Flusser: The Photographer Inside the Programme
Apparatus, programme and functionary, quoted with page numbers. Also the discovery that the English edition everyone cites is a third party translation, not Flusser's own.
Date
Category
Writer
The book was not published by a philosophy press. Für eine Philosophie der Fotografie came out in 1983 from European Photography in Göttingen, a small imprint run out of the photography world, and it is short: the first English edition, published by the same house in 1984, runs to 62 pages. Vilém Flusser, born in Prague on 12 May 1920, living by then in Robion in Provence after three decades in Brazil, was sixty-three. He died in a car accident near Prague on 27 November 1991.
That 62-page book is now quoted constantly by people arguing about generative tools, usually in the form "Flusser said we are just functionaries of the apparatus". He did use that word. The sentence around it is not the sentence people think it is, and there is a second problem underneath: the English edition almost everyone quotes is not the English Flusser wrote.
Four originals, and the one that went out of print
Marc Lenot, writing in Flusser Studies issue 30, counts four original versions of the book, published by Flusser himself: German in 1983 and again in 1989, English in 1984, Portuguese in 1985. The two German texts are nearly identical, differing by about 300 words. The Portuguese is a substantially different book: Lenot calculates "about 2300 words that have disappeared from the German version (or 10.5% of it), while about 1450 new words have appeared in the Portuguese version".
The English edition in every design-school reading list is the Reaktion Books one, translated by Anthony Mathews and published in 2000. It is a translation from the German, not a reprint of the 1984 English. Lenot's complaint is exact: "the only book Flusser wrote in English (and his only book in English published in his lifetime) is no longer available, and English-speaking readers no longer have access to his original text." He asks whether readers should "at least be informed that they are reading a third-party translation while an original Flusser text exists in English". Nothing in the Reaktion edition tells them.
Lenot also lists errors carried into the 2000 text: a misattributed lyric, Einstein credited with a single equation rather than several, a qualifying phrase in chapter 7 dropped. I cannot check these against the 1984 edition, which I have not been able to obtain. There is also a small bibliographic puzzle I could not resolve: Monoskop lists the 1984 European Photography edition as edited by Derek Bennett, while Lenot treats it as one of Flusser's own original versions. Those are compatible if Bennett edited Flusser's English, but I have not seen the book's own statement.
Everything quoted below is from the 2000 Mathews translation, with its page numbers, and should be read as such.
Technical images
The starting move is a distinction between two kinds of image. Traditional images are made by a hand working on a surface. Technical images are not: "Technical images are produced by apparatuses" (p. 21). The consequence Flusser cares about is not aesthetic but epistemic. Because a technical image is produced by a device that appears to be recording, it gets received as though it were a window rather than a statement. A painting is obviously somebody's account of something. A photograph passes as the thing.
This is why the book insists that a technical image is a coded surface requiring decoding, and that the coding is done by the apparatus rather than by a person. Whatever you think of the rest of the argument, this part has aged extremely well, because it identifies the precise reason automated images are dangerous: not that they lie, but that they arrive wearing the clothes of a measurement.
Apparatus is not a synonym for technology
Flusser's apparatus is a narrow technical term and most paraphrases widen it into meaninglessness. He starts with etymology: the Latin word comes from apparare, to prepare, so "an 'apparatus' would be a thing that lies in wait or in readiness for something" (p. 21). Not a thing that does work. A thing that waits, already prepared, for you to arrive.
He then separates it from the two categories it is usually confused with.
Category | Human position | What the human supplies |
|---|---|---|
Tool | Surrounded by it | The work and the intention |
Machine | Standing at it | Attendance and the intention |
Apparatus | Inside it | Selection from what it already holds |
The sentence that sets this up is one of the best in the book: "Unlike manual workers surrounded by their tools and industrial workers standing at their machines, photographers are inside their apparatus and bound up with it" (p. 26). And he is explicit that the difference is categorical, not one of degree: "No-one can own apparatuses in the sense that human beings program apparatuses for their own private purposes. Because apparatuses are not machines" (p. 29).
An apparatus, for Flusser, is a black box. "For they know how to feed the camera (they know the input of the box), and likewise they know how to get it to spit out photographs (they know the output of the box)" (p. 27). Competence at the edges, ignorance in the middle. And its content is a programme, which he treats as a finite combinatorial space: "the camera is programmed to produce photographs, and every photograph is a realization of one of the possibilities contained within the program" (p. 26). He calls the apparatus "a complex plaything, so complex that those playing with it are not able to get to the bottom of it; its game consists of combinations of the symbols contained within its program" (p. 30).
What functionary actually means
The word is not an insult, and this is where the popular paraphrase does real damage. A functionary is someone who operates a function: someone who acts within a set of permitted operations that they did not write and cannot inspect. Flusser's formulation is that people and device "merge into a unity", and that it is therefore "appropriate to call photographers functionaries" (pp. 26 to 27).
The condition he describes is genuinely double-sided. Functionaries "control a game over which they have no competence" (p. 27). They govern the apparatus through its exterior, its input and output, and are governed by it through the impenetrability of its interior. Both halves are true at once. The photographer is not a dupe. Nor is the photographer an author in the pre-photographic sense. The circuit closes in a sentence worth memorising: "the apparatus functions as a function of the photographer's intention, this intention itself functions as a function of the camera's program".
Playing against the camera
The famous move is the chess analogy, and it is more specific than it is usually reported. "Photographers are not working, they do not want to change the world, but they are in search of information" (p. 26). Then the turn: "Yet photographers do not play with their plaything but against it" (p. 26).
Photographers endeavour to exhaust the photographic program by realizing all their possibilities.
Playing against the apparatus does not mean smashing it or refusing it. It means treating the programme as a space to be exhausted rather than a set of recommended paths to be walked, and hunting for the combinations nobody has realised yet: "photographers attempt to find the possibilities not yet discovered within it" (p. 26). The bodily version is delightfully unromantic: "They handle the camera, turn it this way and that, look into it and through it" (p. 26).
Flusser's accounting of this is the part I find most useful. Every photograph either is or is not informative, and the ledger has two columns: "With every (informative) photograph, the photographic program becomes poorer by one possibility while the photographic universe becomes richer by one realization" (p. 26). Images that repeat what has already been realised he calls redundant: "Such images are 'redundant': They carry no new information and are superfluous" (p. 26).
Now apply it to software with a settings panel
The translation to any modern tool is almost embarrassingly direct, and Flusser saw the economics coming. In 1983 he wrote that "it is not those who own the hard object who have something of value at their disposal but those who control its soft program", and observed that "the hard side of apparatuses, the hardware, is getting cheaper all the time, the soft side of them, the software, is getting more expensive" (pp. 29 to 30). That is a decent one-sentence description of the last forty years of the industry, written before most of it happened.
If you work through a piece of software with presets, defaults and a settings panel, you are in the position he describes with unusual precision. You know the input and the output. The interior is not available to you and, past a certain complexity, would not help if it were. Your intention is real and it is shaped by what the interface makes easy. The preset list is a set of realised combinations someone else chose to expose.
Flusser's distinction between informative and redundant work gives a usable test, which I now apply to our own output more or less as written here.
There is a harder consequence of the same idea that I resisted for a long time. If redundancy is defined against the programme rather than against your own back catalogue, then originality within a studio is not the right measure. A layout can be new to us and old to the programme. The question Flusser's accounting forces is not whether we have done this before, but whether the tool has, ten million times, at the point where its defaults sit. That is a much less comfortable question and it is answerable, roughly, by looking at what the tool ships as examples.
Two places where the analogy breaks, and I would rather say so than pretend the fit is perfect.
The programme is no longer finite in any useful sense
Flusser's camera has a combinatorial programme: discrete settings, enumerable states, a space that could in principle be exhausted. A modern generative tool does not have that. Its output space is not a list of combinations in the sense that aperture by shutter by focal length is a list. "Exhausting the programme" stops being a finite project and becomes a direction of travel. That is a weaker claim than his, and anyone who quotes the exhaustion line about a language model should notice they have changed the subject.
There is nobody else in Flusser's picture
The book has an apparatus and a functionary and essentially no third party. Real practice has clients, colleagues, users, budgets and deadlines, and a great deal of what looks like submission to the programme is submission to a schedule. When a studio ships the default, the apparatus is rarely the reason. Friday is the reason. Flusser gives you no vocabulary for that, and borrowing his to describe it is flattering ourselves.
Three things he did not say
He did not say the camera takes the photograph and you do not. He said your intention and the programme are mutually conditioning, which is a harder and more interesting claim.
He did not say apparatus means technology. It means specifically a prepared black box you operate from inside, which is why he takes the trouble to distinguish it from tools and machines.
He did not say that using presets makes you a functionary. Being a functionary is the structural condition of working through any apparatus at all, including when you are fighting it. The choice is not between functionary and free agent. It is between a functionary who realises the programme and one who plays against it.
The part I keep returning to is the smallest. An apparatus is a thing that lies in wait, already prepared. Every default in every tool we use is a decision somebody else finished making before we arrived. Playing against it starts with noticing that it was a decision.
Sources
Vilém Flusser, Towards a Philosophy of Photography, trans. Anthony Mathews, Reaktion Books, London, 2000 (original German Für eine Philosophie der Fotografie, European Photography, Göttingen, 1983). All page references above are to this edition. Scanned text: lab404.com/3742/readings/flusser.pdf
Vilém Flusser, Towards a Philosophy of Photography, selected chapters, Reaktion Books. Second scanned copy used to corroborate the passages on playing against the apparatus and on exhausting the programme. ajourneythatwasnt.org/wp-content/uploads/2022/01/w04_selection-from-towards-a-philosophy-of-photography_flusser-1983.pdf
Marc Lenot, "Critical textual analysis of Towards a Philosophy of Photography by Vilém Flusser", Flusser Studies, issue 30. Source for the four original versions (German 1983 and 1989, English 1984, Portuguese 1985), the word counts of the divergences, and the criticism of the 2000 Reaktion translation. flusserstudies.net/sites/www.flusserstudies.net/files/media/attachments/lenot-critical-textual-analysis.pdf
"Für eine Philosophie der Fotografie (1983)", FlusserWiki. Source for the German first edition and the 1984 English edition, both from European Photography, Göttingen. wiki.flusser.club/doku.php?id=fuer_eine_philosophie_der_fotografie
"Vilém Flusser", Monoskop. Source for birth on 12 May 1920 in Prague, death on 27 November 1991 near Prague in a car accident, residence in São Paulo, Rio de Janeiro, Meran and Provence, and the listing of the 1984 English edition at 62 pages, edited by Derek Bennett. monoskop.org/Vilém_Flusser
Related reading
Latest Articles.
© Moc Mallen Studio
Thoughts, ideas, and perspectives on design, simplicity, and creative process.
