Decay, Persistence, and the Liminal Archive

Methodological Implications of Archaeobytology

Digital Archaeology Reconsidered: Essay III of IV

Unearth Heritage Foundry

Abstract

Essays I and II established the theoretical foundations of archaeobytology: the ontological vacancy of digital archaeology and the processual nature of digital artifacts alongside the dual materiality (forensic and formal) as well as the state-based taxonomy (Vivibyte, Umbrabyte, Petribyte, Nullibyte). The present work translates theory into method. Drawing on principles from traditional archaeological stratigraphy and archival science alongside digital preservation practice, the text articulates the methodological protocols appropriate to archaeobytological inquiry. The essay develops the concept of "digital stratigraphy" as an analogue to physical stratigraphy, examines the distinctive decay mechanisms of digital artifacts, presents protocols for excavation and documentation, and addresses the special challenges of the "liminal archive" (the zone of partial accessibility where most digital heritage now resides). The result constitutes a practical methodology adequate to the theoretical commitments of archaeobytology.

Introduction: From Theory to Method

Theory without method represents speculation; method without theory is technique. The previous essays in this series have established the theoretical foundations of archaeobytology by defining ontological commitments and taxonomic categories as well as the understanding of what digital artifacts are. However, a discipline is constituted not only by what claims are made about the world but by what the discipline does in the world. The following text addresses the question of method: given a theoretical understanding of digital artifacts, how should they be investigated?

The challenge is significant. Archaeological method has developed over centuries in response to the properties of physical artifacts, encompassing spatial distribution and material decay alongside relationship to stratified deposits. Digital artifacts share some of these properties (location and decay as well as temporal layers) but differ in others (infinite replicability and sudden decay alongside ambiguous location). Traditional archaeological method cannot meet the need for wholesale importation; adaptation to the distinctive ontology of the digital is required.

The argument proceeds in four movements. First, the text examines the distinctive decay mechanisms of digital artifacts, distinguishing between substrate decay and ecosystem decay while analyzing implications for preservation. Second, the concept of "digital stratigraphy" is developed as an analogue to physical stratigraphy, adapting the principles of superposition and succession to the digital context. Third, protocols for archaeobytological excavation and documentation are presented to define the practical procedures for investigating digital sites. Fourth, the text addresses the special methodological challenges of the "liminal archive": the vast zone of partial accessibility where most digital heritage currently resides.

Part I: Mechanisms of Digital Decay

The Double Entropy

Essay II distinguished between forensic materiality (physical inscription) and formal materiality (symbolic structure). The forensic-formal divide generates a corresponding distinction in decay mechanisms. Digital artifacts suffer double entropy; decay at the substrate level (physical degradation) and decay at the ecosystem level (interpretive obsolescence).¹ The two processes are independent but cumulative; together they constitute the distinctive thanatology of the digital.

Substrate decay follows the second law of thermodynamics. Magnetic orientations weaken over time as thermal energy randomizes the alignment of domains. Optical media oxidizes as the reflective layer reacts with atmospheric oxygen. Flash memory cells lose charge through quantum tunneling after limited write cycles. The degradation processes are continuous and monotonic as well as ultimately irreversible. Environmental control (temperature and humidity) may slow the processes, and periodic copying (refreshment) may defer them, but stopping them is impossible.²

Rates of substrate decay vary dramatically by medium. Magnetic tape may last decades under ideal conditions but degrades rapidly in heat and humidity. Optical media ranges from a few years (cheap CD-Rs) to centuries (archival-grade gold DVDs). Solid-state storage is remarkably robust to environmental conditions but limited by write cycles. The archaeobytologist must attend to these material properties, for they determine the urgency of preservation efforts and the likelihood that historical artifacts have survived.

Ecosystem decay follows a different logic. The phenomenon is not physical but social, a function of changing standards and discontinued software alongside evolving practices. The bits of a file may be intact, yet if no software can interpret them, the files are lost. Software abandonment creates the "digital dark age" scenario: vast stores of data, intact but wholly illegible.³

Ecosystem decay is neither continuous nor monotonic. A format may be broadly supported for years, then abruptly abandoned when a major vendor changes direction. Flash content remained accessible for two decades, then became unplayable within months when browsers removed support. The transition from Vivibyte to Petribyte can be swift and catastrophic; a "cliff" rather than a slope. The discontinuity distinguishes digital decay from the gradual fading of physical artifacts and requires different preservation strategies.

Preservation Strategies

The dual nature of digital decay requires dual preservation strategies. For substrate decay, the primary strategy is refreshment: the periodic copying of data to new media before the old media degrades beyond readability.⁴ Refreshment must be performed on a schedule determined by the expected lifespan of the storage medium, with safety margins to account for variability. Best practice involves maintaining multiple copies on different media types in geographically dispersed locations, so that no single failure can cause total loss.

For ecosystem decay, two strategies compete: migration and emulation.⁵

Migration involves converting data from obsolete formats to current ones, preserving the information content while changing the representation. A Word 6.0 document is migrated to DOCX; a RealPlayer file is transcoded to MP4. Migration keeps data accessible by adapting the information to contemporary systems. However, migration is lossy. Each conversion may introduce errors, discard metadata, or fail to capture format-specific features. The migrated document is not the same document but a descendant, related by a chain of transformations whose fidelity may be questionable.

Emulation involves preserving not the data alone but the software environment that interprets the data. Rather than converting a HyperCard stack to a modern format, one runs a Macintosh System 7 emulator and opens the stack in its native environment. Emulation preserves the original artifact more accurately than migration. The same bits are interpreted by (functionally) the same software. Nevertheless, emulation is resource-intensive: the strategy requires preserving operating systems and applications alongside hardware specifications, and it requires ongoing maintenance as the emulator itself must be updated to run on contemporary systems. Emulation also raises philosophical questions regarding whether the emulated experience is the "same" as the original or whether the process functions as a simulation that obscures the authentic material context.⁶

The Open Archival Information System (OAIS) reference model, now enshrined as ISO 14721, provides a conceptual framework for digital preservation that accommodates both strategies.⁷ OAIS distinguishes between the Submission Information Package (SIP), received from the producer; the Archival Information Package (AIP), stored by the archive; and the Dissemination Information Package (DIP), delivered to users. The tripartite structure separates the concerns of ingest and storage alongside access, allowing different preservation strategies to be applied at different points in the information lifecycle.

Archaeobytology endorses no single preservation strategy as globally correct. The appropriate strategy depends on the nature of the artifact (simple text files may be migrated; complex interactive applications may require emulation) as well as the resources available and the needs of the designated community alongside the judgment of the practitioner. Archaeobytology does insist, however, that preservation strategies be documented: the chain of custody and the transformations applied alongside the decisions made must be recorded as part of the artifact's provenance.

Part II: Digital Stratigraphy

The Principle of Superposition

Archaeological stratigraphy is founded on the principle of superposition: in an undisturbed sequence, upper layers are younger than lower layers, because each layer must have been deposited on a pre-existing surface.⁸ Superposition enables relative dating: even without absolute dates, the archaeologist can establish that artifacts in layer A are older than artifacts in layer B if layer A lies below layer B. Stratigraphy transforms a chaotic jumble of artifacts into a legible sequence.

Can this principle be applied to digital contexts? Not directly. Digital artifacts do not "deposit" in physical layers; artifacts are created and copied as well as modified or deleted in ways that have no simple spatial analogue. A file created in 2020 may be stored on a hard drive manufactured in 2015, in a directory created in 2018, on a computer first booted in 2019. Where is the "layer" located? What is "below" what?

Yet the logic of stratigraphy, the use of temporal relationships to establish sequence, applies as much to digital as to physical contexts. Digital artifacts possess temporal properties: creation dates and modification dates as well as access dates. Websites have archived snapshots. Software has version numbers. Databases have transaction logs. The artifacts' temporal markers constitute a kind of digital stratigraphy: not a spatial layering but a temporal one, a sequence of states that can be partially reconstructed from surviving evidence.⁹

The Wayback Machine exemplifies digital stratigraphy. For any archived URL, the Machine presents a calendar interface showing when snapshots were captured. Each snapshot represents a "layer," a frozen moment in the site's history. The archaeobytologist can move through these layers, observing changes and dating content while establishing relative sequences. The snapshot from October 1999 is "older" than the snapshot from March 2001, as surely as a Minoan layer is older than a Classical one.

Digital stratigraphy differs from physical stratigraphy in crucial ways, however. Physical strata are complete; all the artifacts deposited in a layer are (in principle) recoverable from that layer. Digital "layers" are fragmentary; the Wayback Machine captures some pages but not others, misses dynamic content, and fails to preserve server-side functionality. The archived snapshot is not the full layer but a sample, and often an unrepresentative one. The archaeobytologist must read digital stratigraphy critically, attending to what is missing as much as to what is present.

Contexts and Cuts

Sophisticated archaeological stratigraphy distinguishes not only layers but contexts: "single events or actions that leave discrete, detectable traces in the archaeological sequence."¹⁰ A context might be a layer of deposited soil, but the event might also be a cut (a pit dug into the soil) or a fill (material deposited in a pit) as well as a structure (a wall built on a surface). Each context has relationships to other contexts because of overlying some, underlying others, and cutting through still others. The Harris Matrix systematizes these relationships into a diagram that represents the complete stratigraphic sequence of a site.¹¹

Digital archaeology requires an analogous attention to contexts. A webpage is not a simple artifact but a complex of contexts: the HTML source (one context), the CSS stylesheets (another), the JavaScript files (another), the embedded images (others still), the server-side scripts (often lost), the database queries (usually irrecoverable). Each context has its own temporal history, its own chain of modifications, its own relationships to other contexts. Understanding a webpage archaeobytologically requires disaggregating it into contexts and reconstructing their relationships.

The concept of cuts is particularly important. In physical archaeology, a cut is an event of removal: a pit dug, a foundation trench excavated. The cut is not itself a physical layer but the absence of layers that were removed. Cuts introduce complexity into stratigraphy because artifacts from lower layers may be redeposited in the fill of the cut, appearing in contexts later than their original deposition.

Digital archaeology has its own cuts: deletions, overwrites, platform shutdowns. The GeoCities extinction of 2009 was a massive cut, removing millions of pages from the live web and leaving only partial captures in the Wayback Machine. The Trump administration's removal of climate data from government websites was a cut. Every platform death and account deletion alongside every hard drive reformatting creates a cut that removes contexts from the accessible archive and potentially redeposits fragments elsewhere.

The archaeobytologist must attend to cuts as carefully as to deposits. What was removed? When? By whom? What traces remain? The cut itself is evidence, including evidence of editorial decisions and corporate strategies as well as political interventions and technological changes. Gaps in the archive are as meaningful as the contents.

Disturbance and Intrusion

Physical stratigraphy is complicated by disturbance: post-depositional processes that disrupt the original sequence. Animal burrowing, tree root growth, human excavation, and erosion can all move artifacts from their original contexts into later ones. An artifact found in layer B may actually belong to layer C if bioturbation has transported it upward. The archaeologist must distinguish in situ artifacts (found in their original position) from disturbed artifacts (moved after deposition).

Digital contexts are similarly subject to disturbance. A file's "creation date" may reflect not original creation but the moment of copying to a new drive. A webpage's "last modified" date may reflect a trivial template change rather than substantial content revision. Metadata can be spoofed and timestamps altered while provenance can be fabricated. The forensic archaeologist must approach digital metadata with skepticism, cross-checking dates against other evidence and remaining alert to signs of tampering.

Format conversion represents a particular form of digital disturbance. When a WordPerfect file is opened in Microsoft Word and saved, the creation date may change and the formatting may shift while metadata may be lost or altered. The converted file is not the original but a disturbed artifact, analogous to a potsherd moved by plowing from its original layer into a later one. The archaeobytologist must document conversion events and distinguish between original artifacts and their transformed descendants.

Part III: Protocols for Excavation

Site Identification

Traditional archaeology begins with site identification: locating places where human activity has left material traces. Some sites are visible (tells rising above the landscape), others are discovered through survey (systematic examination of terrain), still others are revealed by chance (construction work exposes buried remains).

Digital archaeology similarly requires site identification. Where are the digital "tells," the places where significant digital heritage has accumulated and may be recoverable? Several categories of site present themselves:

Personal archives: The computers, hard drives, and cloud accounts of individuals, especially individuals of historical significance. The papers of a literary author now include not only manuscripts and correspondence but also hard drives and email archives alongside social media accounts. The Salman Rushdie archive at Emory University includes disk images of his Macintosh computers, accessible through emulation.¹²

Institutional repositories: The servers and databases as well as digital collections of organizations (governments and corporations alongside nonprofits and universities). These institutional deposits often contain materials unavailable elsewhere and may be subject to preservation efforts or politically motivated deletion.

Web archives: The Wayback Machine and its counterparts (national libraries' web archives, specialized collections like the Archive Team's rescues). The archives function as secondary deposits, not the original sites but captured snapshots, yet they are often the only surviving traces of vanished content.

Platform residues: The artifacts that survive when platforms die. GeoCities shuttered in 2009, but the Archive Team rescued millions of pages; AOL Instant Messenger ended in 2017, but chat logs may persist on users' hard drives; Vine closed in 2017, but videos were archived by enthusiasts. Platform death does not necessarily mean total loss.

Forensic recoveries: Hard drives from estate sales and discarded computers alongside data recovery operations. The finds constitute the digital equivalents of metal detector finds. Such materials are serendipitous discoveries of potentially significant material whose provenance may be unclear.

The archaeobytologist surveys these sites methodically, assessing their content and condition as well as their significance. Not every site merits full excavation; triage is necessary. The Archive functions as reconnaissance, mapping the terrain of surviving digital heritage and identifying priorities for recovery and preservation.

Excavation Procedures

When a site warrants investigation, what are the protocols for excavation? Traditional archaeology has developed elaborate procedures for systematic excavation: gridding the site, removing layers in sequence, documenting finds in place, and preserving context through meticulous recording. Digital archaeology requires analogous procedures, adapted to the medium.

Imaging: Before any intervention, the archaeobytologist creates a forensic image of the digital site, a bit-perfect copy of the storage medium.¹³ As the physical archaeologist photographs and maps a site before excavation disturbs the area, the digital archaeologist preserves the original state before any analysis that might alter it. Forensic imaging uses write-blockers to prevent accidental modification and creates checksums to verify integrity. The image becomes the primary document; subsequent analysis operates on copies, preserving the original.

File system analysis: The forensic image reveals not only existing files but also the file system metadata: directory structures, timestamps, deleted files that have not been overwritten. Tools like Autopsy, FTK, and EnCase allow examination of file systems at multiple levels, from the logical (directories and files) to the physical (sectors and clusters).¹⁴ Deleted files can often be recovered if their data blocks have not been reused.

Format identification: Each file must be identified by format, not merely by extension (which can be misleading) but by signature analysis (examining the file's internal structure). PRONOM, the UK National Archives' format registry, provides unique identifiers for thousands of formats; tools like DROID automate the identification process.¹⁵ Format identification enables appropriate interpretation and flags files that may be at risk of obsolescence.

Contextual documentation: Beyond the files themselves, the archaeobytologist documents the context: directory structures and file relationships as well as metadata and any available information about the site's provenance and history. Contextual records are as important as the files themselves, for they enable future researchers to understand not only what was preserved but where the material originated and how the content related to its environment.

Emulation or migration: For files in obsolete formats, the archaeobytologist must decide whether to access them through emulation (running appropriate period software in an emulated environment) or migration (converting them to accessible formats). Both approaches are documented, with the original files preserved alongside any converted versions.

Reporting: The excavation concludes with a report documenting the site and the procedures employed alongside the artifacts recovered and any challenges encountered. The finalized report becomes part of the archaeological record, enabling future researchers to evaluate the excavation's methods and conclusions.

The specified protocols are not technical procedures; the steps embody epistemological commitments. The insistence on imaging before intervention reflects the principle that evidence must be preserved in its original state. The attention to context reflects the understanding that artifacts derive meaning from their relationships. The documentation requirements reflect the commitment to transparency and reproducibility that distinguishes scholarly investigation from mere data recovery.

Part IV: The Liminal Archive

Between Presence and Absence

Most theorizing about digital preservation imagines a binary: artifacts are either preserved or lost, accessible or inaccessible, Vivibyte or Nullibyte. However, the reality is messier. Much of the digital heritage we inherit exists in a liminal state, partially preserved, imperfectly accessible, neither fully present nor fully absent.

This zone is designated the Liminal Archive.¹⁶ The zone constitutes the realm of the Umbrabyte: the archived webpage whose CGI scripts no longer execute and the interactive CD-ROM whose content is preserved but whose interactivity is lost as well as the email thread whose attachments are broken and the database whose structure is intact but whose application layer is gone. The Liminal Archive is not nothing; it contains genuine evidence and real artifacts alongside recoverable information. Nevertheless, the fragment is not the thing itself; the survivor is a shadow and a trace as well as a partial remainder.

The Liminal Archive poses distinctive methodological challenges. How are artifacts interpreted when only partially functional? How is the distinction made between what is preserved and what is lost? How are the twin errors of overinterpreting the remainder (treating the fragment as the whole) and undervaluing that remainder (dismissing the partial as worthless) avoided?

Interpretive Protocols

Working in the Liminal Archive requires what might be called symptomatic reading: attention not only to what is present but to what its presence indicates about what is absent.¹⁷ The broken link is evidence of a connection that once existed. The failed script is evidence of an interaction that once occurred. The placeholder image is evidence of a visual that once rendered. The archaeobytologist reads these failures as symptoms, reconstructing from the pathology what the healthy artifact once was.

Comparative analysis requires symptomatic reading. If one archived snapshot of a site shows broken images, but an earlier snapshot shows those images intact, the archaeobytologist can recover the images from the earlier capture and reconstruct the later page as it was meant to appear. If the archived HTML references JavaScript functions that no longer execute, documentation of the original JavaScript (perhaps preserved elsewhere, or reconstructable from comments and function names) may allow the archaeobytologist to understand what the interactivity accomplished.

The Liminal Archive also requires what is termed generous reconstruction: the willingness to hypothesize beyond the evidence in cases where the evidence is fragmentary but the pattern is clear.¹⁸ If three webpages from a webring exist and webrings are known to have had dozens of members, it may be logically inferred that the missing members existed even though their content is irrecoverable. If a user's homepage from 1999 exists but not the guestbook, it may be logically inferred that a guestbook existed (standard practice at the time) and that visitors signed the book, even though the content is lost. Generous reconstruction fills the gaps with probabilistic inference, marked as inference rather than evidence.

Generous reconstruction has limits, however. The archaeobytologist must distinguish between what the evidence supports and what might be desired. The temptation to over-reconstruct, to fill in gaps with speculation that serves present agendas, is as dangerous in digital archaeology as in physical archaeology. The protocols of archaeological objectivity require that reconstruction be grounded in evidence, that speculation be flagged as such, and that alternative interpretations be considered.

The Archive and the Anvil

The dual methodology of archaeobytology is invoked here: the Archive (systematic excavation and preservation) and the Anvil (generative creation and intervention).¹⁹

In the Liminal Archive, the Archive function is primary: documenting what survives, preserving it against further decay, and making it accessible to researchers. The Archive does not judge the fragmentary artifact as worthless because of incompleteness; the Archive preserves the fragment precisely because fragments constitute the available record. The Archive is humble before the evidence, accepting what the past has left rather than demanding what might be desired.

The Anvil also has work to do, however. Where the evidence permits, the Anvil may attempt reconstruction, not falsification of the record but creative restoration based on documented evidence. The Internet Archive's Save Page Now feature is Anvil work: creating a record that did not exist before intervention. The Archive Team's rescue operations are Anvil work: preserving endangered content through active effort. The creation of emulation environments is Anvil work: building new tools to access old artifacts.

The Anvil may also create what are termed Digital Monuments; new artifacts that commemorate and interpret as well as respond to lost digital heritage.²⁰ The Digital Monument is not a fake, nor a pretense that the lost artifact has been recovered. The work serves as an acknowledged response, a contemporary creation that keeps the lost artifact present through intentional remembrance. The monument to a vanished homepage is not the homepage itself but a new artifact that documents what the homepage was and interprets its significance while acknowledging the loss.

The relationship between Archive and Anvil is dialectical. The Archive provides the evidence base that grounds the Anvil's creations; the Anvil provides the tools and interpretations that make the Archive's holdings meaningful. Neither alone is sufficient; together they constitute the practice of archaeobytology.

Part V: Toward a Field Protocol

The Archaeobytological Field Guide

The essay concludes by synthesizing the methodological principles developed herein into a practical field protocol for archaeobytological investigation.²¹

1. Site Assessment

Before excavation, assess the site:

• What type of site is it? (Personal archive, institutional repository, web archive, platform residue, forensic recovery)
• What is the estimated scope of the content?
• What state is the content likely to be in? (Vivibyte, Umbrabyte, Petribyte, Nullibyte)
• What resources are available for investigation?
• What is the significance of the site, and does it warrant full excavation?

2. Documentation Before Intervention

Document the site's condition before any intervention:

• Create forensic images of all storage media
• Record checksums for integrity verification
• Photograph physical media and equipment
• Document provenance: how did this material come to be available for investigation?

3. Inventory and Assessment

Create a systematic inventory of the site's contents:

• Identify all file formats present (using signature analysis, not extensions)
• Assess the functional state of each artifact (Vivibyte/Umbrabyte/Petribyte/Nullibyte)
• Flag artifacts at risk of imminent ecosystem decay
• Identify dependencies (what software is required to interpret each artifact?)

4. Stratigraphic Analysis

Analyze the temporal structure of the site:

• Establish relative chronology through timestamps and version numbers alongside archived snapshots
• Identify "cuts" (deletions, platform deaths) and assess their impact
• Document disturbances (format conversions, metadata alterations)
• Create a Harris-style matrix if the stratigraphic complexity warrants it

5. Excavation

Proceed with systematic excavation:

• Work from copies, preserving original images intact
• Extract and preserve all recoverable data, including deleted files where possible
• Document context: directory structures and file relationships as well as metadata
• Apply emulation or migration as needed for obsolete formats, documenting all transformations

6. Interpretation

Interpret the recovered artifacts:

• Analyze artifacts in context, not in isolation
• Apply symptomatic reading to partial artifacts in the Liminal Archive
• Employ generous reconstruction where evidence permits, flagging all inferences
• Consider alternative interpretations and acknowledge uncertainty

7. Reporting and Archiving

Conclude with comprehensive documentation:

• Produce a formal report documenting the investigation
• Archive all forensic images and recovered data alongside documentation
• Make materials available to future researchers through appropriate repositories
• Consider whether Anvil interventions (monuments, reconstructions) are warranted

The outlined framework is not a set of rigid rules but adaptable guidelines. Each site is unique; each investigation must be tailored to its circumstances. However, the principles underlying the protocols (documentation before intervention, attention to context, transparency in method, and humility before evidence) are non-negotiable. The protocols constitute the disciplinary standards that distinguish archaeobytology from data scavenging.

Conclusion: The Discipline Takes Form

The present text has translated the theoretical commitments of archaeobytology into methodological practice. The text has examined the dual decay mechanisms that afflict digital artifacts, developed the concept of digital stratigraphy, presented protocols for excavation and documentation, and addressed the special challenges of the Liminal Archive. The result is not a complete methodology, for no single essay could provide that, but a foundation on which more detailed procedural guidance can be built.

What emerges is a discipline that takes both its "archaeology" and its "digital" rigorously. From archaeology, the discipline inherits the commitment to systematic excavation and contextual documentation alongside stratigraphic analysis and interpretive caution. From digital studies, the field inherits attention to the distinctive properties of digital artifacts; their processual nature and dual materiality as well as their peculiar decay mechanisms, their substrate-independence, and their ecosystem-dependence.

The final essay in this series, "The Politics of Digital Stratigraphy," will turn from method to ethics and politics. Who controls the digital archive? Whose heritage is preserved and whose is lost? What power relations are encoded in the infrastructure of digital preservation? Methodology is never innocent; it embodies values and serves interests. Archaeobytology must be reflexive about its own political situation if the field is to be worthy of the heritage it claims to serve.

Notes

Works Cited

Acker, Amelia. "Emulation Practices for Software Preservation in Libraries, Archives, and Museums." Journal of the Association for Information Science and Technology 72, no. 8 (2021): 1016–1030.

Althusser, Louis, and Étienne Balibar. Reading Capital. Translated by Ben Brewster. London: Verso, 1970.

Carrier, Brian. File System Forensic Analysis. Upper Saddle River, NJ: Addison-Wesley, 2005.

Consultative Committee for Space Data Systems. Reference Model for an Open Archival Information System (OAIS). CCSDS 650.0-M-2. 2012.

Digital Preservation Coalition. Digital Preservation Handbook. 2nd ed. 2015. https://www.dpconline.org/handbook.

Granger, Stewart. "Emulation as a Digital Preservation Strategy." D-Lib Magazine 6, no. 10 (2000).

Harris, E. C. "The Laws of Archaeological Stratigraphy." World Archaeology 11, no. 1 (1979): 111–117.

Harris, E. C. Principles of Archaeological Stratigraphy. 2nd ed. London: Academic Press, 1989.

Hedstrom, Margaret. "Digital Preservation: A Time Bomb for Digital Libraries." Computers and the Humanities 31, no. 3 (1998): 189–202.

Kirschenbaum, Matthew G. Mechanisms: New Media and the Forensic Imagination. Cambridge, MA: MIT Press, 2008.

Kirschenbaum, Matthew G. "The .txtual Condition: Digital Humanities, Born-Digital Archives, and the Future Literary." Digital Humanities Quarterly 7, no. 1 (2013).

National Archives (UK). "PRONOM: The Technical Registry." https://www.nationalarchives.gov.uk/PRONOM/.

Rothenberg, Jeff. "Avoiding Technological Quicksand: Finding a Viable Technical Foundation for Digital Preservation." Council on Library and Information Resources, 1999.

Turner, Victor. The Ritual Process: Structure and Anti-Structure. Chicago: Aldine, 1969.

Unearth Heritage Foundry. The Unearth Lexicon of Digital Archaeology. 2025. https://unearth.wiki.

Digital Archaeology Reconsidered is a four-part essay series by Unearth Heritage Foundry examining the theoretical foundations of digital archaeology. Essay IV, "The Politics of Digital Stratigraphy," will appear next in this series.

Citation: Unearth Heritage Foundry, "Decay, Persistence, and the Liminal Archive: Methodological Implications of Archaeobytology," Digital Archaeology Reconsidered III (2025).

DOI: [To be assigned]