Restrictions and counts have a shape that a date does not show: an order is issued on a day, and the cost of living under it is paid afterwards by people who were not in the room. This chapter reads two of this study's records — the Ordinance of Labourers of 1349 and the Life Esidimeni announcement of 2017 — beside the histories that trace what such measures did to the people they reached. Its question is not whether any act was right. It asks what the records establish at the hour of the act, who the record names as bearing the cost, and what a later reader must supply because the record does not. The series states the claim it tests in this chapter as: "The cost of waiting falls not on whoever decides but on those nearest the point where things stop." The axes, the frame's recurring sentences and the two codings are set out in Chapter 2; this chapter uses them and reports their counts from "The ledger's numbers".
6.1Plague orders and the poor
Two histories stand at the beginning because they ask, in the historian's own terms, a question the records of this study rarely answer: who did a public measure actually reach?
A synthetic scene, marked as such, composed for this study from the two histories and the two records this chapter reads. It adds no date, number, person or outcome to them, and shows only the roles and the waiting those texts place in one place.
A parish street: a house shut against the street, a watchman at the threshold, a searcher called up to view the dead inside. In the parish office a clerk waits with the week's paper, the clerk's hand ready to set down the name and the cause reported for each death. What is waited for in the street is the searcher's report; what is waited for in the office is the line that will enter the weekly bill; what is waited for in the house is the end of the shut-up. Slack's work and the records this chapter reads say who was reached by such orders; neither says who stood in the street bearing the waiting, and the scene does not supply it.
Slack and Carmichael stand at different ends of one chain: Slack on what an order did and whom it reached, Carmichael on what the disease was said to belong to and what followed in who was shut up, watched or refused entry.
Both works are this study's lenses, not its evidence. What they give a reader of the ledger is a vocabulary for what is usually missing from a history record. The records state acts, targets and scope — the Ordinance of Labourers states who must serve and at what wage, and 6.2 reads the record's own sentences on that scope — and they rarely state cost as borne by named groups. This study reads both works as evidence of cost written down after the hour of acting rather than in it: they describe reception and framing, not what a public body wrote down at the moment of acting.
The ordinance puts the poor in the act itself. It addresses the Archbishop of Canterbury, and its provisions bind able-bodied men and women under sixty, without land or a fixed craft; the last of its clauses turns to the poor by name and forbids what piety might excuse, ordering that "let no one, under the aforesaid pain of imprisonment presume, under colour of piety or alms to give anything to such as can very well labour". Here the reach of the measure is written into the act; who among the poor bore it while the ordinance ran is not.
That asymmetry bears on this study's question about time. An order carries a date. Skeptron's record of the Ordinance fixes its time field at June 1349, England; the record of the Life Esidimeni announcement fixes its time field at 2017-02-01, a date 6.4 reads apart from the deaths it reports. The cost an order imposed is not dated in the order; it is dated, if at all, in the records of the courts, the parishes and the households that felt it. The ordinance's own opening clause gives no date for the pestilence it names, only that workmen and servants had now died in it. The time of the act and the time of the account of its cost can therefore be far apart, and the actors' record of the hour may contain neither the cost nor any anticipation of it. Slack's third finding makes the point from the other side: the justice of a measure was disputed at the time — by magistrates, clergy and physicians — and a history record that names only the act and its target preserves none of that dispute.
One further link runs through Slack to another part of this study. Slack's work records that the plague orders kept bills of mortality; Chapter 5 reads Graunt's use of those bills and the criticism he levelled at them. This book reads that, in the English case, a restriction and a count were made by the same apparatus; what neither work says is that the count reached the poor first. Where a record contains a count but no cost, this chapter asks who paid for the count as well as who paid for the order.
Who is named as bearing the cost is 6.5's cell; what a restriction's reason records is 6.6's question.
6.2The Ordinance of Labourers
The ordinance is addressed to the Archbishop of Canterbury and opens with a reason: a great part of the people, and especially the workmen and servants, had died in the pestilence; some of the survivors, seeing the straits of the masters and the scarcity of servants, were, in the ordinance's words, "are not willing to serve unless they receive excessive wages", while others preferred to beg in idleness rather than gain a living by labour. The ordinance says the king deliberated with the prelates, nobles and other learned men, and by their counsel ordained what follows.
Who must serve: every man and woman of the kingdom of England, of whatever condition, bond or free, who is able-bodied and below the age of sixty, not living from trade or a fixed craft, without the means of living of his own or land of his own to cultivate, and not already serving another — bound to serve whoever seeks him out for a suitable service. The ordinance adds that lords are preferred to others when retaining their own bondsmen or land tenants, but that a lord retains as many as are necessary and no more.
At what wages: only the wages, liveries, meed or salary customary in the place where the person sought to serve, in the twentieth year of the king's reign, or the five or six common years next preceding. The ordinance then caps artisan labour at the same customary rates — it names saddlers, skinners, tawers, cordwainers, tailors, smiths, carpenters, masons, tilers, shipwrights and carters — and requires vendors of victuals to sell at a moderate price, judged against prices in adjoining places and the distance goods are carried.
With what penalties: a person sought after for service who refuses is taken on the proof of two faithful men before sheriffs, bailiffs or constables and sent to the next jail in strict custody until he finds surety for serving in the prescribed form. A retained worker who leaves before the end of the agreed term without permission or reasonable cause is imprisoned, and no one may receive such a person under the same penalty. Anyone paying or receiving more than the customary wages forfeits double the sum to the aggrieved party, or to whoever prosecutes, in the court of the lord of the place. Lords who act contrary to the ordinance face the triple penalty. Covenants already made for higher wages are not honoured. Victuallers who sell otherwise pay double, and mayors and bailiffs who fail to enforce the ordinance face the treble of what was sold and grievous punishment on the king's part. Finally, no one may give alms to people who can very well labour, under the penalty of imprisonment, so that they may be compelled to labour for the necessities of life.
Skeptron's record of this act carries its own caution. Its Scenario states, in one sentence, who was reached and on what basis:
Exceptions were placed for one's own livelihood, land, occupation, and so on, possible labour was required from those under 60, and wages were restricted on the basis of customary existing practice.
The record's answer to a later reader's question about whether this law may be classified as a modern minimum wage system sets the boundary: it is a historical coercive provision restricting the refusal of employment and high wages, and it must not be reinterpreted as having the function of a guarantee of minimum living. The record's frame adds the studio's recurring sentence that announcement, decision, implementation and performance are not merged into a single event; Chapter 2 explains what the frame is and why its sentences are not findings. What the frame supplies here is a discipline: the ordinance of June 1349 is one act, and the decades of enforcement that followed are another.
Two readings bear on who bore the cost. Bertha Haven Putnam's The Enforcement of the Statutes of Labourers during the First Decade after the Black Death, 1349–1359 (1908), from the records of the justices, shows that the 1349 ordinance and the Statute of Labourers of 1351 were enforced vigorously and widely in the first decade, with many prosecutions of labourers for taking excess wages or refusing service, and that enforcement grew into the work of the justices of the peace. Samuel Cohn's After the Black Death: labour legislation and attitudes towards labour in late-medieval western Europe (2007) compares the labour laws passed across western Europe and argues that they varied from place to place, often contradicted one another, and expressed elites' fear and contempt of labourers as much as any economic calculation. Neither reading is the record; both are this study's lenses, drawn from the works' reading notes.
This study reads the ordinance as an act that fell on the survivors. The dead appear once, in the ordinance's first sentence, as the reason for the act; they are not its addressees. Its addressees are the living: able-bodied men and women under sixty, without land or craft, and the beggars who decline labour. The cost of the act is therefore not a cost of dying but a cost of continuing — compelled service, capped wages, doubled penalties, and the loss of alms. On the series' reading of this record's cells, the coders agreed that who decided was an institution or state by implication, that who waited and paid was those nearest the stoppage by implication, that whether the act could be undone was recorded as unknown, and that the human's place was left to an implied rule rather than to a named person. They divided on the horizon, which counts as unclear in the verdicts; agreement on the human-place cell across the ledger is one of its weaker ones, and any comparison that rests on it is one on which the coders divided. The record itself adds what the coders could not: the ordinance's scope, its exceptions, its age limit and its penalties, verified only to the extent the record states.
6.3Foucault's plague town
Foucault contrasts this with the exclusion of lepers, and treats the plague model of segmentation and constant surveillance as an image of disciplinary power.
What the note describes is an arrangement rather than an act. The passage comes from a work about disciplinary power, and the town under plague supplies Foucault with one of its images. Read for this study, it supplies no date, no named decider, no horizon, no group bearing a cost and no count. The daily visits, the hierarchical reporting and the shut-in households are the content of the image; who ordered them, when, and at whose expense are questions the passage does not put and does not answer. This study therefore uses the passage only as a lens, and never as evidence about a case.
The image fits some of this study's acts well. Slack's plague orders shut up households, appointed searchers and watchmen, and kept bills of mortality; what a searcher did at a house is treated in Chapter 5, and this chapter keeps to Slack's note. In Foucault's model a restriction and a register are not two measures but one machinery: to shut people in is also to make them countable. The acts of this study that carry both a shut-up order and a count include the plague measures read in Chapter 3 and the English orders whose bills Chapter 5 reads through Graunt. Where a record holds an order and a tally in separate fields with separate dates, the model lets a reader see them as one gesture. This study reads that fit as a reading aid: it makes visible a relation between an order and a count that these records do not themselves state.
The contrast with lepers sorts the acts a second way. Segmentation and registration stand on one side; the casting-out of lepers on the other. This study's records hold both shapes at once — refusing entry to the sick in Florence, and refusing alms to able-bodied beggars in the Ordinance of Labourers, stand near one side; shutting up a household and registering its dead stand near the other. The model does not tell a reader which shape any act belongs to, because the records use neither term, and this study sorts by what the records state.
It fails in three ways that this study's records make plain. First, Foucault's passage is an image of disciplinary power, not a record of any act in this study; no ledger cell rests on it, and it adds no date, no decider and no count. The ledger's decider cell across the eighty-nine cases was agreed in fifty-three, with thirty-six unsettled, and the ledger's summary lists cases with a decider copied into the record's own field as 0 of 89 — so a model that assumes a town knowing everything about every household cannot be checked against these records on that point.
Second, the acts read in these chapters are not all surveillance. The Ordinance of Labourers does not inspect households; it caps wages and compels service. Its machinery is proof by two faithful men before sheriffs, bailiffs or constables, and a refuser sent to the next jail in strict custody, with the double penalty for anyone paying or receiving more than the customary wage — not the daily visit. Its date is June 1349; the prosecutions that make the compulsion visible come later, in the enforcement records that cover 1349 to 1359. The image's continuous inspection has no counterpart in the record, which holds one dated act and leaves its enforcement to other documents.
Third, the records of this study are histories of events, not decision records of the kind Foucault's model assumes. In the Life Esidimeni record the act is an announcement of a provisional count, taken after an inquiry whose findings it publishes; the record's own decider cell is unsettled as well. A model of a town that knows everything about every household and a record of an announcement about a count are different kinds of evidence about different things, and this study keeps them apart.
What the image retains is a vocabulary for what a restriction and a register do to the people subject to them — the subject the ledger codes as the bearer and usually cannot settle. It answers a question the records never pose, about what surveillance is, rather than the question they pose, about what was established at the hour of the act. Chapter 13 returns to Foucault beside the study's own findings on who waits.
6.4Life Esidimeni
The record read here is dated 2017-02-01, and the record keeps that date apart from the deaths it reports. Its Scenario states what the announcement established:
The Ombudsman announced that 94 people died after patients were transferred to 27 NGOs, and stated that this figure was provisional because of unverified deaths and further reports from families. On the same day, the resignation of the provincial health head was also announced.
The record's answer to a later question — whether 94 can be fixed as a final figure that does not change afterward, and whether an administrative resignation can be handled as a personal criminal guilty verdict — separates what the announcement did and did not establish:
The damage confirmed at the time, the provisional status, further reports, administrative responsibility, and the subsequent judicial judgment are separated. Personal medical information of patients is not collected in this record.
Five things are held apart in that answer, and the separation is the record's own work. The damage confirmed at the time. The provisional status of the figure. The further reports that families were still making. The administrative responsibility that the resignation bears on. The subsequent judicial judgment, which the record does not describe and does not fold into the announcement. The record also states, in its answer on chronology, that the adopted date is the publication date of the ombud's inquiry findings and not the date of each individual death; the chronology of the event and the generation and publication of the source material are not merged into the same field.
What the record dates, in order, is a short list: the announcement of the figure on 2017-02-01, the provisional status given to that figure on the same day, and the resignation of the provincial health head, announced on the same day. What it leaves undated is much longer: the transfers of patients to the 27 NGOs, the deaths themselves, the beginning and the coverage of the inquiry whose findings were being published, and any figure that followed the provisional one. The record supplies no interval between any two of those moments. A later reader cannot build one from this record, and this study does not build one.
What the act was is an announcement of a count, taken after the deaths it reports and after an inquiry whose findings it publishes. What was known when it was taken is narrower than that. What the record shows of the hour itself is one thing: the figure was announced together with a statement of its own incompleteness. What it does not show is anything about what stood behind the 94 before the day — whether the deaths had been checked, how many were unverified, how many reports from families were outstanding. The record says these things exist; it does not say how many, and no quantity of them can be recovered from its sentences.
No secondary work on Life Esidimeni could be verified for this study, and the ombud's own report is not among the public-domain source texts available here. What is said of the patients is therefore only what the record says: that patients were transferred to 27 NGOs, and that 94 people died after those transfers. The record states that personal medical information of patients is not collected in it, and this chapter adds nothing to that. The patients are spoken of here only as the record speaks of them, in the numbers it gives, and no further.
The time of the act is 2017-02-01. The time of the deaths is not given. The time of the count is the day the ombud announced it, and the count is provisional on that day — the record names its own incompleteness, naming unverified deaths and further reports from families as the reason. Responsibility is separated into administrative and judicial parts, and the record does not merge them. What a later reader can hold apart, on the strength of this record alone, is the hour of the announcement from the hour of the inquiry findings it publishes; the record's chronology answer says the adopted date is the publication date and not the date of each death. What a later reader cannot hold apart from anything else in the record is the hour of the deaths, because the record gives it no date.
What the record leaves to the reader is the wait. Between the transfers and the announcement there are no dates the record supplies; between the announcement and any later figure there are none either. On the series' own cell for who waited and paid, the codings did not settle a value on this record, and "The ledger's numbers" reports the cell as unsettled, which counts as unclear in the verdicts. This study's own field on the same question reads differently and is reported beside it: on the bearer-named cell, both codings found the record's text naming who bore the harm — the patients transferred to the 27 NGOs, among whom 94 died after the transfers. One reading says the record does not fix who bore the wait; the other says the record names the class who bore the harm. Both are readings of the same sentences, and this study keeps them apart rather than merging them.
On this study's own count cell, the codings read the count as provisional-or-revised, stated in the record's own words. Across the ledger, thirteen cases have a count the codings settled; eight of those thirteen qualify the count, including this provisional-or-revised value, and five give it as settled. Among the twenty-three cases this book reads closely, it appears in this one case and nowhere else on that cell; five more read qualified-by-definition-scope-or-denominator, two read settled, ten read no count, and five are unsettled. On the revision cell, the codings found named, stated: the further reports from families are something the record itself says would change how the figure stands, and a later reader can watch for them without leaving the record. On the gap-named cell, the codings read this case as not naming a gap between the event and its record, even though the chronology answer does state that the adopted date is the publication date and not the date of each death. This study reports both: the record's sentence about the two dates, and the coders' value on its cell.
One caution belongs here. The record's closing answer, which sets conditions for comparing its figures elsewhere, repeats the studio's recurring sentences; Chapter 2 explains how those sentences were found and why they are method, not findings. A frame sentence that appears in this record tells a reader what Skeptron's form requires, and carries no information about Life Esidimeni itself. What carries information is what the record states about the announcement, the provisional figure, the resignation and the separation of responsibilities.
This study reads the case as one in which the act and the harm are far apart in time and close in the record. The announcement is dated to the hour of the ombud's findings; the harm is undated; the count is provisional on the day it is made; and the responsibility is split between an administrative resignation announced the same day and a judicial judgment the record does not report. What the record fixes is the hour at which a number was spoken aloud, and the qualifications that were spoken with it. What it does not fix is the hours the number was made of. A later reader who wants the time of the deaths, or the criteria by which 94 was reached, must go to documents this study does not have.
6.5Who bore the wait in the ledger
The claim tested in this chapter is the series' own, stated by what it says: the cost of waiting falls not on whoever decides but on those nearest the point where things stop. The bearer cell of the ledger bears on it, read by the two codings from what the record says.
Across the eighty-nine cases, the coders agreed on the bearer cell in fifty-two cases, or 58% — far below the 75% to 88% agreement of the act, act-kind, gap, count and revision fields. Their settled values stand as follows, each with what it counts:
| Read value, both codings agreed | Cases |
|---|---|
| Unknown | 38 |
| Those nearest the stoppage, implied from the setting | 8 |
| Third parties or the vulnerable, implied from the setting | 5 |
| Those nearest the stoppage, stated in the record's words | 1 |
Thirty-seven cases are unsettled — the codings differed — and those count as unclear in the verdicts. The shape of the table matters as much as its totals. Only one case among the eighty-nine has the bearer named in the record's own words on this cell; thirteen have the bearer implied from the setting. That is what "implied" means here: the record does not name who waited, and a coder infers it from who the act reaches — workers compelled to serve, patients transferred, households shut up. An inference is not a finding, and this study reports the split rather than merging it.
Among the twenty-three cases read closely, the two codings agreed the bearer is named in five cases, not named in thirteen, and unsettled in five. Of the twenty-two cases in which both codings found a public act, thirteen are read as not named on this field and four as unsettled. The record read in full above, Life Esidimeni, is one of the five named: its text names patients transferred to 27 NGOs and 94 people who died after the transfers. The Ordinance of Labourers is among the thirteen whose bearer the codings read as not named — the record's own sentences state who must serve, but the coders did not find the record naming who bore the cost while an act waited, which is what E5 asks.
E5's strict reading counts only what both coders found stated, and it is unclear — 17 cases support, 6 go against, 49 are unclear, 17 are not applicable, a share of 74%. Its lenient reading also takes what the setting implies, and it supports the hypothesis — 21 for, 9 against, 20 unclear, 39 not applicable, a share of 70%. Read by group, the seventeen medicine-history cases support E5 on both readings: strict, 6 for, 3 against, 8 unclear, a share of 67%; lenient, 9 for, 4 against, 4 unclear, a share of 69%. The discrepancy between the strict and lenient readings is the whole of the finding here: the records frequently imply who waited, and seldom say it.
Two further counts frame the answer. The ledger's summary lists cases with a decider copied into the record's own field as 0 of 89; the decider cell, by contrast, is a coder reading, agreed in 53 of 89 cases, with 36 unsettled. So the series' claim that the cost falls on someone other than the decider rests, on this ledger, on cells where the coders often divided, and on records that are histories rather than decision records: they describe what a public body did, and rarely who stood in the queue, in the ward, or at the gate while it did so. The one record in this chapter that names its bearer does so through a public announcement about deaths after transfers, not through the actors' own account of the wait.
6.6The debate over the justice of restriction
A question sits behind every measure in this study that the records themselves never put in words: was the restriction just to the people it reached? Four works bear on that question, each by its own argument as its reading note gives it. None is evidence about a case in this ledger; each is a lens, and this section keeps the order the chapter has set: Slack on what the dispute looked like while it was being administered, Baldwin on what a state's position requires of it, Foucault on what restriction does to those subject to it, and Richards on what the people a measure reached understood themselves. The records are read after those four, on what their own sentences carry.
Slack's note reports a dispute among magistrates, clergy and physicians over whether isolation was just; it existed among the people who administered the measures, and what survives of it is what a historian found in the documents they left. Slack's reading bears on the question of burden this section keeps, not on any outcome in the records of this study.
Baldwin's contribution on the question of burden is that a measure follows from position and capacity, which relocates the justice question from doctrine to cost: what a state is obliged to do by where it stands, and who bears the result while it does so. Chapter 3 sets out his debate with Ackerknecht over doctrine.
Foucault's plague town, read in 6.3, answers with an image: the shut-up household inspected daily and the death registered and passed upward are the visible form of what restriction does to people who are subject to it. The note calls it an image of disciplinary power, and this study treats it as a lens for what the records show — that a restriction and a count were often one machinery — without letting the image stand in for any act's justification. Its limits for this study are set out in 6.3: it supplies no date, no decider and no group bearing a cost, and the records of this ledger are histories of events rather than the town registers that Foucault's image assumes.
Paul Richards's Ebola: How a People's Science Helped End an Epidemic (2016), from fieldwork in Sierra Leone, argues the other side of the burden. Local communities learned quickly how the disease spread and changed their own practices, notably burial and the care of the sick, and this people's science did much to end the epidemic, while the international response arrived late and at first misread communities as obstacles. Chapter 10 reads Richards beside the Ebola counts; here the point is narrower. Richards's argument assigns competence and cost to the people a measure is imposed on, which is the same claim the ledger's bearer cell tries to code and usually cannot.
Where do this study's records bear on the debate? Two of them are unusually explicit, and both are read in this chapter. The Ordinance of Labourers states its own reason in its first sentence — the mortality among workmen and servants, and the survivors' refusal to serve at customary wages — and adds a second ground in the same voice: "considering the grave inconveniences which might come from the lack especially of ploughmen and such labourers". The reason is written from the side of the masters and of the ploughmen's absence; the workmen and servants appear in it only as the population among whom the death had been. The ordinance names the people it binds, by condition, age and occupation, and the record's own Scenario, quoted in 6.2, gives the exceptions and the age limit that scope carries. What the act's recital does not give is any word from someone who disputed it.
Life Esidimeni's record states that the ombud's figure was provisional and that administrative responsibility is separated from the subsequent judicial judgment; it names no argument, no dissent and no defence, and it names the patients only as people who were transferred and among whom 94 died. Its Scenario, quoted in 6.4, gives the announcement; what it does not give is any account of what was weighed before it.
Across the eighty-nine cases the ledger reports on the cell that asks whether a case's own text records a reason for the act. The coders agreed in 87 of 89 cases: 80 read the reason as stated in the record's own words, 4 as absent by implication, 2 as stated by implication, 1 not applicable by implication, and 2 unsettled. A stated reason in a history record is usually the act's own recital, as in the ordinance; the cell does not ask whether anyone accepted it. The coders also agreed in 62 of 89 cases that no human role is specified in the record — 58 read as absence by implication, 4 as absence stated in the record's words, 27 unsettled — so the person who weighed the burden is not the person these records name, and the dispute these four works describe has to be reconstructed from historians and practitioners rather than read off the acts. This section sets out only what the cell counts; whether a restriction's reason survives in its record, as one of the series' hypotheses asks, is for Chapter 12 to judge.
What the records cannot do is settle the justice question, and the reason is structural rather than moral. Slack's division among magistrates, clergy and physicians, Richards's fieldwork on what communities understood, Baldwin's account of what a state's trade required — none of these appear in a history record whose own sentences state acts, targets and scope. A record can carry a stated reason; it rarely carries the people who disputed that reason, the ones who complied under protest, or the ones the measure did not reach. Where the coders could read a bearer, they usually had to infer it from who an act addresses. The two cases read here show the range: one record whose text names the living people its wage cap and service duty bind, and one whose text names the transferred patients among whom 94 died, while leaving the wait between the transfers and the announcement undated.
This study therefore reports the debate rather than resolving it. Slack, Baldwin, Foucault and Richards each answer a question the records pose without answering; the records answer a different question, which is what was established at the hour of the act and what a later reader can still hold apart. The comparison of the records — restrictions side by side, on the same cells — begins in Chapter 7.