Kept in the dark
The Archaeology of Slaves and Slavery in the Late Roman Republic and Early Empire
Cover illustration
Object type Tag
Museum number: 1975,0902.6 (British Museum, London)
Description: Bronze tag for a dog or possibly a slave, inscribed with information about return. Culture/period: Roman
Date: 4th century AD Material: bronze
Dimensions: Diameter: 5.8 centimetres Inscription Language: Latin
Source
http://www.britishmuseum.org/research/collection_online/collection_object_details.aspx?objectId=46 3968&partId=1. Last visited 10-11-2014
Inscription Content TENE ME NE FUGIA ET REVO CA ME AD DOMNUM EVVIVENTIUM IN ARA CALLISTI Inscription Transliteration
Tene me ne fugia(m) et revoca me ad dom(i)num Viventium in ar(e)a Callisti Inscription Translation
Hold me, lest I flee, and return me to my master Viventius on the estate of Callistus
The interpretation of this tag, of which 36 are known1, as pertaining to a dog or to a slave shows how little is known for certain about slaves in the Roman world. According to F.H. Thompson this bulla was worn by a slave (Thompson 2003, pp. 239-240). The British Museum has kept options open.
Felix
And now you are mine Until the day I die And forever after
I will name and cloth you House and lock and Teach and flog you And when time comes I may even love you For you are mine
I cannot live without you
© René Brandhoff, 2014
the things we hope for may already have happened without us knowing
tell us this is not a dream
Acknowledgements
The years at the University of Groningen have been made unforgettable and possible by the moral support of my children Rob and Anne who gave me the “go for it” in 2008, which was crucial to me. Also special thanks to Rob for spelling and grammar corrections in this thesis. Two people of the Groningen Institute of Archaeology (GIA) need to be mentioned here. Dr. Elisabeth van ‘t Lindenhout, students advisor at the time, was my first contact with GIA in October 2008 and the one to clear away my last bits of doubt to start the whole enterprise of “returning to school”. During the years of study Dr. Bert Nijboer was an inspiration as a teacher in various lectures and at the Crustumerium excavation tours and my supervisor for the bachelor thesis and mentor during the research master.
Studying archaeology is also about working together with your fellow students. I thank all of them for making me feel part of the community. Also here two people stand out. Pieter Swart and Theo Verlaan added an extra dimension to the university years in many discussions on archaeology, facts of life and total nonsense. Thank you for your friendship.
Last but not least I would like to thank my thesis supervisor Dr. Lidewijde de Jong for her patience and guidance and Dr. Tymon de Haas for his co-readership.
Content
Introduction ...7
1. The slave community; the subaltern, identity and humans as objects ... 12
2. Slavery in the Roman society ... 17
2.1 Citizens and slaves ... 22
2.2 Defining slaves the Roman way ... 23
2.3 From slave to freedperson ... 24
2.4 Demography; influx of slaves into Roman society ... 25
2.5 The economic significance of slavery ... 27
2.6 The slave in classic texts ... 29
2.7 The slave in Roman law ... 33
3. The archaeology of slavery ... 36
3.1 Mortuary practice; Roman citizens, slaves, freedpersons ... 40
3.2 Representations in art ... 45
3.3 Tools and restraints ... 50
3.4 Slaves in the rural landscape; villas ... 52
3.5 Slaves in the urban landscape; urban housing ... 55
Introduction to case studies ... 59
4. Settefinestre, a villa in Roman Etruria; location and building ... 60
4.1 Settefinestre; the material ... 69
5. Pompeii, urban housing and slaves ... 73
5.1 The House of Menander ... 79
5.2 Archaeological finds in the servile area ... 81
5.2.1 Room 35 ... 82
5.2.2 Room 37 ... 82
5.2.4 An inscription in the servile area ... 84
5.3 The House of the Labyrinth ... 85
6. Roman slaves; towards a contextual approach. ... 89
6.1 The contextual approach... 91
6.2 Parallels for slavery systems ... 92
6.3 Bio-archaeology ... 93
6.4 From bad to worse? The poor and the slave ... 94
6.5 The scorched city; looking beyond the Empire ... 95
6.6 The freedpersons road to full Roman identity ... 96
6.7 Geophysical research... 96
7. Conclusion ... 98
Introduction
Antique slavery is usually described as a phenomenon driven by economic motives and current knowledge is primary sourced from historic texts and inscriptions. The history of slavery in the Mediterranean world is almost entirely produced by scholars of classics, history and anthropology while archaeology has shown only limited interest in the topic. As will be demonstrated, a history of Roman slavery has been produced, but a history of slaves is missing. This thesis is a search into how the Roman slavery system has been studied up to now and how we can find a way to learn more about the slaves in Roman society.
The historians Moses Finley and Keith Hopkins brought the study of Roman slavery from an element of economic history to a study involving both the economic and the social aspects of the phenomenon. It is only in the last decades that more attention is given to the archaeological data, although many assumptions and interpretations regarding the life of slaves and freedpersons (former slaves) are still predominantly based on literary sources. Classic texts are useful but provide an incomplete picture as these are produced top down by the masters, only very rarely by freedpersons and never by slaves. Inscriptions and (large) grave monuments relate to only a very small section of freedpersons2.
Historical texts leave us with a multi-layered challenge; modern scholars interpreting the ancient writers' interpretations of their experienced reality. Besides that, some of the more important ancient historians left us, often fragmented, texts regarding times well before their own while their sources remain unknown. A so far underdeveloped angle is the one involving archaeological material and its capacity to enhance our picture of slaves and ancient slavery. Can archaeology do better than it has up to now?
Research question
The research question is: How can archaeology contribute to a wider and better understanding of the slave-community in the Roman World, the identity of slaves and the relationship between the Romans and their slaves? For the time frame the 2nd century BC - 1st century AD is chosen and for the region Central Italy. The 2nd century BC is chosen based on the fast rising influx of slavery into the Roman community around 150 BC. The end includes the eruption of Mt. Vesuvius in 79 AD. This made it possible to take data from
2 In this thesis I will use the word "freedpersons" rather than freedmen when manumitted slaves, male and
Pompeii into consideration. Towards the 2nd century AD the Roman Empire consolidated in
size and major conquest campaigns ended, resulting in a diminishing influx of slaves captured in warfare. Although slaves were employed all over the Roman territory, including the provinces and the northern and southern regions of the Italian peninsula would exceed the limited extent of this thesis and time for research. In research up to now, slaves have been neglected and treated as an invisible group of people. This follows the formal Roman view; slaves were no part of society, rather they were objects; commodities without a human identity. Slaves were statistics, inventory, workforce etc. but other than that not very interesting.
Thesis overview
Aim of this study
The aim of this study is to explore the research practice up to now and to find a way to re-interpret archaeological and historical data. This should help to better understand the Roman Republic and early Empire as a slave society but also as a society with many slaves. It should add to our knowledge of the social reality of slaves. Why do we need to research this? The knowledge of slavery is up to now mainly based on information on the actions and writings of slave owners. Scholarly interest was focused on how the system facilitated Roman economy but there is scant attention to those who actually did the work. This results in a picture which insufficiently takes into account the role and position of an estimated up to one-third of the population of the Roman world in Central Italy. In general, both archaeology and history had a tendency to focus on the doings of the upper classes; those in power (usually men) and to large private, ritual and public buildings. This generated an incomplete and askew interpretation of how Roman society functioned. For a part this is the result of more powerful and/or richer classes leaving more traces in as well the historical as in the archaeological record, for another part it is due to research priorities. Archaeology has a tendency to search for any sort of deliberate recordable structure and the prospect of recognisable artefacts (Craddock, 2010, p. 12, George 2010, pp. 143-145). For this, burial grounds, road systems, domestic and ritual structures etc. present more fruitful opportunities than the scarce archaeological material left behind by slaves. The archaeological problem is not only the paucity of material but also the tendency to not research it because it is considered less interesting and/or hard to retrieve. A better understanding of the Roman multicultural society, which it created itself by importing large numbers of people from other regions and how those in control acted and reacted may teach modern society something on how to deal with multiculturalism, migration and the problems this obviously presents. But this cannot be done without understanding all segments of society instead of only the upper classes. It is also obvious that modern society still struggles to find answers to the problems large scale migration presents, problems caused by a multitude of circumstances, among which are the economic needs of the western world.
Contemporary slavery
U.S. and Caribbean the slave population consisted largely of native Africans and their offspring. Slaves in the southern U.S. were identifiable by skin colour. The last country to legally abolish slavery is West-African Mauritania, in 1981. A legal penalty on holding slaves, which is still in practice today and affecting an estimated 155.000 people in Mauritania, was introduced in 2007 but seldom someone is actually fined (Gyuracz 2014, p. 158-159; The Global Slavery Index (GSI) 2014, p. 18). In Mauritania, black Haratin and Abd communities are still enslaved by the Beydan (“white men”). The existence of slavery is formally denied by the Mauritanian government (Gyuracz 2014, p. 159; Lawrence 1999, p. 14). According to the recent GSI surveys conducted by the Walk Free Foundation (WFF) in 2013, globally around 29,8 million people are still living in slavery nowadays. (Gyuracz 2014, p. 156). For 2014 a total of almost 36 million slaves are given by WFF, as they state due to a different method of calculation. This makes the topic of this thesis quite contemporary. The definition used by the WFF is: “modern slavery involves one person possessing or controlling another person in such as a way as to significantly deprive that person of their individual liberty, with the intention of exploiting that person through their use, management, profit, transfer or disposal”. Including the words “or controlling” results in a considerably larger group of people considered to be slaves. The WFF also includes slavery-like practices: debt bondage, forced or servile marriage, sale or exploitation of children (including in armed conflict) and descent-based slavery (The Global Slavery Index 2014). The GSI covers 167 of 195 acknowledged states. In modern western society, slavery is connected to racial discrimination. It is fundamentally different from antique slave societies where the determining factor was predominantly a territorial one: slaves were captured non-citizens from outside the Roman territory3. In the Republic and Empire, a freeborn immigrant may have originated from the same region as an enslaved person. Modern legally based slavery (depending on definitions) no longer exists. In general, modern slavery is partly comparable to antique slavery and it is problematic to use it as a full ethnographic parallel. Nevertheless, some aspects can be used as such. In today’s practice political and social debates are ongoing regarding the Muslim community in the Netherlands (as in other European countries) and in how far they can express their religious identity or if they rather should fully adapt to what is considered the Dutch identity or even be legally forced to do so. For a large part, non-western immigrants in the Netherlands can be regarded as an example of a subaltern class. The Roman practice of stripping slaves of their original identity and creating an underclass of people with
less or no freedom in expressing their identity can teach us more on the effects of these kind of acts4.
1. The slave community; the subaltern, identity and humans as objects
This thesis sets out to better understand the slave community. Therefore we need to know more of their role and position in society, how they reacted to that and how it may have defined or redefined their identity. The slave from a Roman perspective was an object in the legal sense. Socially their role was different from that. We know slaves were conscious, thinking people, acting in society in their own ways within their social context. Slaves belonged to the lowest stratum in society, a layer that has been described as “subaltern”. When this is better understood, we can find a way to recognise the material conditions of existence for slaves, which is the focus of archaeology; understanding the interaction between humanity and material forms (Lesick 1997, p. 37). This includes landscape and built environment. In this chapter the theory of the subaltern will be described, that of persons vs. objects and the significance of identity. Subaltern relates to slaves as a class, identity to slaves as individuals and to being viewed as persons or as objects. It is also important to take into account that there was no slave community of one type. There were worlds of difference between slaves in the various social, legal and economical contexts the Romans placed them in.
Subaltern
Subaltern can be defined as follows: “Subaltern, meaning ‘of inferior rank’, is a term adopted by Antonio Gramsci to refer to those groups in society who are subject to the hegemony of the ruling classes. Subaltern classes may include peasants, workers and other groups denied access to ‘hegemonic’ power” (Ashcroft et al 2013, p. 244)5. The slave class can be included in this definition. The concept is important to understand that lower classes are not only the powerless but they are also those who define the ones in power. You can neither be a slave when there is no master nor can you define yourself as a master when there are no slaves. In essence the notion is applicable to Roman slavery although there was no absolute impossibility of social mobility. In the studied period this was limited to few and limited to a certain level, but nevertheless possible. This is important for freedpersons and for the way the upper strata of freedpersons apparently adapted an almost complete identity change. This remarkable change of identity was noted during this study and is in my view certainly something worth further research. The freedperson upper class is however not the topic of this thesis since they represented only a small percentage of the total slave and freedperson population.
Spivak discussed the complicated social structures in India under British rule in answering the question “Can the subaltern speak?” meaning can they make themselves heard as a group or as individual towards the elite. The ultimate answer to this, under the circumstances, was “no” (Spivak 1988, pp. 271-311). In a lecture for the University of California Spivak described subalternity as a position without identity (Spivak 2008, University of California lecture, YouTube video at app. min. 9-116). My interpretation is that a subaltern position is one that de-identifies you from the viewpoint of others, not necessarily your own, which was the case for Roman slaves. The matter was further discussed by Sharp in a chapter carrying the exact same title Spivak used, where the conclusion is confirmed with the addition that the only way to “escape” subalternity was to act and speak and, if possible, be like the elite. In other words; to fully adapt to the elite and leave behind all that relates to your original cultural identity (Sharp 2009, pp. 109-130). This is exactly what the upper class freedpersons did when they tried to change their original slave status and identity into a full Roman identity. The theory on the place of the subaltern, usually applied to colonial and postcolonial societies, is in line with the assumption that slaves are invisible in the Roman world, other than in general terms written down by the elite. While subalternity is a reality, it is not a problem of silence in the class but rather a listening reluctancy of the hegemonic class; a problem of not speaking the same language. In research after slaves it is crucial to not in advance accept the invisibility and silence of the subaltern but to try to and understand their position and the ways Roman society controlled this. A class invisible for, or rather overlooked by, those in control does not make the class invisible for research.
Identity
Group related identity is complex and in general based on shared beliefs, values, memories, descent, language, place of birth, belonging to families and peer groups. The value and meaning of each of these notions is the outcome of shared opinions within a society. Individuals have different and a more or less powerful influence on these opinions (see the paragraph on subaltern). The total of a person’s roles and social environment in place and time make up a personal, individual identity. This is the set of meanings that define who one is when one plays a particular role in society (as a parent, a teacher, a warrior, a politician or as a slave) and particular characteristics that identify him or her as a unique person (Burke and Stets 2009, p. 3). Each role is a piece of a person’s identity and one role may be more
defining or dominant in a persons’ daily life. It will be clear that a slave working a water wheel in an Iberian silvermine will experience the slave-role within his identity during working hours, although he might still remember being a father, soldier or farmer. Within the group of slaves his identity may be determined by ethnicity or status within his group. Identity is also a function of social relationships. The total of all identities of women, men and children makes up (the identity of) a society. The way others in society react on a person’s roles and identity influences the persons view on his or her identity and changing circumstances change roles and identity. Identity is therefore dynamic and more a process than a fixture (Burke and Stets, pp. 23-25). Slaves may have been formally stripped of identity by the Roman owners but it is unlikely they forgot all about themselves and they created new identity aspects in a new setting, not only depending on the Roman views.
Persons and objects
Janet Hoskins introduced her article “Agency, Biography and Objects” stating that anthropologists since Mauss7 and Malinowski8 made clear that the lines between persons and things are culturally variable. Slaves could have been approached as mere possessions while things could almost seem to act as persons. For the Roman upper class, owning slaves was as much a sign of status as owning land or a large and well decorated house. Objects were given a name, gender, history and ritual function. A sword could bare a name and represent the deeds and reputation of its (former) owner(s) (Hoskins 2006, p.74). The best known example, although not Roman, is the sword Excalibur. The idea of what an object is and how and by whom its value is determined is must be understood, simply because slaves have been pushed into the role of object. Through this we can study the position of slaves in the ideological context of Roman society, re-evaluate textual evidence and recognise the still tangible effects of their actions and presence in a society. When this theoretical framework is applied to the Roman slavery system, it will be better possible to research the total of the material record and within that, to identify the slaves’ presence or absence in this record. How should we define an object? Objects are not only produced and used for their practical capacities like holding food or liquids, to be used in hunting, building, working the land or in combat. They also hold the capacity to reflect abstract values like culture, status, wealth, group membership, memories and value. Some objects lack practical usability but serve ritual purposes or reflect the status of the owner. This can be intrinsic through the basic material or through the
7 Marcel Mauss (1924/1954). The Gift. London: Cohen &West.
connection with the object’s history of use and ownership. But also this supposed intrinsic value of material is assigned by humans and variable, based on valued properties and related to specific places, time and culture (Appadurai 1988, p. 3).
The principles of value were most influentially described by Malinowski, based on the Kula ritual network. Anthropologists asserted that the lines between persons and things are variable and not drawn in the same way in all societies. Kopytoff acknowledged that notably slaves were very close to being regarded and treated as objects (Hoskins 2006, p. 74, Kopytoff 1986, p. 64-65). Where Patterson inclines to look at slaves as being objects for the rest of their lives, regarding them as socially dead, Kopytoff sees slaves as a commodity only during the period between capture and sale (Patterson 1982, p. 8; Kopytoff 1986, p. 64-65). After sale, the slave is indeed stripped of all former aspects of identity but is then inserted in a new social setting in which the slave is re-socialised and given a new name and identity. This aspect of the slave-identity is however always liable to change as the slave can still be resold or when the slave is freed by the owner (Patterson, 1982, p. 8, Kopytoff 1986, pp. 64-65). The difference is quite important since, following Kopytoff, a slave could change from human to object, back to human and then back to object again. Patterson sees the slave as an object during his lifetime as a slave, therefore denying a social life from which we have some proof from graffiti (treated later in par. 2.6). A single qualification cannot describe all slaves. It would more probably have differed depending on the situation; while serving a master a slave might have been more seen as a human, from their appearance, provenance and numbers they could serve as prestige objects (Greek slaves were considered more civilised and better educated), in legal cases they were nothing but an object, as will become clear in the next chapters.
The slave as a person and as an object
shape protest against their situation. All of the above require the human capacity of agency and therefore being a person. Ascribing these acts of resistance are just as well a Roman impression and an implicit acknowledgement of the slaves’ identity and humanity, although possibly not a conscious one. It is not possible to state that Romans experienced slaves as nothing but objects. However person-like an object may be experienced, physically it cannot act in the ways described above. Resistance is part of the whole concept of identity and agency. It shows choices made by slaves in how to deal with their slave status. It could have been anything between full resistance where Spartacus would be the ultimate (but also somewhat mythical) example and choosing to go with the flow as much as possible, for which successful rich freedpersons could serve as an example.
Another example related to identity is graffiti, abundantly found in places like Pompeii and Herculaneum. The texts from which many have been ascribed to slaves, show interhuman contacts, opinions, accusations, poems, declarations of love and hate and so on. All inscriptions are results of thoughts and actions taken by humans in relation to each other. Both the text and the locations are meaningful since they were produced within a Roman context. In which way these can help us enter the slave’s world of thought needs further research. They reveal something of a person’s view on his or her own identity and on that of others; fellow slaves and Roman masters. It can also point to the places slaves (and/or free servants) could tag walls without being caught in the act.
A third example follows from the phenomenon of freedpersons, particularly those who made the best of their situation by being freed by their owner, become successful in business and starting a family. The strong effort made by the top layer of freedpersons to replace a lost former identity9 with one as Roman as possible was mentioned before. It shows that for some an identity which would make you “blend in” was more important than returning to an identity that related to origins. It represents an oppotunity to escape from slavery, slave identity and subalternity. This is however different for captured slaves and for home born slaves since these vernae10 had nothing but the slave identity.
2. Slavery in the Roman society
In this chapter the definition of slaves and freedpersons and their position in society and Roman economy are discussed. Current knowledge of these themes is predominantly based on literary sources, much less on archaeological material, as will become clear in the following paragraphs. The current scholarly view and knowledge regarding slavery evolved based on Roman literature, which tells us much about how the Roman upper class experienced slavery and slaves but is much less informative on slaves themselves. This chapter shows much about the Roman context in which slaves lived. The point is here is to show that our knowledge relates to Romans and Roman culture and slavery but hardly to slaves. The chapters 2 and 3 must be understood with that in mind.
Slavery: a normal fact of life
Slavery in the Roman world was a normal part of everyday reality. Although few would have chosen to be a slave, it was never considered to be a phenomenon that should be abolished throughout the existence of the Roman state11. Slavery was as accepted as part of the rule of the many deities that were thought to determine everyday life. Slavery was rooted so deep in the mental and ideological structures of the Antique that an impulse for change could not and did not occur. We cannot impose our morality on the Romans. This is how Bradley explains his statement “... dass die fehlende Gegnerschaft zur Sklaverei unser Problem ist, nicht ihres...” (Bradley 2010, p. 33). From the early days of Rome, not much is known about the existence of slavery, neither from material- nor from literary sources. The first indisputable written source for slavery within the Roman community is found in the Law of the Twelve Tables, dating to ca 450 BC. Putting slaves to do your work was so commonly accepted that, although legally impossible, even slaves themselves held slaves in their turn, called vicarii (Joshel 2010, p. 193). The Twelve Tables are not the first written indication. On Linear B tablets dated to the 14th and 13th centuries BC, the words doero (male) and doera (female) are
found. These words are equivalent to the later Greek words for slaves; doulos (m) and doulè (f) (Andreau and Descat 2006, p. 17)12.
11 Some, reliable numbers are not available, sold themselves as slaves until this became prohibited by law in 326 BC, see also par. 2.2. This practice, known as nexum, is not treated in this research as it was no longer practised in the discussed period (Cornell 1995, p. 332-333). There are no statistics available on the topic but the practice must have been significant enough to call for legislation, either by numbers or as a phenomenon.
Definition of slaves
The definition of slaves in this research is not race-related since slavery itself was not race related in Roman times (Hopkins 1993, p. 6, George 2010, p. 155). It is likely that black Africans were already enslaved and traded to the Romans by the Garamantes, after North Africa became part of the Roman Empire (Bradley 2010, p. 32). There is however no evidence available that racial considerations were involved in this trade. Slavery was initially related to conflict. These conflicts were at the outset small-scale in a geographical sense. Most of the first slaves were those who were captured instead of killed. The most relevant attribute of Roman slaves is that they were considered legal property and, legally, regarded as objects rather than humans13 (see par. 2.7: The slave in Roman law). In general freeborn Romans could not be slaves. Only those from outside could be slaves. In practice things were a bit different, since exposed and sold children were enslaved as well, although they may in fact have been freeborn but did not have anything or anyone to confirm that status. The definition for Roman slaves used here is: Humans legally regarded as objects and owned by a Roman citizen. This is the only common attribute applicable to all slaves in this time and it does not include the intentions connected to ownership, unlike the WFF-definition mentioned in the introduction. This definition leans more towards a “personal” situation between people as opposed to one social class exploiting another one. Influenced by Finley and Hopkins, most historians discern a “slave society” from a “society with slaves”. A slave society is characterised by a percentage of up to thirty percent and no less than five percent of the population while slaves are put to work in more than primarily domestic service. Finley named only five slave societies: ancient Greece and Italy, Brazil, the Caribbean and Southern United States14. A slave society is also characterised by the enactment of laws and institutions to control slavery since in the eyes of the owners, slaves are hostile and dangerous (Andreau and Descat 2006, p. 15).
A slave society
The designation “slave society” would have been meaningless to the Romans as they considered slavery a normal part of any society and it would probably be on the same level as describing a society as “beef eating”. Important is however that eventually the Roman population for a considerable part was made up of slaves and this had an influence on how
13 There is some flexibility in the Roman mind. The practice of buying a female slave to set free and marry her does not point to a fixed interpretation of slaves as “things”.
people lived together. This research aims at finding ways to reconstruct this “living together”, rather than to confirm a social structure that has been reconstructed from predominantly written sources. This reconstruction is based on a top-down view, registered by the Roman elite and omits the slaves’ reality. A traditional view on the circumstances under which slavery could have developed, is when agriculture and manufacturing scaled up from (extended-) family run enterprises to larger estates and workshops (Thompson 2003, p. 2). This is believed to have created a situation where work could no longer be done by the family or community and additional workforce became necessary. According to Finley, the logic is that demand precedes supply and the economic situation would have had to be a situation involving land ownership and surplus production of commodities and trade (Finley 1977, p. 56). Furthermore, there must be a market where slaves could be bought and paid for and conditions in which internal labour was either unavailable or more costly than employing slaves. Lastly, slaves had to come from somewhere, which initially in essence was warfare. This seems logical and plausible but it still is a predominantly economic analysis which does not take into account social and cultural factors involved like religious beliefs, status and power structures. The idea of putting slaves to work must have become rooted in the collective mind of people too. The economic factors may have triggered the idea, but cannot fully explain how the phenomenon of slavery came to be experienced as a normal fact of life, as it seems to have been by slaves and slaveholders equally. This process is interesting but not the topic of this thesis and will therefore not be discussed in detail. For this research it is accepted that slavery has been a normal fact of life for very long time and developed as a social and economical system in many civilizations independently15. Apparently the step from cooperation with to exploitation of fellow humans is in the nature of the beast and is introduced as an outcome of the development of power structures. There are almost no societies where slavery was not at some point in time a more or less permanent feature of their existence, either as slaves, as slave holders or as traders. Diodorus of Sicily wrote: “Among some strange customs practised by Indians, the most astonishing appears to be one that was produced by their ancient philosophers; the law states that, without exception, no one among them is a slave and that people shall live as freemen with respect for the equality of all” (Library of History 39.5). When the notions “slaves” and “freemen” are indeed used in this law mentioned by Diodorus, it follows that slavery had existed among the Indians.
The Roman jurist Gaius and other Roman lawyers considered slavery as being common in all nations (Institutes VII-5216). Laws of nations applied to human culture and society. Slavery was no crime and slaves were not victims. The system was a social reality, but not considered to be part of natural laws, since these concerned animals and humans, male-female relations and procreation (Joshel 2010, p.6).
Becoming a slave
Becoming a slave was subject to circumstances; essentially fate and bad luck, like defeat in a war or being born from a slave mother. Nevertheless, Romans always understood that the position of a slave was undesirable (see par. 2.1). That many slaves were not thrilled about their fate is illustrated by numerous slave revolts17. The rate of suicide and attempts to flee from the owners was high. Slaves took their own life rather than being subjected to slavery or did so when they already were. How often it happened is impossible to tell but that it was not an exception is shown by the fact that a slave trader in antiquity was obligate to inform the buyer when a slave had attempted suicide (Andreau and Descat 2006, pp. 137-138). Julius Caesar noted, albeit in passing, that it was normal for any man to desire freedom and to hate being enslaved (Bello Gallico, III.10)18. This did not stop him from auctioning off a whole conquered Atuatucan town of 53,000 including all their possessions to traders, after 4,000 had been killed in battle (Bello Gallico II.33). What happened next and how the logistics involved with transport and “redistribution” of 53,000 people were tackled is not described. In my opinion, some exaggeration in Caesar’s war diary is quite likely.
The Roman social strata
The Roman social strata had a clear (legal) structure, evolved from the early Republic where the gens were the leading class into Imperial time where the emperor was on top. It is where we find the tradition of ancestor worship which is mirrored in Roman and pre-Roman grave culture. This also resulted in a major division in society between slaves and free citizens since slaves were not considered to be persons, had no (formal) ancestors and were not part of a
16 Latin text: In potestate itaque sunt servi dominorum. Quae quidem potestas iuris gentium est: Nam apud omnes peraeque gentes animadvertere possumus dominis in servos vitae necisque potestatem esse, et quodcumque per servum adquiritur, id domino adquiritur.
Translation: Slaves are in the power of their masters, and this power is acknowledged by international law, for we know that among all nations alike the master has the power of life and death over his slaves and whatever property is acquired by a slave is acquired by his master.
17 See for an overview of slave revolts; Urbainczyk 2008.
family. The social order in Roman life in Imperial time is shown in table 1, where slaves make up the subaltern.
The Emperor Nobles
Citizens Senators Social elite
Equestrians
Municipal magistrates and senators
Freeborn Roman citizens Freed slave citizens slaves
Table 1. The Roman social order after 27 BC (after Joshel 2010, p. 31)
What is represented by a black horizontal line dividing classes was (and is) usually a zone of grey-shades. Clear divisions occur in legislation, much less in daily life. Social mobility was possible but limited and conditional (the legal position of slaves will be further discussed in par. 2.7.). Roman citizens may have been rich and powerful but many lived in poverty and had no form of power at all. In many cases one would not have been able to recognize if someone was a poor citizen or a slave19 (Andreau and Descat 2006, p. 99). Seneca (4 BC - 65 AD) advised the emperor Nero (ruled the Empire 54 – 68 AD) against a proposal made in the Senate to distinguish slaves from citizens by their clothing. He warned against the danger that would arise if slaves by wearing these "uniforms" would come to realize with how many they really were, which might lead to rebellion. The proposal was denied (Bradley 1998, p. 477; George 2011, p. 399).
The picture painted above is that the majority of slaves and freedpersons in Roman society were confined to the lowest social classes, from a legal point of view and in the Roman citizens' mind20. Marriage and adoption provided possibilities to change ranks (and/or clans).
It was also not uncommon for the more modest Roman men to buy a female slave to then free and marry her (Andreau and Descat 2006, p.153). In this way, some women would have escaped the slave class. It is however a legal point of view. In practice, slaves were active and present from the lowest to the highest classes in a multitude of activities. They cannot be described as one social class with well defined characteristics, other than that all slaves were
19 Andreau and Descat refer to Greek slaves here. It is plausible the same was true for Roman slaves since in both societies it was customary that the master clothed the slave, for which they would logically not use imported or specific garments but cheap local ones, the same that would be worn by the poor.
people, legally seen as objects and owned by other people (Andreau and Descat 2006, p.12). Other than that, individual slave’s lives would have been worlds apart.
2.1 Citizens and slaves
The Roman society had a long tradition of employing slaves. Greek slavery is considered to be partly comparable to Roman. In Greek and Roman Antiquity, slavery was common and slaves participated in household, economy and cultural life on various levels. There are many accounts of people committing suicide rather than become enslaved. Sometimes it is dramatically represented in art, like in the Hellenistic statue “The Galatian Suicide” (fig. 1) which has been interpreted as an act to prevent both of them to be brought into slavery (Pedley 2007, p. 353). Also, here we must take into account that the original Greek bronze and this Roman marble copy are interpretations conveying a message from the victor, not from the (to be) enslaved Gaul. This message was that the Gaulish warriors were an extraordinary strong enemy, logically implying that the winners were even stronger. The image is concerned with the Romans, not with the Gauls. Slavery (including manumission) was also a familiar concept in Etruscan society (Barker and Rasmussen 2000, p. 101). In the period covered in this thesis, slaves and freedpersons supposedly made up for a considerable part of the population of the Italian peninsula. Roman citizens may have owned no slaves at all, a few slaves, but some were said to have owned even up to thousands. Estimates vary from two to seven slaves per household on average but given the wide variation of slave employment this average does not tell us that much more than that the estimates have quite some range. The demographic aspects will be treated further in par. 2.4.
2.2 Defining slaves the Roman way
The definition for slaves in this thesis was defined as humans legally regarded as objects and owned by a Roman citizen. There are two things relevant: firstly the legal status of Roman slaves and secondly the social consequences. What defined a slave in Roman daily life is that he or she was not primariliy considered to be a person but the property of a master or mistress. Varro referred to rural slaves as "speaking tools" (instrumenti genus vocale) (Joshel 2011, p. 216; Mouritsen 2011, p. 13). There was a difference was between a slave and merx, the word for merchandise (Andreau and Descat 2006, p. 10-11). For the slave, the difference would have been hardly notable, although it is from our point of view almost impossible to look into the minds of slaves and to know if slaves knew the difference and cared if they did. What we can be reasonably sure of is that whatever the law stated, slaves were at the mercy of their owners.
Good slaves and bad slaves
2.3 From slave to freedperson
An important aspect of the Roman slave system is known as "manumission"; granting freedom to the slave by the owner. This is, in the discussed period, regulated by laws which over time became stricter and regulated not only the granting of manumission but also the heritage of the former slaves and the span of the won or regained freedom, which had its limits, like not being eligible into the senate (Andreau and Descat 2006, pp. 149-156).
Manumission was part of Greek and of Roman slavery tradition, although it is not known when it was introduced in Rome. A slave could be set free by the owner, could buy freedom or be bought free by paying a ransom or by testamentary manumission. These basic forms could involve many legal constructs and restraints but an important aspect was that the opportunity of manumission is thought to have had a positive influence on the slaves' behaviour and servitude. This would have been the reason for introducing the system in the first place. It was also a legal way out of slavery, the other ways being flight or slave revolts which were frequent but never successful and never led to major changes in the system or thoughts about abolition of slavery. The best known slave revolt is that of Spartacus in the late 1st century BC which ended in defeat of his army and allegedly in 6000 crucifixions of slaves along the Via Appia (Urbainczyk 2008, p. 2). The slavery system functioned with a complex arrangement of rewards and punishments.
Forms of manumission
Manumission had various advantages for the slave. For a part it meant a return to humanity, although not a return to an original identity (for the non-vernae). Freedpersons were no longer objects. They had a new (Roman) name, could have an acknowledged family, be married and the children would be freeborn Roman citizens. Despite this, the ties to the former master were never completely broken. For many freed slaves, their life could continue more or less like it was, working for their former master. In Augustan time, a freedperson at the time of his death owning more than 100,000 sesterces had to leave part of his estate to the former master unless he had three or more children (Andreau and Descat 2006, p. 111). During his or her lifetime a freedperson still owed a defined amount of work to the former master.
2.4 Demography; influx of slaves into Roman society
Table 2. Reported enslavements of war captives. (After Scheidel 2011, p. 295).
Capture in war was one of the major sources for new slaves. From various sources, largely Roman tradition, Scheidel recorded the figures mentioned in table 2. The table covers the period 297 - 167 BC. Again, these numbers are highly speculative but the conclusion that war was an important source for new slaves is justified and it is in line with the estimates made on the totals of slaves living in the Roman heartland, although one seems to prove the other and vice versa. Various Roman writers mention even larger numbers, like Caesars' alleged enslavement of a million prisoners from Gaul in the first century BC. Although these figures seem quite fantastic just considering the logistics involved, there is not much to rely on if one would want to question it and come up with better estimates (Scheidel 2011, p. 294 - 297). In my opinion the numbers suffer from a fair dose of war propaganda and need to be questioned. Considering that, numbers may have been significantly lower but even then substantial. Other sources of enslavement also are thought to have generated large numbers of people brought into slavery (even well before the 2nd century BC) but here statistics are even less present, let alone reliable which is as true for figures delivered from slave markets as for biological reproduction (Scheidel 2011, p. 308). Bradley sums up the numbers of captives brought into slavery as stated by ancient historians (see table 3 below). Here, the numbers are equally unreliable and cannot be verified but even when they are wrong by quite a margin, the numbers must have been impressive (Bradley 2011, p. 246).
Conflict Date Reported number of captives
Third Samnite War 297 - 293 BC 58,000 - 77,000
First Punic War 264 - 241 BC 107,000 - 133,000
Gallic War 225 - 222 BC 32,000
Second Punic War 218 - 202 BC 172,000 - 186,000
Various wars 201 - 168 BC 153,000
Sack of Epirus 167 BC 150,000
Source Location and date Number of mentioned captives Diodorus Siculus Agrigentum 261 BC 25,000
Orosius North Africa 256 BC 27,000
Livy and Polybius Tarentum 209 BC 30,000 Livy and Polybius Carthago Nova 209 BC 10,000
Table 3. Totals of captives as recorded by ancient historians (after Bradley 2011, p. 246)
According to Scheidel at least 100 million people have been enslaved over the 1000 years the Roman Empire lasted (Scheidel 2011, p. 309). However uncertain the (proposed) statistics may be, it is clear that enslavement by warfare and by other means was the fate of many. Another important source of new slaves was birth from a slave mother. Statistics on this are unavailable, but it is thought to have delivered a substantial contribution to the slave population, growing in significance over the centuries.
2.5 The economic significance of slavery
and high ranking officers held whole familiae of slaves and freedpersons (Lo Cascio 2007, pp. 635-636).
Slave jobs
In ancient literature as well as in research, a division is made between urban slaves and rural slaves. From classic literature, inscriptions and documents, it is possible to generate an overview of the work slaves performed. In Cato’s work "De Agricultura" a list of 44 job titles for rural slaves shows a variety of activities, like that of ergastularius; the jailer in charge of the ergastulum21 (slave jail) (Bradley 1994, p. 60). With human history and psyche in mind, it is not surprising that it always would have proved to be possible to find a slave who could be trusted to keep the other slaves imprisoned, when the compensation is sufficient. We must also take into account that although the jobs listed are in many cases described as slaves' jobs by Cato; he is not clear if and when these were all or exclusively performed by slaves. Bodel states that there were no jobs exclusively carried out by slaves and there were few jobs slaves could not undertake. Some jobs could have been performed predominantly by slaves and thus became associated with them. The exceptions were politics, law and military service, which were closed to slaves (Bodel 2011, p. 312). Slaves may have been responsible for more than one task and the mentioned job list is not conclusive. It omits for instance domestic jobs like cook, waiter or launderer. Other overviews offer almost 100 different jobs/tasks (Bradley 1994, pp. 60-61). Slaves, although omnipresent, were no real individuals in the eyes of the master22. They were referred to by the names and jobs the masters had given them. Literary evidence of agricultural (slave) labour is primarily found in the manuals written by Varro and Cato (Joshel 2010, p. 171). Other sources are found in inscriptions, epitaphs, and figurative works like reliefs, wall paintings and mosaics. The historical texts refer to the situation in central Italy and cannot simply be transferred to the situation in the colonies (Bradley 1994, p. 61). The little personality a slave was allowed to carry usually came with the job and this is what is found in inscriptions and epitaphs. An example is seen in the columbarium of the Statilii in Rome, in two inscriptions23:
Zena - cocus (Zena, cook)
Felix - topiarius (Felix, ornamental gardener)
21 The list omits jobs which are described rather than titled, like those of women who spin wool and also work in the field (Joshel 2010, p. 168).
22 Again, this is stereotyped from Roman literature, it may very well not be in general applicable to the Roman mind, see footnote 11.
The inscriptions refer to nothing reflecting a personal identity other than the job: the name was chosen by the owner, the job likewise and maybe also the job title in the inscription was not the slaves’ choice (Hackworth-Petersen 2006, p. 3). An important question is: were slaves in columbaria granted a decent funeral and last resting place or were their remains placed there to make sure that in death, slave owners were still surrounded by slaves as if they were a form of grave goods? Was the objective of pigeon holes for slaves to stress the position of the family as powerful and wealthy enough to keep a considerable number of slaves, rather than to grant them a decent resting place? Given the indications so far, the columbaria served the Roman purpose, like the mass funerary sacrifice of dependent servants in royal tombs, described in detail by Herodotus (Histories 4.71) which practice, although much earlier in time (± 450 BC), has been clearly attested archaeologically (e.g. Parker Pearson 1999, p. 11).
2.6 The slave in classic texts
One of the very few texts in which an ex-slave is presented as an individual is Petronius’ Cena Trimalchionis (Trimalchio’s’ Dinner), a part of the larger work Satyricon, which was not preserved in full length. The fictional character Trimalchio came to play an overly important role in scholarly views and thoughts regarding rich freedmen in general and in interpreting material remains, notably grave monuments. The influence on archaeological and historical interpretations is treated in this paragraph.
The fictional freedman Trimalchio
A specifically influential text is the fictional character of Trimalchio, a rich freedman brought to life by Petronius. Petronius lived and wrote in the 1st or 2nd century AD (Holzberg 2013, pp. 405-409). Despite the fact Trimalchio never existed, the character has heavily influenced, if not determined, the scholarly view on the upper class of libertini. It is virtually impossible to find literature on freedpersons without a prominent role for Trimalchio. The common interpretation of the figure is that Petronius in his satire showed the character of rich freedpersons in general, dismissing them as boorish, tasteless and extravagant. In some passages some authors go as far as citing Trimalchio almost as if the man really existed24. Of
24 For example Andreau and Descat write: “Free men were obviously the principal beneficiaries of […] sexual
course they know better and this may be more style than opinion, but it is exemplary for the way Petronius’ text is used.
The case of Trimalchio and its use for assigning tombs and other monuments to libertini (and in fact to characterise the whole class of rich libertini has been rightfully questioned by Hackworth-Petersen. She challenges that the idea that Petronius constructed a story that probably was a near accurate description of how the top-class of freedpersons were viewed by the Romans and that it characterised the upper class freedpersons correctly (Hackworth-Petersen 2006, pp. 1-11). This interpretation of the Cena Trimalchionis (Trimalchio’s dinner) has received too much emphasis regarding the scholarly view of the former slaves. The use of Cena Trimalchionis is justified by, e.g., the literary quality of the work and the fact we hardly have anything else. The possibility is not imaginary that Petronius’ satire is much more challenging the life style of the upper class Romans, rather than that of libertini. Thus Trimalchio functions merely as the carriage for the denunciation that the upper class developed into a tasteless and prodigal class of Roman citizens. It would have been the safe way to bring up the subject without directly and clearly offending his (Roman elite) public. Hackworth-Petersen suggested that the character is possibly a caricature of the emperor Nero and the same is noted by Holzberg (Hackworth-Petersen 2006 p. 9; Holzberg 2013, p. 409). Interpreting Petronius
not always evidently assignable to slaves, like the bulla from which it is uncertain if it was carried by a slave or by a dog25 (fig. 2, see also cover pages).
It is likely that Petronius should be understood more in the way we interpret the works of Flavius Josephus, whose large account of the history of his time, more or less contemporary to Petronius, is interesting and an important source of information but in no sense an accurate description of historical facts. Much of the text is a representation of opinions on what happened during and after the conflicts between the Romans and the Jews but are also written in a way that aimed to satisfy and entertain the Roman elite. Josephus managed to not be killed by Nero and lived during the reign of Nero and his successors Vespasian, Titus and Domitian (Maier 1999, pp. 7-16). It must be taken into account that it is not the complete and not the original work of Petronius that survived. What we have is a copy of parts and of the complete parts 14-20 of the work, produced “somewhere” in medieval time (Holzberg 2013, p. 355). The problems of copies and translations of ancient literature are well known; mistakes, omissions, misinterpretations and the copiers’ “enhancements” of the original texts are only some of the things to deal with when trying to establish in how far the texts represent the original.
25 Inscription Transliteration:
Tene me ne fugia(m) et revoca me ad dom(i)num Viventium in ar(e)a Callisti Inscription Translation:
Hold me, lest I flee, and return me to my master Viventius on the estate of Callistus See also cover pages.
Figure 2. Bronze tag for a dog or possibly a slave, inscribed with information about return. Source: British Museum.
The effort of Hackworth-Petersen in trying to evade Trimalchio in the analysis of freedperson art is interesting since it tries to interpret the material more in its own right. The work itself, being satire, was never meant to be an accurate account of a social reality relating to freedmen like Trimalchio but rather as a an indictment against Roman pomposity. It might as a literary form be comparable to Jonathan Swifts’ Gullivers Travels (published 1726 AD), a political and social satire in the guise of a phantasy. Both writers would have known better than to directly insult the hand that fed them. Hackworth-Petersen however fails in cutting all ties to Trimalchio. Looking at monuments like the Tomb of Eurysaces in Rome (fig. 3) and the Isis Temple in Pompeii (fig. 4), her conclusion still is that rich freedpersons were foremost interested in creating a position that “aped” the Roman upper class and aimed at becoming a member of that class in a “Trimalchian” way. The dedication to the rebuilding of the Isis temple apparently helped the six year old son of a freedman to enter the higher Roman ranks of the city. This effort in trying to assimilate to the Roman citizens may be true for some freedpersons but even if it is, we are still looking at a tiny fraction of former slaves appearing in the archaeological record. This tiny percentage nevertheless seems to be responsible for the scholarly view on freedpersons.
Figure 4. Temple of Isis, Pompeii. (‘d Ambrosio and Guzzo 2010, p. 124).
Graffiti
A different form of classic texts are graffiti. Except for epitaphs and graffiti26 the written record of slaves’ lives, produced by slaves is absent. The picture is therefore one-sided and incomplete. Although different in character, they are in their own way texts. Graffiti are an interesting phenomenon as it has both textual information and contextual; it was made and stayed in situ (unless stonework was later quarried for building material or used as spolia) and it was composed and executed by a single person. A study of the Palatine Paedagogium delivered information on the slave community in this specific place. With the restriction that the text was inscribed in stone by literate slaves (probably not the majority), it shows a variety of social relations and the sense of being part of a group. Slaves were thus "persons" (Keegan 2013, pp. 69-98). An example of an inscription ascribed to a possible slave will be discussed in par. 5.2.4.
2.7 The slave in Roman law
Roman civil law stated that “slaves are either born or made” (Thompson 2003, p. 1). This is a compact statement reflecting something of the Roman view on slavery. What it shows is that the Romans were the ones in power. They decided who was to be a servus (slave). Gaius (2nd century AD), in his Institutes27, provided us with this simple legal definition; "The primary distinction in the law of persons is this: that all men are either free or slaves. Next, free men are either ingenui or libertini. Ingenui are those born free, libertini those manumitted from lawful slavery (Institutes 1.9-11)”28. This discerns slaves and freedpersons from freeborn
Romans. Laws and most legal documents may be considered factually correct, although the body of evidence is not complete and we need to take into account the possibility of false documents. Another complicating factor is that it is not always known or clear who made the laws, why these were written down and to whom these would have applied. Furthermore there is the problem of incomplete texts and translations and suggested completion of missing parts are not always agreed upon by everyone. The laws concerning the position of slaves and freedpersons are a reflection of their social position but not a description. From these texts it is possible to define how to understand what for the Romans defined slaves and freedpersons.
26 In Pompeii, graffiti and dipinti are found. The first are scratched in, the second painted in red or black on stucco (Overbeck 1884 [1968], p. 463). In this thesis “graffiti” refers to both.
27 Gaius lived outside the researched period but the basic legal situation regarding slaves may be assumed to pre- and postdate him for quite some time, as in most literature studied is assumed.
Debt-bondage or nexum (selling you as, or being convicted to be a slave to pay off a debt) was already banned in the early 6th century BC by Solon in Greece (Thompson 2003, p. 2). In the Roman world in this debt-bondage was legally banned in 326 BC in the Lex Poetelia (Cornell 1995, p. 332-333). The legal causes to be considered (or to become) a slave were capture in war, being born to a slave mother or purchase from a non-Roman. These are considered to be the largest sources of slaves. Initially warfare was the largest source, later this shifted to slave trade and house born slaves. Roman citizens could not be enslaved, with the exception of condemned debtors who failed to repay. They could be sold by the creditor as a slave but only "across the Tiber" which was non-Roman territory at the time of the Twelve Tables (Gardner 2001, p. 415).
The slave as property
Roman law described the position of slaves as being property of Roman citizens. It is one of the ways slaves were put away as objects. Understanding these laws is important since it shows the Roman attitude towards slaves and the slave system. Law defended the interests of the slave owner and the state, hardly that of slaves. It helps defining the place slaves had in society and how this affected their identity. Laws on manumission were specific on the rights and duties of slave owners and on the duties of freedpersons towards their former masters. Legally, a slave was not a person and could therefore not have any rights. It followed that a slave from a legal point could not be held responsible for actions, but the master would be. Of course this did not mean a slave could not be punished. Punishment involved whipping, being chained or imprisoned up to being killed, usually by crucifixion. Although the master had the power of punishment, there were rules to respect. Punishment needed consent of the family and some friends and had to be performed in public, to serve as an example for other slaves. In how far these rules were always followed cannot be known. Under Roman law a slave was not allowed to testify in trial. A slave's word was only accepted if it had been extracted by physical torture. This torture required the consent of the owner. Legal punishment of slaves was performed by physical means. Since a slave did not (legally) own anything, a fine was not possible (Andreau and Descat 2006, pp. 109 - 113).
Roman law for Roman citizens
with regulating rights and duties for everyone (Cornell 1995, p. 245-251). Slaves played a passive role in law, something like a cow or a house. The law thus can tell us more on Roman citizens than on Roman slaves. Only after manumission a slave gained rights and recognition as a human being. These rights were still tied to their former status as a slave and to the former owner. If the owner was a Roman citizen, then the manumitted slave became a Roman citizen (with limits). If the former owner was foreign, then the manumitted slave became a free foreign inhabitant (Andreau and Descat 2006, p. 152). In law, the slave population was marginalised which also meant that the body of legal texts can only marginally reveal the situation of a large part of the total population. Slaves had no legal power and freedpersons only had limited legal power in society. Nevertheless, if only by their numbers and by for a large part originating from other cultures, they must have had some influence on Roman social life. It does not take much imagination to see the same phenomenon in contemporary society where immigrants do not have a high social status and are only limitedly present in political functions. But they bring in different religious and cultural beliefs and are bound to special legal obligations. Many in Western society, like the Romans, fear their character and numbers29. The present situation might reflect something of what the Roman citizens thought and felt about the slaves and freedpersons as a group in society. Likewise, modern immigrants may experience a social position in some ways resembling that of the Roman slaves in their time.
Essential is that Roman society and economy were to a large extent dependant on slaves as a status symbol and as a workforce. This also created the necessity of controlling them. Law is an important instrument in this. Roman law concerning slavery is mainly built around the interests of the owners. Slaves have a more or less passive role in it; they are in law what they are socially (for as far as we know): objects, owned by Romans (Gardner 2011, p. 436). Roman law draws distinct lines between slaves, freedpersons and citizens and foreign free people. The segregations seem to split groups into clear groups of people. In practice, the difference must have been much less clear, proven by historical facts like the discussion on introducing a slave uniform and by material culture which seems to be more a blend than that we see isolated and identifiable artefact sets pertaining to each of the four groups mentioned above.
3. The archaeology of slavery
In this chapter the material evidence identified as probably pertaining to slavery will be discussed. Besides the information gathered from literature, laws and inscriptions, knowledge regarding slaves' lives can be reconstructed from material evidence. The current idea in this area is that as yet there is not that much to go on, in particular due to the conception that the slaves of the Romans produced hardly any artefacts for themselves, never lived in their own durable houses or built other structures, possessed little and therefore did not leave many traces in the archaeological record. This can however only be denied or confirmed if we find ways to recognise which materials or structures can be related to slaves or can be certain of the non-existence of these materials. In general, scholars outside as well as inside archaeology share a rather pessimistic view on the possibilities archaeology and material evidence may contribute to our understanding of the slavery system. The overall attitude is expressed by many. Starting with Thompson: "Any discussion of the archaeological evidence for slaves in Italian agriculture in the last two centuries of the Republic must start from the literary sources, which now become of considerable importance” (Thompson 2003, p. 79). In 2006, Handler and Lange state that “...the archaeological data alone, without the support of documentary evidence, did not reflect an institution or behavioural system identifiable as slavery. This problem presents a major obstacle to any purely archaeological study of slavery and has also confronted others who have worked with the remains of slave cultures” (Handler and Lange, 2006, p. 5). Binsfeld checked the work Bibliographie zur antiken Sklaverei; it includes 10.000 titles but only 125 are concerned with the archaeology of slavery30 (Binsfeld 2010, p. 161). This seems to also point to lack of interest by archaeology, rather than just to a lack of possibilities, as already noted in the introduction.
Many scholars have doubted the contributions archaeology can make to the study of slavery or even to the reconstruction of many aspects of ancient communities at all. Finley’s statement is still repeatedly cited; “I merely wish to make the simple point that archaeological analysis by itself cannot possibly uncover the legal or economic structure revealed by the Oxyrhynchus papyri or the alternative structure in Arezzo, Puteoli, Lezoux or North Africa.” (Emphasis by Finley)31 (Finley 1985, p. 25). The statement was made in 1985 and Finley aimed at the emerging processual “New Archaeology”, the general attitude seems to apply to the discipline as a whole. The idea that archaeology is at best an illustrative supplement but
