The Rent at the Gate

It is a kind of intelligence office for the whole country.

— Henry David Thoreau, letter to Sophia Thoreau, 22 May 1843¹

The notary charged a fee.

There was work to be paid for. An agreement had to be put into a form that others knew how to read; a record had to remain available after the parties went away. In the early modern Roman offices studied by Laurie Nussdorfer, bound and indexed protocols, permanent premises and partial duplicate deposits gave that expectation material substance. A person returning for a copy was returning to an arrangement maintained over time. The original transaction might have been brief. Its documentary life was potentially much longer.²

It is easy, from the receiving end of a charge, to overlook the work whose success has made it uninteresting. The requested instrument is found. Its date and parties can be distinguished from those of another transaction. Someone knows which copy is wanted and what standing that copy has. These are modest achievements until they fail. The notary's authority did not rest on handwriting alone, and the recognition accorded his instruments was not something a cheaper scribe could confer merely by reproducing their appearance.

Yet the custody that made a later visit useful also made it an occasion to charge again. Nussdorfer describes Roman reforms that opened clients' inspection of records and permitted simple copies alongside more formal public copies, dismantling profitable barriers to access. The question was no longer only how to preserve a transaction. It was on what terms the people concerned could return to it. Protection of an archive and protection of its keeper's income had come to occupy some of the same procedures.

The keeper was not necessarily a rich man. An office could be bought or burdened with debt; the investment made in acquiring it was another claim upon its future receipts. A client might therefore encounter, in one modest charge, both the expense of maintaining records and the consequences of a market in the authority to maintain them. This does not tell us how much of that particular charge was excessive. It does make the explanation that the fee simply paid for writing rather inadequate. Someone had paid for the position from which the writing could be sold.

Nor should we make the service larger than it was. In Cuzco on 16 April 1663, María Panti acknowledged receiving coins in a cloth bag from Francisco de Allier before the notary Martín López de Paredes. The coins were not counted in the notary's presence. Panti renounced a claim based on an error in the count. Kathryn Burns found this small accommodation in a record whose official form might otherwise encourage us to imagine a more complete verification.³

The bag had been received and the receipt made, but the amount inside remained outside the notary's observation. It is a detail worth lingering over. The writing did not eliminate this interval between what had been acknowledged and what had been checked; it gave the interval a legal treatment. Panti's renunciation meant that a later discovery of an error was intended to meet words already put on record. We know neither why the coins went uncounted nor whether the number was right. The receipt tells us what the parties arranged to do about that uncertainty.

Here public faith had a legal setting. An instrument's recognized evidentiary force did not make all its assertions true, still less ensure recovery from anyone named in it. Parties could have valid obligations supported by other kinds of evidence. What they purchased from a commissioned office was a particular form of recognition. Its value depended partly on what another person or court would subsequently make of it, and the office could preserve an acknowledgment that later mattered without undertaking to settle every question about the transaction.

I will call the work needed to make a claim usable the coherence fee, and a premium enabled by control of an accepted route to that use the trust tax. The names distinguish questions; they do not divide a surviving tariff into two conveniently labeled columns. Costs can fall, offices can compete, and legal restrictions can be justified or abused. An observed fee is not an estimate of rent. But the return to the archive has already introduced a question that an account of useful work cannot answer for itself: who is entitled to determine the conditions on which others may obtain its benefit?

A great many clerks

Thoreau mentioned Lewis Tappan's new business in a letter otherwise occupied with less systematic forms of intelligence. He had been ill; the garlic in the fields was spoiling the butter; he promised to put his microscope to good use. Then he asked whether his father had heard about the credit agency. Giles Waldo, a clerk there whom he had recently met, thought it might make ninety-nine of every hundred New York stores dispensable. Thoreau also expected the business to employ a great many clerks. Even in this early glimpse, the prospect of dispensing with intermediaries required a considerable office.¹

Tappan founded the Mercantile Agency in New York in 1841. The successful enterprise grew from a familiar commercial difficulty: a seller could know his goods rather well and know almost nothing about the person asking for time to pay. Tappan had been a silk merchant. He turned the work of asking around into a business of its own, drawing on local correspondents whose information could be assembled for subscribers elsewhere. Many of the early reporters were lawyers; referrals for debt collection helped make the unpaid reporting worth their trouble.⁴

The reporter knew a town. The subscriber needed to make a decision about someone from it. Between them stood an institution that could collect many such local accounts, keep them and answer inquiries. Credit no longer had to wait upon a seller's own circle of acquaintance. An unknown buyer could benefit as much as the subscriber if the alternative was a refusal to deal with anyone the seller did not already know. A business selling information about strangers could open relations that personal trust, left to itself, would keep closed.

The early reporting room offered an unusual kind of reading. A clerk delivered the information orally. Subscribers could make notes, though they did not receive copies of the underlying reports. The words could travel again in those notes and in the listener's recollection; the retained report and the account carried away were already different objects. Printed rating books later gave subscribers another way to consult the agency's judgments. Dun's reference book began publication in 1859.⁵

The judgments could be blunt without being of one mind about their subject. On 20 March 1850 a correspondent assessed Charles Dull, a Philadelphia paper dealer, whose personal reputation was poor but who was thought to have money and owned property nearby. The report ended: “if he gives his note he will no doubt pay it.”⁶

Dislike had not been allowed to decide the whole question. The correspondent could pass on an unfavorable social judgment and nevertheless recommend confidence in a promise to pay. We do not know from these lines whether the expectation was fulfilled. We do know that the account retained a distinction a borrower might have been grateful to have preserved: the pleasure of his company was not the security of his note. Commercial calculation can be morally obtuse; it can also decline to make moral approval a condition of every useful relation.

The report is troubling for reasons that its possible accuracy would not remove. Who was entitled to circulate the adverse judgment about the man? What did it add to the estimate of payment? It was information supplied for another person's decision. The relation between agency and subscriber gave the subscriber a reason to ask for it, while supplying the subject with no corresponding place in that exchange. Dull was present as a description. Whether he could discover and answer that description was another matter.

Secrecy protected a saleable resource and the people supplying it. It also made it harder for a subject to discover which report had done harm. The surviving guides and the history of the industry show a business organized around subscribers' inquiries, without a corresponding right of subjects to inspect their own dossiers. Much more survives of the institution's ways of describing people than of their experience of being described. Lauer cautions against taking that archive as evidence that its subjects came to regard themselves as the agency regarded them. The limits of their access are easier to establish than the extent of its command over their minds.⁷

Some did more than complain to the office. John and Horace Beardsley, merchants in Norwalk, Ohio, sued Tappan for libel. The agency had communicated to a customer that John Beardsley's wife was about to seek divorce and alimony, that he had put property out of his hands and that the store would probably close if the suit was brought. Marriage, property and the prospects of the firm had entered the same commercial account. In the litigation that followed, they became harder to separate.⁸

A divorce suit was indeed commenced, months after the alleged libel and after the Beardsleys had begun their action against Tappan. John Beardsley's answer accused Tappan and others of arranging it to substantiate the agency's earlier report. The record contained depositions that appeared to support the accusation; the divorce petition was eventually dismissed. This material was read to the jury in the libel trial, which awarded the merchants $10,000. The Supreme Court's account is vivid enough to tempt a historian into treating the alleged plot as the solution to the whole affair.

That is precisely what the Court would not permit. Tappan had not been a party to the divorce proceedings. He had had no opportunity there to cross-examine the witnesses or answer Beardsley's charges. In 1870 the Court reversed the judgment and ordered a new trial because the record had been admitted in a way that violated those protections. It did not resolve every allegation about the agency by granting that protection. Nor did it pronounce the agency's communications immune from suit.

There is an awkward symmetry here, but the people need not be made symmetrical to see it. The merchants objected to an adverse account circulating beyond their control. Tappan objected to an adverse record being used where he had not been able to challenge its production. An institution criticized for secrecy had a legitimate claim against a particular use of someone else's evidence. To deny him that claim because his own business deserved scrutiny would have made answerability a favor bestowed upon the party we preferred.

Mrs. Beardsley is less readily recovered. In the commercial report, the prospect of her action threatens the firm's assets. In her husband's answer, the proceeding in her name becomes an instrument of someone else's alleged conspiracy. Those accounts do not entitle us to supply her wishes. The difficulty should remain. A dispute over who had misrepresented whom could still leave a person central to both representations strangely hard to hear.

This is not an origin story in which helpless subjects waited for the twentieth century to acquire any recourse. The court existed and imposed demands of its own. Nor is a lawsuit the equivalent of routine access to a report before someone acts upon it. Litigation began after the disputed communication, involved another body of records and produced another occasion for contest. The work of making an account answerable had not been completed when the first clerk wrote it down.

More than a century separates that dispute from a bureau-based credit score. FICO introduced its general-purpose score in 1989; in 1995, Fannie Mae and Freddie Mac strongly encouraged lenders to use credit scores in mortgage underwriting. A Federal Reserve study published the following year found useful relationships between scores and loan performance. The number had something to recommend it beyond the convenience of a rule that could be applied quickly.⁹

It also traveled through several decisions. A reporting agency maintained credit information. A model turned selected information into an estimate. A lender decided what risk it would accept, and purchasers of mortgages influenced which loans that lender could readily sell. The score's mathematical form did not merge these institutions into one. A predictive advantage could justify using it while leaving the proposed use, the cutoff and the handling of an erroneous file open to criticism.

The Fair Credit Reporting Act recognizes much of this division. It provides for disclosure and dispute, and requires a reporting agency to reinvestigate disputed information under specified conditions. When a user takes an adverse action based on a consumer report, the notice identifies the reporting agency and also states that the agency did not make the decision and cannot supply the reasons for it. The distinction is already in the law: the institution furnishing information and the institution acting upon it have different things to answer for.¹⁰

An error in the record and a bad decision from an accurate record are therefore different problems. Correcting one does not necessarily correct the other. Equally, an accurate report that supports a justified refusal is a successful use of information, however disappointing the answer. A serious criticism of the bureau cannot rest on the premise that every person has been wronged whenever information makes a lender less willing to lend.

The obligation to a person described in a file is nevertheless real even when that person is not a customer. It cannot be priced adequately by asking whether the subscriber received good value. No estimate of the agency's monopoly profit is needed to establish that much. The trust tax remains a question about the premium attached to a position; the injury caused by a false report is not another name for that premium. An institution can earn little and exercise consequential power badly.

After the file was opened

When Timothy Garton Ash obtained his Stasi file after the fall of East Germany, he could do something the institution that assembled it had not intended: read its account beside his own diaries and memories. Its 325 pages contained reports, copied material and plans for further investigation. It also contained mistakes. The discrepancy between the date on which the file said he had left East Berlin and the date on which he had actually left was small enough to correct; other discrepancies required him to find people and speak with them.¹¹

The secret file had possessed administrative force without possessing an equal measure of truth. Officers could assemble selective information into a reason for further surveillance. The subject did not have a contemporaneous opportunity to challenge that developing account. Its papers were consequential because a state organization was prepared to act on them, not because their accumulation had finally overcome the uncertainties of observation.

Opening the archive changed who could ask questions. It did not make the answers arrive together. Garton Ash sought out informers he had known and officers who had handled his case; in his later account, all but one of those he sought agreed to talk. These encounters are among the least dispensable parts of the story. The people behind the reports did not disappear when their names became available. They had accounts of what they had done and what it had meant, and the former subject now had the disagreeable freedom to listen.¹²

The archive's own custodians distinguish political conviction, ambition, material advantage, a desire for recognition and fear among motives for collaboration. Coercion mattered, but it was not everybody's explanation. A history that makes every informer either a monster or a helpless victim has disposed of the work of judgment before the meeting begins.¹³

There is no need to assume that a friendship exposed by a file was false in every respect. That is part of the difficulty. An acquaintance may have been helpful, affectionate, interested in the conversation—and also have reported it. To discover the report is to acquire evidence of a betrayal. It is not to acquire a machine for sorting the rest of the relationship into events that count and events that never really happened. The injured person may wish for such a machine. So may the person trying to explain himself. Neither wish gives the archive powers it lacks.

I would be reluctant to call the resulting uncertainty a failure of disclosure. Before the file was opened, the subject had been denied knowledge. Afterward, he could confront something that deserved to remain difficult. The alternatives are not perfect understanding and a kindly ignorance maintained by the state. A democratic reckoning needs access to what was done; it need not pretend that making a document accessible settles what one person owes another after reading it.

Cornelia Vismann gives the administrative conceit its Latin: quod non est in actis, non est in mundo—what is not in the files is not in the world. Read as an account of an office's habits, the maxim is formidable. Read as a claim about everything that exists, it is absurd. A missing event may be outside the official account while remaining painfully present to someone. The danger begins when the office is empowered to proceed as if the account were exhaustive.¹⁴

Even an exhaustive account would leave a further question: what is this record still entitled to do? The secret archive and the opened archive put some of the same papers to different uses, under different powers. Destruction could remove evidence needed to hold an officer accountable; unrestricted renewed use could extend an injury long after the authority that first inflicted it had gone. Remembering and forgetting acquire their political meaning through those particular consequences.

A later republic may need the record to investigate an institution while refusing to let that institution's old classification determine someone's prospects. That is a decision about standing and use, not a discovery that the paper has become inaccurate on a certain date. It also leaves the wronged person with a question the republic cannot answer in their name. Public release of a record, public limits upon its use and personal forgiveness draw on different authority. The file may be necessary to all three inquiries without supplying the answer to any of them.

In a familiar light

The diamond dealer might prefer his own office, where he knew the light. Lisa Bernstein found this preference among large dealers in New York; smaller dealers made use of the secure floor provided by their club. A common place to trade was valuable without being the place in which everyone wished to do all his business.¹⁵

The Diamond Dealers Club brought people together because the work benefited from their proximity and experience. A stone was a portable concentration of value, and the promise to return it or pay for it might depend on a relationship whose force could not be put in the same envelope. Members used industry arbitration and commercial sanctions. Expertise mattered: the person hearing a dispute could know why a description, inspection or delay had become consequential without first being taught the trade.

Publicity was selective. Bernstein describes arbitration whose result could remain confidential if the award was promptly paid; default made publication and exclusion available. The arrangement did not aspire to disclose every quarrel to every potential customer. It gave a member reason to comply while permitting business to continue without a permanent public recital of a dispute already settled. Here privacy was part of the institution's means of securing performance. The same secrecy that would be objectionable if it hid an uncorrected injury could protect a completed accommodation.

Entry was controlled. Membership applicants needed industry experience, photographs were posted for objections, and family connections could ease admission. Space on the floor was itself a limiting condition. An out-of-town dealer had a different route: a member could sponsor his visit, assuming financial responsibility for his acts and liabilities on the premises, with admission also subject to the board and a fee. The club could let someone in by making his presence another member's financial concern. Belonging and being admitted were different arrangements.¹⁶

Bernstein advanced a strong survival proposition: “The private regime must be Pareto superior to the established legal regime in order to survive.” Its force depends on whose alternatives are being compared. Continued use can reveal advantages for participating dealers. It cannot, by itself, tell us that every outsider is no worse off, or that a member's cheapest available alternative is a satisfactory one. Survival is evidence that an institution has found means to continue. The welfare judgment requires more.¹⁷

Barak Richman's later account makes the continuity less impersonal. Family firms could pass a business and a reputation beyond the life of one dealer. The possibility of a last profitable defection had to be set against the business left to children. Other participants depended on dense religious communities whose sanctions reached beyond the income from an individual transaction. Richman distinguishes these arrangements rather than treating all Jewish merchants as governed by one communal motive.¹⁸

A reputation that outlives its present holder is a commercial resource, but it is also something the holder has not made alone. The prospect of damaging what a family has built gives a promise force. It may also make departure costly for someone who has not chosen the family business as enthusiastically as the family has chosen it for him. This is a possibility raised by the arrangement, not a biography we can infer from a model of its incentives. What matters is that the long future which reassures the creditor belongs to other people as well as to the person making the promise.

Writing in 2006, Richman also considered how changes in production, supply and marketing might erode these advantages. A working institution can find some of its services displaced while its members continue to value others. An outside certificate can replace one judgment while leaving credit, custody and the handling of disputes to be arranged elsewhere. The fate of the intermediary depends on which of those tasks its customers still need from it, and on the alternatives they can actually use.

Collective governance offers a real alternative to a proprietor who alone sets the terms. It does not abolish the problem of the boundary. Ostrom's studies of common resources gave organized users a place in making rules, monitoring compliance and resolving disputes with one another and with officials. Her comparative work did not yield a single set of operational rules that worked everywhere; the variation was part of what she had to explain.¹⁹

People who rely on an institution can acquire powers within it. They need not be confined to purchasing its service or leaving. But the participants in a successful arrangement still have neighbors, applicants and others affected by what they decide. The distinction between governing a common undertaking and governing everyone who encounters it remains to be negotiated.

The neighbor's claim

Hernando de Soto approached the boundary from the other side. People could occupy houses, work land and recognize one another's claims while lacking the title that a distant lender or buyer required. His argument about formal property is often reduced to the delivery of a document. In an interview explaining it, he insisted instead that property was a social agreement: paper alone did not decide who owned what.²⁰

That insistence should make the apparent solution less easy. A claim intelligible locally has to be admitted into another arrangement, where different people may have reasons to object. The neighbor's boundary, a relative's interest, an earlier transfer or a competing grant cannot be made harmless by the applicant's need for credit. Nor should their possible existence license an office to demand indefinitely that the applicant return with another document. The work of recognition has both an obligation to discover other claims and a capacity to shelter delay behind them.

What looks like a procedural inconvenience from inside the office can consume travel, wages and time that the person seeking recognition cannot spare. Those costs matter even when the official receives no bribe and earns no monopoly profit. They also need to be distinguished from the expense of hearing a valid objection. A swift title issued to the wrong person may be cheap for the office and ruinous for someone else. Speed is a virtue here only after we have asked what it accelerates.

Formalization changes the reach of a claim. It does not create a willing lender merely by creating a document the lender can read. The security, the borrower's means of repayment, the powers of enforcement and the terms of the loan remain consequential. Equally, failure to obtain a mortgage does not show that the house was previously without economic life. It may have sheltered a family, supported a business and been transferred under locally recognized rules. The problem is which of those relations another institution is willing and able to honor.

The temptation is to resolve this from one end: to dignify every local claim as a sufficient title, or to treat the official registry as the point at which possession first becomes real. Neither position allows enough room for disagreement among the people concerned. Useful recognition carries obligations to those whose claims it displaces. Its price to the applicant is only one part of its cost.

A market worth joining

The platform promises to spare its participants some of these difficulties. Customers can find sellers, payments can be arranged, previous transactions can inform a reputation, and disputes can be handled without each trader first assembling an institution of his own. The services need not be fictitious for the resulting dependence to be substantial. A market that brings customers is worth joining; that is why exclusion from it can hurt.

Rahman and Thelen locate the strength of the platform firm partly in this support from consumers, together with finance and the pursuit of network dominance. Their account is political as well as technological. People who benefit from an intermediary can become defenders of the conditions under which it operates. The attachment need not be foolish or manipulated to complicate reform. A customer may have a good reason to value cheap, convenient service without bearing all the costs imposed in producing it.²¹

Nor does the platform necessarily perform one service at one intelligible price. Market access may arrive with payment processing, reputation and enforcement. A gross commission does not identify what was paid for each, much less how much was a premium attributable to control. To establish that premium requires an account of alternatives and costs. Yet one can investigate suspension rules, withheld explanations or obligations to people described in a rating without first solving that accounting problem. A price inquiry and an inquiry into power can assist one another without becoming the same inquiry.

Departure is similarly uneven. A participant may keep customers, skills and a reputation outside the service, while losing the standing recognized within it. Exporting a list of favorable transactions is useful only to the extent that a receiving institution will identify the participant, understand the record and give it weight. The copy alone cannot compel recognition. There may be good reasons for the receiver to ask further questions; there may also be a commercial advantage in declining to recognize what a rival has recorded. Those possibilities need investigation in the particular market. Calling every difficulty of exit a technical defect would hide the interests that help determine it.

Competition can reduce the force of such interests. Entry restrictions can be removed, records made more portable and decision procedures opened to challenge. These are meaningful changes. But the person who is the subject of a report may still have no purchase to move to a competitor, and two rival institutions can offer their customers equally poor ways of hearing the people they describe. The discipline of customers' choices reaches only so far when consequences fall on someone else.

Barzel made the cost of determining attributes, protecting claims and arranging exchange central to his account of institutions. Certification can spare repeated inspection; the rules of an exchange can make dealing with a stranger possible. These are substantial economic achievements, and they belong to the explanation of why an institution endures. The coherence fee names the work required. The trust tax asks what control over its accepted provision permits the provider to charge.²²

The price can be fair and the customer satisfied while someone else has good reason to object. He may want an error corrected, an account heard before a decision is made, or a limit on what a perfectly accurate record is allowed to determine. Those demands are not all demands for better service to the purchaser. They arise from being made subject to the institution's work. Once we admit that relation, efficiency remains an achievement but ceases to be the institution's complete defense. The person who paid for the report cannot be the only person to whom an answer is owed.

The historical record offers no law that verification must create one kind of gatekeeper. It shows a recurring contest over who performs necessary work, who may recognize the result, and who can charge for access. Computation changes the cost and speed of that work. It does not decide who should control it.

Source notes

Footnotes

1. Henry David Thoreau to Sophia Thoreau, 22 May 1843, transcribed by the Thoreau Institute, citing The Correspondence of Henry David Thoreau, pp. 105–106. The extravagant estimate about stores is attributed to Giles Waldo, identified earlier in the letter; he and William Tappan were agency clerks and young friends of Emerson. It is not a finding about the agency's realized savings. The epigraph reproduces Thoreau's sentence exactly. ↩ ↩²

2. Laurie Nussdorfer, Brokers of Public Trust: Notaries in Early Modern Rome (2009), conclusion, pp. 226–230, especially pp. 227–229; author-uploaded excerpt. These arrangements concern the Roman offices studied, not a uniform medieval profession. The source also recognizes alternatives to notarization. The account of a fee bearing upon an office investment is the chapter's interpretation; no tariff decomposition or measured rent is claimed. ↩

3. Kathryn Burns, “Notaries, Truth, and Consequences,” American Historical Review 110, no. 2 (2005), pp. 350–379, especially p. 353 and n. 14; article. Burns cites Archivo Regional del Cuzco, Protocolos Notariales, protocolo 146, fols. 714v–715, 16 April 1663. The document is known here through her analysis, not an independent archival examination. Legal acknowledgment and the notary's observation must be kept distinct. ↩

4. Josh Lauer, Creditworthy: A History of Consumer Surveillance and Financial Identity in America (2017), pp. 29–33; publisher. Tappan's was the first successful agency, not the first attempted service. The chapter follows Lauer's account of local reporting and collection referrals without inferring an incentive or motive for every correspondent. ↩

5. Harvard Business School, Baker Library, “The Mercantile Agency”, and Lauer, pp. 32–33, on oral delivery and subscribers' own notes. Library of Congress, “About the Dun and Bradstreet Reference Book Collection”, dates Dun's published reference book to 1859. Manuscript reports and later printed ratings are distinct records. ↩

6. Report on Charles Dull, 20 March 1850, as transcribed in Lauer, p. 39. The quoted final clause is the correspondent's expectation. No claim is made here about the subsequent transaction, Dull's actual repayment, or his response to the report. ↩

7. Lauer, pp. 24–25, 42–43, on the limitations of the institutional archive and litigation; Baker Library, “The Reference Book”, on reporting guides and restricted subject access. The chapter does not adopt the exhibition's broad characterization of immunity: the actual judgment below supplies a narrower account of one contest. ↩

8. Tappan v. Beardsley, 77 U.S. (10 Wall.) 427 (1870), pp. 428–433 for the communication, pleadings, record and verdict; pp. 433–436 for Justice Miller's opinion and the order for a new trial. The account distinguishes the allegations in the divorce record from established findings. The Court rejected the use of the record against a nonparty who had lacked the relevant opportunity to challenge it; it did not decide the agency's general claim of privileged communication. The wife's wishes are not established by this reported record. ↩

9. FICO, “History of the FICO Score”, on 1989, read as company chronology. Robert B. Avery, Raphael W. Bostic, Paul S. Calem and Glenn B. Canner, “Credit Risk, Credit Scoring, and the Performance of Home Mortgages”, Federal Reserve Bulletin 82 (July 1996), pp. 621–648, especially pp. 628–629 and 647–648. Their evidence of predictive usefulness and their account of institutional guidance do not estimate every consequence of adoption or establish the adequacy of every particular model use. ↩

10. Fair Credit Reporting Act: 15 U.S.C. §1681g (disclosure, with specified limits); §1681i (reasonable reinvestigation, generally within thirty days, subject to statutory qualifications); §1681m(a) (adverse action based in whole or part on a consumer report). The discussion concerns consumer reporting under the statute, not all commercial reports or a complete account of rights to reasons under other law. ↩

11. Timothy Garton Ash, The File: A Personal History (1997), Part I, book preview. The preview describes a file dating his departure to June 1980 where his own account gives October, among other discrepancies and plans. The narrative uses the book's comparison of file, diary and recollection; it does not treat administrative classification as proof of the suspected activity. ↩

12. Garton Ash, “The Stasi on Our Minds”, New York Review of Books, 31 May 2007. The retrospective confirms the scope of the personal inquiry and the meetings. The subsequent paragraphs develop the chapter's interpretation of the difficulty disclosure leaves; they invent no particular conversation, remembered friendship or act of forgiveness. ↩

13. Bundesarchiv, Stasi Records Archive, “The Unofficial Collaborators (IM) of the MfS”, section on motives. A range of recruitment pressures does not establish any one person's explanation or excuse. ↩

14. Cornelia Vismann, “Out of File, Out of Mind,” in Wendy Hui Kyong Chun and Thomas Keenan, eds., New Media, Old Media: A History and Theory Reader (2006), pp. 97–104, especially pp. 97–98. This is the inspected passage for the maxim. The chapter does not attribute a continuous Roman-to-modern legal rule to it. The distinction between public release and personal forgiveness remains the selected trilogy's existing distinction; the example here does not introduce a new formula or protocol. ↩

15. Lisa Bernstein, “Opting Out of the Legal System: Extralegal Contractual Relations in the Diamond Industry”, Journal of Legal Studies 21, no. 1 (1992), pp. 115–157: offices and light, p. 120; confidentiality, pp. 124 and 126; publicity and suspension or expulsion following nonpayment, pp. 128–129. These are practices described in that study, not an assertion about every present bourse. Public law remains relevant to enforcement; preference for industry adjudication does not prove courts incapable of hearing a diamond dispute. ↩

16. Bernstein, pp. 119–120, especially nn. 5–6: visiting nonmembers and applicants for membership follow different rules. The text retains the practical space constraint and family concessions, without attributing every restriction to ethnic exclusion. ↩

17. Bernstein, p. 117, exact quotation. “Pareto superior” requires specifying whose welfare and which alternatives are included. The chapter questions what survival alone establishes, without denying benefits her account identifies for members. ↩

18. Barak D. Richman, “How Community Institutions Create Economic Advantage: Jewish Diamond Merchants in New York,” Law & Social Inquiry 31, no. 2 (2006), pp. 383–420; Duke record and author manuscript, §§IV.B–D and VI. The manuscript distinguishes family firms' long horizons from other community sanctions. Its pagination differs from the published article. Its prospective discussion of erosion is not evidence that the institutions had already disappeared. ↩

19. Elinor Ostrom, “Beyond Markets and States: Polycentric Governance of Complex Economic Systems,” Nobel lecture, 8 December 2009, in The Nobel Prizes 2009, pp. 408–444, especially pp. 421–423; lecture text. This returns to the research developed in Governing the Commons (1990), without claiming that institutional participants adopted one universal design. ↩

20. Hernando de Soto, interview, The Region, Federal Reserve Bank of Minneapolis, 2001, discussing The Mystery of Capital (2000). His explanation is a proposition about recognized property and economic use; the chapter's qualifications concerning rival claims, credit and procedural cost are not represented as measured outcomes of a new case study. Aggregate asset valuations and precise procedural inventories are not used. ↩

21. K. Sabeel Rahman and Kathleen Thelen, “The Rise of the Platform Business Model and the Transformation of Twenty-First-Century Capitalism”, Politics & Society 47, no. 2 (2019), pp. 177–204, especially pp. 179–181. The following discussion of bundles and portability is conditional institutional analysis, not a claim about one platform's current fees, export interface or suspension procedure. ↩

22. Yoram Barzel, “Measurement Cost and the Organization of Markets”, Journal of Law and Economics 25, no. 1 (1982), pp. 27–48, especially pp. 27–29 and 46–48; Economic Analysis of Property Rights, 2nd ed. (1997), p. 3, edition excerpt. Measurement and institutional arrangements are serious antecedents, not merely distant analogies. The distinction between legal assignment and effective control is already explicit in Barzel. No formula equates a gross fee or every adverse consequence with economic rent. ↩