Consistent prose, intact sources
Prose uses atlas terms; quotations, book titles, and source-page titles retain their edition wording.
Definitions, French and English terms, Chinese translations, and edition references for 69 terms. Search a translation, compare meanings, and continue into related concepts and film analyses.
Prose uses atlas terms; quotations, book titles, and source-page titles retain their edition wording.
Rendered as 现实态 / 潜在态 / 实在. Actual and virtual both belong to the real; virtual here names the virtual state.
Cinema contexts use 影像; 图像 / 图象 remain searchable variants, and external titles retain their published form.
For cinema, sound, and technical media, the atlas uses 影像 across moving and still forms; 图象 remains indexed as a historical and community variant.
actuel / l’actuelactual / the actual现实态 marks a state of the real, distinguishing actuel from réel (实在) and from the everyday senses of 实际 as practical or factual.
virtuel / le virtuelvirtual / the virtual潜在态 pairs with 现实态 as two states of the same real and remains distinct from the digital sense of virtual.
réel / le réelreal / the realThe site fixes réel as 实在 so that both actual and virtual can be said to be real, rather than forcing one Chinese word to carry both actual and real.
image-affectionaffection-image情动 preserves a bodily relation of affecting and being affected, extending beyond named private emotions.
coupure irrationnelleirrational cut切断 better indicates a relation actively severed and a gap becoming operative, rather than merely naming an editing technique.
Search any variant to locate the atlas term. Source fields list French texts, published translations, seminar archives, community materials, and atlas terminology records separately.
69 terms · 24 shown
For cinema, sound, and technical media, the atlas uses 影像 across moving and still forms; 图象 remains indexed as a historical and community variant.
The cinema books follow Bergson’s starting point: matter is the aggregate of images, prior to the split between object and copy.
The term follows the highly recognizable Chinese title tradition of the Cinema volumes; the em dash marks a conceptual compound rather than an ordinary modifier.
The movement-image names how movement and sensory-motor linkage organize images; it is a structural category rather than a period label.
The site retains the established Chinese title of the Cinema volumes while applying its consistent image → 影像 policy.
The criterion is how temporal relations organize movement, editing, sound, and image; slow motion, long takes, waiting, flashbacks, and temporal themes are possible materials.
image-temps directe / présentation directe du temps
direct time-image / direct presentation of time
Direct qualifies the presentation of time: time is no longer inferred indirectly from movement but becomes the primary relation organizing movement.
Direct names time’s organizing role across several forms beyond the crystal, including edited, real-time, and moving images.
The site reserves 知觉 for perception, distinguishing it from 感知体 for percept and 感觉 for sensation.
The perception-image concerns how selection, centres, and viewpoints emerge within the universe of images; the subjective shot is one possible form.
情动 preserves a bodily relation of affecting and being affected, extending beyond named private emotions.
In the affection-image, faces, close-ups, or any-space-whatever carry power-quality. Expression is possible material; the conceptual relation exceeds psychological information.
行动 better names a subject responding to and transforming a situation than mere bodily motion; 动作—影像 remains searchable as a published variant.
The action-image depends on action carrying a situation toward a resolvable new situation; bodily movement is material within that structure.
The atlas uses 冲动—影像 for pulsion as an autonomous intensity extending affection toward action before action becomes organized. Diagnosis first distinguishes four structural terms—originary world/derived milieu and elementary impulse/real behaviour—then two classes of signs: a symptom manifests impulse, while a fetish is the appearance of the seized object-fragment. The psychoanalytic variant 驱力 remains searchable without reducing Deleuze's cinematic type to an inner motive.
The four structural terms must form a relation, but symptom and fetish are two types of signs, not extra structural coordinates, and need not both appear. Stroheim proceeds through decay, descent, and entropy, while Buñuel proceeds through repetition, cycles, and restart; these likewise are not cumulative requirements for every case. Strong desire, violence, or animality alone remains insufficient.
grande forme de l’image-action / S–A–S′
large form of the action-image / S–A–S′
Large form names Deleuze's S–A–S′ structure: an encompassing situation S distributes forces, an action A responds, and the situation becomes S′. The colon marks it as a form of the action-image rather than a separate image class.
Large refers to the situation's encompassing and determining role, not spectacle, feature length, production scale, wide framing, or heroic stature. The test follows how the milieu bears upon bodies, how a duel concentrates forces, and whether action actually transforms S.
petite forme de l’image-action / A–S–A′
small form of the action-image / A–S–A′
Small form corresponds to A–S–A′: an act or behavior A discloses a situation S that was not fully given, and that disclosure redirects the subsequent action into A′. The name specifies a direction of inference, not the work's scale.
Small does not mean a minor character, low budget, short film, tight framing, or modest event. If an already totalized situation governs the characters, the structure may remain large form. The evidence for small form is an action functioning as an index from which character and viewer infer the situation.
反映—影像 preserves the figure as an operation making situation and action relate indirectly. It has three figures only: attraction uses plastic substitution or theatrical rehearsal to miniaturize, rehearse, or prefigure another situation or act and thereby convert between large and small forms; inversion exchanges the functional positions of action and situation, so forceful action with a weakened situation or powerless action with an immense world exposes the large and small forms from their reverse side; discourse takes a problem or relation of thought directly as its object. 反思 over-internalizes the process, while 反射 over-physicalizes it as optics; both remain indexed variants.
The reflection-image mediates the extension of action into mental relations by making the direct S–A or A–S relation indirect through figures of attraction, inversion, and discourse. Mirrors, reflections, self-examination, symbolic dialogue, or a supposedly reflective theme are each insufficient, and the three figures need not always convert one action-form into the other.
心智 emphasizes Cinema 1's problem-field: images take relations, symbolic acts, and intellectual feelings as objects of thought rather than transferring a character's private psychology onto the screen. 心理 narrows the range, while 精神 can import senses of esprit or Geist, so both remain search variants.
The mental image is not a catch-all for point-of-view shots, inner monologue, or psychological subject matter. Thirdness finds its most adequate cinematic representation in relation; relation is not the grammatical subject of that claim. The atlas separates mental-image and relation-image only as reading interfaces for a problem-field and its Hitchcockian realization through Thirdness. It does not claim parallel source-level ranks or place recollection- and dream-images beneath the former.
关系—影像 emphasizes relation as a third term external to and irreducible to any one relatum rather than an attribute possessed by one term; relation in turn determines how the terms are understood. It is the decisive Hitchcockian realization within the mental-image's problem-field: Peircean Thirdness, mark, demark, and symbol implicate the viewer's recognition, comparison, and inference in the image's operation.
The relation-image organizes reasoning through a mark within a habitual series, a demark that breaks from the series, and a concrete symbol bearing abstract relations. Character relationships, multi-strand narration, plot complexity, or viewer confusion are insufficient unless the image makes relation primary and reorganizes its terms.
Crisis retains the structural force of crise: action does not vanish, but sensory-motor linkage loses credibility and neither large nor small form can securely bind situation and action into a transformable whole. The phrasing keeps the crisis subordinate to the action-image rather than treating it as another name for the time-image.
Dispersive situations, deliberately weak links, wandering or voyage, consciousness of clichés, and the exposure or denunciation of conspiracy—the network of control behind the plot—form a constellation of surface symptoms. They are external negative conditions of the time-image, need not all appear in every film or sequence, and are not sufficient conditions. One must still test whether optical and sound situations escape practical use and establish new temporal relations.
The site uses the most recognizable Chinese film-studies form, while retaining 感知—行动链 as an explanatory search alias.
The schema enters crisis when seeing and hearing lose their natural extension into adequate action, even while bodies continue to move.
The 2016 edition varies its wording across chapter 12; 图式 is the site’s stable house term for schème sensori-moteur.
Visually checked against page imagesituation optique et sonore pure
pure optical and sound situation
The paired visual/auditory wording preserves two separable channels; the slash signals that sound and image need not form a unified audiovisual whole.
Pure means that an optical or auditory situation exceeds its extension into action and acquires relations of its own; it describes function, not formal purity.
In chapter 1 of Cinema 2, Deleuze distinguishes still life from empty space: empty space derives its force from the absence of possible content, whereas still life is defined by the presence and composition of objects. Cinematic still life differs from painting or photography because the shot gives it duration; in Late Spring's vase shot, that duration presents the unchanging form of time filled by change.
Do not classify every motionless object as a cinematic still life or begin by asking what it symbolizes. First check duration, adjacent shots, and the sound field, then ask whether these relations give time an organizing role not exhausted by narrative function.
The concise 视符 pairs with 声符; opsigne accompanies first use to mark Deleuze’s cinematic semiotic context.
The opsign marks a pure optical situation; classifying a film still depends on its full set of image-relations, with composition as formal material.
Published form visibly checked in the 2004 edition; house prose uses the compact 视符.
Visually checked against page image声符 forms a stable pair with 视符 without reducing sonore to speech or music.
Voice-over, noise, or silence becomes a sonsign only when it ceases to function merely as a cue for action and acquires relations of its own.
The 2004 edition uses 听觉符号 earlier and 声音符号 in the conclusion; both are recorded as edition-specific forms.
Visually checked against page imageThe preferred form follows the visibly checked 2004 translation and supplies tactisigne on first use, retaining its local context of tactile vision in Bresson.
The tactisign is a local extension in Cinema 2 chapter 1; the conclusion section’s recapitulation begins with opsign and sonsign.
A local sign in the chapter-one Bresson discussion; the chapter-ten recapitulation begins with opsign and sonsign.
Visually checked against page image视见者 preserves the force of voyant beyond ordinary looking: when agency weakens, the character is compelled to see what habitual action concealed.
The seer’s incapacity for action becomes intensified perception or witnessing; staring and observation are surface gestures.
任意空间 is established in scholarship; quelconque here means despecified and open to recomposition.
Any-space-whatever loses the stable coordinates and total continuity required by action; ruins, empty shots, and abstract sets can carry this despecification.
现实态 marks a state of the real, distinguishing actuel from réel (实在) and from the everyday senses of 实际 as practical or factual.
Actual and virtual both belong to the real and remain related through actualization and virtualization.
The visibly checked 2004 form in discussions of the crystal and actualization; the house term adds 态 to distinguish actuel from réel.
Visually checked against page image潜在态 pairs with 现实态 as two states of the same real and remains distinct from the digital sense of virtual.
The virtual is a fully real field of differential relations capable of actualization, distinct from falsity and preformed possibility.
The visibly checked 2004 form in discussions of the crystal and actualization; it names a fully real differential structure.
Visually checked against page image