
about the plight of the poor are unlikely to be an accurate reflection of the
circumstances of his time, the Theodosian and Justinianic codes, too, are
concerned to regulate patronage and labour relations.
It is to the world of late Roman law that Caroline Humfress turns in
the final chapter (11). Humfress examines Marcian’s Novel 4 and asks what
relationship the greater prominence of the poor in late Roman law has
to the changing conditions of the late Roman world. She argues that the
poor of Marcian’s text have to be understood as the relatively poor, the
middling rather than the destitute, and that throughout late Roman law
there is no single category of ‘the poor’ but each reference to a poor person
has to be interpreted in context. Much late Roman legislation which bears
upon the poor was not designed to alter the conditions of the poor but was
concerned with mitigating the effects of poverty (e.g. the killing or selling
of children). The regulation by law of what could and could not happen
to bequests is in fact evidence that ways were often found to divert such
bequests from the poor, and there is much evidence for on-going prejudice
against the poor even within the church. Alongside evidence for men falsely
claiming poverty in order to avoid various duties, there is also evidence of
poverty being administered as a penalty. The relativity of poverty made
20 robin osborne
it difficult or impossible to use ‘the poor’ as a legal category. Right until
the end of antiquity, therefore, the political and moral force of claims to
poverty prevented the formation of a coherent social group of those who
were really destitute.
Walter Scheidel’s chapter concludes with the observation that the ques-
tions which dominate development studies have hardly impinged on studies
of the ancient world. Throughout this volume the contributors have ges-
tured towards lines of enquiry and ways of thought which if pursued further
would make us look very differently at poverty in the Roman world. This
volume is not a collection of definitive studies, but a summary of current
understanding, an attempt to survey and define a territory which has to
date been under-explored. What is claimed here remains open to revision
as Roman historians engage more fully with the lessons that can be learnt
from analysis of more recent and contemporary societies. It is such a pro-
ductive engagement between questions generated by the study of more
recent societies and material derived from the Graeco-Roman world that
has marked Peter Garnsey’s own research. And just as his examination of the
ancient world has served also to sharpen awareness of issues in the modern
world, so we hope that this volume also offers insights into the relationship
between reality and representation, ideas and actions, that will themselves
enlighten contemporary engagement with poverty and the poor.