Chapter 5. Relief and Release: Debts Forgiven, Deferred and Lent in Hunger

5.1Older than banks

The oldest emergency acts in this study's eighty-four cases are acts on debt itself: relief and release taken by a ruler when a harvest fails or a political crisis leaves borrowers unable to pay. No bank holds a reserve, no central bank acts as lender of last resort, and no case was decided under such a rule. The authority is a state or a ruler; the record is a rule, an order or a proposal; the act is directed at borrowers rather than at institutions. On the question of who paid and what the act achieved, these records answer neither.

Of the eighty-four cases, thirty-eight record an act by an authority on money or credit; seven present an act with a stated term, and six name an observable condition on which it would end. None of the four records read here names who bore the cost while the relief stood.

A palace storeroom at the end of a season in which the crop has failed. A clerk's tablet lists names against what each household owes. The ruler reads the list while officials wait for the word that will tell them what grain may leave the granary and when it must come back. What is waited for is the order itself, and with it whether relief will be counted in grain, in a tablet washed in water, or in a year.

The first is Hammurabi's law, set within a reign range of 1792–1750 BC with no individual enactment date. "If any one owe a debt for a loan, and a storm prostrates the grain, or the harvest fail, or the grain does not grow for lack of water; in that year he need not give his creditor any grain, he washes his debt-tablet in water and pays no rent for this year" (The Code of Hammurabi, 1910, translated by L. W. King, law 48). The prologue of the same translation places such a provision inside a stated purpose: Hammurabi was called "to bring about the rule of righteousness in the land", "so that the strong should not harm the weak". The record is careful about what the text is:

The second is Solon's release at Athens in 594 BC: "Solon liberated the people once and for all, by prohibiting all loans on the security of the debtor's person: and in addition he made laws by which he cancelled all debts, public and private" (Aristotle, The Athenian Constitution, Part 6, translated by F. G. Kenyon, 1891). The record keeps three things apart:

The third is an order in the Goguryeo annals for the sixteenth year of King Gogukcheon, 194 CE, in a chronicle compiled in 1145. The classical Chinese of the entry is given here with this study's English: officials were ordered, every year from the third month of spring to the seventh month, to issue official grain, lending it in differing amounts according to the size of each household's people, and to take it back in the tenth month of winter, as a standing rule. The record gives the two dated moments in modern terms:

The fourth is the green sprouts proposal of 1069 in the Northern Song. Its own words cannot be quoted here, because the transcription is uncorrected; this study's English of the schedule is that the state lends money or grain before the harvest, to be repaid in grain by instalments along with the taxes, half at the summer instalment and half at the autumn one, in kind or in coin as the borrower prefers, with payment postponed in a disaster until the next instalment's harvest is abundant. The record keeps proposal and execution apart:

This study reads the four records as agreeing, in different words, that a rule is not an execution. Each names a condition — a failed harvest, a political crisis, a season, a disaster — and each waits on evidence the records do not hold. They hold only the authority's own words about credit and hunger.

5.2Hammurabi's conditional relief

Skeptron's record for the Hammurabi provision opens with what the text is and what it is not:

The record's own answer on the limits of judgment goes further, and its second question is worth its full answer because it is the boundary this chapter works inside:

The record names its own sources: the provision as given in a translation held by Fordham IHSP under "Hammurabi law provision §48 · Harper translation", and the reign chronology as given by the British Museum under the Middle Chronology. The source text used here is a different translation, that of L. W. King published in 1910, from a transcription on English Wikisource. The page gives no printed source for its transcription, and the wording is therefore unverified against the 1910 printing; it is born digital and not optical character recognition. Where this chapter quotes law 48, it quotes King's 1910 rendering, and it says so.

King's translation of law 48 reads: "If any one owe a debt for a loan, and a storm prostrates the grain, or the harvest fail, or the grain does not grow for lack of water; in that year he need not give his creditor any grain, he washes his debt-tablet in water and pays no rent for this year" (The Code of Hammurabi, 1910, translated by L. W. King, law 48).

Three things are visible in that sentence. The first is the condition: a storm, a failed harvest, or grain that does not grow for want of water. The second is the term: "in that year". The third is the remedy: the debtor gives no grain, the debt-tablet is washed away, and nothing is paid for that year. The record's own reading calls the same provision an exemption of interest; King's rendering calls what is not paid "rent". Both are given here, because the two texts differ in the word they use for the charge that is set aside, and this study does not merge them.

The prologue of the same translation sets the code's purpose in the ruler's own voice: Hammurabi was called by Anu and Bel "to bring about the rule of righteousness in the land", to destroy the wicked, "so that the strong should not harm the weak". A provision that stops a creditor from taking a debtor's grain in a failed harvest sits inside that stated purpose, and a reader of the prologue can see the provision as one instance of it.

Two neighbouring laws show how far the relief reaches and where it stops. Law 52 provides that where the cultivator has not planted, "the debtor's contract is not weakened": relief is triggered by a crop that fails, not by a debt that is inconvenient. Law 117, on a man who cannot meet a debt claim and sells himself, his wife, his son and his daughter into forced labour, sets the limit of that service at three years in the house of the man who bought them, or the proprietor, with freedom in the fourth year. Relief of debtors in these laws is bounded by a rule, and the boundary itself is written in years.

On the nature of such a text, the study's reading notes say this. Martha T. Roth's collection, Law Collections from Mesopotamia and Asia Minor (1995), gathers translations of the ancient Near Eastern law collections, including Hammurabi's, with introductions and notes, and is described as a reliable translation of the provisions on debt, loans and pledges and as a standard scholarly edition. Raymond Westbrook's edited A History of Ancient Near Eastern Law (2003) is the reference work for such collections. Neither note attributes to these collections any claim about how often a provision was used. What they establish is the scholarly setting in which a law text is read: as a legal collection, beside contracts and records, and not as a substitute for them.

That is the shape of this case. A rule that names its own condition and its own term, surviving in a legal collection, with no case decided under it in the record. The ledger finds the same shape: the act is there, its kind is debt relief, and both readings leave open whether the record presents the act as temporary and what condition would end it. The condition in the law — the failed harvest — is plain in the text; what the record does not settle is whether a reader may call an act that names a one-year term a temporary act at all.

5.3Solon's release

Skeptron's record for Solon keeps three things apart, and says so in its Scenario:

The record's second answer states the measures as the source states them, and its third refuses to read them as one. The sources the record names are an Athenian constitution held on MIT's Internet Classics Archive and a compilation of ancient references to Solon. The record's own dating answer adds that the notation of the source's composition belongs to a later narrative of the 4th century BCE, and refuses to merge the date of the act with the date of the text that tells of it.

The source texts used here are themselves later works. The Athenian Constitution, attributed to Aristotle or his school, was translated by F. G. Kenyon in 1891; its text is treated here as uncorrected. Plutarch's Life of Solon, translated by Bernadotte Perrin in 1914, was written centuries after Solon's time, as the study's note on it says, and draws on earlier writers whose details are debated.

Aristotle's account gives the measures as one linked set, and its first sentence is worth its full width: "Solon liberated the people once and for all, by prohibiting all loans on the security of the debtor's person: and in addition he made laws by which he cancelled all debts, public and private" (Aristotle, The Athenian Constitution, Part 6, translated by F. G. Kenyon, 1891).

Aristotle's account adds that the measure was called the Seisachtheia, which this study renders as a removal of burdens, and reports a charge against Solon that the text itself rejects: that he had told friends of the coming cancellation, that they borrowed and bought land, and that they profited when the debts were cancelled. Aristotle's judgment on the charge is that the popular party's version is the more probable and the accusation false.

Plutarch's chapters 13 to 15 give the situation before Solon in physical terms. All the common people were in debt to the rich: some tilled the rich men's land and paid a sixth of the increase, and some "pledged their persons for debts and could be seized by their creditors, some becoming slaves at home, and others being sold into foreign countries". Many were "forced to sell their own children (for there was no law against it), or go into exile, because of the cruelty of the money-lenders". In chapter 15 Plutarch gives the measures again, and then gives a second version. On his account "the first of his public measures was an enactment that existing debts should be remitted, and that in future no one should lend money on the person of a borrower". Androtion, as Plutarch reports him, "affirm that the poor were relieved not by a cancelling of debts, but by a reduction of the interest upon them", together with a change in the weight of the mina from seventy-three drachmas to a hundred, so that debtors paid the same nominal sum in money of smaller value. Plutarch's own conclusion is that "most writers agree that the disburdenment was a removal of all debt", and that Solon's own poems fit that version better. Plutarch adds that the friends who bought land before the cancellation were afterwards called debt-cutters, and that Solon remitted a debt he had himself lent, five talents on one account and fifteen on another.

Chapter 16 gives the settlement's reception: "the rich were vexed because he took away their securities for debt, and the poor still more, because he did not redistribute the land". Solon himself, in the verses preserved in the Athenian Constitution, claims the measures and refuses both the tyranny and a redistribution of land; one of those verses, in this study's rendering, has Solon say that he brought back to Athens many a man whom fraud or law had sold into slavery and driven from his land.

The record for this case follows the separation rather than the summary. The ban on loans secured by the person, the handling of debt, and the redistribution of land ownership are named as separate measures. A reader may take that separation as the study's instruction: a release of debts and a change of landholding are not the same act, and one does not prove the other.

One caution belongs here rather than elsewhere. The caution about reading a later political narrative as a ledger of contracts is carried instead by the record's own answer, and in 5.5 by what the notes on Westbrook's and Roth's works establish.

What the record is, then, is a later history. It holds an archonship date, a description of measures as later writers give them, and an explicit refusal to convert narrative into repayment ledgers. The ledger says the same thing in different words: the act is found, its kind is debt relief, the record does not present it as temporary, names no condition of its end, says nothing of what became of it, gives no figure and names no bearer. A cancellation has no term to end and no condition to observe. What it has instead is a ban that runs on until something else changes it, and that something is not written in these records.

5.4Grain lent in spring, taken back in autumn

Two of the four records describe state lending into a shortage, and they can be read side by side because they agree on a structure: grain or money goes out before the harvest, and it comes back after it.

The order of 194 is stated in Skeptron's record as follows:

The record's own limits of judgment add that the interest rate and the execution performance are not presented in its source text and remain unverified, and that the record is a historical source compiled in a later era. The source named is the Samguk sagi, compiled in 1145, on the site of the National Institute of Korean History. The source text used here is the classical Chinese of juan 16 of that chronicle, from a volunteer transcription on Chinese Wikisource, which is not checked against a printed page image and is treated as uncorrected.

The passage that carries the order is given here in the received classical Chinese, 仍命內外所司，博問鰥寡孤獨老病貧乏不能自存者，救恤之，命有司，每年自春三月至秋七月，出官穀，以百姓家口多小，賑貸有差，至冬十月還納，以為恒式，內外大悅, and this study's English of it is: He further ordered the offices within and without the palace to ask broadly after widowers, widows, the solitary, the childless, the old, the sick, the poor and those unable to keep themselves alive, and to relieve them; and he ordered the officials, every year from the third month of spring to the seventh month, to issue official grain, lending it in differing amounts according to the size of each household's people, and to take it back in the tenth month of winter, as a standing rule; and all within and without rejoiced.

The entry in which that order sits gives the frost of the seventh month that destroys the grain and the hunger that follows, and the tenth-month hunting in which the king meets a man sitting and crying without work or food for his mother, and gives him clothing and food before issuing the order; the chronicle is the source for that sequence, and the frost is background rather than a part of the act. The record itself marks two different dated moments: the grain was lent out from March to July and collected in October. The chronicle's phrase for the lending rule is 恒式, a standing pattern. The tenth month is not an end date for the act; it is the month in each year when the loan is collected.

The green sprouts proposal of 1069 is stated in Skeptron's record in its own terms:

The record's limits answer adds that the proposal's provisions are to be distinguished from actual implementation, and that effects can be verified only where there are user costs, evidence of whether coercion was involved, and a repayment ledger. Its source is Wang Anshi's proposal as held by Columbia University's Asia for Educators site. The source text used here is the Song shi of 1345, juan 176, from a volunteer transcription on Chinese Wikisource, treated as uncorrected.

The proposal's own words in that chronicle cannot be quoted here, because that transcription is uncorrected. This study's English of the repayment schedule and the deferral is: to be repaid in grain by instalments along with the taxes, half as the summer instalment and half as the autumn instalment; those who request the principal in kind, or who, when the price at the time of payment is high, wish to pay in coin, may do so as convenient; and if a disaster or crop damage occurs, payment is postponed until the day when the next instalment's harvest is abundant. The English is this study's own. Two features of that proposal are visible in the text itself. The choice between kind and coin is the borrower's; and the deferral is not open-ended, because it runs to a named event, the abundance of the next harvest. Hammurabi's law likewise writes down a term — that year — so the green sprouts proposal is not alone in writing down when its own relief would stop; what it adds is the named event at which payment resumes.

On these loans, the study's note on James T. C. Liu's Reform in Sung China: Wang An-shih (1021–1086) and His New Policies (1959) says that the book studies Wang Anshi's reforms under the Northern Song emperor Shenzong, including the green sprouts loans, in which the state lent grain or money to farmers at a set interest in the spring; that Liu examines the aims of the reforms, the debates with opposing officials and the reasons for the opposition; and that the book is an early English-language study that sets the policies in their political context.

Here the record and the source text part company, and both must be given. Skeptron's record says that the interest rate is not presented in the source text it checked. The Song shi excerpt used here does contain interest figures, but inside the objections rather than inside the proposal as the record states it: Su Zhe's objection reports that the proposal lets borrowers pay interest of two parts, while Han Qi's memorial of 1070 objects that for a thousand the borrower is ordered to repay thirteen hundred and describes the collection as taking a profit of three parts; the English of those phrases is this study's own, and the transcription from which it is taken is uncorrected, so none of them is quoted here. The record checked a different document, and the record's refusal to attach a rate stands for what that record holds. What this chapter can add is narrower: where a later chronicle reports both a proposal and the objections to it, the figures that belong to the objections are not the figures of the proposal.

Neither record gives a result. The order's chronicle says only that all within and without rejoiced; the proposal's chronicle goes on to the objections and ends the excerpt with Wang Anshi falling ill and staying away from court. The ledger says the same: for both cases what became of the act is recorded as not said, and the green sprouts record is the one of the four where both readings settled on a stated, observable condition of ending.

5.5The debate over clean slates

Two recent works read debt cancellation as a recurring institution of ancient statecraft rather than as an exception. This section asks what the four earliest records in this study can bear of that reading and what they cannot reach. What it adds to the debate is a separation: what an authority wrote in advance, what later writers said those words did, and what modern works argue about the whole. The debate asks how often cancellation happened; these records hold acts, not frequencies.

What the earliest texts say themselves comes first. In King's translation the prologue gives Hammurabi's stated purpose as bringing about the rule of righteousness in the land, "so that the strong should not harm the weak", and the provisions on debt sit inside that purpose. The failed-harvest relief is written in one year's terms: "If any one owe a debt for a loan, and a storm prostrates the grain, or the harvest fail, or the grain does not grow for lack of water; in that year he need not give his creditor any grain, he washes his debt-tablet in water and pays no rent for this year" (The Code of Hammurabi, 1910, translated by L. W. King, law 48). The same code draws its boundary in years: where the cultivator does not plant, "the debtor's contract is not weakened" (law 52), and debt bondage runs three years of service with freedom in the fourth (law 117). Relief in this code is a rule with a trigger and a limit, written in advance; nothing in it speaks of cancelling what is owed.

Solon's measure survives in later writers who disagree about it. The Athenian Constitution, in the 1891 translation, says "Solon liberated the people once and for all, by prohibiting all loans on the security of the debtor's person: and in addition he made laws by which he cancelled all debts, public and private" (Aristotle, The Athenian Constitution, Part 6, translated by F. G. Kenyon, 1891); it calls the measure the Seisachtheia, reports that friends of Solon borrowed and bought land before the cancellation and became wealthy, and rejects the charge against him as false. Plutarch, in the 1914 translation, reports that some writers, Androtion among them, held that the poor were relieved not by a cancelling of debts but by a reduction of interest, together with a change in the weight of the mina from seventy-three drachmas to a hundred, so that debtors paid the same nominal sum in money of smaller value. The ancient sources disagree not only about the size of the relief but about what kind of act it was.

The green sprouts proposal survives in the Song shi, whose clause on a failed crop, in this study's English of the classical Chinese, is that if a disaster or crop damage occurs, payment is postponed until the day when the next instalment's harvest is abundant. The same excerpt records the objections. Han Qi's memorial of 1070 says that the rule as settled lets households at the top borrow and orders a thousand to be repaid as thirteen hundred, which he calls the state itself lending at interest against the decree's promise; the English is this study's, from a volunteer transcription treated as uncorrected. The excerpt ends with Wang Anshi falling ill and staying away from court. A chronicle that records a proposal records its figures chiefly through the men who objected to them.

The note on Graeber's Debt: The First 5,000 Years (2011) records a wide-ranging interpretive essay tracing debt cancellation from ancient Mesopotamia to the biblical jubilee, and records that specialists have criticised some of its historical claims. It carries no passage quotable here, so nothing of the book appears beyond what its note states.

The notes on the legal texts give the ground on which a cancellation claim must stand. Roth's Law Collections from Mesopotamia and Asia Minor (1995) is called a standard scholarly edition with a reliable translation of the provisions on debt, loans and pledges. This study reads Roth's collection as establishing what such texts are, and this caution is this study's own: a law collection shows what a collection of laws says, and by itself it is not evidence of how many debts were cancelled, how many loans were made, or whether a provision was used in any particular year. That is why Skeptron's record of Hammurabi's provision insists on what the text is not:

The four records bear on the debate in the shape of what an authority wrote in advance. Two are conditional rules: Hammurabi's law names its condition — a storm, a failed harvest, or grain that does not grow for want of water — and its term, that year; the order of 194 names its condition as a season and its term as the tenth month of each year, standing, in this study's English of the chronicle's phrase, as a standing pattern. One is a cancellation, and its record separates its parts:

One is a proposal whose schedule the record refuses to merge with any result:

In each an authority speaks first about credit and hunger, and the record holds the speaking.

They cannot reach the further question the two modern works pose. Whether cancellations were a recurring institution, whether they were routine or episodic, whether they marked a society's stability or its strain: these are questions about how often acts happened and what followed, answered in works built from many sources, not in four records. None of the four holds a court record, a contract, a granary account or a ledger of what was repaid, and none can show how many people were relieved or at what cost.

The ledger's counts agree with that limit. Across this study's eighty-four cases, none names an observable condition on which its judgment could be revisited, and most cases with an act by an authority do not name in their own words a condition, date or event on which that act would end. Within these four records, one names such an observable condition — the green sprouts deferral — one names none, and the readings differ on the other two; none of the four names who bore the cost while the relief stood, which Chapter 6 takes up in the study's later cases.

One limit belongs to the record itself. Skeptron's cases are short readings of public sources made in 2026 and translated for this study; where an actor's own text survives, this section has put that text first. The works above are lenses, and where a lens fits badly this chapter says so rather than leaning on it. Chapter 13 sets these four records beside the two books that read cancellations as an institution; Chapter 5.6 asks which of the four acts named its own end.

5.6Relief with a term and relief without

The four records in this chapter answer in very different ways, and the difference is visible in what each act names.

Hammurabi's law names its condition and its term. The condition is a storm, a failed harvest, or grain that does not grow for lack of water. The term is "in that year", in King's 1910 translation. The relief is not open-ended: it is attached to a crop failure and it expires with the year. The order of 194 names its term in a different way: the loan runs from the third month of spring to the seventh month, is collected in the tenth month of winter, and is set out 恒式, as a standing pattern. Its end is therefore a month in each year rather than a date on which the act itself stops. The green sprouts proposal names a condition and a term in the clearest terms of the four: if a disaster or crop damage occurs, payment is postponed until the day when the next instalment's harvest is abundant. Solon's release names none. A cancellation of existing debts and a ban on loans secured by the person have no term; the ban runs until something else changes it, and nothing in these records says what that would be.

The ledger reads these four records in the same direction. All four show an act by an authority, and in all four the kind of the act is relief of debts or state lending into a shortage. On whether the acts are presented as temporary, the readings are left open for Hammurabi, for 194 and for 1069, and recorded as not said for Solon. On the condition of ending, the readings are open for Hammurabi and for 194, recorded as not named for Solon, and recorded as an observable condition for the green sprouts, where the deferral runs to the abundance of the next harvest. On what became of the act, all four are recorded as not saying. The order of 194 alone is marked as naming two dated moments, and its record states them: the grain was lent out from March to July and collected in October. On who bore the cost while the relief stood, all four are recorded as naming no one.

Two of those readings need a note of what they mean. The readings left open for Hammurabi and for 194 are not a claim that the acts had no term. In both texts a term is plainly present: a year in one, a month of each year in the other. Both readings were left open because the records present both acts as standing rules rather than as measures with an end to come, and a rule with an annual cycle is not the same thing as a temporary act with a term attached to it. That distinction is this chapter's reading of what the two readings did, not a finding from the ledger.

The one settled case is the green sprouts proposal, and it shows what an observable condition gives a later reader and what it does not. Because the record names the condition — disaster, and the abundance of the next harvest — a reader in principle could check whether the deferral was invoked, how often, and for how long. The record holds no such check. What it holds is a proposal and a refusal to merge proposal with execution.

A conditional relief, then, does what the series asks of an emergency act in one respect: it writes down in advance what would make the act stop. Hammurabi's law, the order of 194 and the green sprouts proposal each name a condition a later reader could, in principle, look for. What none of the four records holds is the evidence of that looking: no contracts, no granary accounts, no ledgers, and no named bearer of the cost while the relief was in force. Chapter 6 takes up that last question in the later cases of this study — Diocletian's edict on prices, the South Sea Company, Mexico's moratorium and the stoppage of Greek emergency liquidity — and asks where those records name a bearer and where they do not.