Note 3 – Fig. 7. Canopy Triptych, 2015. Screenshot from algægraphic 59’ film, animated at 25 images/second. A section of the composition situated to the right of the triptych © Lia Giraud

Life revealed by technics: re-examining the human vision and its models of representation

Our knowledge of the living world is inextricably linked to tools of observation (more specifically, optical and cinematic instruments) which have continually challenged our human vision and our concepts of the surrounding world by the exploration of our “visual unconscious”19. Whilst microscopy heralded the “discovery under what is manifest, of richness under poverty, and of extraordinary things lying beneath the familiar”20, time-lapse or slow motion enable us to access “realities that any natural vision has no knowledge of”. Whilst these tools of augmented perception — be they spatial or temporal — bring us face to face with the limits of our human vision, they also open up new spaces and rhythms that can be shared beyond the human sphere.



Note n°3 – Canopées (Canopies), 2014-2015

Between biological and particulate movements, the algal cells of the Canopées (Canopies) organise themselves dynamically to respond to the variations in light of their environment. This biological matrix is here subjected to different processes of pictorial disruptions. The negative used evokes an artificial imaginary ecosystem installed on the site of Batignolles in Paris, under constructions at the time. These images explore the relationships between the biological, technical and symbolic elements that structure our complex experience of the “milieu”.

The ”live” Canopée (Canopy) installation enables the continual creation of algaegraphs within an aquarium. An associated software programme (exposer/flasher) controls the alternating projection of a negative and a rapid flash of light on the surface of the aquarium: with each flash, a photograph of the position of the algae is taken and is added to an animation that is shown on loop. This accelerated (time-lapse) film enables the formation of the image and its evolution to be visualised virtually in real time, demonstrating the transformations of the canopy over time. The programme also enables the negative to be modified over an extended period of time and thus to tell the story of the evolution of the algaegraphic image: Here we are witnessing the creation of the image, followed by its destruction.
The phases of formation and destruction of the landscape are scripted and reproduced according to three variations in the triptych video Canopy Triptych.

  • Note #3 – Fig. 6. <i>La Canopée</i> (<i>The Canopy</i>), 2015, Inkjet printing on baryta paper, 61x300 cm. Photographic composition from the 70 phases of the disappearance of the landscape formed by the microalgae © Lia Giraud</br>
    Note #3 – Fig. 6. La Canopée (The Canopy), 2015, Inkjet printing on baryta paper, 61x300 cm. Photographic composition from the 70 phases of the disappearance of the landscape formed by the microalgae © Lia Giraud


    La Canopée (The Canopy) is a photographic mosaic, produced with the assistance of a serial shot apparatus equipped with a high magnification lens. The image retraces the 70 phases leading to the gradual disappearance of the figurative space.

  • Note 3 – Fig. 7. <i>Canopy Triptych</i>, 2015. Screenshot from algægraphic 59’ film,  animated at 25 images/second. A section of the composition situated to the right of the triptych © Lia Giraud
    Note 3 – Fig. 7. Canopy Triptych, 2015. Screenshot from algægraphic 59’ film, animated at 25 images/second. A section of the composition situated to the right of the triptych © Lia Giraud
  • Note 3 – Fig. 8. <i>Entropie V (Entropy V)</i>, 2015, Hybrid UV ink printing on plexiglass, 35x60 cm. Photographic triptych created with the help of a scanner; the final image  corresponds to the interpretation of the figurative landscape by the microalgae © Lia Giraud
    Note 3 – Fig. 8. Entropie V (Entropy V), 2015, Hybrid UV ink printing on plexiglass, 35x60 cm. Photographic triptych created with the help of a scanner; the final image corresponds to the interpretation of the figurative landscape by the microalgae © Lia Giraud


    The series of the Entropies reuses five details of the composition. The landscape’s dissipation is carried out using a scanner whose mode of acquisition leads to a behavioural modification of the cells, which reorganise themselves according to their own rules of composition.

    Created by: Lia Giraud
    Scientific collaboration: Claude Yéprémian
    Serial system of “La canopée (The Canopy)”: Olivier Gade
    Production assistant: Olivier Goulet
    Programme “Canopée en live (Live Canopy)”: Benoit Verjat
    Production: EnsadLab, Paris Sciences et Lettres, Laboratoire Cyanobactéries Cyanotoxines, Environnement (MNHN), Espace Pierre-Gilles de Gennes (ESPCI).

Enlargement is often used to access the structure of the algægraphic images. The size of the print of La canopée (The canopy) and its production process, allow two relationships of scale to coexist in the same perceptual space. The macroscopic vision of the landscape and the microscopic vision of the algal grain can thus be perceived by a simple shift in position of the spectator who stands in front of the work. Here, this consists in countering a  ‘separatory power of the eye’ which decreases according to the observation distance. By creating distance between the observer and the image, it quickly becomes difficult for our eye or for a lens to distinguish the grain which composes the image. In the example of the algaegraphs, the cells merge into one uniform mass. This natural synthesis somehow makes us look past the intrinsic complexity of the objects that we observe. Re-establishing continuity between our ‘reductionist’ spatial scale and the infinite richness of the invisible world allows for the presentation of a new reticulated and shared vision of the world.

To observe the life processes of the algægraphic images during their formation, a specific tool, called an “exposer/flasher” had to be developed. In the Canopée en live (Live Canopy) installation, this programme enables the juxtaposition of two temporalities: that of the real time formation of the algaegraphic image exposed to the light of the negative and that of a time-lapse animated film which instantaneously collates the different photographs taken during the experiment (Canopy triptych). By speeding up the formation of the algaegraphic image by around 25 times, the film enables the movement that is virtually imperceptible to the naked eye to be viewed. When one looks at the aquarium in which the image is formed, we are in effect  at the limits of a perceptive state where it is difficult to determine if the image is static or animated by a movement.

In their publication, Teresa Castro, Perig Pitrou and Marie Rebbechi explain how the invention of cinema has contributed to the animation of plant life by enabling “the representation of movements that are for the most part invisible to the naked eye and preserving a spatial memory of their variation over time”21. According to them, the use of this technical apparatus allows for human and non-human rhythms to be synchronised and for the creation of “a homogenous framework” (Perig Pitrou), beyond the “heterotemporality of plants” (Marie Rebecchi). This “technical mediation” appears to be not only indispensable to go beyond the limits of our human vision but it also challenges its predominance. As Teresa Castro notes, “film – through the mediation of a mechanical and non-human agent (the camera) – potentially enables the viewer to reconfigure his or her anthropocentric perspective and open themselves up to non-human beings”.

The form that the image takes here remains scripted by the choice of several negatives: the algal image tends towards the creation of a landscape that has been previously selected and essentially human. Seeking to thwart this mechanism, the Entropies  project reverses these roles: the algægraphic image formed will then be exposed to the uniform light of a scanner, which, while enabling the rendering of the image, results in the negation of this human composition for an image that is specifically algal. Faced with the light, the microorganisms can move freely and regain possession of the pictorial space by offering their own interpretation of it.

This tendency towards abstraction, expressed here by the display of pointillist motifs or abstract volutes, has been a first means of opening myself up to new non-human biological representations.  In what ways can these “natural images” disrupt the parameters of our anthropocentric representations? Does the non-figurative image allow this essence of the medium itself to be rethought, by re-evaluating our view of its fundamental constitution or by demonstrating its production processes22 ? Behind the living forms, how can the notion of representation be reinvented?

The acheiropoietic image: re-examining forms of image through mimetic extrapolation

What characteristics would these images have without human intervention (Acheiropoieta)? Those thinkers who take an interest in the forms produced by nature or by machine suggest hypotheses that are in accordance with the autonomous production of image as it is conceived in the Dialogical Dreaming installation.

Notice n°4– Dialogical Dreaming, 2017

This installation explores the operational and symbolic qualities of the perceptive interplay in order to offer a new insight into the living being/mechanical relationship: these two entities, often thought of as isolated, even opposing, could they be united here in the form of a sentient and autonomous interaction? The installation is made up of three systems that execute the process of the apparition and evolution of a living image. Each image is the result of a “dialogical dream” between a strain of microalgae, photosensitive living cells that give tangible form to the image, and an artificial neural network (Deep-dream) which guides the creation of the produced forms.

  • Note 4 – Fig. 9.<i>Dialogical Dreaming</i>, 2017. Example of a dream produced by the Deep Dream software, from the volutes formed by the microalgae in the aquarium.8_Dialogical Dreaming_Résultat du Deep Dream ©Lia Giraud
    Note 4 – Fig. 9.Dialogical Dreaming, 2017. Example of a dream produced by the Deep Dream software, from the volutes formed by the microalgae in the aquarium.8_Dialogical Dreaming_Résultat du Deep Dream ©Lia Giraud

    The central apparatus (Dialogical dream) consists of a technical-sculptural system which enables the production of the microalgae and their movement towards a vertical aquarium in which the living image is formed. The exposer/flasher programme is used here without an initial negative: the experiment begins when the first photograph of the aquarium is taken and sent to a Deep-dream programme which will interpret the forms produced by the microalgae. This “dreamed” image will thus be transformed into a negative and re-projected onto the microalgae, who, in turn, will interpret the image produced by the machine.

  • Note 4 – Fig. 10. View of the <i>Dialogical Dreaming</i> installation, Presentation of the thesis SACRe/PSL à l’EnsAD, December 2017 © Lia Giraud
    Note 4 – Fig. 10. View of the Dialogical Dreaming installation, Presentation of the thesis SACRe/PSL à l’EnsAD, December 2017 © Lia Giraud

    Another photograph of the aquarium is taken, interpreted by the programme, then projected back onto the aquarium and reinterpreted by the algae. The experience continues endlessly and autonomously via this interplay of feedback, generating a film whose evolution is unknown in advance.

In his study of animal forms23, Adolf Portmann states that a functional analysis cannot explain their diversity. Some forms must be considered as “unaddressed” and as only possessing a “representative value”. The first significant property of the animal form would thus  be that of “revealing the particularity of  such a species, to immediately demonstrate this particularity through its form”. This sense of “self-representation” should be considered as a basic property of life24. For his part, Roger Caillois focuses on mimetism (mimicry)25 and in particular on homomorphism (mimicry of form) by considering it as a veritable three-dimensional photograph, “a sculpture-photograph” of the milieu, in which the living being evolves and projects back onto itself26. This tendency of the living being to imitate its surroundings is considered by the author as an attempt of “assimilation to the milieu”. From copies to reproductions, the world leans towards a certain uniformity which brings to mind some artistic works which focus on the standardisation of digital images27.

In Dialogical Dreaming, the reference image which launches the experiment is that of the microalgae that express their most common “apparition”: Brownian movements, in the form of volutes, constitute in effect their most immediate and most “unaddressed “ mode of representation. The intervention by the  Deep-Dream software, which uses deep learning technology (convolutional neural network) “dreams” the image formed by the microalgae. But as Trevor Paglen notes, 28 “These neural networks cannot invent their own classes ; they are only able to relate images they ingest to images that they’ve been trained on”. It is thus from the directory of the images available on the internet that the software can collect the material of its dream, to finally match the volutes of the microalgae to the form that it deems to be the most similar, the most closely imitated. That is to say that the software primarily selects biological and technical forms that it hybridises as it sees fit. One could say that in a sense, it “assimilates the iconic milieu” to most accurately copy the appearance of the microalgae. In the same manner, the cells contained in the aquarium try to “assimilate the luminous milieu” of the negative to fully imitate it.

In this infinite interplay of reproduction, which mobilises both iconogenesis and mimicry, we can perhaps experience first-hand what is implied by Peter Szendy in his research around the “non-human iconomy”29, which intersects the thinking of Simondon and Caillois. On the one hand, the non-human image would be the result of a process of iconogenesis that leads to the creation of an original image because it is subjectively interpreted by the microalgae or by the machine. On the other hand, its tendency towards imitation, (mimesis of the negative or mimesis of the closest image offered up by Google), maintains the image in an eternal recurrence of the same. This “heterochrony of images” is visible in the video produced within Dialogical Dreaming, where the image seems to be caught in a tension between these two contradictory temporalities: by continually reinventing itself, the representation seems to infinitely recycle the same formal repertoire. The algaegraphic image thereby appears trapped in a “time differential which pulls and stretches it, gives it its tone or its tonality,  according to a dissonance that could never be resolved or resorbed.”.

Autonomous images: re-examining the action, the author and the intended audience of the image

In the algaegraphs installation, the reproducible gesture of the photographic release mechanism, or the   programmatic management of the software, design a second space of production of “non-human” machinic images. By becoming autonomous, the image could lose sight of its purpose and maybe also its author. In a text written in 2016, Trevor Paglen explains that “The vast majority of images are now made by machines for other machines, and it’s pretty rare that humans are in the loop”. For him, it is this change in audience and this system of human invisibility which is revolutionary in the advent of digital images. These “invisible images that are actively watching us, poking and prodding, guiding our movements”, resemble the apparatus described by Michel Foucault, and later on, Giorgio Agamben30. According to Agamben, the apparatus is “anything, which has in some way, the capacity to capture, guide, determine, intercept, model, control and secure the gestures, behaviours, opinions or discourse of living beings.” In this sense, the algaegraphs installations literally put the notion of the apparatus to work, by setting a living being and a technical system against each other. However, Immersion and Dialogical Dreaming partially subvert this relationship of authority, because it is also the microalgae themselves that orient the film editing or shape the dreams of the “vision machine”. These life forms perhaps help us to “unlearn seeing as human beings” by initiating a mimetic dialogue with the machine, that has its own sentient language, and its own informative syntax, based on the quantity of light, the coding of the densities and the reaction to a given event, etc.

Vilèm Flusser considers the photographer as a “functionary of the apparatus” who is denied their creative power by the programmatic nature of the photographic apparatus31. In Dialogical Dreaming,  a certain “inactivity” of the author is indeed apparent: the technical and life processes interact to find a form of autonomy and self-sufficiency. When the algaegraphic experience is initiated, the work is produced in an autonomous manner and the artistic gesture is thus displaced towards the setting-up of the initial conditions. Anybody could subsequently launch the experiment of the image-process and obtain a similar result.

Like Flusser, Henri Van Lier observes that the person using an apparatus “has been subordinated and is now often only facultative”, to an extent where it becomes “difficult to imagine man as the microcosm […]. Indeed, we have left behind anthropocentrism and humanism in favour of a more technical, universal, biological, semiotic and indicial perspective”32. Contrary to Flusser, Van Lier considers this human relinquishment as leaving open the possibility of a “cosmological” shift, although the philosopher remains evasive on this point. In more concrete terms, Marc Lenot analyses these images created without human participation as a method that enables an escape from the grasp of the apparatus. Those experimental photographers who give over the task of producing the images “at random” to the living being or to the machine, thwart the rules of the apparatus by reaffirming a positioning beyond its grasp33 . Returning the image-producing apparatus to its own bio-technical autonomy, pushing the mimetic interplay of the image to the absurdity of a self-reference which loops back on itself, are methods to “thwart the rules of the apparatus”; or, as Agamben puts it, to “profane” the image-producing apparatus. Both Immersion and Dialogical Dreaming can in fact be considered as “anti-apparatus” insofar as the production of the image is no longer, neither oriented nor determined, nor controlled. The “ritual” offered here consists in observing an autonomous and evolutionary life of the image which, like us, is the result of a permanent interaction with the milieu. A dialogical milieu in which living and mechanical beings act, as much as they draw on the imaginative and oneiric matter of their existence.

Lia Giraud
To be continued
Translated from French by Jacqui Chappell

19 Walter Benjamin, L’œuvre d’art à l’ère de sa reproductivité technique (The work of art in the age of mechanical reproduction), Allia, Paris, 2009
20 Gaston Bachelard, “l’obstacle animiste” (The animist obstacle), La formation de l’esprit scientifique (The Formation of the Scientific Mind), Vrin, Paris, 1938..
21 Teresa Castro, Perig Pitrou and Marie Rebecchi (dir.)s, Puissance du végétal et cinéma animiste (The power of plants and animist cinema), Les presses du réel, Dijon, 2020.
22 On this topic, please refer to the catalogue of the triple exhibition at the CPIF, Frac Rouen and Micro-onde, La photographie à l’épreuve de l’abstraction (Photography to the Test of Abstraction), Hatje Cantz, Berlin, 2020
23 Adolf Portmann, La forme animale (Animal Forms and Patterns), La bibliothèque, Paris, 2003
24 He thus writes : “Light constitutes a particular sphere in the domain of sensory effects. It is probable that subsequent research on the relationship between light and life, life and light,  should highlight more clearly this specificity of the elementary nature of light effects and differentiate it from other sensory effects which depend upon molecular effects” Ibid. p.279
25 Roger Caillois, le mythe et l’homme (Myth and Man), Gallimard, Paris, 1938
26 He writes: “Morphological mimicry could be, after the fashion of chromatic mimicry, an actual photograph, but of the form and the relief, a photography on the level of the object and not on that of the image, a reproduction in three-dimensional space with solids and voids : sculpture-photography or better, teleplasty if one strips the word of any metapsychical content.” Ibid. p.10
27 On this question, the work of Corinne Vionnet, Trevor Paglen, Bertrand Dezoteux but also the proximity in form of the Seconde nature (Second nature) installations (Grégory Chatonsky, 2019) and Umwelt (Pierre Huyghe, 2018) which both analyse convolutional neural networks (CNN).
28 Trevor Paglen, “Invisible images (Your Pictures Are Looking at You)”, The new inquiry, 8 December 2016. URL :
29 Peter Szendy, Pour une écologie des images (Towards an ecology of images), Éditions de minuit, Paris, 2021.
30 Giorgio Agamben, Qu’est-ce qu’un dispositif (What is an apparatus?), Rivages, Paris, 2014
31 Vilèm Flusser, Pour une philosophie de la photographie (Towards a philosopy of photograpy), Circé, Belval, 2004
32 Henri Vanlier, Op.cit.
33 Marc Lenot, Jouer contre les appareils (Playing against the apparatus), Photosynthèses, Arles, 2017