Archives and society’s memory: who preserves testimony

Epigraph

Selection criteria must be non-discriminatory and clearly defined.

UNESCO, documentary heritage Recommendation, 17 November 2015,1.2; official English text. A normative requirement, not evidence about every archive’s actual condition.

What this article is about

We encounter a society's past through surviving traces: letters, minutes, photographs, recordings, and people's accounts. A trace is not the whole event, and a preserved document cannot answer every question by itself. Between an occurrence and our reading stand a record's creator, the person who retained it, its describer, and access conditions. This article examines archives through those transitions. Neither every missing document proves deliberate silence nor every official record supplies final truth. An invented music club's history will help us consider provenance, selection, gaps, digital copies, and a shared narrative's limits. Thucydides on testimony, UNESCO's documentary heritage recommendation, and a Library of Congress worksheet supply different orientations while retaining their own status. The practical task is to check one historical assertion's grounds and keep uncertainty visible. This original exercise is neither a professional archival appraisal standard nor a promise to recover all of the lost past.

I. Event, trace, and an account of it

A music club prepares a small anniversary exhibition. Participants bring the first concert's programme, photographs, a letter requesting premises, and a former organiser's recollections. The situation is invented. Each item conveys something about the past in a different way. The programme presents announced performances; a photograph retains a visible moment; the letter concerns an arrangement; the later account conveys a present recollection. Calling everything evidence for one conclusion before asking what each supports reverses the work. A question about the first concert might concern its date, actual performers, access conditions, or meaning for participants. These questions need different connections between materials and assertions, even when the exhibition puts all the objects together under one general title.

Distinguish the event from its trace immediately. A programme does not establish that every announced piece was performed. A photograph shows nothing outside its frame. A letter may precede an arrangement that was subsequently cancelled. These possibilities do not make the objects useless; they locate each conclusion's limit. The programme can evidence the announced intention, while a particular musician's attendance needs another source. The work need not end in one general label, “reliable.” A particular detail's accuracy differs from completeness of the picture. Even a precise record of one moment leaves unknown what it was never designed to convey and what its creator could not observe. That boundary is part of using the object accurately.

A historical narrative emerges when traces are connected into a sequence and explanation. The exhibition might say the club arose from several people's initiative, but that conclusion needs grounds separate from an attractive heading. Proposing a meeting, arranging premises, and later supplying a photograph are different roles. Calling everyone “founders” may make the account more definite than its documents permit. Clarification does not destroy shared memory. It preserves contributions without unsupported equivalence between roles. Sometimes placing several items together with an open question is more honest than producing a smooth sequence in which unexplained transitions disappear with inconvenient captions. A narrative can remain meaningful while showing where its author supplied an interpretation rather than recovered a directly recorded connection.

II. What was the document created to do?

An item's original purpose helps understanding. The organiser's letter requests a hall. It might exaggerate expected attendance to persuade its recipient. This is a possible case, not an established motive in an actual person's history. The request still supports the fact that its author made a particular proposal; it does not automatically establish how many guests arrived. Promotional programmes may emphasise attractive performances, while working notes distribute tasks. Ask each object a question suited to its origin. Otherwise a reader expects an answer to a task its creator never undertook and treats silence as a finished historical conclusion. An object need not have been created as historical evidence to provide useful evidence for a suitably limited question.

Authorship can be more complicated than a signature. One person composed a text, another typed it, a third signed for the club, and a fourth retained the copy. These activities need not conflict, but express different relations to the item. Knowing only the signatory does not warrant naming them as author of every word. Unknown authorship can remain unknown instead of assigning the gap to the best-known participant. A small exhibition can explain which provenance is established and on what grounds. An incomplete document biography does not prohibit displaying it under appropriate permission. It limits assertions for which identity, timing, or creation conditions matter and remain unconfirmed. A careful caption can retain the object without manufacturing a completed history for it.

Distinguish an original, copy, and later reproduction. A photograph of a letter enables reading and examination of some features, without access to every property of the object. A transcription may convey words but omit corrections. A programme reconstructed from recollections is another document, even when styled as an old poster. Use it while identifying who reconstructed its contents and when. The problem is silent substitution: a neat present-day sheet appears as a direct trace of the first evening. Preserving the distinction tells readers what they currently observe and how that object connects with the event. This explanation matters more than paper's resemblance to a familiar image of age. A convincing historical appearance cannot itself determine the status of the information displayed.

III. Witnesses do not transmit an event without remainder

Early in The History of the Peloponnesian War, Thucydides discusses speeches and event reports. Crawley's translation distinguishes difficulty remembering speeches word for word, the historian's rendering of their sense, and checking accounts. Differing memory and partiality among witnesses are identified. This is the historian's account of his work, not independent verification of its entire accuracy. We consulted the selected methodological passage, not the whole work or Greek text. Its limited point is that narrative speeches need not be verbatim records; a source can itself explain a transition from hearing to writing.

The club's former organiser remembers the first evening as full. They may sincerely report an impression without supplying an attendance count. A caption saying every seat was occupied expands that meaning. Preserve the statement as a personal recollection of significance and investigate attendance separately. Lack of a precise count does not invalidate the experience. Equally, one discrepancy cannot establish general memory unreliability or an intention to appropriate credit. Discuss the particular report, its subject, and a possible clarification. Someone can be invited to correct an account without defending their entire life before today's exhibition organisers. The correction concerns what a statement warrants now, rather than whether the person deserves a place in the shared history.

A later interview also has its own provenance. When was the question asked, how was it phrased, what recording was agreed, and which words belong to the interviewer? A question containing an existing version can elicit its repetition without independent confirmation. This is a logical possibility, not a diagnosis of all interviewing. Retain the question with the answer or describe conditions sufficiently to understand the report. Editing should distinguish an excerpt from uninterrupted speech. Reordering replies can create an impression the conversation did not provide. Society's history also develops through this editorial work. Naming it makes today's account's grounds accessible rather than abandoning its meaning. A useful edited account can remain useful while letting readers recognise the steps that shaped what they hear.

IV. Selection becomes part of memory

UNESCO's 2015 documentary heritage recommendation, section 1.2, addresses clearly defined, non-discriminatory selection criteria and contextual materials. This is normative guidance to states and memory institutions, not an inspection of a particular archive. It raises a question for our example: what grounds place an object in the exhibition's narrative? The subsequent reasoning is ours. A famous soloist's photograph may matter, but fame alone cannot establish the irrelevance of other materials. Selection affects the visible account, so its purpose should be intelligible. The requirement does not tell our invented organisers which particular image to choose or establish that any actual institution follows the recommendation.

Organisers might choose only early concerts, musical programmes, or records about the premises. Each can be a limited purpose if named. The problem is presenting one kind of selection as the club's whole history. Ceremonial photographs exclude ordinary work from view. That does not prove deliberate concealment: work may never have been photographed, material may be lost, or exhibition space limited. Titles and captions should fit the actual selection. Readers can understand its boundary, and organisers can explain today's question and what lies outside the accessible narrative. A limitation in display can then remain a limitation rather than becoming an unwarranted historical statement that the unshown contribution did not exist or had no significance for participants.

Selection and destruction differ. Leaving a letter out of an exhibition grants no permission to discard it; another question may need it. Convenience of a typed version likewise does not make the earlier sheet unnecessary. This distinguishes responsibility rather than prescribing retention schedules or professional appraisal. A small initiative can seek appropriate help when an object's condition needs specialist action. Do not turn a reading exercise into clearing out an unfamiliar collection. Understanding today's narrative requires identifying the selection's boundary, retaining its grounds, and avoiding the claim that personal impressions of uselessness establish an item's loss of all value. There can be many ways to organise a display without treating each item outside it as material already authorised for removal.

V. Absence has more than one explanation

The club's selection contains no records of early rehearsals. Perhaps rehearsals did not occur, no record was created, an item was lost, or it is held elsewhere. Other explanations remain possible. An accessible collection's gap does not select among them. Distinguish “we did not find it” from “it never existed.” Absence's meaning depends on expected coverage and how the search was conducted. A complete register recording every rehearsal creates a different question about an omission from a few chance pages. The same visible gap does not provide identical grounds in both situations. It becomes a question about the collection's provenance rather than automatic negative proof. The distinction prevents an incomplete survival pattern from being silently presented as a complete record of events.

Distinguish absence from a collection and absence from an accessible description. An item might survive under another subject, name, or date, or remain outside the displayed selection. These possibilities describe no actual archive's condition; they show the limited meaning of one unsuccessful search. Record where and how you looked and what answer you received. A diary of every movement is unnecessary. Retain conditions that let someone understand the negative result's boundary. If the search was too narrow, the next question may concern description rather than the event's historical absence. A further search can then address a specified limitation instead of beginning from an ungrounded conclusion or merely repeating the same attempt with greater confidence that nothing exists.

Archival silence can matter, but its significance needs grounds. Investigate who could leave records and which information was retained. Do not fill a gap with a morally appealing account and call it testimony. State a hypothesis and suggest a check. A final answer may remain unavailable; an honest narrative can preserve uncertainty and the question's importance. An absent programme name cannot exclude someone from all the club's work, and missing recollections cannot authorise inventing them. Fair representation requires precision especially where there is a wish to make a lost voice audible at any cost. The aim can remain important without upgrading a present interpretation into words or memories that the person never supplied and that surviving materials cannot establish.

VI. Description helps discovery and can guide reading

“Club founders” beneath a photograph tells readers whom they see. When its origin is unknown, the caption simultaneously creates unsupported grounds for a role. Separate the image from its later description: certain people are visible, a named person supplied the title, and involvement is known from another item. Identity may be secure while the date remains approximate. These degrees can coexist without forced uniformity. A good description need not look complete. It helps discovery while distinguishing what was observed, reported, and supposed. Our exhibition's neat labels should retain material uncertainty rather than hide it to make all captions visually identical. Clear presentation can show mixed degrees of knowledge without making either the item or the account impossible to understand.

The one-page Library of Congress Primary Source Analysis Tool distinguishes observation, reflection, questions, and further investigation. Its text was already read in this task; it is a methodological tool, not an efficacy report. We use it narrowly: what is visible, what do we infer, what remains a question, and what could investigate it? This does not turn a reader into an archivist. It helps keep a caption distinct from an image and confidence in interpretation distinct from established provenance. The musical example and its questions are our adaptation, not a worked example taken from the sheet. The comparison supplies a way to organise reading without proving that its use changes memory or understanding.

Descriptions also have histories. A participant later identifies a different person in the photograph, and the caption changes. Retain the earlier record as earlier, state correction grounds, and show today's version. Neither perpetuating an error for stability nor pretending no correction occurred serves accuracy. A small exhibition needs only an intelligible indication of the changed information. Disputed identification should be labelled too. Readers can distinguish new confirmation from a confident new guess. Maintaining description becomes part of preserving testimony: future readers need the current caption and its recoverable grounds. They need not reconstruct every conversation that surrounded it, but should be able to understand the material change and find the evidence on which the revised identification rests.

VII. Preservation and openness differ

A document can survive without being available to every reader. Its condition, storage, an owner's terms, or information about others may limit display. Establish the reason rather than infer it from closed access. UNESCO section 3.5 discusses restrictions alongside privacy, safety, and other legitimate grounds. This is a normative orientation, not a determination of rights to a letter. Archival accessibility does not grant permission to publish every found text. A private club letter can contain significant testimony and information irrelevant to the exhibition. These aspects require separate decisions under the actual working conditions. Neither an item's historical interest nor the fact that one person has seen it settles all later uses or the terms on which others may encounter it.

The letter's owner and people mentioned may have different positions about display. Organisers should not assume that handing over an object means agreement to every later use. Clarify the permitted extent of display and description. This is our transparency proposal, not legal advice or a universal consent form. Without established permission, retain the testimony question without publishing private details. Sometimes a narrative can identify an item's existence and a limited confirmed role; sometimes even that needs a decision. Missing conditions cannot be removed by an attractive appeal to saving memory. Future public interest does not automatically dissolve responsibility toward people now. A collection can acknowledge an important source while still needing to determine an appropriate way of using it.

Conversely, concern for privacy does not prove preservation prohibited. Suitable conditions need the people authorised to determine them, with professional questions referred appropriately. Retaining an object, research access, exhibition display, and internet publication are different actions with different consequences. Naming them can reveal a limited possible step instead of a choice between complete openness and complete disappearance. Agreement is not guaranteed; access may stay restricted. An honest description records the boundary of available knowledge without accusing a custodian or confirming contents the reader never saw. A report of restriction tells us something about present access, but it does not by itself establish either the truth of a document's assertions or a complete explanation of its history.

VIII. A digital copy preserves particular properties

A participant photographs the concert programme on a phone and supplies the file. Words become more accessible, but image quality may make a small date difficult to read. Recognition and transcription could replace one digit, changing the apparent sequence. This is a possible error in our invented case, not a measured recognition error rate. Compare a material detail with the image and distinguish photograph from typed text. An illegible original image should not produce certainty through a clean transcription; a caption can retain the unknown date. Digital tidiness helps work with material without establishing every restored detail. The question is what is actually recoverable from the image and what the transcriber supplied to complete what they thought it must say.

A file may also lose its intelligible relation to an object. It is named “old photo,” several copies differ in size, and its supplier is remembered only by an organiser. Provenance and copying information help future readers. What was photographed? Who provided it? Was an edge cropped? What was reconstructed in the caption? A small note suffices when it separates the observed from later change. A family or club collection need not become a technical system of unknown complexity. Retain what makes your assertion recoverable. If a date comes from a letter, that connection matters more than an elegant filename. If the letter is unavailable, preserve the fact that it supplied the date rather than attributing the information to a photograph that does not display it.

Image processing has limits of application. Rotating a readable copy differs from drawing a missing face or supplying an absent word by conjecture. The latter might be a useful reconstruction when identified as such and kept distinct from the source. Otherwise a future reader mistakes a present addition for a historical trace. We supply no restoration techniques and assess no particular alteration's permissibility. The distinction concerns preserving available information versus creating new information. Making a file persuasive does not establish its historical contents. Retain the source, identify material processing, and keep uncertainty even where software can make it look finished. A finished visual object can serve one purpose while remaining an inadequate basis for a claim about an unrecorded detail of the past.

IX. Several voices do not erase a shared subject

The first concert can mean different things to participants: creative friendship, difficult organisational work, or a rare opportunity to perform. These accounts need not contradict. They describe different aspects of experience and can appear together without selecting one correct feeling. Date, performers, and premises remain separate questions requiring suitable testimony. Multiple meanings do not give every factual answer equal support. Otherwise care for voices becomes refusal to distinguish experience from an event assertion, although participants may value precisely that distinction. A personal account can retain its meaning without being converted into a complete chronology. A shared exhibition can present both this meaningful account and a question about a detail that nobody is yet able to establish.

False balance is also possible. One report has a surviving record; another is a later conjecture. Display both with different labels rather than announce equal testimony for symmetry. A conjecture can raise an important question without becoming established fact. Nor should anyone be forced to supply private material because the selection lacks their side. Participation in shared memory requires no disclosure of everything experienced. Offer an appropriate format, accept refusal, and identify the narrative's boundary. Completeness cannot be obtained through coercion, because a compelled voice changes the conditions the present collection itself records. A missing perspective remains a limitation to acknowledge, rather than a permission to obtain words, personal objects, or public involvement on someone else's behalf.

A shared account remains possible when subject, grounds, and limits stay visible. An exhibition might give the established date of the first announcement, retain disagreement about the performed programme, and separately show the evening's personal meanings. This need not be elaborate. A few accurate captions can convey more than a confident unified legend. One unknown piece does not erase the club's history; an unconfirmed “founder” label does not erase someone's contribution. Make the account sufficiently definite for today's purpose without declaring the final version of the past. Future clarification then belongs to memory's organisation rather than threatening everyone whose material was displayed. Their earlier contribution can remain part of the collection even when a particular description changes with new evidence.

X. A custodian prepares a transition to future readers

In our example, custody concerns more than possessing the box. Someone captions photographs, another gathers permitted copies, and another shapes the exhibition's account. Their responsibilities concern different transitions: retaining the object, explaining provenance, identifying processing, limiting conclusions, and observing established use conditions. This is our description of an invented initiative, not a list of every archive's professional duties. It reveals shared memory as the work of many people. Nobody must perform every part, but material questions need a responsible participant. Otherwise an unfinished caption can seem minor until it becomes an assured history's only foundation. Identifying the responsibility gives the next person a place to seek clarification without assuming that everyone knows what has already been checked.

Thucydides addresses future use of his work; UNESCO offers institutional orientations for preservation and access; the worksheet organises a reader's inquiry. Their tasks differ. A historian is no contemporary archive, a recommendation does not establish every document's accuracy, and a reading table determines no distribution rights. Comparison allocates questions without automatically creating one ancient and modern standard. The next step depends on the object and authorised task. A corrected caption may suffice for an exhibition; a fragile original needs appropriate professional help; uncertain provenance may remain an open question with a recoverable search history. None of those results should borrow the status of another simply because they all concern caring for evidence of the past.

Society's memory retains the possibility of checking its narrative when evidence's provenance and boundaries remain visible. This guarantees neither future agreement with today's meaning of the concert nor recovery of every detail. We can pass on the material, its intelligible description history, and the difference between the established and proposed. The transition allows new questions without erasing previous work. Custodians possess neither the whole past nor authority over the only feeling about it. They help create conditions for encountering a particular trace and making a grounded reading. This limited responsibility connects a personal box, a small exhibition, and a larger archival question without treating them as identical institutions. It concerns what another reader can responsibly understand from what has been preserved and explained.

What this means in practice

Choose an accessible document you are entitled to work with and one historical assertion to examine. Sensitive correspondence and collection clearance are unnecessary. Identify the object as an original, image, transcription, later reconstruction, or account. Record known provenance and retain the unknown separately. Ask whether the material was created for your present question. A programme can establish that a piece was announced without establishing its performance.

Separate observation, interpretation, and questions. What do you actually see? What does the caption or owner report? Which inference is yours? Identify a material detail needing another source and an available check. An appropriate voluntary request to a participant should preserve refusal. Missing records establish no motive, and personal recollection need not become compulsory public testimony. This original exercise has no independently established efficacy and replaces neither archival appraisal, restoration, nor decisions about specific use rights.

Finish with a note or caption preserving status: “This material supports this assertion; this remains unknown; this description was added later; this is the further question.” Identify grounds for corrections. Treat retention, access, and publication separately within existing permission. Do not discard an original because a copy is convenient or fill uncertainty to produce a finished narrative. An open question is an acceptable outcome: future readers receive an honest boundary of present knowledge, rather than a history quietly completed on their behalf.

Key quotation

as a possession for all time.

Closing clause in Thucydides’s description of his work’s purpose, contrasting it with applause of the moment; Richard Crawley translation, selected Book I methodological passage. A stated purpose, not proof of complete preservation.

Connections between traditions

Thucydides / Library of Congress sheet: a transition from observation and reports to inference needs distinctions. The historian explains his account; the sheet organises present reading. Neither automatically confirms our exhibition text.
Thucydides / UNESCO: a future reader connects historical writing and heritage preservation. An author’s narrative purpose differs from institutional documentary guidance; shared concern does not establish identical procedures.
UNESCO1.2 / our club: selection clarity raises a question about a collection’s boundary. The exhibition is invented; normative criteria establish neither an actual collection’s condition nor permission to destroy undisplayed objects.
Worksheet / Thucydides’s speeches: observation, interpretation and questions help distinguish rendering from verbatim recording. This is our comparison, not a claim that the ancient author used the modern sheet or that it supplies a unique interpretation.
UNESCO3 / reading inquiry: access and the limits of available knowledge can be considered together. Guidance retains its domain; an inquiry establishes no private document publication rights. Preservation, access and narrative are different transitions.

Primary sources

About the authors

Andrey Zarinsh is the author and founder of United Wisdom. He explores inner transformation, the unity of traditions, intercultural dialogue, and contemporary spiritual discipline. Instagram: @andrey.zarinsh.

Maria Evilkut is a co-author of United Wisdom. Alongside her research, she seeks to test these ideas in practice and observe how principles of unity and mutual understanding work in real life between people and cultures. Instagram: @ilkutki.