Methods

Pipeline overview

Every visualisation on this site draws on a structured analytical pipeline applied to a normalised plain-text transcription of Gravity's Rainbow. The pipeline proceeds in numbered steps, each producing its own outputs and diagnostics. Only public-safe exports — stripped of source text, quotations, and analyst notes — cross into this site repository.

The pipeline divides broadly into three stages:

1. Segmentation and metadata. The novel is divided into 73 episodes following the conventional section breaks in most editions. Each episode receives word counts, paragraph counts, and a diegetic time estimate.
2. Structural analysis. Character co-presence, plotline intensities, modal registers, historical recurrence patterns, vocabulary registers, foreign-language passages, and song/verse segments are extracted computationally, then reviewed and corrected by hand.
3. Publication export. Analytical outputs are filtered through a public-safety policy that strips source text and private annotations. Only the approved exports are copied to the site.

The Braid

The braid diagram assigns each of the novel's 73 episodes an intensity value (0–3) for each of eight plotlines. These values represent how prominently a given plotline figures in each episode:

0

Absent — the plotline does not appear

1

Minor — a brief reference or passing mention

2

Moderate — a sustained presence within the episode

3

Dominant — the plotline drives the episode's action

The eight plotlines — Slothrop, Blicero, Enzian, Pökler, Pointsman, Counterforce, Tchitcherine, and Film/Body — were chosen to capture the novel's principal narrative strands. They are not exhaustive; several minor threads (Katje, Geli, the Schwarzkommando rank and file) appear within the episodes but are not tracked as separate strands.

Intensities were initially assigned computationally based on character co-presence, mention density, and narrative focus, then manually reviewed and adjusted.

Modal registers

Ten modal registers classify the dominant tonal or thematic atmosphere of each episode. Like plotline intensities, these are scored 0–3:

• Science & Control — Pavlovian conditioning, statistics, operational research
• Pop Song & Comedy — music-hall routines, limericks, slapstick
• Film & Media — cinema, stage, camera consciousness
• Language & Allusion — etymological play, foreign tongues, literary parody
• Occult & Divination — Tarot, séances, sympathetic magic
• Corporate & Military — IG Farben, Shell, bureaucratic machinery
• Rocket & Terminal — V-2 technology, descent, the 00000/00001
• Elect & Preterite — Calvinist theology, chosen and passed-over
• Sexuality & Domination — sadomasochism, power, the body as instrument
• Waste & Grotesque — coprophagy, refuse, the body's lower registers

Categories were derived from close reading and computational keyword analysis. They overlap deliberately — an episode scored high on both “Rocket” and “Elect & Preterite” reflects Pynchon's persistent yoking of technology and theology.

Historical recurrence

Five registers of historical recurrence track how the novel encodes patterns of systemic violence that persist across the narrative:

• Species extinction — the dodo, the Dodoes of the Mauritius, entropy
• Colonial genocide — South-West Africa, the Herero, Rathenau's enterprises
• Holocaust adjacency — the camps, the ovens, I.G. Auschwitz, absence
• Forced labour — the Mittelwerk, Dora, industrial body processing
• Technocratic continuation — Operation Paperclip, corporate afterlives, the postwar

These registers are scored on the same 0–3 scale. They are analytical judgements, not neutral descriptions: the category “Holocaust adjacency” reflects the novel's characteristic approach of circling the event without directly depicting it.

Data and boundaries

All data files used by this site are public-safe analytical exports. They contain episode identifiers, numeric intensities, categorical labels, graph metrics, and structural metadata. They do not contain source text from the novel.

A small number of short quotations appear on the site as part of a curated, manually approved quotation layer. Each is under fifty words and falls within fair-use guidelines for critical commentary. The quotation layer is intentionally sparse.

The underlying text corpora, normalised source files, extraction registries, analyst notes, and diagnostic outputs remain private. Every asset imported into the site repository is logged in the project's provenance record with source path, date, and justification.

Further reading

The Reference page collects scholarly essays, reader's guides, annotations, and community resources for Gravity's Rainbow.