The archive is not the village

Archives do not simply hold a past that waits to be retrieved. Records were made for particular purposes, selected by institutions, described through changing systems, and encountered through unequal conditions of access. This founding editorial sets out a method for distinguishing surviving evidence from the historical worlds an article seeks to explain. It proposes a public source ledger and a seven-question analytical rubric inspired by Marc Bloch.

A village does not enter an archive all at once. It arrives as tax assessments, court depositions, parish registers, maps, contracts, petitions, censuses, photographs, letters, and the later descriptions that make some of those objects findable. Each form answered a practical demand. Each admitted certain people, actions, and categories more readily than others.

That observation does not make the record unusable. It makes provenance part of the evidence. A poor-relief ledger may establish that an officer recorded a payment under a named category on a given date. It does not, by itself, establish the whole economy of need, the recipient’s account of the encounter, or the experiences of people who never appeared in the ledger. The distinction between the statement of the record and the reach of the inference is where analysis begins.

The journal therefore asks authors to identify the exact object used, its creator and administrative purpose, its path into custody, its present description, and the transformations—copying, transcription, translation, digitization, search, or excerpting—through which the author encountered it.

The first question is not only “What does this source say?” but “What had to happen for this source to exist here, in this form, for this question?”

— Founding analytical rule

Marc Bloch’s unfinished Apologie pour l’histoire organized the historian’s craft around observation, testimony, criticism, analysis, and causation.¹ An earlier address, “Critique historique et critique du témoignage,” likewise treated indirect knowledge and the discrimination of truth, error, and plausibility as central problems.²

The useful lesson is not that testimony is always false or that suspicion is sophistication. Criticism is comparative. It tests a statement against the circumstances of its production, independent traces, internal consistency, material features, and alternative accounts. A source may be inaccurate about an event and unusually revealing about the categories through which its maker perceived that event. The historical question determines which feature matters.

This is also why “primary source” is not a permanent honorific attached to an object. The American Historical Association’s professional standards emphasize that primary and secondary status depends in part on the question, and that primary evidence must be critically examined in relation to relevant secondary scholarship.³ A later catalog, edition, or oral recollection can play different evidentiary roles in different inquiries.

“The archive is silent” is not yet an explanation. A missing name may reflect non-participation, deliberate exclusion, lost volumes, inconsistent spelling, catalog structure, a search interface, an unconsulted related series, or the simple fact that the relevant action did not ordinarily produce a record. To argue from silence, an author must explain why a trace would normally be expected, how complete the surviving series is, and which alternative causes of absence have been tested.

Finding aids help establish the architecture of survival. The Archives nationales guide to historians’ private papers, for example, identifies the Marc Bloch fonds and the files containing manuscript and working material for the Apologie. It does not provide the manuscript itself, guarantee access, or authorize reproduction.⁴ A rigorous citation records that distinction instead of turning a catalog description into the object described.

Silence also has a politics. Institutions were uneven in whom they observed, believed, classified, compensated, punished, preserved, and described. Yet an account of power cannot be inferred from institutional bias alone. The author still owes the reader a demonstrated mechanism, comparison, chronology, and engagement with evidence that might resist the proposed interpretation.

The Liberty Articulator’s response is procedural. Every long-form article will carry a source ledger beside its bibliography. The ledger records the exact edition or archival unit used, the locator, the role the source plays, the state of verification, known mediation, and rights or access limits. It binds evidence to a particular article version so that later corrections do not erase the route by which an earlier conclusion was reached.

A second device, the Bloch Lens, asks seven public questions about the inquiry, surviving evidence, silence and mediation, comparison, inference, counterevidence, and revision conditions. Reviewers will not be asked merely whether an article “feels analytical.” They will assess whether each required decision is explicit and adequately supported.

None of this makes error impossible. It makes error easier to locate, contest, and correct. The journal’s scholarly standard is not invulnerability. It is a record strong enough to be examined by someone who was not present when the editorial decision was made.