Off-screen: the liminal dimension of the cinematic image – Cristiano Dalpozzo

This essay problematizes, from a theoretical standpoint, the concepts of off-screen and the cinematic image. Taking into account both Deleuze’s definition of time-image and the concept of limit, it’s possible to compare the filmic image to that of a threshold. Such a threshold is to be experienced by establishing a parallelism with the condition of the spectator. This image has a memory of itself and, at the same time, contains traces of what it is going to become thanks (also) to the off-screen space. It is thus both visible and invisible: indeed, it provides two faces of the same coin in a relationship of mutual implication. The filmic image will then be understood as something provisional, a path, not the final destination;  a mediation in a perpetual dialogue conjured up between the visible and the invisible, between the represented and the representable — a medium that exceeds its own limits, figuring and reconfiguring itself between  excess and  absence. An image that hints to an invisible or, rather, to a not (yet) visible, whose traces have been already placed in it (thanks, for example, to the looks exchanged between the actors, and their movements within the frame). An image that never stops to question us, and to act as a mediator becomes,  simultaneously, both an “instigator” and a figure of our mind. An image which is ambiguous, allusive, hybrid, metamorphic, simultaneous, sometimes conciliatory, as in a given idea of classic cinema, but more often than not, conscious of an absence; conscious of the fact that it ‘tolerates the absence of unity” in a perpetual (and paradoxical) process that, as it posits boundaries, simultaneously overcome them.

A boute de souffle (1960)

A boute de souffle (1960)

As is well known, Deleuze maintains that the cinema can offer both a direct or an indirect image of time, depending on the fact that it is bound to time-image or to movement-image. The time-image, inaugurated by Neorealism, further employed by the Nouvelle Vague and still used today, replaces the sensomotoric situations of movement-image with purely optical images and sounds. Thus, reality and imagination blend so as to create a new dimension of the mind. Actions that are linked to each other by a cause and effect relationship, are replaced by dead time and silence. What easily comes to mind is the loafing of many characters in Godard’s movies (Belmondo in À bout de souffle [1] and in Pierrot le fou [2]) or the temporal loss of so many of Resnais films (Hiroshima mon amour [3],  L’année dernière à Marienbad [4] and Muriel, ou le temps d’un retour  [5])

dalpozzo 2

Hiroshima mon amour (1959)

One can maintain that the film is a false movement resulting from a (love and death) relationship between frames and snapshots, and an abstract and mechanical movement of the camera or of the projector that unrolls those frames. The cinema  reconstructs movement through the motionless sections (frames) and recreates time by giving us an average image which is the resultant of the sum of all the frames. Basically, the picture frame cuts out a portion of an open space, a movable section of the time-duration which it opposes to a whole– at once figure of an eternal time and denial of the process of becoming. It is then through the time-image of a certain examples of modern cinema that time comes to be expressed directly, perhaps through the use of the sequence shot (and its ability to express both continuity and duration) and the depth of field (and its ability to express the simultaneous presence of more than just one actor on the scene).

Professione Reporter / The Passenger (1975)

Professione Reporter / The Passenger (1975)

The finale of Professione Reporter/The Passenger (1975) by Michelangelo Antonioni evolves, in fact, in this direction (it entails continuity and duration of a criminal action, and the presence of several characters and events on the same scene) imprisoning ideally, what’s more, the look (at the same time source and object) in a self reflective mise en abyme. In the meantime there is a kind of cinema that through editing and decoupage (both elements preferring a chain of cause and effect relationships) of classic cinema, favors (besides plane sequences and depth of field) the voids and the scraps of the editing process, thus enhancing the intervals that come into place between one image and the other. Voices and images, unpredictably, slip through in a fragmented style, that declares the final break between the world and the characters represented within.  This is the case, for example, in Le Mépris/Contempt (1963) by Godard, a film in which some of the characters listen to a singer in a club: every time one of them starts talking the song stops abruptly.

From such a viewpoint, then, the concept of off-screen seems to take on a different meaning. How can the time-images be chained to one other, when they do not extend into an action? If, previously, the off-screen communicated “on the one hand to an external world which was actualizable in other images, on the other hand to a changing whole which was expressed in the set of associated images” [6], now the images are no longer linked by rational cuts and continuity, but are relinked by means of false continuity and irrational cuts” [7]. We then achieve a re-assemblage of dis-assembled images through which “there are no longer grounds for talking about a real or possible extension capable of constituting an external world: we have ceased to believe in it, and the image is cut off from the external world” [8]. Think, for instance, to the scene of Pierrot le fou in which Belmondo and Karina are conversing. Suddenly, from off-screen we hear a burp, though the two main characters are absolutely alone on stage. Where does that sound, autonomous and without reference, come from? In the direct time-image,  therefore, an absolute contact between an inside and an outside, which are independent, asymmetric, come to completion.

We move with ease from one to the other, because the outside and the inside are the two sides of the limit as irrational cut, and because the latter, no longer forming part of any sequence, itself appears as an autonomous outside which necessarily provides itself with an inside [9].

The limit and the interstitium–Deleuze tells us–now passes between the visual and the sound image. It is necessary that the latter becomes image in and of itself, thus releasing the status of being merely a component of the visual. So, even though the off-screen remains de facto present even in the time-image, it will be necessary that it

lose all power by right because the visual image ceases to extend beyond its own frame, in order to enter into a specific relation with the sound image which is itself framed (the interstice between the two framings replaces the out-of-field); the voice-off must also disappear, because there is no more out-of-field to inhabit, but two heautonomous images to be confronted […] It is possible for the two kinds of images to touch and join up, but this is clearly not through flashback, as if a voice, more or less off, was evoking what the visual image was going to give back to us [10].

Le Mepris / Contempt (1963)

Le Mepris / Contempt (1963)

Where does the voice of Brigitte Bardot come from when evoked in the finale of Le Mépris when Piccoli reads the letter that anticipates that she will leave him? From the mind of the protagonist? From the past? From the future?

From this point to Deleuze’s assumptions there is a short run. The French philosopher  declares in fact that: “modern cinema has killed flashback, like the voice-off and the out-of-field”. [11] But perhaps not all is lost; moreover, the cinema is not all made up of time-images. Let’s take a step back and confront the concept of limit itself. Peras, limit (in Greek) means “point that delimits something.” The limit defines a form, whatever it might be, and, what happens within it, is not finally that important. It will always stay that precise form. Thus in, De Gaetano’s quote about Deleuze’s thought, he explains:

The limit defines a shape, abstract or sensitive as it might be. All what happens within the line sketching the perimeter of a shape does not matter; whether we fill it with sand or with intelligible matter, the perimeter of a cube or for what matters, that of a circle, it is nevertheless bound to remain always a cube, or a circle [12].

According to De Gaetano, the concept of framing as frame, that is to say as something that is capable of separating an interior from an exterior–a seen from a not seen– represents a static idea of limit. The problem, says the scholar, is that this seen, isolated from the frame, is not identifiable statically and logically. Eisenstein, in fact, basing his theory of editing on the idea of framing as cell, literally “blasts” the limits of the frame; he turns the seen into a force, an inner tension “that leads to the overcoming of its unity and ‘self-sufficiency'”.

Both editing inside the frame, and editing as a conflict/relationship between shots and sounds, just lead to overcoming a static limit in order to offer as a “tension towards to”, a force that allows to pass: “from the shot-framing, from the edges of the frame, that though mobile, nevertheless always remain edges, to the frame-force, tension, power, that goes beyond the edges of the frame; its very perimeter”. The example of the famous scene of the “Odessa steps” in Battleship Potemkin is a good example [13]. The limit becomes a dynamic term; “term of a force endowed with degrees of growth and decrease. And then the force is distinguished from the form, not because it has a limit, but because its limit coincides with its power: “the problem is no longer in the definition the terminal point of a shape. To posit such a question means already to fall into an abstraction, into an artifice. The real question, at this point, is: where  does the action end [14]?

The Latin word limen/limes [15] means threshold, boundary, target, and refers to an idea of demarcation between inside and outside. Basically, whenever we are faced with a threshold, we have three possibilities: to overcome it, not to overcome it, or to stand in the threshold. And, if we were willing to make general assumptions, we could say that these three options could well represent the three ages of cinema itself: classic cinema (not overcome), modern cinema (stand on the threshold), digital cinema (overcome). The idea of limit also refers to the frontier between two worlds, between order and disorder. But frontier also means openness, a possibility of inter-penetration;  it evokes an idea of border-filter or border-membrane, thus allowing us to introduce elements external to the system, after they have been made compatible with what is inside. In the end it corresponds to the dynamic that rules the field/off-screen dialectic: not everything is off-screen, but only what is consistent with what is being shown in the picture. The movie image promises to be an image filter or image membrane that can be gone through, but –at the same time– discriminates, by delimiting what is different from itself. Basically, the edges of the framework resemble the skin: at the same time an epidermal and an identity border too.

Identity, of course, because, talking about the border, one can not avoid to mention the erratic condition implied by this type of image: intermittent, migrant and always on the move (escape or exile as it might be), temporary and precarious by nature. But we should not forget that it is thanks to this very liminality, that it is possible to establish an audiovisual conversation with the viewer. It is precisely such an uncertainty that compels the viewer to fill in the gaps, to explore different possibilities and to become an active counterpart of the game [16]. To inhabit, therefore, this unstable space, is the condition of the viewer, or according to Massimo Dona’s formula [17], to inhabit the threshold, of a cinema experienced, precisely, liminally. The idea of threshold applies to the fruition of the spectator too. Dona, in fact, compares the spectator’s position, to what Odysseus experienced in the famous episode of the sirens. Here the omeric hero is basically put to the test, tempted  with a chance of infinity, the ultimate promise of an “other” dimension.  A temptation to which Odysseus would have succumbed, had it not been for Circe. Only by being on the threshold, therefore, only by making himself bound to the mast of the ship could he really enjoy the melody.

One and only one could therefore be the way forward: to remain within one’s own “limit”, or rather, to keep to pro-ceed from limit to limit, knowing that, if we had been promised that we would experience something as an “infinite”, the latter would have not shown itself, if only as essentially inappropriable […] That is to say, we should know how to find peace in a finitness lived as a “threshold” between something which is finished, and an in-finishdness open to deny in actu signato its own distinction. And this happens, once again, only at the movies. […] that is why, at the movies, each one of us becomes “threshold” rather than “finding” himself, simply, “on” the threshold. Thanks to this, at the movies, each of us experiences, quite simply, his-no- longer-being-the-finite-that-he-is. And he enjoys such a perception [18].

The quote points out a situation parallel to that experienced by the viewer, who sits  motionless in the darkness of the movie theatre, while attending at the unfolding of the world, at the breaking down of the frames. Paraphrasing Donà when he compares this type of threshold to the one described by Kafka in his short story Before the Law we could say that, the filmic image is, for us, in and of itself, a threshold  that makes evident its being other from what it (no longer) is. A type of image that has memory of itself [19], of what it has been and, at the same time, contains the traces of what it is going to be, thanks (also) to off-screen space. A two-faced image exactly like the Roman goddess Janus: placed to watch over the entry and the exit –the way in and the way out– thereby hinting to its characteristic of linking by separating and the vice versa. Paraphrasing Deleuze, when he speaks of the crystal-image, we could say the same of the threshold-image, considering it as a two-faced image, actual and virtual at the same time, so that there would no longer be a chaining of the real to the imaginary, but an indiscernibility of the two in a perpetual exchange. Always about the visible and the invisible, Merleau-Ponty, states:

No thing, no side of a thing, shows itself except by actively hiding the others, denouncing them in the act of concealing them. To see is as a matter of principle to see further than one sees, to reach a being in latency. […] The invisible is the outline and the depth of the visible. The visible does not admit of pure positivity any more than the invisible does [20].

“Reach a being in latency”: how to find better words to describe the “openings” that happen thanks to the exchange between on- and off-screen in the cinematic images? Two sides of the same coin: no radical distinction neither capital opposition, rather, a relationship of mutual implication, A “chiasm” as the French philosopher would name it. As stated in the introduction to the posthumous work of Merleau-Ponty, edited by Mauro Carbone

in the experience of vision, then, together with the visible and the invisible, even the “here” and the “elsewhere”, presence and absence, reality and imagination, up to the very space and time, lose their mutual distinction―a distinction that freezes them in opposites, or, at the most, in juxtaposed and, behind such a distinction, reveal themselves implicated in one another exactly as the invisible is impicated in the visible [21].

One needs only to recall, once again, Hiroshima mon amour and L’année dernière à Marienbad by Resnais, already mentioned at the beginning of this paper. The filmic image will have to be understood as something provisional, a path, not the goal; a mediation. But provisional does not mean empty, if anything “fillable”, in continuous tension, in a perpetual dialogue between the visible and the invisible, between the represented and the representable; a medium that exceeds its own limits, continually figuring and reconfiguring  a surplus and an absence. Or, to use the terms analyzed by Franzini in his Fenomenologia dell’immagine [22] between memory and imagination, perception and fantasy. A means for a viewer who is both, and at the same time, receptive and productive. An image that entails the possibility of mistakes, that sometimes induces them, especially because, if in one’s fantasy it is not possible to make mistakes (“if nothing is true, everything is possible,”  William Burroughs said), in the workings of one’s memory, this is possible: “mistake that arise from the reproduction as such, and, on this ground, plays the filmic image”. Memory and imagination: these are the open terms of the filmic image. To quote a few examples close to us, just think of works like Memento (C. Nolan), Fight Club (D. Fincher) or Eternal Sunshine of a Spotless Mind (M. Gondry).

Eternal Sunshine of a Spotless Mind (2004)

Eternal Sunshine of a Spotless Mind (2004)

An image that alludes to an invisible or not (yet) visible, whose traces it has already placed  whithin itself (through, for instance, the looks and the movements of the actors within the frame). An image that never stops to question us, and to mediate, thus becoming,  simultaneously, “instigator” and figure of our mind. Ambiguous, allusive, hybrid, metamorphic, simultaneous at times conciliatory as in a given idea of classic cinema, but more often (than not) aware of an absence, aware of “tolerating the absence of unity ” (as pointed out by the same Franzini, when he quotes Lyotard about the symbolic function of the artist). And here the references to psychoanalysis could be far too many but, perhaps, they would risk to take us astray. Suffice it to bring underline the analogy between the off-screen image and the unconscious; between the visible and the invisible part, between the exposed part and the dark one: ambiguous and yet capable to interact with, and force, that visible. An image, in any case, forced to be contaminated, syncretic, hybrid and in constant transition. Threshold-image, then, as a transition that entails the coexistence of opposites and, therefore, capable of positing itself as the very limit of meaning. In some ways aesthetic limit as, to a certain extent, it is able to set up a category other than that of a binary logic, “potential mechanism of escape from the grids imposed by meaning”.

Blue (1993)

Blue (1993)

This is the case with images that are able to become autonomous; images that lose their contours, to cross the threshold and  to dissolve, transfiguring themselves in color as in the case with Blue (Blue Film, Derek Jarman, 1993). What are the relationships between on- and off-screen in such a work? Everything happens off-screen or, rather, is it a case of  off-screen re-configuration? Evidently, we are dealing with threshold-images where everything, according to the words of the director, is “a fragment of a work without limits.” But if, beyond the limit, there is often only a void or a threat, in the limit there is also  complexity, exchange, richness. [23].
To experience such a threshold (movie-image or life as it might be ) means, finally, to perceive this space as a continuum, which is in perennial transformation. The idea of limit, finally, can not help but bringing us back to that of the limits inherent in representation itself.

If, as Bazin stated, love and death can not be represented as if not off-screen, Julio Bressane, the Brazilian director, in an interview published on Fata Morgana, recalled that this is not clearly an ethical limit but, rather, a difficulty of method and of time:

the cinema must show, must find pictures to show, the problem is that it is hard to do it […] it is impossible to show the images […] The time of observation, of representation […]  is no longer. […] It is difficult to represent images, not so much to represent with images [24]

Such a limit becomes almost an establishing one, thus reminiscent of the limit analyzed by Wittgenstein in his studies on language, therefore seen as a constitutive limit of man (“man can not get out of language as he himself inasmuch as a linguistic being, constitutes a  limit to himself”):

The limits of my language mean the limits of my world […] What we can not think, that we cannot think: we cannot therefore say what we cannot think [25].

So man himself lives in “liminality” and in liminality he finds his meaning (and his own his limitations). Wittgenstein talks of images as models of reality i.e.,  not as copies but as models mean to reproduce reality as structure, relationship of the parts. Basically between the image and that of which it is the image,  there should be a substantial correspondence (a logical form of correspondence):

That the elements of the picture are combined with one another in a definite way, represents that the things are so combined with one another. […] It is like a scale applied to reality [26].

Wittgenstein was not talking about movies, when he was thinking about images. He was talking of images of the facts: of a language to be applied to the world. It seems to me though, that the fact that these images might either be originated by the mind, or  reproduced through a device (as is the case with film) and in movement, does not change Wittgengstein’s concept.

By the end, every filmic image has a mental origin, and such origin lingers and extends through the device, so as to reproduce itself in the mind of the viewer, in the time of his own fruition, up to the point in which it crosses over to turn into memory (thus returning to be alive in the minds of those who experienced it and, perhaps, producing new images and so on). Although the philosopher claimed that rhetorical states such as tautology and contradiction are not pictures of reality inasmuch as meaningless, the fact that they have no meaning does not signify that they are meaningless (as Odifreddi rightly points out, in his commentary on the Tractatus).

Odifreddi’s statement authorizes us, to some extent, to maintain our parallelism, inasmuch as, if the threshold-image could be a model of logical contradiction (“p and not p”, that is to say, the image is both –at once– what we see and what we do not see) it will be equally acceptable to think of it as the model of the liminality of man. By the end, it does not matter if the image is true or false, it has meaning in and of itself,  as structure, logical concatenation of elements. It is a “demonstration” of meaning (it show a sense, it shows a logical form): the subject exhibits itself, does no say itself, Wittgenstein admits. What the filmic image shows, then, is its structure and its own limits. But, this is the paradox, if it is true that, by definition, it has, constitutionally an open structure, it cannot but overcome its limits. Threshold-images are then tools/models that thematize man and his limitations, in a perpetual self-reflective game and in a perpetual mise en abyme. Threshold-images as model-images to capture the correspondences and the limits of a research, which is,  at the very bottom, truly  tautological. To show, then, is tantamount to grasping this limit and, simultaneously, fathoming its possibilities. This is partially similar to what J.L. Godard did in his Histoire(s) du Cinema (1988-1998), where, through the implications of resemblance and approximation, he shows images as thresholds, through which to penetrate an osmotic space. But the same could be said of Vertov’s Chelovek’s kino-apparatom [27].

The Truman Show (1998)

The Truman Show (1998)

From what I argue above, it seems, therefore, that it is plausible to give a reading of off-screen space as a mimetic model in comparison with the experience of the senses (in effect it does not amount to a merely cinematic procedure). We indeed, see a portion of space, but hear sounds that come from outside of what we see (the passing of cars on the roads, the chirping of birds, etc.). We establish the boundaries and, at the very same time overcome them, and such a procedure is also the one generally followed by thought. The establishment of boundaries and their trespassing: this is art, this is  thought. This is man, this is cinema.


This article is an excerpt  from:
Dalpozzo, C. Fuori campo. Dentro e oltre l’immagine cinematografica, Edizioni, Padova, 2012.

[1] J.-L. Godard, 1960.
[2] J.-L. Godard, 1965.
[3] A. Resnais, 1959.
[4] A. Resnais, 1961.
[5] A. Resnais, 1963
[6] Deleuze, G. Cinema 2. The Time-Image, University of Minnseosta Press, Minneapolis, 1989, p. 179.
[7] Ibidem, p. XI
[8] Ibidem, p. 277.
[9] Ibidem, p. 292.
[10] Ibidem, p. 278.
[11] Ibid.
[12] De Gaetano, R. “Tre idee di limite, tre stati del cinema”, in Fata Morgana, n.5, 2008. My translation.
[13] S.M. Ejzenstejn, 1925.
[14] Deleuze, G. Cosa può fare un corpo? Lezioni su Spinoza, Verona, Ombre Corte, 2007, p. 130. In De Gaetano, R, cit. My translation.
[15] For this short survey on the concept of limit reference to the acts of the project Margin, threshold, boundary, limit, institutions, practices, theories (University of Siena) and published at: htm.
[16] Known the criticism formula that “every film is a thriller.”
[17] Donà, M. Abitare la soglia. Cinema e filosofia, Milano-Udine,Mimesis, 2010.
[18] Ivi, pp. 39-42. My translation.
[19] Not surprisingly, the memory has been defined by many as its prototype of the “threshold” as pointed out by the same Dona.
[20] Merleau-Ponty, M. Signs, Northwestern University Press, Illinois, 1964, pp. 20-21.
[21] Merleau-Ponty, Segni, Il Saggiatore, Milano, 2003, p. 14. My translation.
[22] Franzini, E. Fenomenologia dell’immagine. Al di là dell’immagine, Milano, Raffaello Cortina Editore, 2001. My translation. On the subject of the visible and the invisible, see, among others, the essay quoted Merleau-Ponty The Visible and the Invisible, Northwestern University Press, Illinois, 1968. For a cinematic perspective refer to De Gaetano, R. Il visibile cinematografico, Roma, Bulzoni, 2002. On the subject of transcendent, see, among others, to the study of Paul Schrader, Trascendental Style in Film, University of California Press, 1972.
[23] In biology these liminal spaces are called ecotones, dynamic spaces and essential to the production of life.
[24] Roberti, B.- Canadè, A. (a cura di) “Il limite come intervallo. Conversazione con Julio Bressane”, in Fata Morgana, n.5, 2008. My translation.
[25] Wittgenstein, L. Tractatus logico-philosophicus, Kegan Paul, Trench, Trubner & Co, London, 1922, 5.6; 5.61.
[26] Ibidem, 2.15; 2.1512.
[27] D. Vertov, 1929.

AA.VV. Margine, soglia, confine, limite: istituzioni, pratiche, teorie.
De Gaetano, R. “Tre idee di limite, tre stati del cinema”, in Fata Morgana, n.5, 2008.
De Gaetano, R. Il visibile cinematografico, Roma, Bulzoni, 2002.
Deleuze, G. Cinema 2. The Time-Image, University of Minnseosta Press, Minneapolis, 1989.
Deleuze, G. Cosa può fare un corpo? Lezioni su Spinoza, Verona, Ombre Corte, 2007.
Donà, M. Abitare la soglia. Cinema e filosofia, Milano-Udine, Mimesis, 2010.
Franzini, E. Fenomenologia dell’immagine. Al di là dell’immagine, Milano, Raffaello Cortina Editore, 2001.
Merleau-Ponty, M. Signs, Northwestern University Press, Illinois, 1964.
Merleau-Ponty The Visible and the Invisible, Northwestern University Press, Illinois, 1968.
Roberti, B.- Canadè, A. (a cura di) “Il limite come intervallo. Conversazione con Julio Bressane”, in Fata Morgana, n.5, 2008.
Schrader, P. Trascendental Style in Film, University of California Press, 1972.
Wittgenstein, L. Tractatus logico-philosophicus, Kegan Paul, Trench, Trubner & Co, London, 1922,

À bout de souffle, (1960, J. – L. Godard)
Pierrot le fou, (1965, J. – L. Godard)
Hiroshima mon amour (1959, A. Resnais)
L’année dernière à Marienbad (1961, A. Resnais)
Muriel, ou le temps d’un retour (1963, A. Resnais)
Professione Reporter (1975, M. Antonioni)
Le Mépris, (1963, J. – L. Godard)
Bronenosec Potëmkin, (1925, S. M. Ejzenstejn)
Memento (2000, C. Nolan)
Fight Club (1999, D. Fincher)
Eternal Sunshine of a Spotless Mind (2004, M. Gondry)
Blue (1993, Derek Jarman)
The Truman Show (1998, P. Weir)
Histoire(s) du cinema (1988-1998, J. – L. Godard)
Chelovek s kino-apparatom, (1929, D. Vertov)

Cristiano Dalpozzo teaches “Media History” and “Film History and Criticism” at the Salesian University of Venice (IUSVE). He works with “Osservatorio tv” (independent research project about contemporary tv series) and he is a member of the Scientific Committee of the journal studies “IUSVEducando.” Among his publications in volume include Introduzione al linguaggio cinematografico (2013), Fuori campo. Dentro e oltre l’immagine cinematografica (2012), Michel Gondry. Il gioco e la vertigine (2011).