Saying that Kubrick watched a film, admired it, and used it as a reference makes three different claims. A reliable attribution preserves the source and does not turn one claim into another without evidence.
Start with what the source actually says
The selection published in 1963 allows us to attribute a choice of titles to Kubrick within the context of a specific survey. Its strength comes from preserving a dated public statement. It does not need to be supplemented with later films to be useful. Adding them under the same heading would alter the document and make it impossible to know what belonged to the original selection.
A letter can document a different kind of appraisal. The Ingmar Bergman Archive identifies one sent by Kubrick on February 9, 1960, with the reference K:636. In it, he expresses admiration for Bergman’s work. This testimony allows us to speak of Kubrick’s appraisal of the filmmaker at that time; by itself, it does not provide a ranked list of each of his films.
The distinction between a person, their work, and their approach matters. Praising a director does not prove that someone approves of all their decisions. Admiring a performance is not the same as considering the entire film a favorite. When the original statement or a verifiable summary is available, it is best to preserve the exact object of the praise rather than expand it into a more striking claim.
Interviews add the context of a question. An answer about visual effects may highlight a technical solution without expressing a general preference. An answer about children’s movies may refer to the experience of watching them with family. The meaning of an attribution changes when that context is removed and only the mentioned title is kept.
A recollection also identifies the person telling it
Accounts from family members and colleagues provide information that may not appear in documents written by Kubrick. Their value depends on what they witnessed and how they describe it. Watching a screening together, ordering a print, and hearing a recommendation are different experiences. The witness should be named, and that distinction preserved when writing the attribution.
The recollections of Jan Harlan can be understood in the context of his professional and family relationship with Kubrick. That does not mean treating them as an infallible record of every preference. A memory of enthusiasm has one scope; a hypothetical reconstruction of what Kubrick might have chosen many years later has another. A conditional statement in someone’s account should not be turned into a list endorsed by someone who never made one.
The date of a recollection and the date of the event are not necessarily the same. An interview may take place decades after the screening being described. That distance does not make the testimony false, but it is a reason to preserve expressions of uncertainty and verify details when other sources are available. A detail added by someone summarizing the account can create a level of certainty the original speaker never had.
The practice of watching films should be kept separate from attributing favorites. Finding a print in a collection does not prove it was watched; watching it does not prove it was liked; recommending it may be about a specific aspect. These are separate steps. A source may document one without allowing us to infer the next steps through a chain of assumptions.
Comedy titles offer a useful example because they are sometimes grouped into lists meant to surprise readers with an allegedly unexpected taste. Documented references to Chaplin or Fields make it possible to study this connection, developed in comedy cinephilia. They do not justify adding other comedies based on affinity or assuming that every shared laugh reflects a lasting preference.
Cross-check without manufacturing new certainty
When two sources appear to disagree, first check whether they address the same question. A statement of admiration at one point in time and a recollection of less interest years later may be compatible. A comment about one film does not necessarily contradict an appraisal of another film by the same director. Before concluding that someone is mistaken, a comparison must preserve both the subject and the timing.
A lack of recollection also does not prove that something never happened. A colleague may not have been involved in a screening, just as a repeated claim may lack independent support. The conclusion should match the scope of the evidence: documented, attributed to a witness, disputed among accounts, or not sufficiently confirmed.
The Kubrick Archive may contain correspondence, work documents, or materials from a collection. Consulting a catalog is not the same as examining every item. An archival description can identify a document and provide its general context; to attribute specific content to a letter or note, you need to know what that item says, not infer it from a file title.
It is also worth tracing how publications depend on one another. Ten lists that copy the same interview do not amount to ten independent accounts. Repetition increases the circulation of a claim, not necessarily its support. The key is to locate the original source and check whether later versions preserved its nuances or added a level of certainty that was not there.
The method for evaluating interpretations calls for the same proportionality. A useful attribution identifies the film, type of appraisal, source, and date when known. It can preserve uncertainty without filling in the gaps. The result may be less dramatic than an all-time ranking, but it will clearly distinguish what Kubrick expressed, what another person remembers, and what cannot yet be claimed.
Sources: Cinema, The Directors Choose the Best Films (1963); Ingmar Bergman Archives, letter from Stanley Kubrick to Ingmar Bergman, February 9, 1960, K:636; University of the Arts London, description of the Stanley Kubrick Archive. Accounts from colleagues should be read with attention to attribution and publication context.
View all articles in Kubrick’s Admired Films and Cinephile Habits