Edition or Variant Collation Table¶

Collation tool — instantiates Textual Close-Reading Mode

Aligns the versions, translations, amendments, and witnesses side by side so a reading can't mistake version-contingent wording for a stable property of the text.

You cannot closely read a text until you know which text you are reading. The Edition or Variant Collation Table settles that. It lines up the available versions — editions, translations, amendments, manuscript witnesses, successive drafts, alternate renderings — in parallel, mark against mark, so that every place where they agree and every place where they diverge is visible before any interpretation begins. Its single defining job is to keep a reading from resting on wording that only one version happens to carry: when the whole argument turns on a word, the table is what tells you whether that word is stable across the tradition or an artifact of the copy on your desk. It produces the stable substrate every other close-reading mechanism silently assumes.

Example¶

An editor preparing a scholarly Hamlet reaches the line on which the prince wishes his flesh would melt. The Second Quarto prints sallied, the First Folio prints solid, and generations of editors have emended toward sullied — three different words, three different plays of meaning, one line. Put into a collation table, the witnesses sit in adjacent columns: Q1, Q2, Folio, and the major emended editions, each cell showing exactly what its source reads, with sigla marking which is the copy-text and which are departures.

The table does not decide the crux. What it does is prevent the mistake underneath every hasty reading of the line — treating solid flesh, the Folio's word, as simply what Hamlet says — by making the variation itself part of the evidence. A reader who wants to argue for sullied now has to argue against the witnesses in the table, in the open, rather than quietly picking the version that suits the reading.

How it works¶

Assemble the witnesses in scope — the editions, translations, amendment states, or manuscript copies the reading could depend on — and designate one as the base text.
Align unit by unit and record every divergence, separating substantive differences (a changed word) from accidentals (spelling, typography) so the two are not confused.
Attach provenance to each reading — which witness, which siglum, original or editorial emendation — so no variant floats free of its source.
Flag the version-contingent claims — the interpretive points that would change if a different witness were chosen as authoritative.

Tuning parameters¶

Witness scope — how many versions to admit. A wider net is safer but has sharply diminishing returns; a narrow one is fast but can miss the witness that overturns the reading.
Substantive-vs-accidental threshold — where to draw the line between a meaningful variant and mere spelling or typography. Set it loose and the table drowns in noise; set it tight and a meaningful accidental slips through.
Copy-text choice — which witness anchors the comparison. It quietly sets the "default" wording and determines what everything else looks like a departure from.
Normalization — how much to modernize or regularize spelling before aligning. Convenient for reading, but every normalization erases a small piece of the evidence the table exists to preserve.

When it helps, and when it misleads¶

Its strength is that it dissolves a whole class of false confidence: the reading that quietly depends on one translation's phrasing, one amendment's addition, or one printing's comma is exposed the moment the witnesses sit side by side. For any text that exists in more than one form — scripture across translations, a statute across amendments, a play across quartos — it is what turns the text says into this witness says.

Its own failure is the opposite kind of false confidence. A clean table can imply that one column simply is the text, when choosing it was an editorial act; and heavy normalization can erase the very accidentals a reading might turn on. Its classic misuse is picking the copy-text that yields the wanted reading — or, more quietly, closely reading a convenient reprint as though it were authoritative. The discipline that guards against this is to choose the copy-text on stated principle rather than convenience, preserve accidentals alongside substantives, and mark every emendation as an emendation.^[1]

How it implements the components¶

textual_substrate_and_version_record — the table is the version record: the authoritative version, its rendering, and its boundary, held explicitly rather than assumed.
edition_translation_and_collation_control — it operates the collation itself, aligning editions and translations and controlling which readings are admitted and which are flagged as variant.

It fixes the object but does not interpret it: the formal analysis is the Diction, Syntax, and Punctuation Trace, and the reading-in-sequence is the Line-by-Line Annotation Protocol, both of which consume the stable text this table establishes.

Instantiates: Textual Close-Reading Mode — the table supplies the stable substrate the whole appraisal rests on.
Sibling mechanisms: Parallel-Passage Comparison · Line-by-Line Annotation Protocol · Diction, Syntax, and Punctuation Trace · Close Reading Table · Motif Concordance and Recurrence Map · Sequence and Placement Map · Omission and Negative-Evidence Audit · Micro-to-Macro Claim Matrix · Paraphrase-Difference Test · Counterreading Stress Review · Contextual Reentry Memo · Interpretive Saturation Log

Notes¶

Collation is not the same as parallel-passage work. This table compares versions of the same passage to fix which wording is stable; the Parallel-Passage Comparison compares different passages to judge whether a wording is exceptional or routine. Reach for this one when the question is "is this really the text?" and for that one when the question is "is this word unusual?"

References¶

[1] W. W. Greg's "Rationale of Copy-Text" — the editorial principle of selecting a base text for its accidentals while admitting substantive readings from other witnesses on their merits. It is the standard discipline for why one column anchors a collation, and it exists precisely so the choice is argued rather than assumed. ↩