← all systems · see it in The System
The archive and the historical record
An archive is an accumulation of historical records or materials, in any medium, or the physical facility in which they are located (src:wp:archive): records generated as a by-product of legal, commercial, administrative and social activity, selected for long-term keeping, arranged by provenance, preserved and read. Walked as archivists and historians describe it: how the record is made, captured, appraised, kept, lost and read, up to the historical record a society can know its past through.
Properties published scienceour proposal
Sourced fields, and our own readings.
| born at | third millennium BC (clay-tablet archives: Ebla c. 2500-2250 BC); its administrative precursors (tokens, seals) and proto-cuneiform bookkeeping by the end of the fourth millennium BC |
| born precision | centuries |
| boundary | Inside: records once they are kept as records (captured, appraised, held in custody), the people and rules that keep them (archivists, records managers, finding aids, standards), the repositories (buildings, vaults, servers) and the reading done in them. Outside: the activities that make the records (government, firms, households), the materials they are written on, the forces that destroy them (fire, damp, wear, faults), and the histories, courts and sciences that read them. |
| nested level | Part of a society's institutions (Human society L5) and, for state archives, of government (L5); runs on writing (Writing system L7) and on people (Human L7); its L7 output, the historical record, is the L1 history reads. |
| scale | social |
Measures (34) published science
| Share of US federal records kept forever | 1-3 % |
| Share of US federal records 1789-1982 that was or would become permanent archives (1985 count) | 1.39 % |
| NARS holdings destroyed after accession, 1938-1983 | about 275,000 cubic feet (16% of records accessioned) |
| Time for 1.39% of federal records to fill the main National Archives building (1985) | under 12 years |
| Survival ratio of medieval chivalric and heroic WORKS (six vernaculars) | 68.3 (CI 63.2-73.5) % |
| Survival ratio of medieval chivalric and heroic DOCUMENTS (manuscripts) | 9.0 (CI 7.5-10.7) % |
| Document survival by vernacular | 4.9 (English) to 19.2 (Irish) % |
| Medieval narrative works estimated / surviving | 1,170 / 799 works |
| Medieval narrative manuscripts estimated / surviving | 40,614 / 3,648 documents |
| STM articles suffering reference rot | 1 in 5 (7 in 10 of those citing web resources) articles |
| Webpages from 2013 no longer accessible by October 2023 | 38 % |
| Webpages existing 2013-2023 no longer accessible (October 2023) | 25 % |
| Wikipedia pages with at least one dead reference link | 54 % |
| Half-life of objects in a digital library | nearly 23 years |
| Time for text-bearing pieces to go missing at average use (The National Archives, Kew) | ~55 (DP 300), ~160 (DP 400), ~450 (DP 500) years |
| Equivalence of RH and temperature in paper degradation rate | +20% RH ~ +4 degC ratio |
| Share of acidic paper in a typical Western library or archive | 70-85 % |
| Long-term planning horizon for collections | 500 years |
| Paper degradation experiments behind the damage function | 121 experiments |
| Text restoration accuracy: Ithaca alone; historians alone; historians with Ithaca | 62; 25; 72 % |
| Ithaca geographical attribution accuracy; dating error | 71%; under 30 years mixed |
| Cairo Geniza fragments | some 400,000 fragments |
| Individuals indexed by Goitein from the Geniza (950-1250) | about 35,000 people |
| Nineveh library tablets in the British Museum database; texts originally | 30,943; some 10,000 tablets; texts |
| Ebla archive | up to 1,800 complete tablets, 4,700 fragments |
| Military personnel files destroyed in the 1973 NPRC fire | about one-third of 52 million files |
| Herculaneum papyri carbonised in 79 AD | more than 1,800 scrolls |
| Web pages held by the Internet Archive (5 September 2024) | over 866 billion pages |
| Expected life of the Arctic World Archive film | 500-1000 years |
| Design life of archival disc media | 50 years |
| Daily wind records quantified from whaling logbooks, 1820-1900 | ~81,000 from ~200 logbooks records |
| Items collected by A Journal of the Plague Year by 30 November 2020 | more than 11,000 items |
| Wait from excavation to reading of the En-Gedi scroll | 1970 to 2016 years |
| Degradation of a mixed paper collection | two waves (acidic papers, then the rest) pattern |
Open contentions (7) open contention
Where sources or readings disagree. Both stay on the map.
- C-1822 Do archives grow naturally out of activity, or are they made by choices about what to keep?
booked: The map keeps both: records as by-products of work, and the archive as shaped by selection and power.
the other reading: organic: archives are natural and necessary by-products; the archivist a steward (Jenkinson 1922); made: archivists reshape, reinterpret and reinvent the archive; selection is power over memory (Schwartz and Cook 2002, via Schwartz 2006) - C-1841 How old is the burnt En-Gedi scroll?
booked: The map keeps the radiocarbon date and the handwriting date side by side.
the other reading: first or second century CE (palaeography, Yardeni) - C-1856 Do records move through a life cycle, or along a continuum?
booked: The map books both models, each where it fits.
the other reading: life cycle: creation to disposition in stages; continuum: 'a time/space construct not a life model'; elements pass into each other - C-1858 What share of medieval manuscripts survives?
booked: The map books about 9% of manuscripts and 68% of works for chivalric and heroic tales, and keeps other estimates; it is the same question as C-1857, booked from the archive side.
the other reading: 7% overall for general-purpose manuscripts, 20% for higher-end codices (book history estimates); earlier survival-study figures that met criticism (Kestemont et al. refs 16-17) - C-1867 What share of US government records is kept forever?
booked: The map books 1-3% today and keeps the older counts beside it.
the other reading: 1.39% for 1789-1982 (Dancis 1985); two to five percent (older estimate); two to three percent (federal archivists, 1985) - C-1869 Where did the archival rule of keeping records together by their creator begin?
booked: The map books the French circular of 1841 and keeps the earlier Naples claim beside it.
the other reading: Naples, three decades earlier (Duranti); adopted in Belgium and France about 1840 (src:wp:archival-science) - C-1870 How old are some disputed Athenian decrees?
booked: The map keeps both dates; an AI study's predictions side with the later one.
the other reading: pre-446/5 BC (three-bar sigma convention); 420s BC (recent re-dating)
Partners and parts published science
Built from / builds (7)
- Archival records
- Archivists
- Fonds (record groups by origin)
- Finding aids and description
- Records management
- Repository (the physical place)
- Access against preservation
Sits inside / holds (3)
Trades with (valence) (23)
- Photography via sys:written-record · The archive and the historical record gain
- Accounting (the field and practice) via sys:written-record · The archive and the historical record gain
- Human (Homo sapiens) via sys:written-record · The archive and the historical record mixed
- Fire via sys:thermal-energy · The archive and the historical record mixed
- Government via sys:written-record · The archive and the historical record gain
- Law (rules and law) via sys:written-record · The archive and the historical record gain
- Internet via sys:internet · The archive and the historical record gain
- Human society (from local community up to societies) via sys:written-record · The archive and the historical record gain
- Writing system (a script plus the rules binding it to a language) via sys:written-record · The archive and the historical record gain
- Book via sys:written-record · The archive and the historical record even
- Computer via sys:written-record · The archive and the historical record mixed
- Libraries and collections via sys:written-record · The archive and the historical record gain
- Human culture via sys:written-record · The archive and the historical record gain
- History (the field) via sys:written-record · The archive and the historical record gain
- Historiography via sys:written-record · The archive and the historical record gain
- History of science (the field and the story it tells) via sys:written-record · The archive and the historical record gain
- Art history (the field) via sys:written-record · The archive and the historical record even
- Museum via sys:written-record · The archive and the historical record gain
- The Holocaust (the Shoah) via sys:written-record · The archive and the historical record gain
- William Shakespeare via sys:written-record · The archive and the historical record gain
- Linguistics (the field: the practice and the languages it studies) via sys:written-record · The archive and the historical record gain
- Old English literature via sys:written-record · The archive and the historical record gain
- Middle English literature via sys:written-record · The archive and the historical record gain
expressed by (21)
- Templating / Standardization
- Niche Construction (alter environment)
- Interoperability/Transduction (Bridging)
- Encapsulation & Modularity (Abstraction)
- Amplification/Scale-Up (Manufacture & Distribution)
- Internal Coordination/Regulation
- Sensing/Scanning
- Search/Foraging
- Uncertainty Reduction/Learning
- Multi-Scale Data Fusion / Triangulation
- Trail-Marking / Externalized Memory (Stigmergy)
- Frontier Outposts / Sensing Extension
- Social Learning / Public Information Use
- Physical/Virtual Barrier
- Self-Monitoring
- Validation of Identity/Intention
- Resilience & Repair
- Redundancy & Diversity
- Triage & Prioritized Allocation
- Quality Control & Error Correction
- Bridging Supply Gaps (Store & Dormancy)
floor filled by (23)
- Photography
- Human (Homo sapiens)
- Government
- Law (rules and law)
- Paper (and the paper mills)
- Internet
- Writing system (a script plus the rules binding it to a language)
- Collective memory
- Computer
- Clay tablet
- History (the field)
- Archival records
- Archivists
- Fonds (record groups by origin)
- Finding aids and description
- Records management
- Repository (the physical place)
- Diplomatics
- History of science (the field and the story it tells)
- Art history (the field)
- William Shakespeare
- Linguistics (the field: the practice and the languages it studies)
- Old English literature
represents (2)
Kinds (11)
- Government archive (national archives)
- Business archive
- Academic archive
- Film archive
- Church archive
- Web archive
- Dark archive
- Community and counter-archive
- Genizah
- Personal archive
- Oral archive
wired (2)
Notes published science
- Fire is the paper archive's worst enemy and the clay archive's best friend: the blaze that destroyed Ebla's palace around 2250 BC fired its unbaked tablets 'almost as thoroughly as if in a kiln', and Nineveh's burning in 612 BCE baked Ashurbanipal's library, while an explosion in Dublin's Four Courts in 1922 wiped out Irish records reaching back to the 13th century (src:wp:ebla-tablets, src:wp:library-of-ashurbanipal, src:wp:public-record-office-of-ireland).
- The archive is mostly a record of what was thrown away: only 1-3% of what the US federal government writes is kept forever (a 1985 count gave 1.39%, so 98.61% destroyed), and ecologists' unseen-species maths says about 9% of medieval chivalric and heroic manuscripts survive, though about 68% of the stories do, because copies were spread over several books (src:nara-about, src:dancis-2020-one-percent, src:kestemont-2022-forgotten-books).
- Reading is wear: at the UK National Archives' average use, weak paper loses pieces carrying text in about 55 years, yet a collection consulted once a decade barely suffers; an archive literally trades access for life (src:strlic-2015-damage-function-3).
- History zoomed in, then back out: US history dissertations covered about 75 years in 1900, only about 30 by 1975, and 75-100 again in the 21st century, as a glutted job market pushed new PhDs toward ever narrower archives (src:guldi-armitage-2014-history-manifesto).
Layers
The ways it takes in its world, where they meet, and how it acts back, each from L1 (where it meets the world) up.
the archive and the historical record as archivists and historians describe it: from the activities that leave traces up to the past a society can read (reader's view)
- L1: The world of activity and its carriers: legal, commercial, administrative and social acts, written on clay, papyrus, parchment, paper, film, disks and the web
- L2: Document creation: the trace (a by-product, 'the secretions of an organism', not a message to posterity)
- L3: Capture as record: registration, linking, retention and disposition data; the creator's records management through the life cycle
- L4: Appraisal: the doorway every record passes; 1-3% of US federal records kept forever
- L5: Custody: arrangement by provenance and original order, description, preservation of paper and bits
- L6: Access and reading: authenticated records read critically by historians, genealogists, lawyers and scientists
- L7: The historical record as collective memory: what a society can know of its past, and the power over memory that comes with it
the boardroom (shared floors)
- L4: Appraisal and acquisition: the doorway all records pass; value judged by provenance, content, authenticity, condition and cost; 1-3% of US federal records kept forever
- L5: Arrangement, description and preservation in custody: provenance (respect des fonds, original order), description standards, storage environment, conservation, digital preservation (OAIS, fixity)
- L6: Access and reading: authenticated records read critically (diplomatics, source criticism), now also by imaging and machine restoration
- L7: The historical record as collective memory: records pluralised into social, cultural and historical memory, with power over memory and identity
way in: Community documentation of a crisis
- L1: People's own experience of an event (the COVID-19 pandemic)
- L2: Rapid-response collecting: more than 11,000 items by 30 November 2020
- L3: Robust metadata for each item
way in: Fire, war and disaster
- L1: Fires, explosions, eruptions, war
- L2: Paper and parchment burn; unbaked clay is fired as in a kiln
- L3: Records destroyed (a third of 52 million files, 1973) or carbonised (Herculaneum, 79 AD)
- L4: Fire-damaged records kept by a preservation office
way in: Handling by readers
- L1: Readers' hands on the documents
- L2: Wear and tear: tears, missing pieces
- L3: Text-bearing pieces go missing (~55 to ~450 years at average use)
- L4: Readers judge fitness by missing pieces
way in: Law, policy and standards
- L1: Records law and codes of practice (UK Freedom of Information Act 2000, s. 46)
- L2: Life-cycle rules: legal hold or destruction
- L3: Description standards (ISAD(G))
way in: Media, formats and faults
- L1: Storage media with a design life (archival disc: 50 years)
- L2: Format obsolescence
- L3: Faults from hardware, software, humans and organisations
- L4: Fixity: integrity checks on content
way in: Private papers and collections offered
- L1: Individuals' letters, papers, photographs, computer files, diaries
- L2: Collectors and donors offer them
- L3: Repositories funded by donors take in papers of people and places
way in: Readers and their questions
- L1: Historians, genealogists, lawyers and others come with questions
- L2: Search across collections through catalogues
- L3: Access by appointment to the reading room
- L4: Imaging and digitisation: virtual unwrapping, spectral imaging
- L5: Handling under conditions: supervised access or conservation first
way in: Relative humidity
- L1: Relative humidity of the store
- L2: Water taken up by the paper (absorption isotherms)
- L3: Degradation speeds: +20% RH as much as +4 degC
- L4: Condition read against the fitness threshold
way in: Reuse of the carrier
- L1: Durable parchment worth reusing
- L2: Scraped, washed and overwritten (a palimpsest)
- L3: Cut into binding waste, boxes, measures
- L4: Fragments not yet recovered because uncatalogued, as against destroyed
way in: Storage temperature
- L1: The temperature of the store
- L2: Cellulose degrades at a rate set by the Arrhenius equation
- L3: Strength falls; below DP 500 handling damage accumulates fast
- L4: Condition read against a fitness-for-use threshold: text evidently missing
way in: The creators' activity: transactions and administration
- L1: Legal, commercial, administrative and social activity in an office, firm, church or household
- L2: Document creation: the trace left by the activity
- L3: Capture as record: registration, linking, disposition data (the registry)
way in: The live web
- L1: Web pages and the links between them, moving or vanishing
- L2: Crawlers collect pages automatically
- L3: Snapshots stored (over 866 billion pages)
way out: Copies, replicas and new carriers
- L1: Film reels in a permafrost vault, expected to last 500-1000 years
- L3: Independent replicas and dark copies held elsewhere
- L5: Preservation planning across a digital object's life cycle (OAIS)
way out: Disposition: records destroyed
- L1: 98.61% of federal records destroyed (1789-1982)
- L2: Defensible destruction
- L4: Appraised as not of enduring value
way out: Evidence for rights, law and accountability
- L1: Courts, claimants and governments act on them
- L3: Records produced for compliance, legal hold, claims
- L7: A legal record for those concerned about their human rights
way out: History written from the record
- L1: Histories written from whatever archive exists (historians of the Nakba from the Israeli state's archives)
- L4: Stories told from what the archive kept: some privileged, some marginalised
- L7: Collective memory built, recalled and disseminated
way out: Old records reread as scientific data
- L1: Data handed to climate and solar science
- L3: Daily wind records quantified from whalers' logbooks
- L6: Legacy records re-examined and homogenised
way out: Records delivered to readers
- L1: The page in light before the reader's eyes
- L3: Digitised items and catalogues served for search
- L6: Records read in the reading room or online
Sources (46) published science
Archival science (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
Respect des fonds (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
Records management (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
Chatzistergos T (2026) Historical records in solar physics. Living Reviews in Solar Physics, doi:10.1007/s41116-026-00044-9 (behind its links)
About the National Archives. US National Archives and Records Administration (behind its links)
Chapekis A, Bestvater S, et al. (2024) When online content disappears. Pew Research Center, 17 May 2024 (behind its links)
Link rot (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
Cairo Geniza (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
Library of Ashurbanipal (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
Ebla tablets (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
National Personnel Records Center (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
Herculaneum papyri (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
Internet Archive (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
Baker M, Shah M, Rosenthal DSH, et al. (2005) A fresh look at the reliability of long-term digital storage. arXiv:cs/0508130 (EuroSys 2006) (behind its links)
Schwartz JM (2006) "Having new eyes": spaces of archives, landscapes of power. Archivaria 61:1-25 (quotes Schwartz and Cook 2002, Archival Science 2:1-19) (behind its links)
Archival appraisal (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
Damerow P (2006) The origins of writing as a problem of historical epistemology. Cuneiform Digital Library Journal 2006:1 (MPIWG preprint P114) (behind its links)
Williams PKG 2025. DASCH: Bringing 100+ Years of Photographic Data into the 21st Century and Beyond. arXiv:2501.12977 (behind its links)
History of accounting (Wikipedia, rev 1379621520, via Wikimedia Enterprise cache) (behind its links)
History (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01, revision 1382189844) (behind its links)
Moxham N, Fyfe A (2018) The Royal Society and the prehistory of peer review, 1665-1965. The Historical Journal 61(4):863-889 (open-access repository record, abstract read) (behind its links)
Provenance (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
Jaff DQ. 2025. CORHOH: text corpus of holocaust oral histories. Data in Brief 59:111426. doi:10.1016/j.dib.2025.111426 (PMC11927712) (behind its links)
Shakespeare authorship question (Wikipedia, via Wikimedia Enterprise) (behind its links)
Language documentation (encyclopedia article, WME snapshot) (behind its links)
Cotton library (Wikipedia, via Wikimedia Enterprise) (behind its links)
Sir Gawain and the Green Knight (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
Upward F (1996) Structuring the records continuum, part one: postcustodial principles and properties. Archives and Manuscripts 24(2):268-285 (behind its links)
Strlic M, Grossi CM, Dillon C, et al. (2015) Damage function for historic paper. Part I: Fitness for use. Heritage Science 3:33, doi:10.1186/s40494-015-0062-1 (behind its links)
Diplomatics (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)
Source criticism (Wikipedia, via Wikimedia Enterprise snapshot 2026-11-01) (behind its links)