7. Diagnostic Method

7.0 How to use the protocol

The ten tests below are for revision, not composition. Applied while drafting a prologue they produce paralysis and, worse, produce the checklist prologue (§4.0). Applied to a finished draft — and ideally to a finished manuscript, since half these tests require knowing what the rest of the book does — they are an efficient triage instrument.

Five procedural notes, of which the last is specific to this chapter.

• Run them in order. Tests 1–4 are positional, and their outcomes determine whether the rest are worth running. There is no point tuning the register of a prologue that should be Chapter One’s third paragraph.
• Record the answer, not the verdict. The output of each test is a sentence describing what the prologue does, not a pass or a fail. “The distance is a viewpoint the book never resumes” is useful. “Fails test 6” is not.
• A failed test is a question. Every one of these tests is failed by at least one work discussed in this chapter, several of them canonical. The test’s job is to make a choice visible so that it can be confirmed as a choice.
• Half the tests need the whole manuscript. Closing, plants, register consistency and genre contract cannot be assessed from the prologue alone. A workshop that critiques prologues in isolation is, structurally, unable to run this protocol — which students who have been told their prologue is unnecessary by readers who have not read the book should hear.
• The protocol’s output is a three-way decision, not a two-way one. Not keep or cut but keep separate, absorb, or relocate. Most prologues that fail this protocol should be absorbed or relocated. Very few should be deleted, because the material is usually good; it is the compartment that is wrong.

7.1 The ten tests

Test 1 — The higher-bar test

Could this material be delivered inside Chapter One, or folded into the main timeline, without loss?

The chapter’s first principle in operational form. The procedure is to write a single sentence beginning “This must sit outside the main narrative because…” and to judge the sentence rather than the prologue.

Acceptable completions name a structural impossibility: because no character in the main narrative could have witnessed it; because the main voice is a fourteen-year-old’s and this needs three centuries; because the vantage belongs to someone who is dead before Chapter One; because the book must state where its own account came from.

Unacceptable completions name a preference: because it’s a stronger opening; because Chapter One is slow; because it feels right there; because the book needs a hook. Each of these is a real problem correctly identified and incorrectly located. A slow Chapter One is a Chapter One problem (§8.10).

Interpretation. If the sentence cannot be completed structurally, the prologue is not earning separation, whatever its quality. Proceed to test 2 to determine what to do about it.

Test 2 — The absorption test

What specifically breaks if this is folded into Chapter One?

Test 1 asks whether separation is justified in principle; test 2 asks what the practical consequence of removing it would be, and it is the more useful of the two because it produces an inventory rather than a judgement.

The procedure is literal. Attempt the absorption. Take the prologue’s material and write it into Chapter One — as a paragraph, a memory, a document the protagonist finds, a line of dialogue, a scene relocated to its chronological place. Then list what could not be carried across.

Typical inventory entries, and what each means:

Interpretation. In this edition’s judgement the commonest honest result is the second row, and it is the hardest to act on, because dread is real value and is nonetheless the value most easily faked. Cross-check against test 10.

Test 3 — The label test

Would anything change if this were headed “Chapter One”?

The purest test in the protocol and the fastest to run. Delete the word Prologue, print Chapter One in its place, renumber everything after it, and read.

Three outcomes:

• Nothing changes. The passage was Chapter One. This is failure pattern §8.5, and the remedy is the relabelling you have just performed.
• Something is lost, and you can name it. Usually: the reader would have assumed the prologue’s viewpoint character was the protagonist, or would have expected the register to continue, or would have taken the flash-forward for the present. The label was preventing a specific misreading, which is a real job.
• Something is gained. This is the interesting outcome and it happens more often than students expect. Removing the label sometimes produces a stronger, stranger book, because the reader is no longer told to hold the material at arm’s length. §9.2 (Wuthering Heights) and §9.5 (Their Eyes Were Watching God) are both books that took this option, and it is worth asking of any draft prologue whether the label is protecting the reader from a productive confusion.

Interpretation. Note that this test can be passed by a prologue that fails test 1. A passage may need the label (to prevent a misreading) without needing the separation (because the material could have been elsewhere). Where that combination occurs, the correct move is usually to relocate the material and drop the label.

Test 4 — The distance-axis test

Which of the four axes — time, place, viewpoint, register — is actually doing the work?

Name the axes on which the prologue is distant from Chapter One. Then strike them out one at a time and ask whether the prologue still functions.

The result is almost always that one axis is load-bearing and the others are incidental. A prologue set in 1893 in Vienna in the voice of a doctor may seem to be distant on three axes, but if the same material would work in 1953 in Lisbon in the voice of a lawyer, then time and place are decoration and viewpoint is the axis.

Interpretation. Three uses:

1. The load-bearing axis tells the return what to deliver. A viewpoint distance is closed by a viewpoint event; a time distance by arrival; a register distance by recurrence. Mismatched closings — a viewpoint gap “closed” by the narrative merely reaching the right date — are a common and hard-to-see defect.
2. Incidental distance is cost without benefit. Every unnecessary axis adds orientation expense (Chapter 5’s orientation budget applies) for no purchase. Collapsing a prologue onto its load-bearing axis usually shortens it by a third.
3. No axis at all means no prologue. See test 3.

Test 5 — The register test

Is the prologue’s tone, vantage or distance doing something Chapter One’s register structurally cannot?

Sharpened by the substitution procedure of §4.3: rewrite the prologue’s content in Chapter One’s voice and see what happens.

Interpretation. The third row is the strongest result available in the protocol, and it is the reason mythic overtures and editorial notes survive centuries of hostility: a close third-person narration simply cannot say for three hundred years the valley was empty, or the papers are here set forth in sequence, without ceasing to be what it is.

Test 6 — The closing test

Does the main narrative close, explain, echo or deliberately hold open the distance the prologue opened?

Expanded by the taxonomy at §6.5.

Procedure: list every element the prologue introduces — persons, objects, places, phrases, claims, questions. For each, record the page on which the book next touches it. Then classify the overall pattern as total closure, fragmentary closure, recontextualisation, or non-closure.

Interpretation.

• If the answer is “never” for the load-bearing axis, the prologue is attached to the book rather than integrated into it. This is §8.4, in this edition’s judgement the commonest failure after §8.5.
• If the answer is a single page number very late in the book, run §6.6’s argument: consider redistributing the closing into three or four earlier fragments.
• If the answer is “deliberately never”, apply the three signals in §6.5 that distinguish design from oversight, and be honest about the result.

Test 7 — The plant-audit

List every specific in the prologue. Which are redeemed later, and which are decoration?

Per Rabinowitz (§3.9), a prologue occupies a privileged position twice over, so every detail in it is read as significant. The audit makes the resulting promissory notes visible.

Procedure: tabulate every named or specified thing — every object, name, place, number, gesture, phrase — against the pages on which it recurs. The table typically has three regions:

Interpretation. The remedy for the third region is nearly always deletion rather than invention. A writer who discovers eight unredeemed specifics should not add eight payoffs; they should cut six specifics and strengthen two. Chapter 9 (§7, test 7) makes the same argument for opening pages; the prologue’s version is more urgent, because the density of significance in a separated unit is higher.

Test 8 — The skip test

What happens to a reader who skips this?

Genette’s observation that the preface is the most easily bypassed zone of a book (§3.1) applies with full force to the prologue, and the skip test is the only place in this protocol where the writer is asked to design for a reader who declines to read.

Procedure: hand the manuscript to a reader with instructions to begin at Chapter One, and interview them at page 100 and at the end. Four questions:

1. Was anything incomprehensible?
2. Was anything retrospectively incomprehensible — a scene that made sense at the time but turned out to depend on something you did not have?
3. At the end, read the prologue. What did you now understand differently?
4. Did the book feel as though something was missing?

Interpretation. The intended profile is: nothing incomprehensible, and something lost. A skipping reader should be able to follow the book completely while missing an irony, a dread or a recognition. That profile is the practical form of the definition in §2.1.

Two failing profiles:

• Nothing incomprehensible and nothing lost. The prologue is inert. Absorb or cut.
• Something incomprehensible. The prologue is carrying load-bearing plot information that only it supplies. This is not automatically a fault — some books legitimately require their prologue — but it should be a decision, and it means the prologue is no longer skippable, which contradicts the licence its label grants. The usual remedy is to restate the essential fact somewhere in the main narrative, so that the prologue supplies first knowledge rather than only knowledge.

Test 9 — The removal test

Cut the prologue and start cold at Chapter One. What is lost — and is what is lost more than a hook?

It differs from the absorption test (2) in that nothing is carried across: the material is simply gone.

Procedure: produce the manuscript without the prologue and read the first thirty pages. Then ask, in this order:

1. Does Chapter One now open the book adequately by the standards of Chapter 9 (The Opening Pages)? If not, the prologue may have been concealing a defect (§8.10).
2. What does the reader no longer know?
3. What does the reader no longer feel?
4. Would the reader notice the absence?

Interpretation. Question 4 is the sharp one. A prologue whose removal produces a book that is slightly worse in ways no reader would name has not earned a compartment; it has earned a paragraph. A prologue whose removal produces a book that reads correctly but differently — the same events, no dread — has earned its place if the dread is what the book is for.

Test 10 — The genre-contract test

Does the prologue promise the book that actually follows?

The most expensive failure in the repertoire (§8.2). It is the exact analogue of Chapter 9’s promise test and of Chapter 16’s final-note test, and the three should be run together on any manuscript.

Procedure: read the prologue, then read pages 40–60, then read the last chapter. Ask of each pair whether they belong to the same book, in four respects:

Interpretation. A mismatch in any of the four is a false-advertising risk, and the correct direction of repair is nearly always to change the prologue, not the book. Chapter 9 states the reason and it applies here with more force: the prologue is a thousand words and the book is a hundred thousand. Revise the smaller object.

The important qualification: a deliberate mismatch is a legitimate technique. A comic novel with an ominous prologue, an epic with a domestic one, are both available and both effective — provided the reader can tell the mismatch is designed, which normally requires the book to acknowledge it. The test’s real output is therefore: is this mismatch legible as a choice?

7.2 Reading the results

A warning about test 10. When a prologue and a book disagree, writers instinctively revise the book towards the prologue, because the prologue is more finished, more exciting, and was often written last with full knowledge of the whole. This is almost always the wrong direction, and it is more tempting here than in Chapter 9’s version of the same warning, because a prologue’s separateness makes it feel like a statement of intent rather than a passage of text. It is a passage of text.