Barry Lyndon scripts: versions, annotations and limits of the surviving material

Barry Lyndon’s written materials make it possible to investigate adaptation and production decisions. A draft, however, does not prove that a scene was filmed or that a lost cut exists.

A catalogue identifies materials, not their full contents

Archives may preserve drafts, typed versions, annotations and planning documents for Barry Lyndon. A catalogue entry can tell us that a particular item exists and may provide dates or general descriptions. It is no substitute for reading the documents in full, nor does it justify attributing specific changes to them unless these have been verified. To study a version, its status, provenance and relationship to other copies must be identified.

The film adapts a novel by Thackeray, but a full literary comparison belongs in another study. The adaptation analysis examines that relationship. Here, the focus is the trajectory of the working text: how a scene may appear in one draft, be modified in another or be left out of the final copy. Each state provides different information and calls for precise terminology.

An archive catalogue may indicate that several versions of a screenplay exist, but its description is usually shorter than the document itself. To study actual changes, the texts must be examined and their chronological relationship established. Two copies with similar titles may differ slightly or substantially; an annotation may come from a later stage. Without that comparison, it is not possible to reconstruct a precise sequence of revisions.

The date of a document also requires interpretation. It may refer to when it was typed, revised or added to the archive. It does not necessarily indicate when each scene was first written. Research must record these limitations to avoid creating a fictional chronology of the creative process.

Written, planned, filmed and edited

A scene included in a screenplay may have been planned without ever being filmed. A filmed scene may have been discarded during editing, and footage edited into a working version may not have appeared in the release. These are four different claims. The existence of a passage in writing proves only that it appears in that document. To prove filming, production reports, photographs, filmed footage or specific testimony would be needed; to speak of an edited-out scene, there must be evidence of an actually edited version.

Annotations can also be ambiguous. A marginal mark may indicate a suggestion, a correction or a decision already made, depending on its context. It is best not to turn every deletion into a deleted scene. The work of Tony Lawson helps place the editing phase without confusing it with screenplay development. Research must preserve the distinction between a proposal and the result.

The difference between a scene being written and being filmed is especially important when discussing deleted material. A passage may disappear before a location is prepared or the cast is called in. Only evidence of filming would allow us to claim that filmed images existed. Even then, it would be necessary to show that they were included in some cut before claiming that a different edited version existed.

Production photographs, shooting reports and work prints can provide additional information. None should be interpreted outside its context. A posed photograph may show costume and set without proving that the action depicted appeared in a take. The history of deleted scenes requires the same care in distinguishing preparation, filming and editing as is applied to screenplays.

What questions can the documents answer?

Written versions can help examine changes to names, the distribution of dialogue or the intended position of a voice-over. They may also reveal how episodes were grouped before the final organisation into two parts. The analysis of intertitles and structure describes the released film; the documents allow it to be compared with earlier states, provided they are available and dated.

A catalogue description can guide future enquiries, but it does not justify inventing direct quotations or reconstructing a supposed alternative cut. Best practice is to record which document has actually been examined, what information it provides and what remains unverified. Barry Lyndon is a useful case for remembering that production history is written using evidence of different kinds, not by automatically turning every draft into a missing scene.

Drafts can also help investigate where the voice-over was intended to appear. A written passage may change position or disappear before the final mix. But the text does not prove how Hordern’s performance would have sounded or which shot would have accompanied the line. Reconstructing that relationship would require audio or editing materials, not just a typed page.

Barry Lyndon’s division into two parts raises another documentary question: what structure appears in the surviving versions and how it relates to the intertitles in the released print. The study must be based on documents that have actually been consulted. When their existence is known only from a catalogue, the matter should be framed as a line of enquiry, not presented with an invented answer.

Documentary caution also guards against the temptation to announce discoveries that have not been made. A folder titled with the name of a scene may contain planning notes rather than filmed images. Accurately describing the type of material available is more useful than presenting every archival reference as proof of a lost cut.

View all articles in Barry Lyndon: characters, scenes and the construction of the 18th century