Power in Recommendation and the Jurisprudence of Co-Experience - Who Has the Right to Interpret, Contest, Transmit, and Reproduce Shared Experience?

Transcript

Abstract

An account of a shared experience is composed by one participant in a relation and used by a third party who did not participate in it, while the person it describes is ordinarily absent. This paper asks what rights attach to such an account and to whom, treating interpretation, contestation, transmission, reproduction, and ownership as five aspects of one object on the ground that the scope of the last four is fixed by the allocation of the first. It argues along two routes kept separate throughout: a doctrinal survey of four legal systems, and derivation from a set of principles of relational reproduction. The survey finds one allocation repeated across provisions enacted in different systems for different purposes, namely that the person described is granted access to an account of themselves and is withheld authority over its content; access and interpretive authority are shown to be separately allocable, and the convergence is explained by the prior location of interpretive authority in whichever party composes the account. Three further findings are reported: no located doctrine confers an integrity interest upon the person a work describes; positive law names the operative force of instruments whose force it has itself constituted and supplies no name for the force of an instrument working without legal constitution, with the consequence that exchanges of such instruments have no description in property law; and the events an account reports lie outside copyright entirely. Where the two routes meet directly, they conflict: copyright denies that authorship of a jointly produced past is shared, and denies it for a stated reason the paper accepts as sound. The framework’s principle is therefore amended rather than defended, separating a claim to participate in settling an interpretation, which survives, from a claim to own accounts of the experience, which is abandoned. Two questions are stated and left open, with the conditions an answer would require.

Keywords: interpretive authority; co-experience; recommendation; joint authorship; data-subject rights; moral rights; contestation.

Discussion Paper Note

This paper is a preliminary discussion paper intended to share an evolving idea and invite further dialogue, criticism, revision, and independent development. Its definitions, distinctions, and constructions remain provisional. Circulation across scholarly and practical communities is part of the purpose of releasing the manuscript at this stage.

The author treats the viewpoints, concepts, and lines of reasoning presented here as contributions to a shared field of inquiry. Similar or related ideas may have appeared in other intellectual, cultural, and disciplinary traditions. The manuscript therefore states its known antecedents, separates the researcher-origin proposal from later formal reconstruction, and leaves historical priority open pending a systematic originality review.

The arguments should be understood as provisional and historically situated. Readers are encouraged to question, test, revise, extend, reinterpret, or independently develop the ideas presented here. Where appropriate, acknowledgment of this paper as one point of encounter in the development of a related idea is appreciated. Such acknowledgment records an intellectual route; the ideas themselves remain available for criticism, revision, and independent development.

Responsible Use and Rights Reservation

This section separates requested scholarly conduct from the legal permissions stated on the following page. It records an ethical request for responsible use and then defines the narrower scope of retained legal rights.

The author encourages good-faith discussion, criticism, independent inquiry, and responsible use of the material in this work. Separately from the licence’s terms, the author asks users to consider foreseeable harms when adapting or applying the arguments made here. This paper argues that a person described in an account of a shared experience holds interests in how that account is interpreted, contested, transmitted, and reproduced. An argument of that shape can be used to suppress accurate adverse accounts, including accounts that warn of conduct posing a foreseeable risk to third parties. The law of negligent hiring exists because such warnings are sometimes owed. Nothing argued here supports withholding them, and the paper states at the point the question arises which interests yield where a risk to others is in issue.

A second foreseeable misuse concerns the ownership limb. That copyright in an account vests in the party who wrote it is a description of positive law and not an endorsement of the allocation. It should not be cited as authority for the proposition that a person has no standing regarding accounts of their own conduct, which the paper denies.

The author reserves all rights not granted by the licence stated in the notices below.

Notices

Status.

Preliminary discussion paper. Not peer reviewed. Not submitted for publication at the time of circulation.

Licence.

This work is made available under the Creative Commons Attribution–NonCommercial 4.0 International licence (CC BY-NC 4.0).

Language-model statement.

The author used a large language model as a drafting and verification aid. All argumentative claims, all doctrinal readings, and all decisions about scope and concession are the author’s own. Every source cited was verified against a publisher, journal, statutory, judicial, or institutional record before it was admitted to the manuscript.

Declared interest.

The author has been the subject of institutional judgments of the kind this paper examines. The paper’s object is general and its arguments are stated for that general object. The author’s own case is not a source of evidence in the paper and is not discussed in it.

Related research programme.

This paper belongs to a series on trust, neutrality, and the transmission of shared experience. It takes as settled the ontology of the judged subject established in the fourth paper of that series and the duties of the issuing party established in the fifth. It cedes distribution, unequal generative conditions, and testimonial injustice to the injustice paper, and cedes circulation, accumulation, and exchange to the political-economy volume.

Suggested citation.

Huang, Wanhong. Power in Recommendation and the Jurisprudence of Co-Experience: Who Has the Right to Interpret, Contest, Transmit, and Reproduce Shared Experience? Preliminary discussion paper, 2026.

Introduction

A person who has worked closely with another is asked to give an account of them. The request comes from a third party that will decide something about the person described: whether to admit, hire, promote, fund, license, or trust them. The account is written by one participant in a relation, concerns an experience that two people produced together, and is used by a party who stood outside that relation while it was occurring. The person the account is about is typically absent when it is written, is often unaware of what it says, and in the ordinary case has no occasion to answer it. This paper asks what rights attach to such an account and to whom, and it treats that question as a question of jurisprudence rather than of measurement, professional ethics, or institutional practice.

The object is general. Letters of recommendation are its most familiar instance, but employment references, performance appraisals, peer review reports, clinical formulations, teacher evaluations, and structured risk assessments share the feature that generates the question: an experience constituted jointly by two parties is interpreted unilaterally by one of them for the use of a third. Where the argument turns on a feature specific to one instrument, the paper says so. Where it turns on the general structure, it states the general claim and identifies the domains in which the structure holds.

Five rights are examined and they are examined together: the right to interpret a shared experience, the right to contest an interpretation of it, the right to govern its transmission, the right to govern its reproduction, and the ownership of the account in which it is recorded. They are treated as aspects of one object rather than as five topics, on the ground argued in Section 5, that authority to interpret is prior to the remaining four and that the others follow from its allocation. That architectural claim is the paper’s principal structural commitment and its principal exposure. If interpretive authority does not order the remaining rights, the arguments of Section 6 stand as five independent analyses and the paper’s unity is lost, though its individual results are not.

The paper proceeds by two routes and keeps them separate throughout. The first is doctrinal: it asks what positive law already allocates, in which jurisdictions, and on what stated reasoning. The second derives consequences from the generative relational framework in which the author works, and in particular from principles of relational reproduction that framework states. Section 4 sets out both routes, the jurisdictions and instruments selected, and the conditions under which each result would fail. Section 8 then reports where the routes converge, where a result depends on one route alone, and where they contradict each other. The last of these is not a formality. On the question of whether authorship of a jointly produced past is shared, positive law and the framework reach opposite conclusions, and the law reaches its conclusion for a stated reason that the framework must answer.

Much of what follows is already owned, and the paper concedes each owner at the point the corresponding claim arises rather than in a closing note. Six concessions are large enough to state here.

First, that evaluative assessments of a person constitute that person’s own data, and are subject to rights of access and rectification, is settled in European law and is not claimed here. The Court of Justice held in Nowak (Court of Justice of the European Union 2017) that written answers submitted by an examination candidate together with the examiner’s comments upon them are personal data, subjective assessments included, an interpretation prepared by the concept of personal data developed in Opinion 4/2007 (Article 29 Data Protection Working Party 2007) and now carried by Articles 15 and 16 of the General Data Protection Regulation (European Parliament and Council 2016). The limit of that holding matters more to this paper than the holding. Rectification reaches the accuracy of the record of an opinion, requiring that it reflect what was said, by whom, and when; it does not reach the correctness of the opinion recorded. Where the substance is disputed, the remedy is a supplementary statement rather than an alteration. The paper’s question begins where that limit falls.

Second, the person described already holds a default right to inspect a recommendation in at least one jurisdiction, and the practice of that jurisdiction is waiver. Under the regulations implementing the Family Educational Rights and Privacy Act, an institution need not permit inspection of a confidential recommendation if the student has waived the right and the recommendation concerns admission, employment, or an honour; the waiver may not be required as a condition of a benefit, and the names of those who wrote must be supplied on request (United States Department of Education, n.d.). The right exists, and its ordinary exercise is to give it up.

Third, a legislature has considered whether the person described should see a confidential reference and has answered that they should not. Schedule 2, paragraph 24 of the Data Protection Act 2018 removes the listed data-subject rights from personal data consisting of a reference given in confidence for employment, education, appointment, or the provision of a service (Parliament of the United Kingdom 2018). The exemption covers the reference in the hands of the giver and of the recipient alike. This is the most direct legislative answer against the position developed here, and Section 6 meets it rather than noting it.

Fourth, on the confrontation of adverse evidence the paper claims nothing new and inherits a threshold it cannot avoid. Goldberg (Supreme Court of the United States 1970) establishes an opportunity to defend by confronting adverse evidence where a statutory entitlement is at stake, and Roth (Supreme Court of the United States 1972) recognises a liberty interest where an official act imposes a stigma that forecloses other opportunities. Paul v. Davis (Supreme Court of the United States 1976) then holds that reputation alone, unaccompanied by a more tangible interest, is neither liberty nor property. Constitutional process therefore does not reach a private evaluative account, and no argument here depends on the supposition that it does. In the law of references proper, the duties that exist run to the accuracy and fairness of what is written rather than to the participation of the person written about: a duty of care in preparing a reference (House of Lords 1994), a requirement that the reference not give a misleading overall impression (Court of Appeal of England and Wales 1998), and a requirement of investigation before an allegation of misconduct is passed on (Court of Appeal of England and Wales 2001). A statutory dispute procedure of the kind provided for consumer files (United States Congress, n.d.) shows what a rectification right looks like when a legislature builds one, and shows equally that it is built around factual accuracy.

Fifth, the events underlying an account cannot be owned. Facts do not originate in an act of authorship and are therefore not original and not copyrightable (Supreme Court of the United States 1991); the statute excludes ideas, procedures, and concepts from protection regardless of how they are described (United States Congress 1976b); and the exclusion has been extended to historical events and to interpretations advanced about them (United States Court of Appeals for the Second Circuit 1980). A shared experience, considered as what happened, holds no intellectual property. Only an expression of it does.

Sixth, the moral rights that protect the integrity of an account protect its author. Attribution and integrity are vested in the author by the Berne Convention (World Intellectual Property Organization 1971), by the narrow federal provision available in the United States (United States Congress 1990), and by the civil-law traditions that state them more strongly (National Diet of Japan 1970). No doctrine surveyed for this paper vests an integrity interest in the person a work describes.

Taken together these concessions disclose a structure that no single one of them displays, and the paper’s first substantive claim is about that structure rather than about any right taken alone. Across every limb examined, positive law grants the person described some measure of access to an account of them and withholds authority over its content. Assessments are their data and are disclosable, and the opinion within stands. Inspection is the default and waiver is the practice. The reference is exempted outright in one jurisdiction. Integrity belongs to the author, ownership to the writer, and reputation alone falls below the threshold at which process attaches. Six provisions, arrived at independently and in different centuries, converge on one allocation: interpretive authority over a jointly produced experience rests with whichever party writes it down.

Against that allocation the framework’s principle of co-reproduction holds that a shared past may not be interpreted, defined, or revised unilaterally, since authorship of it is shared. Copyright law denies precisely this. A joint work requires two or more authors intending that their contributions merge into inseparable or interdependent parts of a unitary whole (United States Congress 1976a); that intent must exist at the time of creation (United States Court of Appeals for the Second Circuit 1991); and authorship is required and differs from making a valuable and copyrightable contribution, with control usually the most important factor (United States Court of Appeals for the Ninth Circuit 2000). The person described satisfies none of these conditions. The denial is moreover reasoned rather than inadvertent: a looser test was rejected on the ground that it would expose authors who had talked with others about their work to claims from those conversations. The paper takes that ground seriously in Section 6 and states in Section 9 what the framework must give up if the ground is good.

Two adjacent literatures are relied upon without being extended. That the past recollected is organised within social frameworks rather than retrieved intact is Halbwachs’s (Halbwachs 1992); that credibility and interpretive resources are unequally distributed, and that the inequality is an injustice rather than a misfortune, is Fricker’s (Fricker 2007). Neither is applied by its author to instruments of institutional selection, and this paper does not develop either, since the distribution of interpretive resources and the injustices arising from it belong to the injustice paper in this series.

The paper is bounded in three further directions. It takes as settled the ontology of the judged subject established earlier in the series and does not re-argue what a judgment formed inside a relation warrants outside it. It takes as settled the duties of the issuing party. It cedes circulation, accumulation, and what is exchanged for an instrument to the political-economy volume, and enters the economics of these instruments only so far as is needed to show why property concepts fail to reach what is exchanged. One correction to the earlier work is discharged here rather than there: the claim that no forum exists in which the person judged could contest is too strong, and the accurate statement is that forums exist which reach the record of an evaluative judgment and stop at its substance.

Sections 2 and 3 set out the object and the existing treatments of authority over an account of a person. Section 4 states the two routes and the conditions of disconfirmation. Section 5 gives the framework’s account of a shared experience and argues the priority of interpretive authority. Section 6 examines the five rights and closes with the structure common to them. Section 7 argues that the force of such an instrument has no name in doctrine and that property concepts consequently cannot reach its sale. Sections 8 and 9 report the standing of each result and what follows for the framework, and Sections 10 and 11 state the limits of the account and conclude.

Two general limitations are stated at the outset because they qualify everything that follows. The doctrinal route surveys a bounded set of jurisdictions and instruments, and the absence of a located prior owner for a claim is not a demonstration that none exists. And where the paper reports that positive law is silent on a question, the silence is evidence about the law rather than about whether the question is well posed.

Background and Preliminaries

This section fixes the object of the paper and the vocabulary used to describe it, states which domains and jurisdictions the argument covers, records the results inherited from earlier work that are taken as settled here, and marks the questions reserved to later work in the same series. Its method is stipulative and expository rather than argumentative: it defines and delimits, and defers every contested claim to the sections that argue for it. The section proceeds in four parts, describing the practice and defining its terms, bounding the domains and jurisdictions examined, listing the inherited results together with the sections that depend on each, and stating the reservations.

Accounts of Shared Experience in Institutional Selection

Three parties figure in the practice examined here, and the relations among them generate the paper’s question. Two of them participated together in some course of dealing: they worked together, studied together, supervised and were supervised, treated and were treated. That course of dealing produced observations available to both and available in that setting alone. The third party stood outside it. At some later moment the third party must decide something about one of the first two, and asks the other for an account of what they saw.

Definition 1 (An Account Of A Shared Experience). A written or spoken characterisation of one person, produced by a second person, drawn from a course of dealing in which both participated, and supplied to a third party who did not participate in it and who will use the characterisation in reaching a decision about the first person.

Definition 2 (The Parties To Such An Account). The issuer is the party who produces the account. The person described is the party the account characterises and who participated in the course of dealing from which it is drawn. The receiving institution is the party who requests, holds, or acts upon the account and who did not participate in that course of dealing.

Four features of this arrangement recur across every instrument examined and together constitute the object. First, the experience characterised was produced jointly. Neither party could have generated it alone, and each was a condition of what the other did within it. Second, the characterisation is produced unilaterally. One participant selects what to report, in what terms, and under what emphasis, and the selection is not negotiated with the other. Third, the characterisation is used outside the relation that produced it, by a party whose own purposes govern what is done with it. Fourth, the person described is ordinarily absent from the transaction: absent when the account is composed, frequently unaware of its contents, and in the standard case without occasion to answer it.

The fourth feature is a matter of ordinary arrangement rather than of necessity, and Section 6 examines the instruments in which it is relaxed. The first three are constitutive. Where an account is drawn from no shared course of dealing, or where the characterisation is jointly composed, or where it is used only within the relation that produced it, the paper’s question does not arise in the form put here.

Scope of the Domains Considered

The domains examined are those in which all three constitutive features hold. They include the letter of recommendation in academic and professional selection, the employment reference, the performance appraisal used in promotion and retention decisions, the peer review report, the clinical formulation transmitted to a party other than the patient, the teacher evaluation entering a pupil’s record, and the supervisory assessment entering a professional licensing or disciplinary process. In each, a party who observed another from within a working, therapeutic, or instructional relation renders that observation to a party who did not.

Three neighbouring practices fall outside the object and are excluded by the definition rather than by choice. Assessments composed from records by a party who never encountered the person, including credit and actuarial scoring performed on documentary inputs alone, involve no shared course of dealing and so raise no question of divided authority over a jointly produced past. Standardised test results occupy the same position: whatever difficulties they present, the difficulties concern measurement and the interpretation of scores and arise between a person and an instrument rather than between two participants in a relation. Self-reports and personal statements are composed by the person they describe and therefore lack the second feature. Each of these may raise adjacent questions about accuracy, fairness, and standing. None raises the question of who holds authority to interpret an experience two people made.

One general treatment bears on the object as a whole and is conceded before the jurisdictional bounds are set. That institutions render populations governable by imposing simplified and standardised descriptions upon them, and that the descriptions serve the institution’s need to act rather than the described party’s circumstances, is Scott’s (Scott 1998). The instruments examined here are instances of that simplification, and nothing in the present paper adds to the general claim. What Scott’s account addresses is the relation between an institution and a population it must administer; what it leaves untouched is the question pursued here, which arises between two particular parties before any institution receives what they produced.

The doctrinal route surveys four jurisdictions and no others. It examines federal law of the United States on copyright, procedural due process, student records, and the correction of consumer files; the law of England and Wales on employment references and on data protection; the law of the European Union on personal data and on the rights of data subjects; and Japanese copyright law where moral rights are at issue. Statements about what the law provides are statements about these systems. Where a provision surveyed here has an analogue elsewhere, the paper says so only where the analogue has been verified, and it makes no claim that the pattern reported in Section 6 extends to systems it has not examined.

Results Carried from the Preceding Papers

Four results established earlier in this series are taken as settled and are not re-argued. Each is stated here together with the sections of this paper that rely upon it, so that a reader who rejects an inherited result can identify what falls with it.

The first is the ontology of the judged subject: that an attribute exhibited within a relation depends for its exhibition upon that relation, so that the validity of an observation is indexed to the relation in which it was made and extrapolation beyond that relation is an error of scope rather than of accuracy. Section 5 builds upon this and Section 6 uses it wherever the correctness of a characterisation is distinguished from its warranted reach.

The second is the account of what such an instrument does in the receiving system: that it alters the relational conditions available to the person described rather than conveying a quantity of information about them, so that the characterisation offered functions as a justification for altering those conditions rather than as the substance transmitted. Section 7 depends upon this result and would fail without it. The result is contested by accounts on which the instrument is informative through what its issuer stakes rather than through what it says, and the contest is being pursued in separate work. This paper cites the result as it stands, does not re-argue it, and confines its use to Section 7.

The third is the set of duties borne by the issuing party: that judgments of persons formed inside one relation and used inside another are governed by duties appropriate to an exercise of discretionary power rather than by norms of assertion, that the primary obligation runs to the person described rather than to the party who asked, and that refusal to issue is the act requiring the stronger justification. Section 6 takes these as given wherever a duty of the issuer is in question and adds nothing to them.

The fourth is the set of principles of relational reproduction stated by the framework in which the paper works, together with two extensions to them: that the conditions making later revision possible must themselves be preserved rather than merely permitted, and that where an act confers interpretive authority the capacity to contest that authority must be generated by the same act rather than supplied afterwards. Section 5 states these principles for the present object and Section 6 derives from them; Section 9 states what must be surrendered among them if the doctrinal route is right where the two routes conflict.

Boundaries Observed toward the Later Papers in the Series

Three bodies of question adjacent to this one are reserved and are not developed here, and the reservations are stated so that their absence is read as a boundary rather than as an oversight.

Questions of distribution belong to the injustice paper in this series: how unequally the capacity to obtain a favourable account is distributed, how differences in the conditions under which people generate observable performances produce differences in what can be said of them, and how the suppression of adverse accounts propagates. The unequal distribution of credibility and of interpretive resources belongs there also, and the present paper uses that literature only to locate its own question against it.

Questions of circulation and accumulation belong to the political-economy volume: what accrues to those who issue such instruments, what is exchanged for them, and how a capacity to issue functions where it is held unevenly. Section 7 enters this territory to the extent needed to show why concepts of property do not reach what is exchanged when an instrument is sold, and stops at that point.

Questions of institutional design at scale belong to the governance paper: what apparatus would administer the rights argued for here, at what cost, and with what failure modes. This paper states the institutional conditions a remedy would require, in Section 6, and does not propose a regime.

Existing Treatments of Authority over an Account of a Person

This section identifies the existing owners of each claim the paper will make and states how far each owner goes, so that the contribution claimed in Sections 6 and 7 can be measured against what is already held. Its objective is concession rather than survey: each body of work is presented for the claim it forecloses to this paper, and the limit of that claim is stated in the same paragraph. The section treats in turn the theories of interpretive authority, the procedural standing to contest an adverse record, the control given over personal data, the allocation of authorship and ownership in copyright, and the analytical apparatus of correlative rights and duties, closing with the boundary that remains. The method throughout is doctrinal and textual: provisions are read for what they allocate, and theoretical works for the claim they establish rather than for their wider programme.

Interpretive Authority in Legal and Hermeneutic Theory

The claim that a past is not simply retrieved by whoever reports it is Halbwachs’s. Recollection is organised within social frameworks and reconstructed in the present according to the concerns of the group within which the remembering occurs (Halbwachs 1992). This forecloses to the present paper any suggestion that a shared experience possesses a determinate content which a faithful account would transmit and a careless one would distort. What it does not supply is any allocation of authority. Halbwachs describes how a past is constituted socially and says nothing about which participant may render it for institutional use.

The claim that interpretation is never the neutral recovery of a meaning lying ready in its object belongs to philosophical hermeneutics, and in its best-known form to Gadamer, for whom understanding proceeds through the interpreter’s own situation rather than despite it, so that no interpretive standpoint is available from which an object could be read without the reader entering into what is read (Gadamer 1989). Two limits attend the inheritance. Gadamer’s object is a text or a tradition rather than a person, and his account of understanding as a meeting of horizons presumes a reciprocity that the instruments examined here withhold, since one party composes and the other is absent. Critics have pressed the second point, observing that the model has little to say where the parties are unequal and the disagreement is not resolvable by better conversation.

The nearest existing owner of the specific claim that a jointly produced past is jointly authored is Dworkin’s chain novel (Dworkin 1986). An author writing a further chapter of a work composed serially by many hands is constrained twice, by fitting what precedes and by making the whole the best work it can be, and no single contributor holds authority over what the work means. The construction is close to this paper’s object and must be conceded as prior. Its distance is equally clear. Dworkin’s contributors write in sequence and each contributes text; the parties to a shared experience produce that experience jointly and only one of them writes anything. The constraint of fit governs an interpreter facing a record, whereas the question here is whether a participant may make the record at all.

The claim that institutions constitute persons as describable objects through the documentary apparatus by which they assess them is Foucault’s (Foucault 1977). On that account the examination combines observation with judgment and fixes its result in a file, and the resulting categories do not merely record subjects who existed independently of the procedure but supply the terms through which such subjects are afterwards governed. This forecloses to the present paper any suggestion that the practice of writing accounts of persons for institutional use is a neutral reporting apparatus onto which questions of power are later imposed. Two limits attend the inheritance and both matter here. Foucault’s analysis is directed at the formation of a technique and its historical spread, and it allocates no rights, proposes no remedy, and identifies no party with standing. And its unit is the apparatus rather than the relation, so it does not reach the feature that generates this paper’s question, which is that the experience being documented had two producers.

The claim that authority to be believed, and the interpretive resources needed to make sense of one’s own experience, are unequally distributed, and that the inequality is a wrong rather than a misfortune, is Fricker’s (Fricker 2007). It is the strongest existing statement that interpretive standing is a matter of justice. Its object is epistemic conduct and the ethics of credibility; it is not addressed to instruments of institutional selection, and it allocates no rights. The distribution of interpretive resources is in any event reserved to the injustice paper in this series.

Procedural Standing to Contest an Adverse Record

Where a state deprives a person of a statutory entitlement, an opportunity to defend by confronting adverse evidence is required (Supreme Court of the United States 1970), and where an official act imposes a stigma foreclosing other opportunities a liberty interest is engaged (Supreme Court of the United States 1972). The reach of both is settled by a third holding: reputation standing alone, unaccompanied by a more tangible interest, is neither liberty nor property (Supreme Court of the United States 1976). The consequence for this paper is a limit rather than a resource. Constitutional process does not reach a private evaluative account, and no argument in Section 6 supposes that it does.

The law of references imposes duties, and their direction is instructive. An employer owes a duty of care in preparing a reference and may be liable to the subject for economic loss caused by carelessness (House of Lords 1994); a reference must not convey a misleading overall impression (Court of Appeal of England and Wales 1998); and an allegation of misconduct requires investigation and reasonable grounds before it is passed on (Court of Appeal of England and Wales 2001). Each duty runs to the quality of what is written. None gives the person written about any part in the writing of it, and the duty of care in Spring is discharged by taking care rather than by consulting.

A legislature building a right to dispute a record produces something of a different shape. The statutory procedure for consumer files requires reinvestigation upon dispute and deletion of information that is inaccurate or cannot be verified (United States Congress, n.d.). This is the clearest model available of what a rectification right looks like when one is designed deliberately, and it is built around ascertainable accuracy.

Control over Personal Data in Information Law

The most developed body of law bearing on the paper’s object holds that evaluative material about a person belongs, as data, to the person evaluated. The concept of personal data was given its broad construction in Opinion 4/2007, whose treatment of what it is for information to relate to a person (Article 29 Data Protection Working Party 2007) was carried into the holding that an examination candidate’s answers together with the examiner’s comments upon them are the candidate’s personal data, subjective assessments included (Court of Justice of the European Union 2017). Rights of access and rectification follow from Articles 15 and 16 (European Parliament and Council 2016).

The limit is the part that matters here, and it is not incidental. Because those rights are tied to the purposes for which data are processed, and because the accuracy principle requires that a record faithfully represent what was expressed, by whom, and when, rectification reaches the record of an opinion and stops at its content. A disputed assessment is answered by appending a statement, not by amending the assessment. The law thus grants the person evaluated a right to see and a right to annotate, and withholds any authority over what was concluded.

One theoretical owner must be conceded alongside these provisions, since it holds the claim on which Section 6 rests when it describes an account travelling beyond the relation that produced it. That information carries with it the norms of the setting in which it was disclosed, and that the wrong done by its movement consists in a breach of those norms rather than in the sensitivity of what is disclosed, is Nissenbaum’s (Nissenbaum 2004, 2010). Her analysis specifies a flow by its sender, its recipient, the subject of the information, the attributes conveyed, and the principle governing the transfer, and this paper’s account of circulation adds nothing to that specification. What her framework settles is the appropriateness of a flow; what it leaves open is the question pursued here, namely who holds authority to compose the description that flows. The two are distinct: an account may travel wholly appropriately, to the recipient the originating setting would sanction, and the question of who was entitled to write it remains untouched.

Two further provisions bear directly. In the United States a student holds a default right to inspect a recommendation, subject to a waiver whose ordinary execution removes it, with the institution obliged to disclose on request the names of those who wrote (United States Department of Education, n.d.). In the United Kingdom the position is the opposite and was legislated: personal data consisting of a reference given in confidence for employment, education, appointment, or the provision of a service are removed from the listed data-subject rights, and the exemption covers the reference in the hands of the giver and of the recipient alike (Parliament of the United Kingdom 2018).

Authorship and Ownership in Copyright Doctrine

Copyright vests initially in the author, and where a work is made for hire the employer is treated as the author (United States Congress 1976c). Ownership of the copyright is distinct from ownership of the object in which the work is embodied, and transferring the object conveys no rights in the work (United States Congress 1976d). A joint work requires two or more authors intending that their contributions merge into inseparable or interdependent parts of a unitary whole (United States Congress 1976a), an intent that must be present when the work is created (United States Court of Appeals for the Second Circuit 1991), and authorship is a further requirement that a valuable and copyrightable contribution does not by itself satisfy, control being usually the most important factor (United States Court of Appeals for the Ninth Circuit 2000).

The events an account describes lie outside copyright altogether. Facts do not originate in an act of authorship and are therefore neither original nor copyrightable (Supreme Court of the United States 1991); the statute excludes ideas, procedures, and concepts however they are expressed (United States Congress 1976b); and the exclusion has been applied to historical events and to interpretations offered of them (United States Court of Appeals for the Second Circuit 1980).

Where a legal system protects the integrity of a work it protects the author of it. The attribution and integrity rights of the Berne Convention (World Intellectual Property Organization 1971), the narrow federal provision available in the United States (United States Congress 1990), and the stronger civil-law statement of the same two rights (National Diet of Japan 1970) are alike in their beneficiary. Among the doctrines surveyed for this paper, none confers an integrity interest on the person a work describes.

Correlativity of Rights and Duties in Analytical Jurisprudence

The apparatus by which a right is analysed into a relation between two parties, so that a claim in one is matched by a duty in another and an assertion of a right is incomplete until its correlative duty and its bearer are identified (Hohfeld 1913, 1917), supplies the discipline this paper applies in Section 6. The apparatus is used and is not extended. Its value here is diagnostic: it makes visible those of the five rights for which no duty-bearer has been identified, and it converts a claim that someone ought to have a say into a question about who would owe what to whom.

Boundary of the Present Contribution

Three things follow from the survey in this section. First, each component of the paper’s position has an owner, and the owners are conceded above and again at the point each claim is made. Second, the strongest existing owner of the paper’s territory grants access and annotation while withholding authority over content, so that the question of who may interpret survives the grant. Third, a search of the legal and philosophical literature conducted for this paper found no treatment of the instruments described in Section 2 as objects of rights, authority, and standing. The existing literature on them divides between the liability of those who issue them and the accuracy of what they contain, and the ethical literature adjacent to them concerns the conduct of the issuer. The framing adopted here is accordingly the paper’s own, subject to the qualification stated in Section 1 that a bounded search establishes the absence of a located owner rather than the absence of one.

Method, Routes, and Conditions of Disconfirmation

This section states how the paper argues, over what material, and what would show it to be wrong. Its objective is to make the standing of each later result inspectable: a reader should be able to tell, for any claim in Sections 6 to 9, whether it rests on positive law, on the framework, or on both, and what finding would defeat it. The section describes the two routes and the reason for keeping them apart, records the material over which the doctrinal route runs, explains how route dependence is marked in the text, and sets out the conditions under which each principal claim would fail. Table [tab:routes] collects the last two together.

Doctrinal Analysis Alongside Derivation from the Framework

The paper argues along two routes and keeps them separate to the end.

The doctrinal route reads statutes, regulations, and decided cases for what they allocate. It asks, of each provision, which party is given which power over an account of a person, and on what reasoning the allocation was made where reasoning is available. It treats a legislature’s considered choice as evidence of what a legal system has decided rather than as an authority on what is right, and it treats a court’s stated ground as something to be answered where the paper’s position runs against it. The route is descriptive in what it reports and evaluative only where it says that a stated ground succeeds or fails on its own terms.

The framework route derives consequences from the principles of relational reproduction stated in Section 5, applied to the object defined in Section 2. It asks what follows for authority over an account if a shared experience is jointly produced in the sense the framework gives that phrase. The route is deductive relative to its premises and inherits whatever standing those premises have.

Separation matters because the routes can disagree, and where they do the disagreement is the finding. A result carried by both routes is stronger than either taken alone, since a normative derivation and a body of positive law have converged from different starting points. A result carried by one route alone is weaker in a way that depends on which route carries it: a doctrinal result unsupported by the framework may reflect an allocation the framework would reject, and a framework result unsupported by doctrine may reflect a demand no legal system has seen reason to meet. Merging the routes would hide both possibilities, and would in particular allow a normative conclusion to borrow the authority of a statute that was enacted for other reasons.

Jurisdictions and Instruments Selected

The doctrinal route runs over the four systems named in Section 2 and over the instruments defined there. Within those systems it examines the provisions that allocate authority over an account of a person: copyright authorship and ownership; procedural requirements attaching to adverse official action; data-protection rights of access and rectification, together with the exemptions that withdraw them; statutory inspection rights over institutional records; the tort duties owed by those who supply references; and the moral rights that attach to a work.

Selection was governed by two criteria. A provision was included where it allocates authority over an account of a person between the parties defined in Section 2, and excluded where its subject is the quality of the account rather than the standing of the parties. The tort duties are included as a limiting case, since they allocate nothing to the person described and their inclusion is what permits the comparison drawn in Section 6. Provisions governing the wider handling of personal information, where they bear on accounts of persons only incidentally, were excluded.

Two consequences of this selection are stated here so that they are visible later. The paper reports what these systems allocate and generalises no further; a provision in an unexamined system that granted authority over evaluative content would qualify the structural claim of Section 6 without being inconsistent with anything reported here. And the reasoning attributed to a legislature or a court is the reasoning stated in the instrument or the judgment, since the paper has no access to the deliberations behind either.

Dependence of Each Result on Its Originating Route

Each principal claim is marked in the text with the route that produces it, and Section 8 collects the markings and reports where the routes converge, where a claim rests on one route, and where they conflict. Three conventions apply. A claim derived from the framework alone is stated as a consequence of the principles and is identified as such at the point it is made. A claim resting on positive law alone is stated with the provisions that carry it. A claim reached by both is stated once, with the convergence noted, and is relied upon more heavily in later sections than claims of the first two kinds.

Inherited results carry the same discipline. The four results taken as settled in Section 2 were established elsewhere, and the sections depending on each are named there rather than here, so that a reader rejecting an inherited result can see what falls with it.

P3.6cmP2.5cmY Claim & Originating route & Finding that would defeat the claim
Access granted and authority over content withheld, across the provisions examined & Doctrinal & A provision in one of the systems examined that permits the person described to alter, veto, or require agreement to the substance of an evaluative account
Priority of interpretive authority over the remaining rights & Framework & A stable allocation in which contestation, transmission, reproduction, or ownership is held by one party while interpretive authority rests with another, and the arrangement is coherent
Denial of shared authorship of a jointly produced past & Doctrinal & A reading of the joint-work conditions on which a participant contributing no expression may be a joint author
Sufficiency of the ground given for that denial & Both & A demonstration that the exposure the ground identifies is avoidable on a narrower test
Absence of a doctrinal name for the operative force of an instrument & Doctrinal & A doctrine attaching consequences to the force of such an instrument rather than to the object embodying it or the expression it contains
Absence of an integrity interest in the person described & Doctrinal & A doctrine conferring an integrity or attribution interest on the subject of a work who did not author it

Conditions of Disconfirmation

Table [tab:routes] states, for each principal claim, the finding that would defeat it. Four features of that list deserve comment.

The defeating findings are of two kinds, and the difference matters for how each claim should be read. Four of the six would be defeated by the discovery of a provision or a doctrine, and are therefore claims about a bounded survey rather than claims about what any legal system could contain. Their standing is accordingly limited in the way stated in Section 1: the survey locates no owner, which is weaker than establishing that none exists. The remaining two would be defeated by argument rather than by discovery. The priority of interpretive authority fails if the remaining rights can be coherently allocated apart from it, which is a question about the structure of the rights rather than about what any system provides. The sufficiency of the ground given for denying shared authorship fails if the exposure that ground identifies can be avoided on a narrower test, which is likewise a matter of argument.

The architectural claim is the most exposed, and it is exposed by design. It organises Section 6 and the whole paper depends on it for unity. Its defeat would leave the individual results standing and would cost the paper its structure, a consequence stated in Section 1 and repeated here so that a reader may hold the two apart.

The claim about the sufficiency of the stated ground is the only one on which the paper reaches a conclusion adverse to its own framework, and its route is marked as both because it requires a doctrinal reading of the ground and a normative assessment of what follows. Section 9 states the consequence for the framework should the ground hold.

Three kinds of claim are outside the paper’s scope and are therefore outside the table. The paper makes no estimate of how often the instruments examined are issued, how often they are contested, or what effects they produce, since it argues about allocation rather than about incidence. It makes no prediction about whether any allocation it identifies as available would be adopted. And it proposes no regime, so questions of administrative cost and institutional failure belong to the governance paper named in Section 2.

The Generative Relational Account of a Shared Experience

This section supplies the framework route with its premises. Its objective is to state, for the object defined in Section 2, what a shared experience is on the account used here, which principles of relational reproduction govern its rendering, and what follows from those principles for the allocation of authority. It proceeds by giving the sense in which the experience is jointly produced, stating the governing principles in a form applicable to this object, arguing the priority of interpretive authority over the four remaining rights, and deriving the criteria any regime of rights would have to satisfy. The method is derivation from stated premises; the premises belong to the framework and carry whatever standing it has, which is why every result derived from it is marked in Section 8 as resting on this route until positive law is shown to converge with it.

Co-Production of the Experience by Both Parties

Joint production means something specific here and something stronger than joint presence. Two parties who happen to occupy the same room produce nothing together. The claim is that in a working, instructional, or therapeutic relation each party is a condition of what the other does within it: the supervisor’s questions shape which problems the student attempts, and the student’s attempts shape which questions the supervisor asks next. What either party did in that relation is therefore a product of the relation rather than a display, within the relation, of something each brought to it complete.

Two consequences follow for an account of such an experience, and they run in opposite directions.

The first constrains what an account can be about. If the conduct observed was generated in the relation, an account of it describes a joint product. The issuer’s own participation is among the conditions of what is being reported, and an account written as though it reported an independent performance misdescribes its object even when every sentence in it is accurate. This is the inherited result recorded in Section 2, applied here to the account rather than to the judgment it supports.

The second constrains who is positioned to give one. The issuer occupies a position within the relation and observed it from there. The person described occupied the other position and observed it from there. Neither position is privileged, and neither exhausts what occurred. An account rendered from one position is a rendering from a position, and its authority is the authority of that position rather than the authority of a view from outside.

Together these yield the claim that generates the paper’s question. The experience an account of this kind reports had two producers, and the account has one author.

The Ethical Principles of the Framework Stated for Accounts of Shared Experience

The framework states principles governing the reproduction of shared experience. They are stated here in the form they take for the present object. The first five are ordered, each presupposing those above it; the sixth governs how the others change over time.

Principle 3 (Generativity). A reproduction of a shared experience should preserve the possibility of continued becoming for those it describes. What is required is that the field of what a person may still become remain open, rather than that any particular course continue.

Principle 4 (Co-Reproduction). A shared experience may not be interpreted, defined, or revised by one of its producers acting alone. Authorship of a shared past is shared.

Principle 5 (Non-Exploitation). No party to a relation may be instrumentalised or sacrificed for the benefit of another. This is a floor rather than an aspiration.

Principle 6 (Counter-Power). Every reproduction of a shared experience confers interpretive authority upon the party who makes it. A just reproduction therefore preserves the capacity to question, revise, and renegotiate what has been reproduced. Because the authority is conferred by the act of reproduction itself, the capacity that answers it must be generated by the same act rather than supplied later as an appeal.

Principle 7 (Subject Preservation). No relation achieves a good result by eliminating the other party as a subject. Agreement is not the goal, and unity purchased by the disappearance of one party is a failure of the relation rather than its completion.

Principle 8 (Historical Revisability). No interpretation should become historically irreversible, at the epistemic, institutional, or normative level. The standing to reopen belongs to the party bearing the cost of the interpretation. What is required extends beyond permitting revision to preserving the conditions that make revision possible, since a liberty to revise is empty where the materials needed to exercise it have been discarded.

The clause in Principle 6 requiring simultaneity, and the clause in Principle 8 requiring that conditions be preserved rather than merely permitted, are extensions to the principles as first stated, and both are load-bearing in Section 6.

One element of this apparatus is inherited and is identified as such. The proposition that value should return to those who generate it, and in particular that generators hold a claim to shape the conditions under which they generate, is Eglash’s (Eglash 2016). Of the limbs of his account, the one operative here is the second: the claim of generators to participate in setting the conditions of production. Applied to the present object it yields the observation that the person described has no part in determining who writes an account of a relation they helped produce, when it is written, in what form, or under what confidentiality. The limb concerning communities and their circulation of value is not taken, since the object here is an act between particular parties. The prohibition on instrumentalisation stated in Principle 5 is older than Eglash’s account and is not inherited from it. The transposition of his principle to this object belongs to the framework, and any error in the transposition rests there rather than with him.

Priority of Interpretive Authority over the Remaining Rights

The paper treats five rights as aspects of one object rather than as five subjects. The ground is that four of them are fixed in scope by the allocation of the fifth. The argument is stated here and its exposure was recorded in Table [tab:routes].

Consider what each of the four governs. A right to contest is a right to contest an interpretation, so its reach is settled by what counts as open to interpretation and by whom. Where interpretive authority rests exclusively with the issuer, a contestation right can reach the fidelity of the record and can go no further, which is precisely the shape the strongest existing instrument takes, as Section 3 reported. A right to govern transmission governs the movement of an interpretation, so a party holding it without interpretive authority may restrict where a characterisation travels while having no purchase on what it says. A right to govern reproduction governs the re-assertion of an interpretation and stands in the same position. Ownership follows the same allocation for a structural reason rather than a contingent one: copyright vests in the author, and the author of an account is whichever party rendered the shared experience into expression. The party exercising interpretive authority is the party in whom ownership vests.

The four are therefore derivative in scope. Each governs an interpretation, and none of them determines who may produce one. This explains a pattern that would otherwise look like coincidence. The provisions surveyed in Section 3 were enacted at different times, in different systems, for different purposes, and they converge on one allocation because each of them addresses a right whose scope was already fixed once interpretive authority had been located in the writer.

One qualification limits the claim and is stated because the evidence requires it. Rights of access and knowledge are separable from interpretive authority in a way the four derived rights are not. A student may inspect a recommendation without acquiring any authority over what it says, and a legislature may withdraw access without altering who may interpret. Access is a precondition for exercising the derived rights rather than one of them, and the priority claim extends to the four and not to access. This qualification is what allows the structural finding stated in Section 1 to be stated in the form it takes there: access is granted and authority over content is withheld, and the two are allocated independently because they are independent.

Criteria the Account Imposes on a Regime of Rights

The principles yield criteria rather than institutions. What follows states what any allocation would have to satisfy to be defensible on this account, and stops there; the design of an apparatus meeting them belongs to the governance paper named in Section 2.

An allocation satisfying Principle 4 locates interpretive authority over a shared experience in more than one party, or supplies some substitute for shared authorship where a single author is unavoidable. An allocation satisfying Principle 6 attaches a capacity to contest at the moment an instrument issues, so that the contestation is a feature of the instrument rather than a remedy sought afterwards from a different body. An allocation satisfying Principle 8 preserves the materials that later revision would require, which for the present object means the record of what was said, the identity of who said it, and any divergence between the participants that was known at the time. An allocation satisfying Principle 7 keeps the person described within the relation whose product is being reported, rather than treating them as its topic. An allocation satisfying Principle 3 leaves the field of what the person may still become open, which bears on the duration of an interpretation as much as on its content. And an allocation satisfying Principle 5 refrains from treating either party as a means to the receiving institution’s convenience.

These six criteria are applied in Section 6 to each of the five rights in turn, and Section 8 reports which of them positive law already satisfies.

Rights over a Shared Experience

This section examines the five rights and the structure they share. Its objective is to establish, for each right, what positive law allocates, what the framework requires, and where the two diverge. Each of the first five subsections follows one order: the practice as it stands, the doctrinal allocation with its owners conceded at the point each arises, the requirement derived from the principles of Section 5, and the conditions a remedy would have to meet. The sixth subsection reports the structure common to all five. The method combines the two routes of Section 4 and marks which carries each result, so that Section 8 can collect the markings.

Interpretation of a Shared Experience

Unilateral Interpretation in Current Practice

In the ordinary case the issuer decides alone what the shared experience amounts to. The issuer selects which episodes to report and which to omit, chooses the terms in which conduct is characterised, fixes the emphasis, and settles what the whole adds up to. The person described participates in none of these decisions. In the ordinary case they do not see the result, and in the common academic arrangement they have signed away the entitlement to see it.

The arrangement is not treated as irregular. It is the standard form of the instrument, and the parties who use it regard confidentiality as the condition of its usefulness rather than as a defect in it. The paper’s question is therefore not whether the practice departs from its own norms. It does not. The question is what authority the practice presupposes, and whether that authority is available to be presupposed.

Doctrinal Allocation of Interpretive Authority

Positive law allocates interpretive authority over a shared experience to whichever party renders it into an account, and does so consistently across the provisions surveyed.

Copyright makes the allocation explicit. Rights vest in the author of the work (United States Congress 1976c), and the author of an account is the party who composed it. Data-protection law makes the allocation visible in a different way. An evaluative assessment is the assessed person’s own data, subjective judgments included (Court of Justice of the European Union 2017), so the person described holds rights of access and rectification over it (European Parliament and Council 2016). Yet the evaluation itself remains the evaluator’s, and the rights operate upon the record rather than upon the judgment recorded. The law of references allocates nothing to the person described at all: the duty to take care in preparing a reference is discharged by taking care (House of Lords 1994), the requirement that a reference avoid a misleading overall impression is a requirement upon its author (Court of Appeal of England and Wales 1998), and the obligation to investigate before repeating an allegation likewise binds the party who would repeat it (Court of Appeal of England and Wales 2001).

An owner must be conceded before the framework’s claim is stated. That a jointly produced work is jointly authored, and that no single contributor holds authority over what the whole means, is Dworkin’s (Dworkin 1986). His construction and this paper’s object diverge at one point, stated in Section 3 and repeated here because the argument turns on it: his contributors each write, in sequence, and the constraint of fit governs a later writer facing an existing text. Here two parties produce an experience and one of them writes. The question is prior to Dworkin’s, since it concerns who may compose the first account rather than how a later one must respect what precedes.

Co-Reproduction and the Denial of Unilateral Interpretation

Principle 4 holds that a shared experience may not be interpreted, defined, or revised by one of its producers acting alone. Applied to the object of Section 2, the principle bears on the instrument at the moment of issue rather than upon its quality.

The reasoning runs as follows. The experience was produced by both parties in the sense given in Section 5: each was a condition of what the other did within the relation. An account of that experience is an interpretation of a joint product. Principle 4 withholds from either producer the authority to settle that interpretation alone. The ordinary instrument presupposes exactly that authority in the issuer, since its composition, its contents, and its conclusion are the issuer’s alone. The instrument is therefore inconsistent with Principle 4 as a matter of its structure rather than of its execution. A more careful issuer, a better-designed form, and a fairer characterisation leave the inconsistency where it was.

Two things must be said immediately, and the second is the harder.

The first concerns the strength of the result. It is a strong conclusion reached from a premise the framework supplies and positive law rejects, so its route is the framework route alone, and Table [tab:routes] records that Section 8 will report it as such. Nothing in the surveyed law supports it.

The second is that positive law rejects the premise for a stated reason. Copyright’s conditions for joint authorship exclude a participant who contributed no expression, and they were framed to exclude such participants deliberately (United States Congress 1976a; United States Court of Appeals for the Second Circuit 1991; United States Court of Appeals for the Ninth Circuit 2000). That denial and the ground given for it are examined in Section 6, at the subsection on ownership, and Section 9 states what Principle 4 must concede if the ground holds. The principle is stated here in the form the framework gives it, and it is not insulated from what follows.

Derivation of the Remaining Rights from Interpretive Authority

Section 5 argued that four of the five rights are fixed in scope by the allocation of interpretive authority. The doctrinal material assembled above lets the argument be checked against what the provisions actually do.

Where interpretive authority is exclusive to the issuer, a right to contest reaches the fidelity of the record and stops, which is the shape of the strongest existing rectification right and the subject of the next subsection. A right to govern transmission operates on the movement of the account and leaves its contents untouched, which is why the provisions governing transmission take the form of exemptions and inspection rules rather than of controls on content. A right to govern reproduction governs re-assertion of an interpretation the holder had no part in forming. And ownership vests in the party who rendered the experience into expression, which is the same party in whom interpretive authority was located to begin with.

The pattern is therefore not a coincidence among unrelated provisions. Once interpretive authority is located in the writer, the scope available to each of the other four rights is settled, whatever the enacting body intended. This is the paper’s reason for treating the five as one object, and it is why the subsections that follow report allocations rather than argue each right afresh.

Contestation of an Interpretation

Forums Available and Unavailable in Current Practice

A person who objects to an account of themselves has more recourse than the practice suggests, and less than the objection requires.

Recourse exists in three forms. Where the account contains a false statement of fact causing harm, the ordinary remedies for defamation and for negligent misstatement are available, and in the case of references a duty of care is owed to the subject (House of Lords 1994). Where the account is held as personal data, access and rectification may be sought and a complaint may be brought to a supervisory authority (European Parliament and Council 2016). Where the account sits in an institutional record covered by statute, a hearing may be sought to challenge the record as inaccurate or misleading, with an entitlement to insert a statement of disagreement where the challenge does not succeed (United States Department of Education, n.d.).

A correction is owed at this point to an earlier paper in this series, which recorded as an open gap that no forum exists in which the person judged could contest. That statement is too strong and is corrected here. Forums exist.

What the forums have in common is where they stop. Each reaches the accuracy of a statement of fact, the fidelity of a record, or the care taken in composing it. None reaches the evaluative judgment itself. And the two provisions bearing most directly on the instrument at issue withdraw even that much: the statutory hearing does not extend to a confidential recommendation over which inspection has been waived, and one jurisdiction has removed the confidential reference from the data-subject rights altogether, in the hands of the giver and of the recipient alike (Parliament of the United Kingdom 2018).

Rectification of a Record and the Limit Reached at Evaluative Content

The most developed contestation right available over an account of a person is also the clearest demonstration of the limit. Because an examiner’s comments upon a candidate are the candidate’s personal data (Court of Justice of the European Union 2017), and because access and rectification attach to personal data (European Parliament and Council 2016), the person evaluated may obtain the assessment and may seek its correction.

The correction available is of a particular kind. Rights of access and rectification are tied to the purposes for which data are processed, and the accuracy required of a record is that it faithfully represent what was expressed, by whom, and when. A record stating that an assessor concluded something is accurate if the assessor concluded it. Where the person evaluated disputes the conclusion rather than the record of it, the remedy is the appending of a statement, which leaves the assessment standing and adds a second voice beside it.

This is worth stating precisely because it is easy to misread in either direction. The right is real and it is more than the practice of the instruments examined here would suggest. It is also, exactly and by design, a right over the record and not over the interpretation. The person evaluated may establish what was said and may say something back. They may not participate in settling what the shared experience amounted to.

Confrontation of Adverse Evidence and the Reputational Threshold

Constitutional process supplies a model and withholds its application. An opportunity to defend by confronting adverse evidence is required where a state deprives a person of a statutory entitlement (Supreme Court of the United States 1970), and a liberty interest arises where official action imposes a stigma foreclosing other opportunities (Supreme Court of the United States 1972). The instruments examined here would appear to engage the second, since foreclosing opportunities is precisely what an adverse account does.

The appearance is defeated by the threshold. Reputation standing alone, unaccompanied by a more tangible interest, is neither liberty nor property (Supreme Court of the United States 1976). A private account that closes doors for a person without depriving them of any status they held falls below the line. No argument in this paper depends on constitutional process, and the doctrine is reported here for the shape of the model rather than for its reach.

Counter-Power and the Requirement of Simultaneity

Principle 6 requires that the capacity to question and revise a reproduction be generated by the same act that confers interpretive authority, rather than supplied afterwards through a separate proceeding. The requirement follows from the ground of the principle: the authority is conferred by the act of reproduction, so what answers it must arise from that act.

Applied here, the requirement identifies a defect that the forums described above share, and identifies it as a defect of timing rather than of scope. Each of those forums is a body a person may approach after an account has issued and after it has been acted upon. The account travels first. By the time a supervisory authority is engaged or a hearing convened, the reconfiguration the instrument effected in the receiving system has already occurred, and what the forum can restore is a record rather than a position. On Principle 6 an instrument that carries no capacity for contestation is defective when issued, and remains so however good the subsequent remedy.

Two consequences follow, and each is a framework result rather than a doctrinal one. A remedy located in a separate body at a later time satisfies the principle only if it operates before the instrument is relied upon. And an instrument satisfying the principle would carry its own contestation channel, so that the capacity is a feature of the instrument rather than a service sought elsewhere.

Institutional Conditions a Forum Would Require

Four conditions follow from the foregoing, stated as requirements rather than as a design, since the design belongs to the governance paper named in Section 2.

A forum answering Principle 6 operates at issuance rather than after reliance. A forum answering the limit identified above reaches the evaluative content and not the record alone, which requires some standard by which an interpretation may be assessed other than its fidelity to what was said. A forum answering Principle 8 preserves the materials revision would need, which here means the account, the identity of its author, and any divergence between the participants known at the time. And a forum answering Principle 7 admits the person described as a participant rather than as the subject of the proceeding.

The second condition is the demanding one. A standard for assessing an interpretation, on an account holding that no position yields a view from outside, cannot be correspondence to what really occurred. Section 10 returns to this and argues that the requirement is better stated as a condition on the mode of participation a forum affords than as a demand for a standard of correctness, which leaves a smaller difficulty in place rather than removing it.

Transmission of an Account

Circulation beyond the Originating Relation

An account composed within one relation travels. It reaches the party who requested it, and thereafter it may reach selection committees, personnel files, successor institutions, and parties who consult the file for purposes unrelated to the original request. Each stage of that movement takes the account further from the conditions in which the observation was made, and the conditions do not travel with it. What arrives at a later reader is a characterisation detached from the relation that produced it, and the later reader has no means of restoring the connection.

The person described has no part in the movement. They did not settle who received the account, for what purpose, or for how long it would be held, and in the ordinary case they learn of none of these things.

The mechanism by which this occurs is owed to Latour, on whose account an inscription acquires its force from being mobile and stable at once: it can be carried anywhere while remaining unaltered, and it can therefore be combined with other inscriptions at a centre far from where it was made, by parties who never witnessed what it records (Latour 1986). An account of a shared experience is an inscription in that sense, and its capacity to travel intact while the conditions of its making stay behind is the property Latour identified. What his account describes is how such a document acquires power in a network; what it does not address, and does not purport to, is who among the parties present at the making of the inscription was entitled to make it.

Evaluative Assessments Treated as Data of the Person Assessed

The provision that most nearly answers this holds that an evaluative assessment is the assessed person’s own data. An examination candidate’s answers together with the examiner’s comments upon them relate to the candidate and are the candidate’s personal data, subjective assessments included (Court of Justice of the European Union 2017), a construction of what it is for information to relate to a person that was established earlier (Article 29 Data Protection Working Party 2007). Rights of access, rectification, erasure, and objection follow (European Parliament and Council 2016), and where those rights apply the person described acquires purchase on the circulation of the account: they may learn that it exists, learn what it says, and object to processing.

The purchase is over movement rather than over content, which is the pattern Section 5 predicted for a right derived from an interpretive authority held elsewhere. A person may discover an account and may constrain where it goes. What it concluded remains what it concluded.

Statutory Exemption of the Confidential Reference

One legislature has considered whether these rights should extend to the instrument at issue and has determined that they should not. Personal data consisting of a reference given in confidence for the purposes of employment, education or training, appointment to an office, or the provision of a service are removed from the listed data-subject rights (Parliament of the United Kingdom 2018). The exemption reaches the reference in the hands of the party who gave it and in the hands of the party who received it, so that the route by which a subject might formerly have obtained a reference from a prospective employer is closed along with the route to the author.

This is the most direct answer positive law gives to the position developed in this paper, and it deserves to be met on its own ground rather than recorded as an obstacle. The reasoning behind such an exemption is that candour is the condition of usefulness: an issuer who expects the subject to read what is written will write something else, and the instrument that results will inform no one. The exemption therefore purchases the informativeness of the instrument at the price of the subject’s access to it.

Two observations bear on that trade, and neither disposes of it. The first is that the trade presupposes what Section 5 denies, namely that the instrument’s value consists in the information it conveys about an independently existing quality; if what it conveys is an interpretation of a joint product, candour secures the frankness of one participant’s reading rather than the accuracy of a report. The second is that the exemption is directed at access, and Section 5 argued that access is separable from interpretive authority. A legislature may therefore withdraw access, as this one has, without having decided anything about who may interpret. The exemption is an answer to a different question than the one this paper asks, and it is a considered answer to that different question.

Confidentiality and the Elimination of the Subject

Principle 7 holds that no relation achieves a good result by eliminating the other party as a subject. Applied to a confidential account of a shared experience, the principle yields a result stronger than a complaint about secrecy.

The relation whose product is being reported had two participants. The report of it is made to a third party, and the participant who is its subject is placed outside the transaction entirely: unable to see what is said of the relation they helped produce, unable to answer it, and unable to be heard by the party deciding on the strength of it. They are present in the transaction as its topic and absent from it as a party. On Principle 7 this is the characteristic form of the failure the principle names, and it is a feature of the standard instrument rather than of its misuse. The result rests on the framework route alone.

Scope Limits on Onward Transmission

An inherited result bears here. An observation made within a relation is valid for that relation, and extending it beyond is an error of scope rather than of accuracy. Transmission is the mechanism by which that error is committed at scale, since each onward movement carries the characterisation without the conditions that qualified it.

A remedy consistent with the principles would attach the conditions to the account and would let them travel with it, so that a later reader receives the observation together with the relation it was drawn from and the period it covered. This restates, for transmission, a proposal argued at length in an earlier paper of this series, and adds only that on Principle 8 the conditions have to be preserved rather than merely stated once, since a reader two institutions later can act on them only if they are still attached.

Reproduction of a Fixed Interpretation

Persistence and Restatement in an Institutional Record

That an institutional file assembles a person’s history into a form that supports the institution’s handling of them, and that the assembled record then governs how the person is subsequently treated within it, is Goffman’s (Goffman 1961). The observation is prior to anything argued here and the paper claims no part of it. Its limit for present purposes is that it concerns the operation of a record within one institution, whereas the instruments at issue are composed in one setting and act in another.

An account entering an institutional record acquires a durability the underlying relation lacks. The relation ends; the record persists. Later decisions consult it, later accounts are written in awareness of it, and a characterisation formed at one moment in one relation is re-asserted at moments and in relations that have nothing to do with it. Each re-assertion presents the interpretation afresh, and none of them re-examines it.

Moral Rights of Attribution and Integrity Vested in the Author

Where a legal system protects a work against distortion, it protects the author. The right to claim authorship and to object to modification prejudicial to honour or reputation is the author’s (World Intellectual Property Organization 1971); the narrow federal provision available in the United States confers attribution and integrity on the author of a work of visual art (United States Congress 1990); and the civil-law statement of the same two rights vests them in the author likewise (National Diet of Japan 1970). The interest protected is the interest of the party who made the work in the work remaining as they made it.

Absence of an Integrity Interest in the Person Described

The person a work describes holds no corresponding interest. Among the doctrines surveyed for this paper, none confers upon the subject of a work any right to object to its distortion, to require that it be attributed, or to resist its alteration.

The gap is easily mistaken for an oversight and is better understood as a consequence of how the adjacent doctrines are organised. Two bodies of law do protect a person against what is said of them, and each is keyed to something other than integrity. Defamation is keyed to falsity and to harm, so it reaches a statement of fact that is untrue and leaves an evaluation alone. Personality and publicity interests are keyed to appropriation, so they reach the commercial exploitation of a person’s identity and leave a non-commercial characterisation alone. Between the author’s integrity interest on one side and these two on the other lies the case of a person accurately and non-commercially characterised in a work they did not write, about a relation they co-produced. That case has no doctrinal home.

This is the clearest opening the survey conducted for this paper located, and its claim is correspondingly bounded: no located doctrine confers the interest, which Section 4 distinguished from a demonstration that none could.

Historical Revisability and the Preservation of the Conditions of Revision

Principle 8 holds that no interpretation should become historically irreversible, and requires beyond permission to revise the preservation of the materials revision would need. Applied to a persistent record, the principle bears on two things at once.

It bears first on the record’s duration. An interpretation formed within a relation that ended some years ago, re-asserted indefinitely, has become irreversible in the institutional sense the principle names, whatever formal liberty to challenge it may exist.

It bears second, and this is the extension doing the work, on what would be required to challenge it. A person seeking to reopen a characterisation needs the account, the identity of its author, the period it covered, and any divergence between the participants that was known at the time. Where the account is confidential, the first of these is withheld; where the reference is exempted from access, all four are. The liberty to challenge survives and the conditions of exercising it have been removed, which is the situation the extension was formulated to name.

Standing to reopen belongs, on Principle 8, to the party bearing the cost of the interpretation, which here is the person described rather than the institution holding the record.

Expiry and Reopening of a Recorded Interpretation

A remedy follows from the foregoing that is unusual in one respect worth stating. Proposals to make judgments of persons more accurate must be tested against outcomes, and the outcomes available are selected by the judgments themselves, since those excluded are never observed. A proposal that interpretations expire requires no such test. Its ground is that the conditions an observation was drawn from recede, and that a characterisation of a relation which ended long ago warrants less as time passes. That ground is available whether or not the original interpretation was sound.

Expiry is accordingly a remedy the epistemic situation permits where remedies aimed at correctness are unavailable. It is stated here as a consequence of Principle 8 rather than as a proposal for a regime, and it carries the framework route.

Ownership of an Account

Vesting of Copyright in the Issuer

Copyright in an account of a shared experience vests initially in its author (United States Congress 1976c), and the author is the party who composed it. Where the account is written by an employee in the course of employment the employer is treated as the author and holds the rights (United States Congress 1976c). In neither case does anything vest in the person described.

Ownership of the copyright is moreover distinct from ownership of the document, and transferring the document conveys no rights in the work embodied (United States Congress 1976d). The law thus divides the artefact into two layers and allocates each separately. Section 7 returns to what this division leaves out.

Exclusion of the Person Described from Joint Authorship

The person described is excluded from joint authorship three times over, and each exclusion is independent of the others.

A joint work requires two or more authors intending that their contributions merge into inseparable or interdependent parts of a unitary whole (United States Congress 1976a), and that intention must exist at the time the work is created (United States Court of Appeals for the Second Circuit 1991). The person described formed no such intention, and neither did the issuer. Authorship is additionally required, and a contribution that is valuable and copyrightable does not by itself establish it, control being usually the most important factor (United States Court of Appeals for the Ninth Circuit 2000). The person described exercised no control. And the contribution they made was to the events, not to the expression, which the statute places outside copyright altogether.

Principle 4 asserts that authorship of a shared past is shared. On the law just stated, it is not. The contradiction announced in Section 1 is located here.

The Ground Given for the Exclusion, and the Answer Owed to That Ground

The exclusion is reasoned rather than inadvertent, and the reasoning must be answered rather than noted. A looser test for joint authorship was rejected on the ground that it would expose authors to claims from research assistants, editors, and others with whom they had discussed their work, so that anyone who talked with another about what they were doing would be at risk of a claim to their work (United States Court of Appeals for the Ninth Circuit 2000).

The ground is a good one, and its strength lies in what joint authorship carries. A joint author is a co-owner with power to license the whole and to share in its proceeds. Admitting conversational contributors would create blocking positions over expression and would make discussing work in progress hazardous. Any position that would produce that consequence is defeated by this reasoning, and the framework’s principle, read as a claim to co-ownership, produces exactly it.

Two replies are available and they differ in strength.

The weaker reply distinguishes the remedy from the recognition. What generates the exposure is the exclusionary power attached to authorship rather than the acknowledgement of a contribution. A claim to participate in settling an interpretation, carrying no power to license, block, or take proceeds, does not create blocking positions and is therefore untouched by the reasoning. This reply preserves the framework’s substance at the cost of its wording, since Principle 4 speaks of authorship, and authorship in law is an ownership concept. Section 9 takes up what the principle must concede.

The stronger reply meets the ground on its own terms by identifying a narrower test that avoids the exposure. The parties the reasoning protects authors against are those who discussed the work. The person described here did something else: they co-produced the events the work is about. A test keyed to participation in the subject matter reported, rather than to contribution to the expression, does not reach an editor, a research assistant, or a former spouse who talked about a manuscript, since none of them co-produced what the manuscript describes. The exposure the ground identifies is therefore avoidable on a test narrower than the one rejected.

The strength of this reply is bounded and the bound should be stated plainly. It applies only to works whose subject matter is a real relation between the author and an identified other person, since only there does participation in the subject matter pick out a determinate party. For a novel, a history, or a scientific paper the test would identify nobody, which is a feature rather than a defect but also a considerable narrowing. The class of works to which the reply applies is small, and the instruments examined in this paper fall inside it.

Section 8 records this result as one reached by both routes, since it requires a doctrinal reading of the ground and an assessment of what follows, and Table [tab:routes] states the finding that would defeat it.

Unprotectability of the Underlying Events

What two people did together is outside copyright. Facts do not originate in an act of authorship, are therefore not original, and are not copyrightable (Supreme Court of the United States 1991); ideas, procedures, and concepts are excluded by the statute however they are expressed (United States Congress 1976b); and the exclusion has been applied to historical events and to interpretations advanced about them (United States Court of Appeals for the Second Circuit 1980).

Two consequences follow. The shared experience itself can be no one’s property, so a claim by the person described to own what happened has no purchase and is not made here. And the exclusion cuts both ways: the issuer’s account is protected in its expression and confers nothing over the events, so a second account of the same relation, written by the person described, infringes nothing.

Divergence of Ownership from Force in the Instrument Issued without Observation

The instrument that separates ownership from effect most sharply is the one in which the person described drafts the text and another party signs it. This arrangement occurs, and the parties who use it do not regard it as a forgery.

Copyright in such a document vests in whoever composed the expression, which is the person described. The operative effect of the document belongs to the party who signed it, since the receiving institution acts upon the signature. Two distinct entitlements over one page therefore rest in two different hands, and the one that does the work is the one copyright does not name. The statutory division between the work and the object embodying it (United States Congress 1976d) separates two layers and leaves this third unaccounted for. Section 7 takes up what follows.

Structure Common to the Rights Examined

Grant of Access Alongside Withholding of Authority over Content

The five subsections above report one allocation six times.

An evaluative assessment is the assessed person’s data, and the assessment stands (Court of Justice of the European Union 2017; European Parliament and Council 2016). A student may inspect a recommendation by default, and the ordinary execution of the waiver removes the entitlement (United States Department of Education, n.d.). A legislature has withdrawn access to the confidential reference on both sides of the transaction (Parliament of the United Kingdom 2018). Integrity in a work belongs to its author (World Intellectual Property Organization 1971; United States Congress 1990; National Diet of Japan 1970). Copyright vests in whoever composed the expression (United States Congress 1976c). And reputation standing alone falls below the threshold at which process attaches (Supreme Court of the United States 1976).

That the interests grouped under privacy are better handled as a set of distinct activities and harms than as one concept, and that among them are the aggregation of information, its distortion, and the exclusion of a person from knowledge of records held about them, is Solove’s (Solove 2006). That taxonomy names several of the harms the provisions above bear upon, and the present section adds no category to it. The finding here concerns something the taxonomy does not classify, since it reports not a harm but the shape of an allocation: what the provisions grant and what they consistently withhold.

These provisions were enacted in different systems, at different times, and for different purposes. They converge because each addresses a right whose available scope was settled once interpretive authority had been located in the party who writes, for the reason argued in Section 5. Where the provisions differ is in how much access they grant, and access varies from a default inspection right to a complete exemption. Where they agree is that none of them gives the person described any part in settling what the shared experience amounted to.

Bearers of the Correlative Duties

A claim is incomplete until the party owing the correlative duty is named (Hohfeld 1913, 1917). Applying that discipline to the rights examined identifies the duty-bearers positive law has settled upon.

The issuer owes duties, and they run to the quality of what is written (House of Lords 1994; Court of Appeal of England and Wales 1998, 2001). The party holding the account as data owes duties of access, rectification, and lawful processing (European Parliament and Council 2016), owed to the person described. The institution holding a covered record owes duties of inspection and hearing (United States Department of Education, n.d.). In each case the duty-bearer is identified and the content of the duty is determinate.

Rights Without an Identified Duty-Bearer

Two of the claims argued above have no duty-bearer, and the absence is instructive rather than fatal.

A claim to participate in settling an interpretation would be owed by the issuer, who is identifiable, and its content would be a duty to seek and consider the other participant’s rendering before issuing. No surveyed provision imposes it. A claim that an interpretation expire would be owed by whoever holds the record, who is likewise identifiable, and its content would be a duty to cease relying after a stated period. No surveyed provision imposes that either.

The claims are therefore addressable in the sense the analytical apparatus requires: each has a determinate bearer and a determinate content, and what is missing is a source rather than a subject. This distinguishes them from claims that fail because no one could be identified to answer them.

Institutional Placement of the Duties

Where such duties would sit follows from the timing requirement of Principle 6. A duty to seek the other participant’s rendering is discharged before issue and therefore belongs to the issuing occasion rather than to a later forum. A duty to preserve the materials of revision belongs to whoever holds the record, for as long as it is held. A duty to cease relying belongs to the party who would rely.

Placing the first of these on the issuing occasion is what distinguishes the account from an appeal right, and it is also what makes it burdensome, since it converts an act one party performs alone into one requiring an exchange. The weight of that burden, and whether an institution could carry it, belongs to the governance paper named in Section 2.

The Force of an Instrument and the Limits of Existing Doctrine

This section examines what an instrument of the kind defined in Section 2 does, as distinct from what it says and from the document in which it is written, and asks which body of doctrine attaches consequences to that third thing. Its objective is to establish that the operative effect of such an instrument falls outside the layers positive law identifies, and to draw the consequence for transactions in which instruments are exchanged. It proceeds by describing the separation the law does perform, identifying the layer it leaves unnamed together with the closest doctrinal counterexample, describing the practice of exchange, and stating why concepts of property fail to reach what is exchanged.

One dependency is repeated from Section 2 because the whole section rests upon it. The account of what such an instrument does in a receiving system, namely that it alters the relational conditions available to the person described rather than conveying information about an independently existing quality, is inherited from earlier work in this series. That account is contested by readings on which the instrument is informative through what its issuer stakes rather than through what it says, and the contest is pursued elsewhere. If the inherited account fails, this section fails with it.

Statutory Separation of the Material Object from the Work

The law already divides such an artefact into layers and allocates them separately. Ownership of a copyright is distinct from ownership of the material object in which the work is embodied, and transferring the object conveys of itself no rights in the work (United States Congress 1976d). The two are treated as entirely separate things: a person may hold the paper and hold nothing in what is written on it, and a person may hold the rights in what is written and have no claim to the paper.

The separation is deliberate and was legislated to reverse a presumption running the other way. Its existence establishes something useful for the argument here, which is that legal systems are capable of identifying distinct layers within a single document and assigning each to a different party. The question is therefore not whether such a division can be made, but whether the division the law has made captures what matters about this instrument.

Absence of a Doctrinal Name for the Force of an Instrument

Three things may be distinguished in an account of a shared experience. There is the document, which copyright treats as the material object. There is the expression, which copyright treats as the work. And there is what the document accomplishes when it arrives: the alteration it effects in what the person described may thereafter do within the receiving institution. Call this third thing the instrument’s force.

The force is what the parties transact for. A receiving institution requests an instrument in order to act upon it, and an issuer who declines to supply one withholds the alteration rather than a quantity of prose. Yet neither of the layers the law identifies is the force. Copyright in the expression is held whether or not the instrument is ever sent and whether or not anyone acts upon it. Property in the document is likewise indifferent to its effect. The force arrives and operates without any doctrine attaching consequences to it as such.

One body of law appears to contradict this and requires examination, since Table [tab:routes] names precisely such a doctrine as the finding that would defeat the claim. The law of negotiable instruments does attach consequences to what an instrument does rather than to the paper or the prose. A negotiable instrument is an unconditional promise or order to pay a fixed amount of money satisfying stated formal requirements, and the term is confined to a signed writing that orders or promises the payment of money (American Law Institute and Uniform Law Commission 1990). A holder in due course takes rights that the underlying dealings between earlier parties cannot disturb. Here the operative effect of a document is named, regulated, and made transferable.

The counterexample narrows the claim rather than defeating it, and the reason is the criterion of entry. What negotiable instruments law names is a force the law itself constitutes: the obligation to pay exists because the legal system creates it, and the doctrine regulates an entitlement of its own making. The force of an account of a shared experience is of a different kind. No legal system creates it, no obligation follows from the instrument, and the receiving institution is under no duty to act upon what it receives. The instrument operates because an institution treats it as a reason, which is a social fact rather than a legal one.

A second concession is owed before the claim is stated, and it concerns the description of the force rather than its legal treatment. That an object or an utterance acquires a status it does not possess intrinsically, that the status carries powers, obligations, and entitlements, and that it holds only so long as the relevant community continues to accept it, is Searle’s (Searle 1995). An instrument of the kind examined here is an institutional fact in exactly that sense: a page acquires a status within a selection procedure, and it holds that status because the participants treat it as holding it. The philosophical description of the force is therefore not this paper’s, and no claim is made to it. What Searle’s account supplies is a vocabulary in which the force can be described; what it does not supply, and does not attempt, is an account of which such statuses a legal system has equipped itself to name. The observation of this section concerns the second, and it is consistent with the first rather than an alternative to it.

A third concession is owed on the specific case rather than the general form. That an instrument may operate without describing anything, deriving its efficacy from the position of the party who issues it and from the delegation that position carries, is Bourdieu’s account of the instituting act (Bourdieu 1982). The certificate that consecrates does not report a quality and does not need to. That observation stands prior to anything said here, and the present paper adds no part of it. One distinction between the instituting act and the force at issue here is argued in separate work of this series and is only noted: the rite Bourdieu describes assigns a determinate status, whereas what an instrument of selection alters is the set of conditions available to a person, leaving open what is made of them. Nothing in this section turns on that distinction, which is why it is recorded rather than argued.

The claim of this section is therefore the narrower one, and it is stated in the narrower form. Positive law names the operative force of instruments whose force it has itself constituted, and supplies no name for the operative force of an instrument that works without legal constitution. The instruments examined here are of the second kind. Between the paper and the prose lies the thing that does the work, and it has no legal description.

Sale and Exchange of Instruments in Current Practice

Instruments of this kind are exchanged, and the forms the exchange takes follow from the separability just described.

The most documented form is the instrument whose text is composed by the person it describes and signed by another. The practice is common enough in professional settings to have drawn ethical commentary, which treats it as raising conflicts of interest, misrepresentation, and an erosion of trust between the signing party and the institution receiving the document (Klufas et al. 2025). What the arrangement demonstrates for present purposes is that the force of the instrument is available separately from the account it contains, since here one party supplies the account and another supplies the force.

Other forms follow the same logic. An instrument may be issued in return for an undertaking about future dealings rather than in return for an occasion to report an observation. An issuer may supply one as a favour to a third party rather than on the strength of anything observed. In each case the instrument retains its effect while the account within it has ceased to be the reason for issuing.

Where law does regulate commodified endorsement, it does so on a different footing and for a different reason. Rules governing endorsements and testimonials in advertising address the disclosure of material connections between endorser and advertiser (United States Federal Trade Commission 2023), and a later rule prohibits the buying, selling, creating, and disseminating of fake or false consumer reviews and testimonials (United States Federal Trade Commission 2024). The interest protected is the consumer’s, against deception in a market for goods. No comparable provision addresses the instruments examined here, and the reason is visible in the framing: those rules reach representations made to consumers about products, and an account of a person supplied to a selection committee is neither.

Failure of Property Concepts to Reach What Is Exchanged

When such an instrument is exchanged, the property concepts available describe everything except the transaction.

Copyright describes the expression, and the expression is not what the receiving party wanted. Where the person described drafted the text, copyright in it vests in them (United States Congress 1976c), so the party who supplied what was actually transacted for holds no copyright at all. Property in the document describes the paper, and transferring the paper conveys nothing in the work (United States Congress 1976d). The events reported are outside copyright entirely (Supreme Court of the United States 1991; United States Congress 1976b), so nothing there is available to transfer. What passed between the parties was the force, and the force is what no doctrine names.

Two consequences follow, and the second matters more.

The first is descriptive. An exchange of this kind cannot be characterised in property terms, so the vocabulary in which such transactions would ordinarily be analysed produces an account in which nothing of substance changed hands.

The second is remedial. Where a legal system wishes to address a practice, it ordinarily proceeds by attaching consequences to the thing transacted. Here the thing transacted has no legal description, which is why the responses available address adjacent matters instead: misrepresentation where a false statement can be identified, professional discipline where a rule of conduct applies, institutional integrity policies where an institution chooses to have them. Each of these reaches something near the practice. None reaches the practice itself. This is the sense in which the absence identified in Section 7 is more than a taxonomic observation.

Boundary Observed toward the Political-Economy Volume

Three questions arising here are reserved and the reservation is stated so that their absence is read as a boundary.

What accumulates to those who issue such instruments, how a capacity to issue functions as a holding, and what its uneven distribution produces, belong to the political-economy volume named in Section 2. The present section establishes that the force is separable and that it is exchanged, and stops before asking what the exchange yields to whom.

Whether the practices described in Section 7 should be prohibited, and by what apparatus, belongs to the governance paper. The argument here identifies why the ordinary vocabulary fails to reach them and does not propose a substitute.

Whether the force of such an instrument could be given a legal description, and what a legal system would need in order to hold an object of that kind, is a question this paper raises and leaves open. Section 10 states what would be required for an answer.

Convergence and Divergence of the Doctrinal and Framework Routes

This section collects the route markings made in Sections 6 and 7 and reports the standing of each result. Its objective is to let a reader who accepts one route and rejects the other determine what survives for them. It proceeds by listing the results carried by both routes, those carried by one, and the points at which the routes conflict, and it distinguishes apparent conflicts from the one that is real. Table [tab:standing] collects the listing. The method is inspection of results already established; nothing new is argued here.

Results Reached by Both Routes

Three results are reached from the framework and are matched by something in positive law.

The separability of access from interpretive authority is argued in Section 5 as a structural point about the four derived rights, and is exhibited by two provisions that vary access while leaving interpretive authority untouched: a default inspection right subject to waiver (United States Department of Education, n.d.), and a statutory withdrawal of access in the same class of instrument (Parliament of the United Kingdom 2018). A legal system may move access in either direction without settling anything about who may interpret, which is what the framework argument predicts.

The preservation of the conditions of revision, required by Principle 8 as an extension beyond a bare liberty to revise, is partly met by the accuracy requirement attaching to records of opinion, which demands that a record represent what was expressed, by whom, and when (European Parliament and Council 2016; Court of Justice of the European Union 2017). Identity and timing are two of the four materials Section 6 identified as necessary for reopening, and a legal system has required their preservation for reasons of its own.

The sufficiency of the ground given for excluding the person described from joint authorship is assessed in Section 6 along both routes, since the assessment requires a doctrinal reading of the ground and a normative judgment about what follows from it.

Results Reached by a Single Route

Four results rest on the framework alone and nothing in the surveyed law supports them. The ordinary instrument is inconsistent with Principle 4 as a matter of structure. Confidentiality places the person described outside the transaction as a party while retaining them as its topic, which Principle 7 names as a failure. An instrument carrying no capacity for contestation is defective when issued rather than when a later remedy proves inadequate, which follows from the simultaneity requirement in Principle 6. And the expiry of a recorded interpretation is available as a remedy where remedies aimed at correctness are not, which follows from Principle 8.

Four results rest on the doctrinal route alone and are not derived from the principles. Positive law grants access and withholds authority over content, across provisions enacted in different systems for different purposes. No located doctrine confers an integrity interest upon the person a work describes. Positive law names the operative force of instruments whose force it has itself constituted and supplies no name for the force of an instrument working without legal constitution. And the events an account describes lie outside copyright, with the consequence that a second account of the same relation by the person described infringes nothing.

The asymmetry between these two lists is itself informative. The framework results are all requirements the law has left unmet; the doctrinal results are all findings about the shape of what the law does. The framework contributes demands and the survey contributes structure, and the paper’s contribution lies in the fit between them rather than in either alone.

YP2.4cm Result & Supporting route
Separability of access from interpretive authority & Both
Partial preservation of the conditions of revision & Both
Sufficiency of the ground for excluding the person described & Both
Structural inconsistency of the ordinary instrument with co-reproduction & Framework
Elimination of the person described as a party under confidentiality & Framework
Defectiveness at issuance of an instrument carrying no contestation & Framework
Availability of expiry where correction is unavailable & Framework
Grant of access alongside withholding of authority over content & Doctrinal
Absence of an integrity interest in the person described & Doctrinal
Absence of a name for a force the law has not constituted & Doctrinal
Unprotectability of the underlying events & Doctrinal

Points at Which the Routes Contradict Each Other

Two apparent conflicts dissolve on examination and one is genuine.

The first apparent conflict is between the framework’s requirement that the person described have standing over an account of themselves and the statutory exemption withdrawing access to confidential references (Parliament of the United Kingdom 2018). Section 6 showed that the exemption addresses access, which Section 5 established is separable from interpretive authority. A legislature deciding that a subject should not read a reference has decided something the framework also has views about, but it has not decided who may interpret a shared experience. The conflict is between a legislative judgment about candour and a framework judgment about access, and it is a disagreement rather than a contradiction: each position is intelligible on the other’s premises.

The second apparent conflict is between the framework’s requirement of contestation and the constitutional threshold holding that reputation alone supports no process (Supreme Court of the United States 1976). This is not a conflict at all. The threshold governs when a state must supply process before acting, and the framework’s requirement addresses what a private instrument must carry when issued. The two speak to different actors about different occasions.

The genuine contradiction concerns authorship. Principle 4 holds that authorship of a shared past is shared and that no producer may settle its interpretation alone. Copyright holds that authorship of an account belongs to whoever composed it, that a participant contributing no expression is excluded, and that the exclusion is required rather than incidental (United States Congress 1976a; United States Court of Appeals for the Second Circuit 1991; United States Court of Appeals for the Ninth Circuit 2000). Both propositions concern the same question, are stated about the same object, and cannot both be maintained without qualification.

Three features make this contradiction different in kind from the two above. The doctrine is not silent on the framework’s question but has addressed it and answered in the opposite direction. The answer is supported by a stated ground rather than resting on the shape of a category. And the ground identifies a real cost, since a looser authorship test would expose anyone who discussed work in progress to a claim over it.

Section 6 offered two replies. The weaker preserves the framework’s substance by distinguishing recognition from the exclusionary remedy that generates the exposure, at the cost of the principle’s wording. The stronger identifies a test keyed to participation in the subject matter reported, which avoids the exposure entirely but applies only to works whose subject is a real relation between an author and an identified other person. Section 9 states what the framework must concede on each reply, and does so rather than treating the contradiction as an objection the framework has survived.

Implications for the Generative Relational Framework

This section states what the framework must give up in consequence of the findings of Sections 6 to 8. Its objective is to settle the contradiction identified in Section 8 by amending the framework rather than by defending it, and to record what the paper hands forward. It proceeds by stating the revision required to the principle of co-reproduction, the narrower revision required to the principle of historical revisability, and the results transmitted to later work. The method is application of the paper’s own findings to its own premises.

Revision Required to the Principle of Co-Reproduction

Principle 4 holds that a shared experience may not be interpreted, defined, or revised by one of its producers acting alone, and adds that authorship of a shared past is shared. The second clause cannot stand in that form.

The reason is the one given in Section 6. Authorship in law is an ownership concept, and a joint author holds power to license the whole work, to block its exploitation, and to share in its proceeds. A principle asserting shared authorship of a shared past therefore asserts, when read as a claim about legal authorship, that every participant in a relation acquires an exclusionary interest in every account written of it. That consequence is what the exclusion of the person described from joint authorship was framed to prevent, and the reasoning behind the exclusion is sound: a rule producing it would make discussing work in progress hazardous and would multiply blocking positions over expression. The framework should not want the consequence and should therefore give up the formulation that produces it.

The revision separates two claims the original formulation ran together.

The first is a claim to interpretive standing: that each producer of a shared experience holds standing to participate in settling what that experience amounted to. This is what the principle was for, it survives the reasoning above, and it creates no blocking positions, since standing to participate in an interpretation confers no power to license, to exclude, or to take proceeds.

The second is a claim to proprietary authorship: that each producer holds an ownership interest in accounts of the experience. This the framework should abandon. It was never required by anything the principle was formulated to protect, and it is the source of the whole difficulty.

Principle 4 is accordingly to be restated so that a shared experience may not be interpreted, defined, or revised by one producer acting alone, that each producer holds standing to participate in settling its interpretation, and that this standing carries no power to exclude others from using, licensing, or reproducing an account of it.

One further consequence follows and it bears on how the revised principle should be read. Section 6 offered a stronger reply, keyed to participation in the subject matter reported rather than to contribution to the expression, and that reply applies only to works whose subject is a real relation between the author and an identified other person. The principle accordingly admits two readings whose reach differs, and a statement of it should indicate which is intended. Read as a general ethical claim about shared experience it applies broadly and makes a demand that no legal system is presently equipped to meet. Read as a claim capable of legal expression it applies to a narrow class of works, of which the instruments examined here are members. Both readings are available and they are not interchangeable, so the register should be fixed wherever the principle is relied upon.

Revision Required to the Principle of Historical Revisability

Principle 8 requires that no interpretation become historically irreversible and that the conditions of revision be preserved. The findings of Sections 6 and 8 support the principle in part and require one qualification.

The support is recorded in Section 8: a legal system already requires that a record of an opinion preserve what was expressed, by whom, and when, so two of the four materials revision would need are preserved for independent reasons. The extension requiring preservation rather than mere permission is thereby met halfway by positive law, which is a stronger position than the principle occupied before this paper.

The qualification arises from Section 6. A forum able to revise an evaluative interpretation requires a standard against which the interpretation may be assessed, and on the account this framework gives, no position yields a view from outside against which correctness could be measured. Revision of an interpretation is therefore unlike revision of a statement of fact. A record may be corrected against what occurred; an interpretation cannot be corrected against what the experience really amounted to, because the account denies that there is such a thing standing apart from the positions occupied.

Principle 8 must accordingly distinguish two operations that its present formulation runs together. Correction, which restores a record to accuracy, remains available for matters of fact. Supersession, in which a later interpretation displaces an earlier one without either being measured against an independent standard, is what revisability can mean in the evaluative domain. The principle’s demand is then a demand for the availability of supersession, and its operative institutional form is the expiry argued for in Section 6 rather than a right of correction. Section 10 records what this leaves unresolved.

Results Transmitted to the Later Papers

Four results are handed forward, and one correction.

To the injustice paper: access and interpretive authority are separately allocated, and a legal system may grant the first while withholding the second. Remedies keyed to transparency therefore leave the underlying allocation untouched, which bears directly on any proposal to address unequal treatment by disclosure alone.

To the political-economy volume: the operative force of these instruments is separable from the account they contain, is exchanged, and has no description in property law. The volume may take the separability as established here and need not re-argue it, subject to the dependency recorded in Section 2.

To the governance paper: four institutional conditions are stated in Section 6, of which the timing condition is the demanding one, since a duty discharged at issuance converts an act one party performs alone into an exchange requiring two. The burden this imposes is the governance paper’s to assess.

To the framework generally: the two revisions stated in this section, and the observation underlying both. On each occasion where this paper’s principles met positive law, the law was found to have considered the question and answered it for reasons. A framework that states its principles without engaging those reasons will keep producing conclusions that look radical and are merely under-informed.

The correction concerns an earlier paper in this series, which recorded that no forum exists in which the person judged could contest an account of themselves. Section 6 shows that forums exist and that they stop at the evaluative content. The earlier claim should be read as corrected, and the narrower statement is the one this series should carry forward.

Limits of the Account

This section states what the paper has left unresolved and what would be needed to resolve it. Two questions were forwarded here from earlier sections and are answered first, in the sense that the paper states plainly that it cannot answer them and says what an answer would require. The remaining limits concern the scope of the survey, the dependencies of particular results, and the range of the object.

The first forwarded question concerns the standard by which an interpretation could be assessed. Section 6 put the difficulty in terms of a standard, and Section 9 proposed that revisability in the evaluative domain means supersession rather than correction. Two things must be said, and the first is that the difficulty was posed in the wrong terms.

An account of adjudication is available on which what distinguishes it is the mode of participation it affords the affected party, namely participation through the presentation of proofs and reasoned arguments, and on which the test of whether a task may be adjudicated is whether the meaning of that participation can be preserved when the task is assigned to it (Fuller and Winston 1978). On that account a forum does not require an external standard of correctness in order to adjudicate, and a party before it has no right to a particular outcome. What is required is that reasons offered be answerable by reasons. Section 6 treated the absence of a standard of correspondence as disqualifying, and that inference does not hold: an interpretation of a shared experience can be met by an interpretation supported by reasons, and the exchange retains its meaning without either account being shown to correspond to what really occurred.

The dispute also survives the limit that account places on adjudication. Tasks unsuited to it are those in which the elements are so interdependent that a determination at one point redistributes the others, so that proofs and arguments addressed to any one point lose their purchase. A disagreement between two participants about a single shared experience is not of that shape. It has two parties, one object, and reasons on each side that bear directly on the same thing.

What remains after this reframing is smaller than the problem stated in Section 6 and is still unresolved. An account of adjudication tells us when a forum can operate; it does not tell us what the forum should conclude, and the framework supplies no criterion for preferring one supported interpretation to another. Combined with the distinction drawn in Section 9, the available position is that such a forum tests whether the reasons offered for an interpretation survive the reasons offered against it, and that supersession follows where they do not. Whether that is enough to constitute a remedy, rather than a procedure, is not settled here.

The second forwarded question is whether the operative force of such an instrument could be given a legal description. The paper establishes that positive law supplies none and identifies why: the doctrines that name the force of an instrument name a force the legal system itself constituted. It does not establish that a force the law did not create is beyond description. Three conditions would have to be met by any attempt. The force would need to be individuable, so that it could be said when two instruments have the same effect and when an effect has occurred at all. It would need to be attributable to an act, so that a party could be identified as having produced it. And a remedy would need to be available that is not terminal, since a remedy consisting only in prohibition would remove the instrument rather than govern it. Whether these can be met is left open.

Four limits follow from how the paper was conducted.

The doctrinal route surveyed four legal systems and a bounded set of provisions within them. Every claim of the form that no doctrine does something is a claim about that survey. Section 4 distinguished the claims that would be defeated by discovery from those that would be defeated by argument, and the first kind carry this limit in full. A provision in an unexamined system granting the person described authority over evaluative content would qualify the structural finding without contradicting anything reported here.

Section 7 rests on an inherited account of what these instruments do in a receiving system, and that account is contested by readings on which the instrument is informative through what its issuer stakes. The contest is pursued in separate work. A reader who rejects the inherited account should treat Section 7 as suspended, and should note that Section 6 and Section 9 stand independently of it.

The architectural claim that four rights are fixed in scope by the allocation of the fifth organises the paper and is defeasible by argument rather than by discovery. Its defeat would leave the individual findings standing and would cost the paper its unity, a consequence stated at the outset and repeated here so that the two may be assessed separately.

The definition adopted in Section 2 makes shared experience constitutive of the object, which places outside the paper a class of cases a reader may care about most. An assessment composed from records by a party who never encountered the person raises questions of accuracy, standing, and contestability, and it raises none of them in the form treated here, because nothing was jointly produced. The paper’s findings about interpretive authority therefore do not transfer to that class, and any attempt to transfer them would have to establish first that something answering to joint production is present.

Three further limits concern what the paper does not attempt. It makes no estimate of how often the instruments examined are issued, contested, or acted upon, so no claim here rests on the practical significance of the practice. It proposes no regime, so the cost of the institutional conditions stated in Section 6 is unassessed, and the timing condition in particular converts an act one party performs alone into an exchange requiring two, which is a substantial burden that this paper does not price. And the premises of the framework route are stated in Section 5 rather than defended there, so results resting on that route inherit whatever standing those premises have and no more.

Conclusion

An account of a shared experience is written by one of the two people who produced it and is used by a third who was not present. This paper asked what rights attach to such an account and to whom, treating interpretation, contestation, transmission, reproduction, and ownership as five aspects of one object rather than as five subjects.

The survey of positive law returned one allocation six times. Evaluative assessments are the assessed person’s own data and the assessment stands. A student may inspect a recommendation by default and the ordinary execution of the waiver removes the entitlement. One legislature has withdrawn access to the confidential reference on both sides of the transaction. Integrity in a work belongs to its author. Copyright vests in whoever composed the expression. And reputation standing alone falls below the threshold at which process attaches. These provisions were enacted in different systems, at different times, for different purposes, and they converge because each addresses a right whose available scope was settled once interpretive authority had been located in the party who writes.

Against that allocation the framework yields four requirements the law has left unmet, and the paper reports them as requirements rather than as findings. Where the two routes met directly, on the question whether authorship of a shared past is shared, the law was found to have considered the question and answered in the opposite direction for a stated reason that the paper accepts. The framework’s principle has accordingly been amended rather than defended, by separating a claim to participate in settling an interpretation, which survives, from a claim to own accounts of the experience, which does not.

Two questions were raised and left open: what standard a forum could use to assess an interpretation when no position yields a view from outside, and whether the operative force of such an instrument, which the law did not create and does not name, admits of legal description at all. Both are stated in Section 10 with the conditions an answer would have to meet. The paper’s contribution lies in the fit between what the framework demands and what the survey found, and neither half would carry the argument alone.

American Law Institute and Uniform Law Commission. 1990. “Uniform Commercial Code § 3-104: Negotiable Instrument.”

Article 29 Data Protection Working Party. 2007. “Opinion 4/2007 on the Concept of Personal Data.” WP 136. European Commission.

Bourdieu, Pierre. 1982. “Les Rites Comme Actes d’institution.” Actes de La Recherche En Sciences Sociales 43 (June): 58–63.

Court of Appeal of England and Wales. 1998. “Bartholomew v London Borough of Hackney [1999] IRLR 246.”

———. 2001. “Cox v Sun Alliance Life Ltd [2001] IRLR 448.”

Court of Justice of the European Union. 2017. “Peter Nowak v Data Protection Commissioner, Case c-434/16.” ECLI:EU:C:2017:994, Judgment of the Court (Second Chamber).

Dworkin, Ronald. 1986. Law’s Empire. Cambridge, MA: The Belknap Press of Harvard University Press.

Eglash, Ron. 2016. “An Introduction to Generative Justice.” Revista Teknokultura 13 (2): 369–404. https://doi.org/10.5209/rev_TEKN.2016.v13.n2.52847.

European Parliament and Council. 2016. “Regulation (EU) 2016/679 (General Data Protection Regulation).” Official Journal of the European Union.

Foucault, Michel. 1977. Discipline and Punish: The Birth of the Prison. New York: Pantheon Books.

Fricker, Miranda. 2007. Epistemic Injustice: Power and the Ethics of Knowing. Oxford: Oxford University Press. https://doi.org/10.1093/acprof:oso/9780198237907.

Fuller, Lon L., and Kenneth I. Winston. 1978. “The Forms and Limits of Adjudication.” Harvard Law Review 92 (2): 353–409.

Gadamer, Hans-Georg. 1989. Truth and Method. 2nd revised. New York: Crossroad.

Goffman, Erving. 1961. Asylums: Essays on the Social Situation of Mental Patients and Other Inmates. Garden City, NY: Anchor Books, Doubleday.

Halbwachs, Maurice. 1992. On Collective Memory. Edited by Lewis A. Coser. Chicago: University of Chicago Press.

Hohfeld, Wesley Newcomb. 1913. “Some Fundamental Legal Conceptions as Applied in Judicial Reasoning.” The Yale Law Journal 23 (1): 16–59.

———. 1917. “Fundamental Legal Conceptions as Applied in Judicial Reasoning.” The Yale Law Journal 26 (8): 710–70. https://doi.org/10.2307/786270.

House of Lords. 1994. “Spring v Guardian Assurance Plc [1995] 2 AC 296.”

Klufas, T., A. Ajmani, S. Hanggodo, A. E. Zhou, S. B. Sloan, and J. M. Grant-Kels. 2025. “Can ‘You’ Write Me a Letter of Recommendation: Ethics of Self-Authored Letters of Recommendation.” Journal of the American Academy of Dermatology 93 (5): 1372–73. https://doi.org/10.1016/j.jaad.2024.12.006.

Latour, Bruno. 1986. “Visualisation and Cognition: Thinking with Eyes and Hands.” In Knowledge and Society: Studies in the Sociology of Culture Past and Present, edited by Henrika Kuklick, 6:1–40. JAI Press.

National Diet of Japan. 1970. “Copyright Act, Act No. 48 of 1970, Articles 19 and 20.”

Nissenbaum, Helen. 2004. “Privacy as Contextual Integrity.” Washington Law Review 79 (1): 119–57.

———. 2010. Privacy in Context: Technology, Policy, and the Integrity of Social Life. Stanford, CA: Stanford University Press.

Parliament of the United Kingdom. 2018. “Data Protection Act 2018, Schedule 2, Paragraph 24: References.”

Scott, James C. 1998. Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed. New Haven, CT: Yale University Press.

Searle, John R. 1995. The Construction of Social Reality. New York: The Free Press.

Solove, Daniel J. 2006. “A Taxonomy of Privacy.” University of Pennsylvania Law Review 154 (3): 477–564.

Supreme Court of the United States. 1970. “Goldberg v. Kelly, 397 u.s. 254.”

———. 1972. “Board of Regents of State Colleges v. Roth, 408 u.s. 564.”

———. 1976. “Paul v. Davis, 424 u.s. 693.”

———. 1991. “Feist Publications, Inc. V. Rural Telephone Service Co., 499 u.s. 340.”

United States Congress. 1976a. “Copyright Act, 17 u.s.c. § 101: Definition of a Joint Work.” United States Code, Title 17, Chapter 1.

———. 1976b. “Copyright Act, 17 u.s.c. § 102(b).” United States Code, Title 17.

———. 1976c. “Copyright Act, 17 u.s.c. § 201: Ownership of Copyright.” United States Code, Title 17, Chapter 2.

———. 1976d. “Copyright Act, 17 u.s.c. § 202: Ownership of Copyright as Distinct from Ownership of Material Object.” United States Code, Title 17, Chapter 2.

———. 1990. “Visual Artists Rights Act of 1990, 17 u.s.c. § 106A.”

———. n.d. “Fair Credit Reporting Act, 15 u.s.c. § 1681i: Procedure in Case of Disputed Accuracy.” United States Code, Title 15.

United States Court of Appeals for the Ninth Circuit. 2000. “Aalmuhammed v. Lee, 202 f.3d 1227.” 9th Cir.

United States Court of Appeals for the Second Circuit. 1980. “Hoehling v. Universal City Studios, Inc., 618 f.2d 972.”

———. 1991. “Childress v. Taylor, 945 f.2d 500.” 2d Cir.

United States Department of Education. n.d. “Family Educational Rights and Privacy Act Regulations, 34 c.f.r. § 99.12.” Code of Federal Regulations.

United States Federal Trade Commission. 2023. “Guides Concerning the Use of Endorsements and Testimonials in Advertising, 16 c.f.r. Part 255.”

———. 2024. “Trade Regulation Rule on the Use of Consumer Reviews and Testimonials, 16 c.f.r. Part 465.”

World Intellectual Property Organization. 1971. “Berne Convention for the Protection of Literary and Artistic Works, Article 6bis.”