Slavery
in a Hurry

The Whole Thing in One Page

The familiar image of slavery is an iron chain, a ship at sea or a plantation under a hot sky. Each image is true. Together they are still too narrow. Slavery existed long before Atlantic plantations, far beyond them and in forms that did not always look like field labour. Enslaved people cooked, mined, rowed, kept accounts, raised children, served rulers, fought in armies and sometimes governed free populations. The work varied. The condition underneath it was the loss of an enforceable claim to direct one's own life.

Slavery began when a person could be placed outside the protection normally owed to a member of a family, community or state. War captives, kidnapped strangers, debtors, criminals and children born to enslaved women could all be pushed across that boundary. The route changed by place and period. The result was a status in which someone else could command labour, movement, punishment, sexuality, family formation and, in many systems, sale or inheritance.

That reach is why slavery must stay distinct from other forms of unfree labour. A serf might be tied to land. A debtor might owe service for a fixed term. A forced labourer might be threatened into one job. Slave status could gather several such powers around the person and make them persist when the task, place or owner changed.

Capture alone could not build an institution. Other people had to keep the captor's claim valid after the weapons were put away. Courts, officials, households, religious authorities and neighbours did that in different combinations. Fugitives were returned, transfers recorded, children inherited and people priced in markets. Trade carried enslaved people through the Mediterranean, across the Sahara and Indian Ocean, around the Black Sea and, on a vast Atlantic scale, from Africa to the Americas. Yet trade was a flow, not the whole system. Many slave populations were sustained locally through birth.

This is where empires enter. Slavery converted conquest and exclusion into usable capacity. It supplied farms and mines, palaces and ports, workshops and armies. Some rulers recruited outsiders whose advancement depended on a court; owners sought claims reaching beyond work into children, skills and time. No empire rested on slavery alone, and some used it far more than others. Still, slave systems repeatedly turned political power into labour, service and wealth that could be moved, accumulated and inherited.

The conversion never became complete. Enslaved people formed families, preserved knowledge, bargained, fled, sued, sabotaged, rebelled and bought or were granted freedom. Their agency made slavery harder to run. It did not make the condition consensual. Every system required force because the people inside it had reasons to resist.

Abolition changed the legal boundary. Revolutions, wars, escape, testimony, campaigning and state action made ownership in people unenforceable across most of the world. That victory mattered. It did not return stolen land, wages, safety or political power, and coercion often reappeared as debt, penal labour, servitude or trafficking. To understand slavery is therefore to distinguish legal status from harsh work, freedom on paper from power in practice, and a trade from the institution it supplied.

That is the book.

Why You Should Care

In the British Museum is a small copper-alloy tag once riveted to a collar. The inscription orders whoever encounters the wearer to prevent escape and return them to an estate in Rome. It gives an owner and an address. It does not give the name of the person forced to carry it.

The object is easy to understand and difficult to absorb. A human attempt to leave has been translated into a property problem. A stranger is recruited as an unpaid guard. The city, its roads and anyone able to read become part of the recovery system. The hardware was metal, but most of its force lay elsewhere: in the shared assumption that one person's claim over another should be recognised after the owner is out of sight.

That is why slavery matters beyond the history of cruelty. Cruelty can be private and chaotic. Slavery made domination durable, legible and transferable. It gave households, merchants and states a way to preserve control through absence, sale, inheritance and distance. An empire did not have to chain every captive every hour. It needed enough people, records and institutions to treat escape as theft and return as order.

The distinction still matters because the word slavery is now used for several different things. People call a punishing job slavery. Governments and charities use modern slavery as an umbrella for forced labour, forced marriage, servitude and trafficking. International law keeps related categories separate because the powers involved are not identical. A worker threatened into labour, a migrant moved for exploitation, a child transferred through a servile marriage and a person held under powers resembling ownership may face overlapping harms, but evidence about one category cannot automatically count the others.

A joint global estimate published in 2022 placed 49.6 million people in situations grouped under the report's umbrella term modern slavery in 2021: 27.6 million in forced labour and 22 million in forced marriage. The scale is appalling. The arithmetic still needs care. These are modelled estimates from current definitions and data, not a count of people sharing one historical legal status or a number comparable with Roman or Atlantic records.

Precision is not a favour to slaveholders. It tells you what has to be dismantled. Pay can be raised. A contract can be cancelled. Debt can be discharged. Slave status reaches further because the controlled object is not one task but the person who performs it. The relevant questions concern movement, punishment, family, sex, children, property, testimony, sale and exit. A privileged palace official can therefore be enslaved while a starving free labourer is not. Conditions matter. Status matters too.

The subject also changes the map. The Atlantic crossing deserves its place at the centre of modern memory, but slavery was never confined to one ocean or one racial order. Ancient Mediterranean cities, Islamic courts, African kingdoms, Asian households and European colonies built different systems from the same dangerous move: deciding that some people could be placed beyond ordinary protection and made available to others. Comparison reveals both the repeated mechanism and the damage done by pretending every case was the same.

Finally, slavery explains something uncomfortable about institutions. Markets did not invent it alone. States did not impose it alone. Families did not reproduce it alone. Each could pass responsibility to the others while benefiting from the result. The collar's owner did not need to stand beside the wearer because an entire social world stood there for him.

Once you see that machinery, the subject stops being a gallery of past horrors. It becomes a test you can apply to any claim about freedom: what powers can one person exercise over another, who makes those powers enforceable, and what would happen if the controlled person tried to leave?

The Core Ideas

Slavery Was a Status Before It Was a Job

The first distinction is the one later arguments tend to blur. Slavery is not a synonym for exhausting work, low pay or dependence. It is a status or condition in which another party can exercise powers over a person resembling powers of ownership. The exact bundle differed across societies. Sale and transfer were common, but not universal in the same form. What joined the systems was control that reached beyond the purchase of a task.

A free labour contract, even an unequal one, is supposed to specify work. Slave status could govern where a person lived, whether they could leave, whom they could marry and whether their children were free. It could also determine punishment, property, testimony and whether another person could inherit or sell the claim. Labour was one use of the controlled person. It was rarely the limit of the claim.

The distinction also works at the level of a whole society. Historians sometimes separate a society with slaves, where slaveholding exists but does not organise most institutions, from a slave society, where politics, wealth, household order and ideas of citizenship are built around the division. The line is disputed and degrees matter. Still, the question prevents a handful of court slaves from being treated as equivalent to a plantation colony whose land, law and ruling class depend on mass enslavement.

That distinction explains cases that otherwise look contradictory. A chained mine worker was enslaved. So was a literate household manager trusted with money, an imperial clerk acting on behalf of a city, a concubine in a palace or a recruit acquired for military training. Some enslaved people lived better than poor free people. Some held authority over free subjects. Privilege could change the conditions of slavery without removing the dependence that made privilege revocable.

This is why a scale from comfortable to miserable cannot define the institution. A starving tenant might suffer more than a well-fed court slave, yet retain a legal family, transmit free status to children and leave property to heirs. The court slave might possess wealth and influence while remaining unable to secure those claims against the will of a ruler. Freedom and welfare overlap, but they are not the same measure.

The distinction also separates slavery from forced labour. International labour law describes forced labour as work or service exacted under menace of a penalty without voluntary offer. That can include a person under slave powers, but it also covers coercion inside a formally free status. Conversely, a child born into slave status was enslaved before performing work. A ruler's concubine was controlled in ways that cannot be reduced to an employment dispute.

No definition removes hard cases. Historians still argue about dependants who could be sold only with land, debtors serving for a fixed term, household retainers with limited rights, and people recruited through military slavery whose legal status changed during training or service. The right response is not to call every dependency slavery or to reserve the word for one Atlantic form. Ask which powers could be exercised, how enforceable they were, whether the person could exit, and what status passed to children.

A person's formal power could therefore be high in one direction and absent in another. An enslaved steward might command workers yet be unable to secure a marriage or protect savings against the owner. Status was a structure of vulnerability, not a prediction of every daily interaction.

Once those questions are visible, slavery stops looking like one kind of labour. It becomes what it was: a system for assigning command over human capacities.

Enslavement Began by Making an Outsider

Nobody is born with a natural mark saying another person may own them. Durable slave systems had to produce a category of people who could be denied protections that insiders expected for themselves. They did it through conquest, distance, stigma and inherited status.

War was one recurring route. A captured enemy stood outside the victor's political community, and killing, ransom or enslavement could be presented as alternatives. Raiding and kidnapping extended the same logic beyond formal battle. Debt, punishment and sale by relatives created other routes. In hereditary systems, the decisive event might have happened generations earlier: the child entered the status because a parent, often the mother, had already been placed there.

Supply often followed the map of protection. A community might forbid the enslavement of its members while buying outsiders captured elsewhere. Traders could search for frontiers where religion, language or sovereignty made claims contestable. Distance helped because a captive removed from witnesses and kin arrived with fewer people able to prove free status or demand return. Many markets required no single doctrine that outsiders were less human. They required buyers, officials or communities willing to treat outsiders' claims as weaker.

The boundaries used to identify the enslavable changed. Greeks and Romans enslaved people from many regions and did not build their systems around one skin colour. Medieval Christian and Muslim traders often treated religious frontiers as lines beyond which captives could be taken, while rules against enslaving co-religionists remained imperfect and frequently broken. In Korea, the category nobi, commonly rendered in English as slavery, could pass by descent within a population without the racial boundary that later structured the Americas. Atlantic colonies then made African ancestry and inherited slave status reinforce each other with unusual reach and rigidity.

Once an outsider category became hereditary, a system no longer needed to remake foreignness in each generation. Law could declare a child born inside the territory to be permanently outside the political community. This is one reason descent rules mattered as much as capture routes.

Orlando Patterson's phrase social death captures part of the operation. Enslavement could sever recognised kinship, strip honour and place a person inside the master's household without full membership in it. Natal alienation, the weakening of enforceable claims through birth and descent, helps explain why removal from home mattered so much. A captive transported far away could remember a family while losing the practical means to make others recognise its claims.

The image needs limits. Enslaved people were not socially blank, and no metaphor should make them disappear twice. They formed marriages, adopted children, joined religious communities, transmitted languages, built neighbourhoods and sometimes acquired recognised rights. Some public servants and people recruited through military enslavement entered institutions that depended on their skill. Manumission made the boundary permeable in some systems. Social death names a pressure towards isolation and dishonour, not a universal description of every life.

Outsider status was also manufactured after capture. Clothing, names, collars, residence rules and restrictions on movement marked people as available for control. Ideologies then converted convention into necessity. Aristotle's natural slave, religious claims about legitimate captives and racial theories in the Atlantic world told beneficiaries that hierarchy reflected nature, faith or civilisation rather than organised force.

That story could be comforting precisely because it hid the decision. A durable slave system required an answer to the question: who may be placed outside ordinary protection? The answers differed, but none was discovered in nature. Each was made, enforced and taught until exclusion looked like a property of the excluded.

Law Turned Capture into a Durable System

Violence can seize a person. It cannot by itself make the seizure survive the captor's absence, death or sale. For slavery to become an institution, other people had to recognise the claim.

Durability came from wider recognition. Law might decide whether an enslaved person could be sold alone or only with land, and whether a child followed the status of the mother or father. It could govern marriage, punishment, testimony and what happened when an owner died. Elsewhere custom, household authority, religion, neighbours or officials carried part of the claim. Fugitives were returned when enough people treated departure as the loss of someone else's property or service.

Recognition also made slave claims financial. A person could be listed in an inventory, pledged against a loan, divided among heirs or seized for debt where the governing rules allowed it. Price did not measure the person's human worth. It measured what buyers expected to control: age, skill, health, fertility, location and legal security. The market therefore rested on an answer to a prior question. What, exactly, would the buyer be able to enforce after payment?

The resulting position was never a clean conversion from person into thing. Enslaved people remained human beings capable of intention, crime, skill and obligation, so legal systems had to recognise enough personality to govern them while denying enough autonomy to control them. Roman slaves could conduct business through a peculium, a fund managed with a master's permission, yet the assets remained vulnerable to that master. Enslaved people could be punished as responsible actors and sold as assets within the same legal order. The contradiction was functional, not accidental.

Different rules produced different systems. The law collection associated with Hammurabi imagines a debtor, spouse or child serving for three years and being released in the fourth. Whatever its relation to daily practice, the rule shows that temporary debt service and permanent slave status could be distinguished. Roman law allowed manumission on a significant scale and created a large freed population, while preserving stigma and obligations to former owners. Other systems restricted sale, attached dependants to estates or made freedom depend on conversion, ransom, military service or an owner's decision.

Hereditary slavery altered the calculation. When status passed through birth, control could reproduce without a new raid or purchase. Atlantic colonies made this principle central and increasingly racial, often assigning children the enslaved status of their mothers. Reproduction then became part of property law. An owner's claim reached into pregnancies, children and future labour, while sale could divide the relationships through which people might have resisted isolation.

A descent rule was an administrative choice with intimate consequences. It settled disputes over mixed-status unions in favour of owners and converted sexual coercion into future property. The law reached the next generation before that generation could act.

Law could restrain owners as well as assist them. Some codes limited arbitrary killing, recognised routes to manumission, allowed self-purchase or accepted petitions against unlawful enslavement. Such rules mattered to people who used them. They did not make the system benign. Regulation often protected the institution by defining acceptable mastery and separating legal abuse from legal domination.

This is the machinery behind the trade. A buyer paid for more than a body in front of them. The purchase depended on a chain of recognition extending into the future. Slavery scaled when a claim first made through force could travel through documents, inheritance and public enforcement as though it were ordinary title.

Empires Converted People into Capacity

The subtitle says slavery built empires. Taken as a single-cause claim, that would be false. Empires also depended on free peasants, taxpayers, merchants, soldiers, tenants, wage workers, conquered allies and administrators. The sharper claim is that slavery gave rulers and elites a recurring way to convert conquest and exclusion into controlled capacity.

The capacities were diverse. Enslaved people cultivated estates, quarried stone, dug ore, built roads, worked docks, rowed ships, cooked, cleaned, nursed children, kept books and made goods. Women and girls faced labour demands alongside sexual access and reproductive control. Public slaves maintained infrastructure and accounts. Court slaves carried messages, guarded rulers and managed households whose private organisation was inseparable from government.

Some uses produced goods that entered markets. Others reproduced the people and institutions that made markets possible. Childcare sustained elite households. Concubinage and coerced reproduction shaped dynasties and future workforces. Eunuchs controlled access to rulers in several court traditions. Domestic service converted intimacy into labour, placing food, illness, children and private information in the hands of people denied an equal claim over their own lives. Economic accounts that count only crops or wages miss much of what owners sought.

Ancient Athens offers the familiar contradiction. Citizen politics rested beside slave labour in households, workshops and the silver mines at Laurion. Rome expanded the range and scale, absorbing captives from war into agriculture, domestic service, commerce, skilled work and imperial administration. Conquest could supply slaves, slaves could support the economy of conquest, and law carried the gains into private fortunes. The loop was powerful without being the whole Roman economy.

Military and court systems reveal why slavery cannot be reduced to cheap muscle. In several medieval and early modern Islamic polities, rulers obtained young outsiders through purchase, capture or coerced levy, converted and trained them, and assigned them to military or palace service. Legal status could change during that sequence, and terms such as mamluk or Ottoman kul did not map neatly onto one unchanged condition. Men of slave origin nonetheless commanded armies, held high office and, in Egypt and Syria, formed a ruling regime. Exceptional careers did not make initial removal voluntary or typical.

Some rulers saw an advantage in disconnection. A recruit removed from a local lineage might have a weaker hereditary base among provincial rivals and depend more on the household controlling training and advancement. Palace women, eunuchs, military recruits and administrators could become intimate agents of power because their position came through the ruler. Dependence could encourage service, but armed corps and palace networks also acquired interests of their own, bargained over succession and sometimes took power.

Private owners made a parallel calculation. Buying a person could secure years of work, specialist knowledge, domestic service and, under hereditary rules, claims over descendants. Slaveholding turned political inequality into an asset that could be mortgaged, inherited or displayed. Prestige mattered alongside output. A large household announced the owner's rank even where the work might have been obtained in other ways.

The word built should therefore be read as supplied and organised, not single-handedly caused. A monument raised by enslaved quarry workers also required engineers, taxes, transport and political command. A plantation empire needed markets and navies as well as coerced labour. The institution mattered through the systems it entered.

Atlantic slavery also fed commercial capitalism through plantation returns, shipping, insurance, credit, customs and markets for commodities and manufactured goods. Historians dispute its scale and causal weight, especially in British industrialisation. The defensible position is neither that slavery invented capitalism nor that its proceeds were marginal. It organised major sectors and shifted wealth through identifiable firms, ports, families and states. Wider growth also depended on domestic agriculture, energy, technology, institutions and free labour.

Slavery therefore built selectively. It helped raise crops and monuments, staff courts, move armies and enrich ports. It also consumed resources in purchase, surveillance, punishment and defence against revolt. An institution designed to create dependable command repeatedly produced fear of the people it controlled. Empire acquired capacity by manufacturing enemies inside its own walls.

Trade Moved People, Reproduction Made Systems Last

The sale of people is slavery's most visible transaction, which is why trade so easily becomes the whole story. It is better understood as one method of supplying a stock of enslaved people.

The distinction between stock and flow prevents several mistakes. A trade figure counts movement during a period. It does not count everyone already enslaved, every local sale, every child born into the status or every person moved without surviving records. A society could import many captives and still hold a smaller slave population if mortality was high. Another could import fewer and expand through birth. Comparing the two with one voyage total would confuse different processes.

Trade changed more than location. Transport assembled strangers, separated families and moved people into jurisdictions where their earlier status was difficult to prove. Markets sorted captives by buyers' expectations and created prices that could encourage further raiding or deliberate reproduction. Yet price should not be mistaken for a complete explanation. Political defeat produced supply, governing rules secured title, ships and caravans carried people, and demand at the destination made repetition worthwhile. Remove any link and the route could contract or move.

Slave trades linked frontier zones to centres of demand. Captives moved through the Mediterranean and Black Sea, across the Sahara, down the Nile, through the Red Sea and Persian Gulf, around the western Indian Ocean, across Eurasian land routes and within African and Asian regions. The Atlantic system later forced millions from Africa towards American plantations and generated an archive large enough to tempt historians into treating it as the normal model. It was distinctive in documented scale, racial heredity and plantation concentration, but it belonged to a longer history of transport, sale and forced incorporation.

Routes did not require one neat triangle. Traders handed people across chains of brokers, ports, caravan stops and markets. Some voyages carried human beings as the principal cargo; others carried small numbers beside textiles, metals, animals or food. Tribute, diplomatic gifts, household transfers and state levies could move enslaved people without an open market. Domestic trades redistributed people after an international route closed, as happened on a vast scale inside the nineteenth-century United States.

Long-distance movement also selected who appeared in surviving records. Ports counted taxable cargo and ships recorded deaths at sea. A child inherited within a household might leave only a name in a will. The archive therefore makes trade easier to count than reproduction.

Reproduction determined whether constant imports were necessary. Sex ratios, family recognition, work, disease, nutrition, mortality and owners' incentives all changed the outcome. Many Caribbean plantations depended heavily on continued Atlantic arrivals for long periods. In the United States, the enslaved population grew through natural increase after legal transatlantic importation ended, while a vast domestic trade tore families apart to supply expanding slave regions. Elsewhere the balance between import and birth is often uncertain because records count purchases more readily than births.

Hereditary status made children part of the system's future. It also made family formation politically dangerous to owners. Families sustained life, memory and mutual aid, yet gave owners relationships they could threaten through sale. Slave systems alternated between encouraging reproduction, separating kin and denying marriages legal force according to their labour needs and fears.

Trade built empires when movement delivered controlled people to places where land, mines, armies or households demanded them. Reproduction made that command persist after the caravan or ship had gone. The trade was the circulation. The status was the structure.

Enslaved People Lived Inside Systems They Did Not Consent To

A history written only from laws, prices and owners' records can make slavery look more complete than it was. The documents describe what masters claimed. They do not prove that every command was obeyed or every relationship severed.

Enslaved people made lives under conditions they had not chosen. They formed partnerships and families, cared for children, worshipped, exchanged news, developed work cultures and carried skills between places. Some accumulated money, cultivated private plots, traded in markets or held delegated authority. These activities could make survival possible and create room for action. They could also increase an owner's profit. The same garden might feed a family and reduce the rations a plantation had to provide.

Family was both refuge and exposed surface. Kin could share food, information and care, preserve names and create obligations stronger than an owner's categories. The same bonds gave owners leverage through threatened sale, forced separation and control over children's status. Women carried a disproportionate part of this conflict because slave systems could claim their labour, sexuality and offspring at once. Accounts that celebrate family resilience without keeping those powers visible turn endurance into a sentimental repair the institution never made.

Resistance took forms matched to opportunity. People slowed work, broke tools, feigned illness, preserved forbidden practices, negotiated, stole back time and goods, escaped temporarily or permanently, petitioned courts, sought ransom, purchased freedom, joined maroon communities and rebelled. Soldiers of slave origin used their organisation to shape regimes. Literate enslaved people turned skills acquired for owners into testimony against ownership. Frederick Douglass made literacy part of his escape from control and later used authorship to attack the institution that had tried to govern his mind.

Even apparent compliance could contain judgement. Skilled workers decided how much knowledge to reveal, domestic servants managed information, and parents taught children which commands were dangerous. Owners noticed the uncertainty and responded with rewards, surveillance and collective punishment. Control had to anticipate minds it could not possess.

None of this proves that slavery was negotiable in an ordinary sense. Bargaining under a power to sell a spouse, beat a worker or claim a child is not a free contract. Nor does a route to manumission cancel the years during which freedom depended on someone else's decision. Elite ascent is an especially poor measure of the whole. A ruler or officer recruited through military enslavement demonstrates one possible trajectory; he does not represent a galley rower, a domestic captive or a plantation labourer.

Agency must also be kept separate from abolitionism. An escape might aim to reunite a family rather than destroy slavery everywhere. A revolt could target one owner, labour regime or ethnic hierarchy. An enslaved trader could own or sell other people. Human beings acted within inherited moral worlds and unequal choices; they were not required to hold a modern political programme before their actions count.

The Haitian Revolution shows what enslaved action could produce at the scale of a state. Beginning in 1791 in the French colony of Saint-Domingue, enslaved rebels fought through civil war, imperial invasion and shifting alliances until slavery was destroyed and Haiti became independent. The revolution was not a moral gift from an enlightened metropole. People classified as property became military and political actors who broke the classification.

Every smaller act mattered for the same reason. Slavery promised owners command over human capacities. Enslaved people kept using those capacities for purposes of their own. The gap between the claim and the person is where family, culture, negotiation and revolt entered the system, and why domination always required maintenance.

Abolition Changed the Legal Boundary, Not Every Power Relation

Slave systems lasted because they were normalised, profitable to particular groups and supported by law, custom and organised power. Their destruction was never the automatic result of economic progress or moral awakening. Enslaved people's flight, revolt, military service and testimony combined with religious organising, political campaigning, war, revolution and changing state interests. Different coalitions ended different systems.

Abolition therefore required more than closing markets. Legislatures had to cancel title, courts had to reject ownership claims, armies and police had to defend emancipation, and local officials had to recognise movement, marriage and contracts. The same sequence rarely happened at once. A trade could be banned while people already enslaved remained in status. A proclamation could declare freedom where owners still controlled land and violence. A treaty could prohibit slavery while administrators lacked the will or reach to enforce it.

In Haiti, abolition came through revolution. Britain abolished its Atlantic slave trade in 1807 and most colonial slavery under the 1833 Act, then required many emancipated people to pass through apprenticeship until 1838 while providing £20 million in compensation to slaveholders. The United States abolished slavery constitutionally in 1865 after civil war, with an exception for punishment following conviction. Cuban slavery ended in 1886; Brazil followed two years later. Elsewhere legal abolition extended into the twentieth century, and implementation often lagged behind the decree.

Ending status also altered the meaning of family and movement. A marriage no longer existed by an owner's permission alone. A child was no longer an increment to an estate. Leaving work ceased to be flight from ownership, even where poverty, contracts or violence made departure perilous.

What changed was neither symbolic nor minor. An owner could no longer invoke ordinary law to sell a person, inherit a child, return an adult fugitive or enforce labour because of slave status. Families gained claims that had previously depended on an owner's tolerance. People could contract, move, own and testify on a new legal footing, though discrimination sharply limited the use of those rights.

The settlement rarely reversed accumulated advantage. Former owners retained land, capital and political connections. Emancipated people often began without property, wages owed for past work or protection against violence. Apprenticeship, indenture, debt peonage, tenancy, pass laws and penal labour could impose severe coercion while preserving formal freedom. These institutions must be studied in their own terms. Calling all of them slavery can hide the legal victory of abolition and the different machinery used to restrict it.

Modern law therefore maintains several categories. The 1926 Slavery Convention centres powers resembling ownership. The 1956 Supplementary Convention addresses debt bondage, serfdom, servile marriage and the transfer of children alongside slavery. International labour law defines forced labour through involuntary work exacted under menace of penalty. Trafficking law concerns recruitment, transport, transfer, harbouring or receipt for exploitation through specified improper means; when a child is involved, those means need not be proved. One person can fall within several categories, but the words are not interchangeable.

The causal loop closes here. Slavery began by making control over a person enforceable beyond one task. Abolition removed that status from legitimate law. Where land, debt, policing, migration rules and political power remained unequal, coercion could return without restoring the old title. Freedom on paper is therefore both decisive and incomplete: decisive because legal ownership matters, incomplete because a person needs the practical power to use the rights that law declares.

How It Actually Works

A debt with an end

A household in Babylonian Mesopotamia that could not meet a debt might surrender labour rather than land or silver. The law collection associated with Hammurabi, compiled in the eighteenth century BCE, imagines a debtor giving a wife, son or daughter into service for three years, with release in the fourth. The clause is severe by modern standards. In its own setting, it marks a limit.

That limit tells us that ancient people could distinguish temporary debt service from slave status. Mesopotamian records used several categories for dependants, captives, household slaves and people pledged against debts. The categories shifted across languages, cities and centuries. Some enslaved people worked for temples and palaces; others belonged to private households. They could be bought, inherited, adopted, married, freed or pursued after flight under rules that did not form one neat code.

War remained a major source. Early states generated captives at the same time as they generated grain stores, taxes and armies. A defeated population could be killed, resettled, ransomed or distributed. The decision converted foreign defeat into labour and status inside the victor's society. Yet debt produced a different problem. If members of the community could be lost permanently to creditors, households, military service and tax revenue disappeared with them. Rulers therefore had reasons to cancel debts or limit service that had little to do with equality.

Clay tablets also reveal why institutional slavery became visible to historians. Palaces, temples and households recorded rations, births, purchases, fugitives and transfers because administration required categories. The surviving person often appears first as an entry in someone else's account, a pattern that will continue for millennia.

The first large written archives already show the ingredients that would recur: an outsider who could be transferred, a local dependant whose service might be bounded, a household able to claim labour, and a public authority deciding which claim counted. There was no single invention called slavery. There were repeated decisions about when a person could be treated as the durable object of someone else's command.

Greece and Rome make the boundary visible

Greek city-states sharpened the contrast between citizen freedom and slave status. Athens celebrated the citizen who spoke and voted while excluding women, resident foreigners and enslaved people from the assembly. Slaves worked in homes and workshops, performed public tasks and dug silver-bearing ore at Laurion. The democratic citizen body did not experience this as an anomaly requiring abolition. Freedom acquired part of its meaning from the existence of people who lacked it.

The Greeks also supplied one of slavery's longest afterlives: a theory that hierarchy could be natural. Aristotle argued that some people lacked the rational capacity for independent rule and were slaves by nature. The argument solved a moral problem for slaveholders by turning conquest and purchase into recognition of a prior fact. It could not explain why Greeks captured in war became slaves, why enslaved people displayed the skills owners relied upon, or why manumission changed a supposedly natural condition. Its weakness did not prevent later readers from reusing it.

Rome built a wider slave system as conquest brought captives into an expanding Mediterranean market. Enslaved people appeared in agriculture, mines, households, workshops, shops, shipping, entertainment and the imperial administration. Their conditions ranged from lethal extraction to trusted management. The state itself owned slaves. Exact population totals are unrecoverable, and the balance changed across regions and centuries, but slavery became embedded in property, status and the organisation of elite life.

Roman law exposed the person-property contradiction. A slave could manage a peculium, conduct business or carry an owner's authority, yet the master retained the superior claim. Manumission was common enough to create a large and visible freed population. Freedom did not erase past status: freedpeople could owe duties to former owners and faced social stigma. Still, the possibility mattered. It created incentives, family strategies and a route by which a person legally classed under another's power could acquire a recognised self.

Supply did not come only from imperial victories. Birth, exposure, kidnapping and regional trade continued after major conquests slowed. One demographic model suggests that birth may have supplied a larger share of the early imperial slave population than older conquest-centred accounts allowed. The evidence cannot establish a percentage, and the balance differed by time and place. The small collar tag now in the British Museum belongs to this later world. Its command to return the wearer shows an institution maintained in streets and estates, long after the battlefield that may or may not have begun that person's enslavement.

Frontiers feed connected seas

The political break-up of the western Roman Empire did not make slavery vanish. Forms of dependence changed, and in parts of Europe serfdom and tied tenancy became more important, but enslavement, sale and household slavery continued around the Mediterranean. The dividing lines often followed religion and political membership rather than colour.

Christian legal and political authorities often restricted the enslavement of Christians by Christians; Muslim jurists protected free Muslims from lawful enslavement. Practice repeatedly violated principle. Conversion could alter status in some settings and leave it untouched in others. Traders and raiders therefore searched beyond protected communities or reclassified victims as legitimate outsiders. Frontiers became reservoirs of people made available by war, piracy and disputed identity.

Judaism, Christianity and Islam all developed in societies where slavery already existed. Scriptures, jurists and religious institutions regulated status, treatment and manumission without producing an immediate general ban. Slaveholders later cited religious authority; abolitionists drew on deliverance, common humanity and moral law. The traditions supplied arguments on both sides because interpretation moved inside political conflict rather than above it.

Between the thirteenth and fifteenth centuries, merchants connected the Black Sea to Mediterranean markets. Captives from the Caucasus, the Eurasian steppe and eastern Europe passed through ports controlled or used by Genoese, Venetian, Byzantine, Muslim and other traders. The same maritime world moved grain, cloth, metals and spices. Enslaved people travelled within commercial systems rather than along a route reserved for them.

Other networks crossed the Sahara, followed the Nile, linked the Red Sea and Persian Gulf, and joined eastern Africa to Arabia, India and western Indian Ocean islands. These trades varied in scale, sex balance, destination and work. Some caravans and ships carried many captives; others moved a handful among mixed cargoes. Household service and concubinage created demand unlike plantation agriculture or military recruitment.

Piracy and naval war added another supply line. Captives could be ransomed if families or governments paid, exchanged in diplomacy, or sent to galleys and households when they did not. The same person might move between captive, hostage, slave and freed status as negotiations changed.

No centre directed the whole. Wars, debts and kidnappings supplied people; brokers assembled them; rulers taxed or protected routes; merchants carried them; buyers assigned new uses. Distance weakened access to kin and law while giving each intermediary room to deny responsibility for the chain. The network survived a port or dynasty's fall because its stages were replaceable.

Bought for the ruler

In the ninth-century Abbasid world, caliphs recruited Turkic military slaves from beyond the empire's settled political communities. They were purchased or otherwise acquired, converted, trained and attached to rulers or commanders. One aim was to build forces less entangled in local factional loyalties, although recruitment and attachment varied. Military skill and proximity to power soon gave some men leverage of their own.

Military enslavement spread and changed. Men recruited through it commanded armies and founded dynasties across western and central Asia and India. In 1250, officers of mamluk origin took power in Egypt and created a sultanate that ruled Egypt and Syria for more than two and a half centuries. Purchase, training, manumission and office were distinct stages whose timing varied. The evidence does not support a rule that every recruit was freed when training ended. A man could command free subjects after manumission while the regime obtained new recruits through enslavement.

The Ottoman system developed a related but distinct form. Through the devshirme, officials levied selected Christian boys from parts of the Balkans and converted and trained them for palace service or the Janissary corps. Some later held senior office within the sultan's household. The category kul could mark legal dependence on the ruler despite wealth and command. The household also included enslaved women, eunuchs, artisans and attendants whose positions ranged from forced intimacy to political influence. Exceptional careers cannot turn removal into consent or describe most Ottoman slavery.

Why recruit outsiders? Some rulers expected trained dependants to have weaker hereditary bases among provincial rivals and stronger ties to the household controlling advancement. The arrangement could supply disciplined service while preserving dependence. It also created armed corps and palace networks able to bargain over access, revenue and succession. Abbasid guards destabilised caliphs; officers of mamluk origin became sovereign; Janissaries acquired corporate interests and family connections that recruitment had been intended to limit.

These careers do not invert slavery's moral meaning. They reveal its range and the need to separate legal phases. The capacity sought might be strength, literacy, fertility, access or administrative skill. A palace could give an enslaved dependant influence over thousands, or elevate a man after manumission, while preserving recruitment founded on other people's inability to refuse.

No continent contained one system

There was no single institution called African slavery before or during Atlantic expansion. Political communities used captives, pawns, dependants and enslaved people in varied ways. Some systems allowed incorporation into a lineage or household over time. Others preserved foreignness, allowed sale across distance or made status hereditary. Enslaved people farmed, traded, served courts, joined armies, worked in mines and households, and could sometimes acquire dependants of their own.

External trades interacted with these institutions rather than meeting an empty market. Demand across the Sahara, Red Sea, Indian Ocean and later Atlantic encouraged warfare and commercial capture in some regions, while political authority and supply conditions constrained outsiders in others. African rulers, merchants and soldiers made consequential choices. Captives were also African actors, and the category of shared continental identity tells us little about the political loyalties of people living in rival states and communities.

The nineteenth century brought further changes. Suppression of Atlantic imports did not end slavery within Africa. In several regions, slaveholding and coerced agricultural production expanded as rulers and merchants redirected labour towards local or export production. Colonial conquest later abolished some legal statuses while imposing taxes, compulsory crops and forced labour under different rules. A decree could change the name of domination without giving former slaves land or protection.

East Asian systems add another correction. Korean nobi could be hereditary, and holders could buy, sell and bequeath people within a society that did not divide slave and free by race. Yet conditions ranged from household service to separate residence and payments resembling tenancy. The category is often translated as slavery, but the English label needs qualification rather than automatic rejection or acceptance. Chinese states and households also used slaves, bondservants and penal dependants whose status varied by dynasty, region and owner.

An early nineteenth-century narrative attributed to Wange Hendrik Richard van Bali gives one local rule a human shape. On Flores, his free father could not prevent three sons being claimed and sold separately because their mother and maternal ancestors had been enslaved. The account is not a regional template. Across Indian Ocean and Indonesian archipelago routes, movement and destinations varied, and newer research complicates any region-wide sex ratio.

Comparison works only if difference remains visible. Incorporation into a household was not the same as freedom. Separate residence was not proof of free status. A route to manumission did not make original enslavement voluntary. The global pattern lies in the powers exercised and the boundary enforced, not in one uniform job or legal vocabulary.

The Atlantic system changes the scale and rule

Portuguese expansion along the African coast in the fifteenth century joined older Iberian, Mediterranean and African practices to Atlantic islands and, after 1492, to American conquest. Europeans enslaved Indigenous people, and that enslavement persisted in many colonies. Disease and colonial violence devastated communities. Plantation expansion and deepening Atlantic commerce increasingly drew captives from Africa, but timing and balance differed sharply by colony. There was no single replacement.

Sugar plantations on Atlantic islands and in Brazil helped assemble a model later expanded across the Caribbean and the Americas: concentrated land, export crops, high capital needs, armed discipline and a workforce treated as alienable property. European and American merchants and states dominated oceanic carriage, finance and colonial demand. African rulers, brokers and raiders were major suppliers at the coast, within political systems that the trade altered unevenly. Coercion and commerce crossed those categories, and no group occupied one role everywhere. Colonial governments wrote racial and hereditary rules that reached into marriage, motherhood and children.

The ocean mattered because forced distance weakened claims that might have protected a captive at home. Law in the Americas then converted isolation into inheritable title. A person could be inspected, insured, mortgaged, sold and moved again after landing. Plantation owners did not buy labour for one season. They acquired a legal claim over time, family and descendants.

The crossing itself was a labour and control regime. Captives were confined, fed at calculated levels, exposed to disease and sexual violence, forced to exercise, punished and watched against revolt. Millions survived the crossing only to enter slave systems in the Americas; many others died before embarkation, at sea and after arrival. The complete Atlantic chain, its routes and numbers belong to The Slave Trade in a Hurry. Here its comparative importance is the transformation: older tools of enslavement were joined to oceanic finance, plantation demand and racial heredity at a scale capable of reorganising continents.

Brazil and the Caribbean received the largest shares of Atlantic arrivals, although later United States memory often pulls the map north. Different American slave societies then developed different balances of imports, birth, manumission and domestic sale. Across them, enslaved people created families, religions, languages, markets and political movements that owners could exploit but never fully command.

Abolition becomes a political programme

Criticism of enslavement existed long before modern abolition. Enslaved people escaped and resisted wherever slavery existed. Religious and legal traditions sometimes protected insiders, limited abuse or encouraged manumission. What emerged from the late eighteenth century was more demanding: organised movements and revolutions seeking to end slave trades or slave status as institutions.

The Haitian Revolution broke the assumption that emancipation would be granted from above. The uprising that began in Saint-Domingue in 1791 grew through war among enslaved rebels, free people of colour, white factions and competing empires. French abolition in 1794 was followed by Napoleon's attempt to restore control. Haitian forces defeated the expedition, and independence in 1804 created a state from the destruction of the plantation slave regime.

British abolition combined Black testimony, religious organisation, mass petitioning, parliamentary work, geopolitical calculation and resistance throughout the Caribbean. Parliament prohibited the British Atlantic trade in 1807. The 1833 Act ended slavery in most British colonies from 1834, but converted many emancipated adults into apprentices required to continue labour. Apprenticeship ended early in 1838. Slaveholders received £20 million in compensation. The formerly enslaved did not receive wages for the years taken from them.

In the United States, slavery's expansion and political protection produced secession and civil war. Emancipation became a war measure and then constitutional law through the Thirteenth Amendment, ratified in December 1865. Its punishment exception would later sit beside systems of criminalisation and coerced penal labour. Cuba abolished slavery in 1886 and Brazil in 1888. Across Africa, Asia and the Middle East, abolition came through local reform, imperial pressure, anti-colonial politics and international agreements, often with long gaps between law and practice.

The common sequence was not moral progress running on schedule. Slave systems ended when resistance changed costs, coalitions acquired power and states withdrew enforcement from owners. The same states could then preserve landholding, racial hierarchy or labour coercion in other forms.

After ownership

Emancipation removed the lawful status that made one person inheritable or saleable by another. It did not supply land, education, security, credit or equal political power. Former owners could offer contracts in markets they still dominated. Governments could criminalise movement, impose passes, enforce debts or lease prisoners. Colonial states that condemned private slave trading could compel labour for roads, mines and crops.

Freedom could also be lost through reclassification. Vagrancy laws and contract offences made movement punishable in several post-emancipation settings. Employers and officials could use debt, recruitment fees or document control to make a formally voluntary migration hard to reverse. None restored the old inheritance rule by definition, but each tested whether the state would protect an individual's right to leave.

These arrangements should not be collapsed into chattel slavery. Their legal mechanisms, routes of exit and effects differed. The distinction is the means by which continuity can be proved rather than assumed. A labourer bound by debt, a prisoner compelled to work, a migrant whose documents are confiscated and a person sold under an ownership claim may all face extreme coercion. The evidence must show which powers are present.

Twentieth-century international law reflected that problem. The 1926 Slavery Convention defined the institution through powers attaching to ownership. The 1930 Forced Labour Convention addressed work exacted under threat without voluntary offer. A 1956 supplementary treaty covered institutions including debt bondage, serfdom, servile marriage and child transfer. Later trafficking law focused on recruitment, movement or receipt for exploitation. The categories overlap because coercive systems combine methods. They remain separate because ending one method may leave another intact.

How we know

Slavery left abundant records written for owners, states and markets: sale contracts, wills, tax lists, court cases, ship papers, plantation accounts, laws and notices for fugitives. They reveal transactions and rules with unusual precision while reducing people to categories useful to power. Material evidence, including collars, housing, tools and burials, can recover parts of life that clerks ignored.

First-person narratives, petitions, letters, songs and testimony restore intention and experience, but no survivor can represent every setting. The Atlantic trade supports large quantitative reconstructions because voyage records are unusually dense. Other routes and periods leave fragments that cannot be added into a compatible world total.

Comparison therefore works through bounded questions: how status began, which powers an owner could exercise, how people moved, what passed to children and how freedom was gained. The labels in sources are translated cautiously. A global account can identify recurring mechanisms. It cannot turn one legal code, household, voyage or empire into the template for all.

What People Get Wrong

“Slavery was always racial”

Race became one of slavery's strongest organising principles, not its timeless definition. Greeks enslaved Greeks and foreigners. Roman markets drew captives from across Europe, Africa and Asia. Medieval Christian and Muslim systems often drew legal boundaries around religion and political membership. In Korea, nobi status could pass by descent among people without the racial division familiar from the United States.

The myth is persuasive because the most familiar slave societies are modern Atlantic ones, and their racial categories survived emancipation. Public memory begins near the end of a much longer history and mistakes the final dominant form for the definition.

The Atlantic system changed this history. European colonists increasingly joined African ancestry, presumed enslavability and hereditary status through law. Skin colour became evidence of legal condition, and racial ideology made a recent political arrangement look biological. This order was neither the first slavery nor one example among equals. Its scale, duration and influence on the modern Atlantic world make it a distinct transformation. The correction matters because saying slavery was always racial makes Atlantic racism appear inevitable, while saying slavery was usually non-racial can minimise what the Atlantic system built.

“Every enslaved person worked on a plantation”

Plantations dominate the visual record because Atlantic sugar, cotton and coffee created large, concentrated workforces and extensive commercial archives. Many slave systems across history were not organised around that one workplace.

The plantation also offers a clear visual geometry: owner, overseer, field and export crop. Household and sexual slavery happened behind doors; administrative slavery can disappear into job titles; military slaves upset the expected ranking of master above slave. Memory selects the form easiest to picture.

Enslaved people worked in households, workshops, mines, quarries, ports, ships, armies, courts and administrations. They served as nurses, cooks, artisans, accountants, guards, concubines and agricultural labourers. Some lived under direct surveillance; others travelled, handled money or managed property. People in public service or recruited through military enslavement could exercise authority over free people, sometimes after their formal status had changed. None of these roles made the original coercion harmless. Looking only at plantations hides the uses that made slavery attractive to states as well as landowners.

“Slavery and serfdom were the same”

Both could bind people to hard labour, inherited disadvantage and a superior's authority. The difference was not comfort. It concerned the structure of the claim.

A serf was usually tied to land, estate or jurisdiction and could owe labour, dues and restrictions on movement. A slave was more commonly subject to powers exercised over the person, including separate sale or transfer, though historical boundaries blurred. Some dependants labelled slaves could be sold only with land. Some serfs were transferred with estates and had little practical exit. Local vocabularies rarely line up cleanly with modern English.

Translation keeps the confusion alive. Words rendered servant, slave, serf, dependant or bondsman carried local histories that English cannot reproduce in one label. The answer is not to abandon categories. It is to state which claims travelled with the person, which stayed with the land and how freedom could be acquired.

The distinction still matters because people fought over it. Freedom papers, marriage rights, inheritance and children's status could turn on which category a court recognised. Treating all unfreedom as one condition hides the legal changes people sought and the different tools owners used.

“Europeans invented slavery in Africa”

African societies contained slavery, pawnship, captivity and other dependencies before Atlantic expansion. African rulers, merchants and soldiers captured and sold people. Denying this removes African political history and turns a continent into scenery.

The claim often appears in an argument about blame rather than explanation. One side imagines an untouched continent corrupted from outside. The other cites African participation as though a seller's existence dissolved the buyer's demand, shipping, law and plantation system. Both remove the chain of distinct decisions.

The opposite conclusion fails as well. Pre-existing institutions did not make Atlantic demand irrelevant or distribute responsibility equally. European and American merchants, colonies and states created an oceanic market, supplied ships and finance, bought captives on a growing scale and built plantation laws that made African ancestry hereditary property. Demand altered incentives for warfare, punishment and sale unevenly across regions. Some African states profited; others resisted, lost population or were attacked.

“Their own people” is a modern continental phrase imposed on societies organised around kingdoms, towns, lineages and enemies. It neither excuses sellers nor explains captives. Responsibility belongs to identifiable actors and powers, not to a colour-coded collective.

“The Atlantic trade was the whole history”

The Atlantic crossing moved millions of people by force and sits at the centre of slavery's modern racial and economic inheritance. It was not the only slave trade.

Long-distance routes crossed the Mediterranean and Black Sea, the Sahara, the Nile corridor, the western Indian Ocean and Eurasia, while domestic trades moved people within African, Asian, European and American regions. Domestic sale could remain huge after an international route closed. Birth could sustain a slave population without any continuing import trade.

The Atlantic also leaves an unusually dense numerical trail. Ships, customs houses, insurers and plantations produced records that can be converted into routes and estimates. A caravan crossing a desert, a household transfer or a child inheriting status may leave far less. Evidence density can become historical importance by accident.

The correction changes measurement. Atlantic voyage estimates cannot be added casually to caravan fragments, household inventories or demographic reconstructions from other periods. Nor does a trade total equal the number living in slavery. Keeping the Atlantic system in global context enlarges the history without shrinking its significance. It shows which features were older tools and which were Atlantic innovations in scale, racial heredity and plantation concentration.

“Enslaved people had no agency”

Slaveholders claimed command, not success. Enslaved people chose among constrained options every day: forming families, exchanging information, negotiating tasks, saving money, petitioning, escaping, preserving religious practice, damaging property and organising revolt.

Agency does not mean freedom, consent or equal bargaining power. A person deciding how to protect a child from sale is acting; the threat that shapes the decision remains. Someone recruited through military enslavement may later gain office while remaining tied to a court founded on forced recruitment. A garden plot can support a family and subsidise the owner who reduced rations.

A second distortion turns any adaptation into collaboration. Learning an owner's language, accepting delegated authority or purchasing freedom can look like approval when read without the threat structure. People used available institutions because those were the institutions they had to survive, not because the menu of choices was fair.

The myth survives partly because the archive records ownership more reliably than resistance. It also survives because replacing passive victims with heroic resisters feels like a correction. Most lives contained neither passivity nor constant rebellion. They contained strategy under pressure. That is enough to make people historical actors without making the system less coercive.

“Abolition ended coercion”

Abolition ended something exact and important: the lawful power to hold a person in slave status. It did not distribute land, repay stolen wages or erase the institutions built around ownership.

Former slaveholders often retained property, credit and political influence. Emancipated people could face apprenticeship, coercive contracts, debt, tenancy, racial terror, pass controls or penal labour. Colonial governments sometimes condemned slave trading while compelling labour for public or private projects. These arrangements could reproduce severe exploitation without restoring the same legal title.

National stories encourage a clean ending: a law is passed, chains fall and the next chapter begins. Real transitions involved apprenticeship, delayed implementation, illegal re-enslavement, compensation to owners and battles over land, contracts and citizenship. Dates mark changes in law, not the instant arrival of equal power.

Calling every successor slavery is tempting and analytically weak. It obscures both the victory that made sale and inheritance of persons illegal and the new mechanisms that replaced them. The useful question is not whether coercion feels like slavery. It is which powers remain, which have changed, who enforces them and what route to exit exists.

Use It

Ask which powers can be exercised

Labels are unreliable. A government may ban slavery while tolerating a condition in which an employer controls documents, movement and exit. A historical source may call someone a servant while recording sale, inheritance and punishment. Another may call a dependant a slave even though the claim was temporary and could not be transferred.

Begin with powers. Who can decide where the person lives and works? Can the relationship be refused or ended? Can the person marry, keep earnings, hold property, testify or protect children? Can another party sell, transfer, inherit or return them after escape? What happens when an order is disobeyed?

Add time to the test. A threat attached to one disputed shift differs from a condition designed to persist across years, owners or generations. Durability is often where exploitation becomes status.

This method prevents two opposite errors. It stops polite language from hiding domination, and it stops one harsh feature from turning every unequal relationship into slavery. The closer the claim reaches towards the whole person, and the more law, custom or organised power enforces it beyond one task, the closer the condition comes to slavery's central structure.

Separate status from work

Do not infer freedom from occupation. Two people can perform the same task under different statuses. A free sailor, an indentured worker, a convict and an enslaved rower may pull the same oar while possessing radically different claims over movement, family, pay and exit.

The reverse matters too. Enslaved people did not occupy one economic class. A mine worker, household child, court administrator and recruit held under military slave status could share a legal disability while living under conditions too different for one labour history. Measuring food, housing or authority alone will miss the common vulnerability. Measuring status alone will miss the difference between a palace and a mine.

Then add a third column for what passed to children. Hereditary status can turn a labour arrangement into a social order without changing the task performed today.

Carry both columns. Ask what the person did and what could be done to the person. The first explains production and daily life. The second explains why opportunity, skill or apparent privilege might remain conditional on another's will.

Separate stock from flow

Whenever a large number appears, ask what it counts. A slave trade total usually measures people moved along a route during stated years. A census measures people present at one moment. A sale register records transactions, which may include the same person more than once. A household list omits people outside the record's purpose. Births into hereditary status may leave no trade entry at all.

This distinction changes historical interpretation. Heavy imports can signal expansion, high mortality, an imbalanced sex ratio or owners preferring purchase to family formation. Population growth after imports fall can signal natural increase, domestic trade or wider enumeration. No one number answers all four questions.

Check the denominator as well. People embarked, people arriving, voyages documented and transactions recorded answer different questions. A percentage can be accurate and still compare unlike populations.

The same discipline applies beyond slavery. Flows of migrants are not migrant populations. New debts are not outstanding debt. Prison admissions are not the prison population. Confusion between movement and stock can turn a dramatic statistic into a false model.

Follow the public machinery behind private control

A slaveholder appears to exercise private power, but private ownership cannot travel far without public help. Someone recognises the sale, records inheritance, restricts testimony, returns fugitives and decides whether violence is punishment or crime. Ports, courts, patrols, armies and tax offices turn a household claim into an institution.

This lens also clarifies abolition. A moral change inside owners' homes was not enough. Governments had to stop enforcing sale and return, recognise marriages and contracts, accept testimony, punish re-enslavement and make freedom portable between jurisdictions. Where officials remained hostile or absent, legal emancipation could be undermined in practice.

Informal enforcement belongs on the map too. Neighbours, employers, relatives, clergy and merchants may recognise a claim before any official arrives. Law is strongest when social expectation does part of its work.

Follow enforcement before assigning explanation to markets, culture or prejudice alone. Ask who wrote the rule, who could invoke it, who bore the cost of enforcement and which authority could refuse. Domination becomes durable when institutions make one party's claim ordinary administration.

Track replacement, reproduction and exit

A coercive system has to maintain its workforce and status order. Ask where new people come from, how long they remain, whether children inherit the condition and how anyone leaves it.

Imports, raids and purchases are costly. Hereditary status can reproduce control locally, but requires rules about children, sex and family. High mortality creates renewed demand. Manumission can reward service, release ageing dependants, create loyal clients or absorb outsiders. Escape and revolt impose security costs. Each route changes owners' incentives and enslaved people's strategies.

Exit routes reveal whose interests a system served. Manumission controlled by owners differs from freedom secured by law, escape or collective revolt. The same outcome can carry a different distribution of power.

This lens makes comparison sharper. A plantation dependent on continual imports, a Roman household weighing manumission, a Korean estate inheriting nobi status and a court acquiring military recruits through enslavement may all use related forms of slave status while solving replacement differently. Their differences are part of the mechanism, not inconvenient detail.

Recognise agency without mistaking it for consent

People act under coercion even when every available option is constrained. They bargain, conceal, comply, delay, care, learn, flee and fight. Some actions reduce harm. Some improve status inside the system. Some help an owner. Some attack slavery directly.

Do not make resistance a purity test. An enslaved person who never rebelled was still denied freedom. A person who gained influence might defend the institution that elevated them. A parent who accepted one demand to protect a child made a choice under a threat, not a free bargain.

Apply the same separation when reading praise from owners. Loyalty produced under dependence may be sincere, strategic or both. The relationship remains coercive even when every feeling inside it cannot be recovered.

Agency asks what people did with the options they could perceive. Consent asks whether the relationship itself was entered and could be left without coercion. Keeping the two apart restores enslaved people as actors without turning survival into approval.

The limits

A global comparison gains range by losing texture. Slavery in a Roman port, a barracks filled through military enslavement, a Korean household, an African farming community and a Caribbean plantation cannot be compressed into one lived experience. The book has compared powers, routes and institutions, not declared every slave system equivalent.

The evidence is also tilted. Owners generated contracts, accounts and laws because control required records. Enslaved people left fewer sources, and those who became literate, reached courts or survived to testify were not a random sample. Silence cannot tell us that family, resistance or suffering was absent. It can tell us which lives an archive was not built to preserve.

“The trade that built empires” must remain a bounded claim. Slavery supplied labour, soldiers, administrators, reproduction, prestige and wealth to many imperial systems. It did not build every empire, explain every conquest or cause capitalism by itself. Free and coerced labour often worked beside each other. Taxes, land, technology, disease, ecology and political organisation still matter.

Modern analogy has the same limit. Historical understanding can tell you which powers to investigate. It cannot prove that a present institution is slavery because it feels exploitative or resembles one feature of the past. The classification requires evidence about control, menace, transfer and exit.

The one thing to keep

Keep one question from the Roman collar: whose claim does the surrounding world recognise?

The metal mattered because it marked a person for return. The inscription mattered because strangers could read it. The estate, road and city mattered because the owner's reach extended through them. Slavery was strongest when command no longer looked like a contest between two people and became the routine response of everyone else.

Freedom therefore requires more than the absence of a visible chain. It requires an enforceable claim to one's body, movement, family, work and future, carried beyond the room in which power is challenged. Ask who can leave, who will be returned and whose account the law treats as the first person. That is what should now be different in how you see the world.

Terms

A working vocabulary for comparing systems without pretending that one language, period or legal order supplies universal categories. Each term is a question to apply, not a label that settles the evidence.

Slavery

The legal or social condition created when another party can exercise powers resembling ownership over a person. The bundle varies by jurisdiction, so control, transfer, inheritance and exit must be examined directly.

Enslavement

The process by which someone is placed into slave status, through capture, kidnapping, debt, punishment, birth, sale or another route. It names an action rather than a natural identity.

Slave trade

The transfer and commercial movement of enslaved people between sellers, brokers and buyers. Trade measures flows and transactions, not the total population living under slave status.

Chattel slavery

A form in which enslaved people are treated as movable property that can be sold separately from land. Atlantic plantation slavery made chattel status racial and hereditary with unusual rigidity.

Alienability

The capacity of a claim to be transferred to another holder. In slavery, alienability means control over a person can pass through sale, gift, inheritance or seizure.

Status or condition

The wording used by the 1926 Slavery Convention. Status points towards legal recognition; condition allows comparable control to be identified even where domestic law avoids the label.

Powers of ownership

A functional legal test asking whether another person is treated as able to possess, use, manage, profit from, transfer or dispose of someone's capacities. The pattern matters more than one formal title.

Social death

Orlando Patterson's model of slavery as natal alienation, dishonour and exclusion from recognised social belonging. Powerful in some contexts, but too uniform for every slave status.

Natal alienation

The weakening or denial of enforceable claims arising from birth, ancestry and kinship. It helps explain why slave systems could control marriage, children and inherited status.

Hereditary slavery

A system in which slave status passes to children under a rule of descent. It reproduces the enslaved population without a new capture or international purchase.

Slave society

A society in which slavery is central to economy, politics and social order, rather than a society that merely contains some enslaved people. The boundary remains debated.

Unfree labour

A broad analytical category for work constrained by status, law, debt or coercion. It can include slavery, serfdom, forced labour and other relationships that should still be distinguished.

Forced labour

Following ILO Convention No. 29, labour required under a threatened penalty from someone who did not offer it voluntarily. It overlaps with slavery but does not require legal ownership powers.

Servitude

A condition of severe domination in which a person is compelled to live and work for another and cannot realistically change that condition. Legal definitions vary by jurisdiction.

Debt bondage

A condition in which personal service secures a debt without fair valuation, defined duration or genuine reduction of what is owed. Debt becomes a mechanism for indefinite control.

Pawnship

The transfer of a person to secure an obligation, often within kinship and credit systems. Rights and duration varied, so pawnship should not automatically be treated as chattel slavery.

Corvée

Labour demanded by a state or lord for specified public, military or agricultural tasks. It can be coercive without making the worker a slave in legal status.

Serfdom

A form of inherited dependency usually tied to land, estate and dues rather than complete alienability of the person. Practice could approach slavery in some times and places.

Captivity

Control over a person taken in war, raid or kidnapping. Captivity may end in ransom, release, incorporation or enslavement; the categories are connected but not identical.

Domestic slavery

Enslavement within a household, covering cooking, cleaning, childcare, personal service, craft and management. Physical proximity to owners could bring trust, surveillance and sexual vulnerability together.

Concubinage

A recognised or tolerated sexual and domestic relationship below marriage in status. Where one party was enslaved, intimacy operated inside an unequal power over consent, children and freedom.

Plantation

A large agricultural enterprise producing crops for sale through concentrated land, capital and labour. Some plantations used slavery; plantations are neither universal to slavery nor inherently identical with it.

Peculium

Property or money managed by a Roman slave with the master's permission. It could fund business or self-purchase, while remaining legally vulnerable to the master's superior claim.

Manumission

The legal release of an enslaved person by an owner, will, purchase, service or public act. It creates individual freedom through the same legal order that previously recognised ownership.

Freedperson

Someone released from slavery. Freedom could coexist with stigma, duties to a former owner, restricted citizenship and insecure family claims, depending on the legal system.

Maroon

A person or community that escaped slavery and established life beyond direct owner control. Maroon societies ranged from small refuges to durable, armed political communities.

Mamluk

A person recruited through military enslavement in several Islamic polities. Purchase, training, manumission and later service differed by period; some men of mamluk origin became officers or rulers.

Abolition

The removal of a slave trade, slave status or related institution from lawful recognition. Trade abolition and emancipation are separate changes and may occur decades apart.

Emancipation

The act of freeing people from slave status, through self-liberation, revolution, war, legislation, court action or manumission. Legal freedom changes enforceable status but does not guarantee material equality.

Trafficking

Recruitment, transport, transfer, harbouring or receipt of people for exploitation through specified improper means, including force, deception or abuse of vulnerability. For children, those means need not be proved. Trafficking is a process that can lead to forced labour, servitude or slavery.

Go Deeper

Four books take the subject in different directions. None can cover the whole world evenly, which is part of the lesson.

The accessible overview

David Brion Davis, Inhuman Bondage: The Rise and Fall of Slavery in the New World (Oxford University Press, 2006). Davis gives a clear intellectual and political history of slavery and antislavery, moving from ancient precedents into the Atlantic world and the United States. His strength is explanation: he shows how religious ideas, law, racial thought, economic interest and resistance interacted without reducing abolition to one cause. The title's New World boundary matters, and the book carries the moral architecture of Davis's long career. Use this as the friendliest serious route into the modern system, then widen the map with Miller and the Cambridge global volumes.

The primary voice

Frederick Douglass, Narrative of the Life of Frederick Douglass, an American Slave, Written by Himself (1845; Penguin Classics expanded edition, 2014). Douglass explains slavery through control of literacy, movement, violence, family and self-command, then turns autobiography into abolitionist evidence. It is concise, deliberate and designed for readers who might suspect an escaped slave's testimony. Read the editorial material as well as the narrative. The book describes one remarkable life in the nineteenth-century United States, not a representative sample of global slavery. Its power lies in showing how an institution reaches into thought and how a controlled skill can become a route out.

The comparative theory

Orlando Patterson, Slavery and Social Death: A Comparative Study (Harvard University Press, 1982). Patterson compares slavery across sixty-six societies and offers a portable model that reshaped the field: violent domination, natal alienation and dishonour. The scale is its achievement and its danger. The model makes slavery more than labour exploitation, but later scholars have shown that social death fits some public, military and permeable statuses poorly. Read it to acquire a powerful hypothesis, then test where the metaphor breaks. Benedetta Rossi's later work on public slaves provides a concise challenge to keep beside it.

The methodological correction

Joseph C. Miller, The Problem of Slavery as History: A Global Approach (Yale University Press, 2012). Miller resists treating slavery as one object that appears unchanged in every civilisation. He follows slaving as a changing historical process shaped by vulnerability, political strategies and local institutions. The prose is denser than Davis or Douglass and the argument is aimed partly at historians. It is the best next step for understanding why comparison should search for mechanisms without flattening difference, and why the verb can sometimes explain more than the noun. Expect an argument about method rather than a chronological world survey.

Notes and Sources

Scope, definitions and comparative method

This book treats slavery as a comparative status and process rather than a timeless package of identical practices. The legal starting point is Article 1 of the 1926 Slavery Convention: slavery is a status or condition in which powers attaching to ownership are exercised. Jean Allain explains why the phrase covers both formally recognised status and comparable control in fact. The 1956 Supplementary Convention and the ILO Forced Labour Convention No. 29 support the distinctions among slavery, debt bondage, serfdom, servile marriage, child transfer and forced labour. Trafficking follows the process-based definition in the 2000 UN protocol.

Orlando Patterson's Slavery and Social Death supplies the concepts of natal alienation, dishonour and violent domination. Joseph C. Miller's The Problem of Slavery as History supplies the corrective emphasis on changing processes of slaving rather than one institution moving unchanged through time. Benedetta Rossi's article on public slavery supports the book's decision to retain a clear slave-free distinction while rejecting the claim that social death describes every enslaved person's lived position equally well. Noel Lenski and Catherine Cameron's What Is a Slave Society? supports the distinction between societies containing slaves and systems in which slavery organises major institutions.

No compatible evidence supports a single total for all people enslaved across world history. Trade flows, population stocks, legal categories, births, deaths, local transfers and periods of observation differ too sharply. The manuscript therefore uses no synthetic world figure.

The Whole Thing in One Page

The global range and variation follow the four volumes of The Cambridge World History of Slavery, supplemented by Patterson, Miller and the regional works listed below. The claim that enslaved people served in households, mines, agriculture, armies, courts and administrations is comparative, not a statement that every system contained every use.

“The trade that built empires” is treated as a bounded subtitle claim. Slavery supplied controlled labour, military service, administration, domestic service, sexual access, reproduction, status and wealth within many imperial systems. It did not cause every empire or operate apart from taxation, free labour, technology, ecology, disease, markets and military organisation. The capitalism paragraph follows David Brion Davis, Robin Blackburn, Jennifer L. Morgan and the Cambridge volumes while preserving disagreement about scale and causal weight.

Why You Should Care

The collar tag is British Museum object G 1975,0902.6, a copper-alloy tag riveted to a collar and inscribed with instructions to prevent escape and return the wearer to the estate of Viventius in Rome. The text paraphrases rather than reproduces the inscription. The object identifies an owner and place but supplies no recoverable biography of the wearer.

The current figures come from the ILO, Walk Free and International Organization for Migration report published in 2022 from estimates referring to 2021. It estimates 49.6 million people in situations grouped under the umbrella expression modern slavery: 27.6 million in forced labour and 22 million in forced marriage. These modelled components cannot be treated as one historical legal status or compared directly with ancient or Atlantic counts. The reference period, publication date and final verification date are kept separate.

The Core Ideas

Status before work. The 1926 Convention, Allain, Rossi and the Cambridge histories support the powers-based definition and the distinction between status and occupation. ILO Convention No. 29 supports the forced-labour definition. Rossi's examples of public slaves with legal or administrative capacity support the point that privilege and dependence could coexist. The book does not infer legal freedom from welfare or welfare from legal status.

Making an outsider. Patterson supports the discussion of natal alienation and dishonour. Hannah Barker shows Christian and Muslim participation in a connected Black Sea and Mediterranean slave trade drawing on frontier populations. Davis and the Cambridge histories support the changing roles of war, religion, political membership and race. Kyung Moon Hwang supports the Korean nobi example, including hereditary status and transfer without the racial division characteristic of the Atlantic United States. The text calls nobi status often translated as slavery because local categories and conditions did not map perfectly onto English. Distance or outsider status facilitated enslavement but did not cause it without wider recognition and enforcement.

Law and durability. Martha T. Roth's translation of the laws associated with Hammurabi supports the three-year service and fourth-year release in law 117. The text presents a rule in a law collection, not a measured description of everyday Babylonian practice. Keith Bradley and the ancient volume of the Cambridge history support Roman peculium, manumission, public slaves and the person-property tension. Jennifer L. Morgan supports the Atlantic importance of descent, gender, reproduction and inherited status. The language avoids saying enslaved people were legal non-persons in every system, since many legal orders recognised limited personality while subordinating it to an owner's claim.

Imperial capacity. The ancient and medieval Cambridge volumes support the range of public, domestic, productive, military and administrative slavery. Stephan Conermann and Koby Yosef support the Mamluk discussion, including the need to separate purchase, training, manumission, service terminology and later office. Peter Jackson supports military slavery and elite office in early Muslim India. Ehud Toledano and the completed Ottoman series manuscript support the bounded account of devshirme, kul dependence and palace slavery. These cases show that power over free subjects could coexist with coerced recruitment or slave origin; they do not make elite careers representative.

The capitalism discussion follows Davis, Blackburn, Morgan and the Atlantic chapters in the third and fourth Cambridge volumes. It distinguishes plantation returns, shipping, insurance, credit, customs, consumption, regional development and national industrialisation. No material claim states that slavery alone created capitalism or the Industrial Revolution. Nor does the book treat identifiable wealth and sectoral dependence as economically trivial.

Trade and reproduction. The second and third Cambridge volumes support the Mediterranean, Black Sea, trans-Saharan, Red Sea, Persian Gulf, Indian Ocean, Atlantic and domestic routes. Barker supports the shared Black Sea-Mediterranean commercial culture. Gwyn Campbell's collection supports the variety of Indian Ocean slavery and abolition. The completed The Slave Trade in a Hurry owns Atlantic voyage totals, destination shares, Middle Passage operations, abolition and inheritances in depth; this manuscript retains only the comparative transformation needed for a global model.

Walter Scheidel's demographic article supports the statement that natural reproduction may have contributed more to early imperial Roman slave supply than older conquest-centred accounts assumed. His result comes from models because ancient texts cannot establish the shares directly. The wording therefore says suggests and may, not proves. The contrast among Caribbean import dependence, United States natural increase and domestic sale follows the Atlantic scholarship listed in the bibliography.

Life and action under coercion. Frederick Douglass supplies the literacy and authorship example. Stephanie Camp supports attention to family, space, everyday resistance and enslaved women's strategies. Patterson, Lenski and Cameron, and the regional Cambridge chapters support bargaining, litigation, self-purchase, maroonage, military action and varied routes to freedom. Agency is defined as action under constraint and is never offered as evidence of consent.

Laurent Dubois supports the Haitian chronology and the destruction of the Saint-Domingue slave regime. The claim is narrower than the common slogan that Haiti was the only successful slave revolt. Haiti was the revolution in which enslaved insurgents and their allies defeated attempts to preserve or restore slavery and established an independent state in 1804.

Abolition and its limits. Seymour Drescher and Manisha Sinha support the interaction of enslaved resistance, Black testimony, religious organisation, campaigning, war and state power. The British dates and £20 million compensation commitment follow the Slavery Abolition Act 1833, parliamentary records and the Legacies of British Slavery research. Emancipation began in most covered British colonies in 1834; apprenticeship ended in 1838. The United States date follows the Thirteenth Amendment, passed by Congress in January 1865 and ratified on 6 December 1865. The punishment exception is stated without claiming that all later penal labour was legally identical to chattel slavery.

Cuba's 1886 abolition and Brazil's 1888 abolition follow the fourth Cambridge volume and regional scholarship. Rebecca J. Scott supports the distinction between legal emancipation and the contested construction of rights, labour and citizenship afterwards. The modern categories and estimates follow the UN and ILO instruments cited above.

The historical sequence

Early states. Roth and the ancient Cambridge volume support the distinction among war captives, household slaves, temple and palace dependants, debt service and other statuses in Mesopotamia. Surviving administrative records show categories used by institutions; they do not provide one continuous code or one experience across ancient Near Eastern societies.

Greece and Rome. The ancient Cambridge volume, Bradley, Patterson and Scheidel support the account. Aristotle's natural-slavery argument is paraphrased from the Politics and treated as ideology, not a description of enslaved people's capacities. The book makes no exact claim for the proportion of Roman inhabitants enslaved. Manumission was significant and varied by time, place, owner and type of work.

Religious and commercial frontiers. Barker and the medieval Cambridge volume support the Black Sea-Mediterranean network, mixed Christian and Muslim participation, and the importance of political and religious boundaries. Davis supports the statement that Jewish, Christian and Islamic traditions developed within worlds where slavery existed, regulated it and later supplied arguments used by defenders and abolitionists. Rules restricting the enslavement of co-religionists were never perfectly observed.

Military and court slavery. Conermann, Jackson, Toledano, Yosef and the medieval Cambridge volume support the Abbasid, Mamluk, north Indian and Ottoman examples. Yosef challenges the convenient rule that Mamluk recruits were automatically manumitted at the end of training, especially in the earlier sultanate. The text therefore treats purchase, training, manumission, service terminology and later office as distinct and variable. Devshirme was selective, changed over time and did not recruit every Christian population or operate throughout Ottoman history in one fixed form.

Africa and Asia. Paul Lovejoy and Miller support the diversity of African institutions, incorporation, hereditary status, regional trade and nineteenth-century transformations. The text neither imagines an Africa without slavery before Europeans nor treats Atlantic demand as a continuation without change. Hwang supports the nobi example. Titas Chakraborty and Matthias van Rossum support the Wange Hendrik Richard van Bali narrative and the caution that Indian Ocean movement was multidirectional and that no region-wide sex ratio should be treated as settled. The source presents the narrative as attributed to Wange, and the manuscript preserves that qualification. The broader Chinese wording remains limited to variation by dynasty, region and owner; no disputed population proportion is used.

Atlantic transformation. Davis, Blackburn, Morgan and the third Cambridge volume support the connection among Portuguese expansion, plantation production, oceanic trade, racial heredity and American law. Indigenous enslavement remains visible. The text avoids assigning Europeans, Africans or Americans one universal role at every stage. It also avoids a global claim based on one plantation, port or national economy.

Abolition and aftermath. Dubois, Drescher, Sinha and Scott support the sequence from revolution and campaigning to legal emancipation and post-emancipation conflict. The text separates abolition of trade, abolition of slave status, effective enforcement and material equality. Apprenticeship, indenture, debt, tenancy, pass controls and penal labour are named as distinct mechanisms rather than relabelled automatically as slavery.

How we know

The source problem is structural. Slaveholders, merchants and states produced most surviving contracts, inventories, laws, accounts and notices because ownership and enforcement required records. Enslaved people appear inside categories designed for other people's purposes. First-person testimony, petitions, court records, material objects, archaeology and oral traditions correct that angle but do not form a representative global sample.

The Atlantic evidence supports unusually detailed quantitative reconstruction. Other routes and earlier periods leave less compatible traces. For that reason, the book compares mechanisms and bounded examples rather than adding incomplete figures into a world total. Every anecdotal object or named life is documented; no invented scene, dialogue, weather, private thought or composite character is presented as fact.

What People Get Wrong

The race correction follows Patterson, Hwang, Barker, Davis and Morgan. It distinguishes premodern enslavement across political or religious boundaries from the systematic racial heredity constructed in Atlantic colonies. The plantation correction follows the occupational range in all four Cambridge volumes. The slavery-serfdom correction follows the 1956 Supplementary Convention, Rossi and the medieval volume, while acknowledging translation and continuum disputes.

The Africa correction follows Lovejoy, Miller and the Atlantic scholarship. African political agency is retained without equalising power or responsibility across captors, sellers, shipowners, colonial governments, buyers and enslaved people. The global-trade correction follows the four Cambridge volumes and Campbell. The agency correction follows Douglass, Camp and comparative work on manumission, petitions, flight and revolt. The abolition correction follows Scott, Drescher, Sinha and current legal instruments.

Use It and Terms

The six lenses are analytical deductions from the evidence rather than historical claims about one place. Powers, status, stock and flow, enforcement, replacement and exit, and agency versus consent are kept separate because each answers a different question. Present institutions require present evidence; resemblance to one historical feature is not enough for classification.

Terms draw from the UN and ILO instruments, Patterson, Allain, the Cambridge histories and regional scholarship. Historical words such as peculium and mamluk retain their setting. Broad English categories such as servitude, pawnship and serfdom are defined cautiously because legal meanings vary.

Data and source verification

Current legal instruments, institutional descriptions, modern global estimates and book metadata were rechecked on 4 September 2026. The newest important global estimate retained was published in 2022 and refers to conditions in 2021. No later completed authoritative global estimate with comparable forced-labour and forced-marriage components was found during the production search.

Bibliography

Primary sources, objects and legal instruments

Aristotle. Politics. Translated by C. D. C. Reeve. Indianapolis: Hackett, 1998.

British Museum. Copper-alloy tag from a collar, collection object G 1975,0902.6. London.

Douglass, Frederick. Narrative of the Life of Frederick Douglass, an American Slave, Written by Himself. 1845. Expanded ed. Edited with an introduction by Ira Dworkin. New York: Penguin Classics, 2014.

International Labour Organization. Forced Labour Convention, 1930 (No. 29). Geneva, 1930.

Roth, Martha T. Law Collections from Mesopotamia and Asia Minor. 2nd ed. Atlanta: Scholars Press, 1997.

United Kingdom. Slavery Abolition Act 1833. 3 and 4 Will. IV c. 73.

United Nations. Slavery Convention. Geneva, 1926.

United Nations. Supplementary Convention on the Abolition of Slavery, the Slave Trade, and Institutions and Practices Similar to Slavery. Geneva, 1956.

United Nations General Assembly. Protocol to Prevent, Suppress and Punish Trafficking in Persons, Especially Women and Children, Supplementing the United Nations Convention against Transnational Organized Crime. 2000.

United States. Constitution of the United States, Amendment XIII. Ratified 1865.

University College London. Legacies of British Slave-Ownership Database. Accessed 4 September 2026.

Modern works

Allain, Jean. Slavery in International Law: Of Human Exploitation and Trafficking. Leiden: Martinus Nijhoff, 2013.

Barker, Hannah. That Most Precious Merchandise: The Mediterranean Trade in Black Sea Slaves, 1260-1500. Philadelphia: University of Pennsylvania Press, 2019.

Blackburn, Robin. The Making of New World Slavery: From the Baroque to the Modern, 1492-1800. London: Verso, 1997.

Bradley, Keith. Slavery and Society at Rome. Cambridge: Cambridge University Press, 1994.

Bradley, Keith, and Paul Cartledge, eds. The Cambridge World History of Slavery. Volume 1: The Ancient Mediterranean World. Cambridge: Cambridge University Press, 2011.

Camp, Stephanie M. H. Closer to Freedom: Enslaved Women and Everyday Resistance in the Plantation South. Chapel Hill: University of North Carolina Press, 2004.

Campbell, Gwyn, ed. The Structure of Slavery in Indian Ocean Africa and Asia. London: Frank Cass, 2004.

Chakraborty, Titas, and Matthias van Rossum. “Slave Trade and Slavery in Asia: New Perspectives.” Journal of Social History 54, no. 1 (2020): 1-14.

Conermann, Stephan. “Slavery in the Mamluk Sultanate.” In The Cambridge World History of Slavery. Volume 2: AD 500-AD 1420, edited by Craig Perry, David Eltis, Stanley L. Engerman and David Richardson, 383-405. Cambridge: Cambridge University Press, 2021.

Davis, David Brion. Inhuman Bondage: The Rise and Fall of Slavery in the New World. New York: Oxford University Press, 2006.

Drescher, Seymour. Abolition: A History of Slavery and Antislavery. Cambridge: Cambridge University Press, 2009.

Dubois, Laurent. Avengers of the New World: The Story of the Haitian Revolution. Cambridge, MA: Harvard University Press, 2004.

Eltis, David, and Stanley L. Engerman, eds. The Cambridge World History of Slavery. Volume 3: AD 1420-AD 1804. Cambridge: Cambridge University Press, 2011.

Eltis, David, Stanley L. Engerman, Seymour Drescher, and David Richardson, eds. The Cambridge World History of Slavery. Volume 4: AD 1804-AD 2016. Cambridge: Cambridge University Press, 2017.

Hwang, Kyung Moon. Past Forward: Essays in Korean History. London: Anthem Press, 2019.

International Labour Organization, Walk Free, and International Organization for Migration. Global Estimates of Modern Slavery: Forced Labour and Forced Marriage. Geneva: International Labour Organization, 2022.

Jackson, Peter. “The Mamluk Institution in Early Muslim India.” Journal of the Royal Asiatic Society 122, no. 2 (1990): 340-358.

Lenski, Noel, and Catherine M. Cameron, eds. What Is a Slave Society? The Practice of Slavery in Global Perspective. Cambridge: Cambridge University Press, 2018.

Lovejoy, Paul E. Transformations in Slavery: A History of Slavery in Africa. 3rd ed. Cambridge: Cambridge University Press, 2012.

Miller, Joseph C. The Problem of Slavery as History: A Global Approach. New Haven: Yale University Press, 2012.

Morgan, Jennifer L. Reckoning with Slavery: Gender, Kinship, and Capitalism in the Early Modern Black Atlantic. Durham, NC: Duke University Press, 2021.

Patterson, Orlando. Slavery and Social Death: A Comparative Study. Cambridge, MA: Harvard University Press, 1982.

Perry, Craig, David Eltis, Stanley L. Engerman, and David Richardson, eds. The Cambridge World History of Slavery. Volume 2: AD 500-AD 1420. Cambridge: Cambridge University Press, 2021.

Rossi, Benedetta. “Reflections on Public Slavery and Social Death.” Bulletin of the Institute of Classical Studies 64, no. 2 (2021): 92-104.

Scheidel, Walter. “Quantifying the Sources of Slaves in the Early Roman Empire.” Journal of Roman Studies 87 (1997): 156-169.

Scott, Rebecca J. Degrees of Freedom: Louisiana and Cuba after Slavery. Cambridge, MA: Belknap Press of Harvard University Press, 2005.

Sinha, Manisha. The Slave's Cause: A History of Abolition. New Haven: Yale University Press, 2016.

Toledano, Ehud R. Slavery and Abolition in the Ottoman Middle East. Seattle: University of Washington Press, 1998.

Yosef, Koby. “The Term Mamluk and Slave Status during the Mamluk Sultanate.” Al-Qantara 34, no. 1 (2013): 7-34.