A Flusserian Interpretation

When looking towards the future of society, Vilém Flusser anticipated a society preeminently characterised by the “dominance of technical images.” Based on their properties, Flusser envisioned two very different possibilities for the character and structure of this future society shaped by the technical image: “Taking contemporary technical images as a starting point, we find two divergent trends. One moves toward a centrally programmed, totalitarian society of image receivers and image administrators, the other toward a dialogic, telematic society of image producers and image collectors.”¹⁾ It is this prophetic sense that is without doubt an essential part of the contemporary appeal to Flusserian thought. And in fact, Flusser ́s late works did appear on the verge of a series of technical developments that have to be understood as major changes to our societal communication structures. Yet our present reality, the Internet, does resemble neither of the two potentials envisioned in Flusserian theory. Instead, when observing the present-day global Web and its culture, the supposedly “divergent” tendencies are actually found in a more intertwined way that is difficult to disentangle. On the one hand, we find a communication structure enabling dialog between the most diverse kind of people and communities, and giving access to almost unlimited amounts of information; while on the other, confronting us with a mainstream culture that, compared to the times of linear media, has not lost any of its dominance, it’s just truly global now. A user on YouTube does not need to be a great cultural pessimist to get a feel for Flusser ́s concept of informational entropy, as they are being confronted with a virtually endless stream of recommendations like “Lamborghini Vs World ́s Largest Shredder” or “Hydraulic Press Vs Lamborghini” by one of YouTubes most successful creators Mr.Beast. The dominating technical images arising from the web did not open society to new horizons. As cultural theorist Mark Fisher put it:“It’s easier to imagine the end of the world than the end of capitalism.”²⁾ And not only in a metaphorical cultural sense has capital been able to keep its reign in the digital. It is still power that structures the Web. The agenda of the most cynical television producers from yesterday is greatly surpassed by the feed of technical images collectively created by the users of platforms like TikTok.

How does the collective dialogue of society culminate in such an entropic mass culture? From a Flusserian perspective, this contradiction of our present situation seems difficult to explain. After all, Flusser emphasises the priority of the technological as the only culturally formative factor and expresses himself dismissively against approaches like those of the Frankfurt School. Therefore, the writer Richard Woodall concludes that in our modern ambiguous situation it is the blind spots in Flusserian thought that show their importance.³⁾ Even if this criticism is justified, this text still represents an attempt to explain internet culture and its collectively created images in a technological, thus Flusserian way. In the same way, the thesis of this text claims that the entropic character of the images created by the dialogues of the web does lie in an increasing coordination of the dialogues through apparatuses. In the form of a stage model typical of Flusserian style, I illustrate here the history of the Internet as a dialogue increasingly coordinated by apparatuses.

A decentralised network enabled via common protocols

In its basic architecture as a global distributed Computer network whose communication is enabled by common protocols, the advent of the Internet represents in fact a technological leap towards the utopic Flusserian vision of society characterised by telematic dialogue. It constitutes an infrastructure enabling two-way telematic communication in all forms of technical images. As such, what phenomena of web-culture (technical images) can be derived from this basic architecture, coming from a Flusserian perspective?

By enabling all of society to share technical images of any form, early internet technology and culture did definitely put former authorities such as tv broadcasters or media publishers into question. Notable for this conflict are especially the debates and fights concerning file sharing sites, web spaces serving the free exchange/ dissemination of copyrighted material. Yet, as in the case of pirated movies, the technical images shared on these pages still originated from the centres of pre- telematic mass culture. The major liberation of telematics, however, lay for Flusser not only in the liberation from the previous broadcasters, but also from the entropic technical images transmitted by them. If the free network is only used to spread the images of mass media even further, society has not gained any freedom. Telematics should break the feedback loop of mass culture and so enable truly novel images. An internet-cultural phenomenon that probably comes closest to the Flusserian vision of images emerging from telematic play is that of the early internet meme. Images that can be the product of a long chain of creators, a product of a dialogical consensus. In them, a communication structure is depicted in which people dialogue with each other through images to produce ever new information, ever more improbable situations. As such, we can summarise the early web culture as a culture still impacted by the images of the pre-telematic culture, but where the mood of society is fundamentally shifting. Previously silent consumers are increasingly becoming active “players” engaged in the creation of the new. Society’s excitement is directed less and less towards the images of mass culture and more towards a collective creation of new images.

Dialogue coordination apparatuses


Thus far, we could observe an early internet culture, captivated by creative enthusiasm, whose parallels to Flusser’s vision of a telematic society are hard to oversee. For Flusser, the telematic society, caught in the enthusiasm for creation, represents an attempt to oppose entropic mass culture through collective creation. Flusser notes that


the telematic society would distinguish itself from earlier societies only insofar as its cerebral-net character has become conscious, enabling us to start consciously manipulating the net structure. The telematic society would be the first to recognize the production of information as society’s actual function and so to systematically foster this production: the first self-conscious and therefore free society.⁴⁾

A society striving towards the creation of “images and music we never dreamed of, drawing on a wealth of information never dreamed of”⁵⁾ , all done by synthesising its information through telematic networking. It is the central thesis of this text, that this societal process of systematically synthesising information, has indeed taken place in some form, but also has provoked technological developments, not fully foreseen by Flusser in their importance. So does the telematic mission confront a society that, according to Flusser, has been zero-dimensionally dispersed by the previous mass media. Only networking can save the dispersed masses of postmodernity. “They can be ‘informed’.” From isolated people “new groups can be formed”.⁶⁾ In zero-dimensions, where dialogue between every individual is technologically equally possible, a society striving for the informative, therefore improbable images, consequently, desires also the informative, therefore rather improbable dialogues, the improbable combinations. However, how can the zero-dimensional dispersed society coordinate itself into such informative dialogues? I argue that it was apparatuses, that only first allowed for an efficient coordination of this dispersed society into informative dialogues.

It is true that all combinations are foreseeable in principle, but some are more probable than others. It was the improbable combinations (the informative ones) that were wanted, and they only occur by blind chance after very lengthy—astronomically lengthy— computation. So, second, the play of pure chance had to be accelerated to secure the desired combinations within a human time frame. This, then, is automation: to build an apparatus that speeds up chance events and to prescribe (program) it to stop when the desired coincidence has occurred.⁷⁾

The great dialogues of the Web are neither the product of random chance nor of a collective consensus between those involved, but of apparatuses built to coordinate society into informative dialogues. Internet forums can be regarded as an early version of such a coordination apparatus. A program that can coordinate the dialogue about a certain topic, bringing together people who are interested in it, and giving this dialogue a unified structure. On the one hand, the individual contributions like comments or images can be understood as technical images that are exchanged in the forum dialogue. At the same time, the forum as a whole can be understood as a collectively created technical image.

The struggle towards new information, thus, also becomes a struggle for new coordination apparatuses. Apparatuses that enable the collective creation of technical images ever more efficiently and on an ever larger scale. For example, if we only look at Google Maps as a global restaurant map, we see a technical image that is only made possible by countless participants with their contributions such as reviews, pictures and further data. From this perspective, the Internet has given rise to truly undreamed-of images, yet less through the product of a collective consensus and more through the guidance of the apparatus.

This guidance of the apparatus is becoming more dominant with each new generation of apparatuses. In order to enable ever larger and more improbable dialogues, the apparatuses are becoming ever more complex in their structure. Along with this increase in structural complexity goes a loss of functional complexity. The more efficiently the exchange of information is coordinated by a coordination apparatus, the less the dialogising actors coordinated by it can play against the coordination apparatus and create an image that differs from the image intended by the program. Especially apparatuses that coordinate the exchange between massive numbers of individuals are almost impossible to control. For instance, it is hard to imagine that with the Google Maps algorithm a fundamentally different image could be created than the one that already exists.

But the ever stronger guidance of the apparatus does not only increasingly restrict collective creation, above all it has the potential to fundamentally challenge the prior dynamic of the Flusserian Internet, as the collective dialogical creation of new information. From a Flusserian perspective, the ever more uncontrollable apparatuses open up the possibility of developing a dramatic momentum of their own. Thus, for Flusser, the loss of control over an apparatus means precisely that the apparatus can no longer be stopped: “[The] program becomes independent of human intention. It becomes autonomous and rolls on.” ⁸⁾ However, what dynamics can be recognized in the specific case of the increasingly autonomous apparatuses coordinating the Web’s dialogues? Where is the increasingly autonomously controlled Internet rolling? With the autonomy of the apparatuses “the original terms human and apparatus are reversed, and human beings operate as a function of the apparatus.”⁹⁾ Therefore, it seems obvious that the originally inspiring motive of telematic dialogue, a creative involvement, is offered to an ever lesser extent by the increasingly automated dialogues, which seem more and more pre-determined in their outcome. So, how can the increasingly autonomous apparatuses still motivate the zero-dimensionally dispersed individuals to participate? Contrary to the concept of “schöpferische Begeisterung” – an enthusiasm for the active creation of novel information – thus, an enthusiasm for the image yet to be created, Flusser describes for the pre-telematic society a passive form of enthusiasm between already existing image and human: “the images become more exciting the more excited the receivers are, and the receivers that much more excited the more exciting the images are.” ¹⁰⁾ Where communication on the Internet is becoming less exciting for its creative potential, it has to become increasingly exciting for its stimulating images and enthusiastic audience. Users on platforms such as Instagram no longer collectively strive for new information, their excitement is for the images already present on the platform and the audience they can reach through them. With the growing uncontrollability of the apparatuses, the active, creatively enthusiastic mood of the telematically networked society increasingly reverses itself into its passive opposite. The Web, dominated by autonomous apparatuses, no longer strives towards new information, but towards ever greater passive “happy communication.”

Recommender Systems

A technology that is both an outgrowth and a manifestation of the new dynamic is the recommender system, which reveals the apparatus rule over internet communication in all its drastic significance. As already discussed, in order to motivate participation, the increasingly autonomous apparatus is forced to offer individuals an ever more passive state of happiness. However, in the context of internet technology, which theoretically allows anyone to send and receive, a particular problem arises. According to Flusser, it is precisely dialogs that can endanger the state of passive contentment: “every dialogue is dangerous because it could awaken the unhappy consciousness from its sleep.” ¹¹⁾ If we consider an online multiplayer videogame, like the match of an online first-person shooter, as an exceptionally happy, meaning exceptionally passively submitted to the program, telematic dialogue, we can see clearly how it is in particular the behaviour of another player that can endanger one’s own happy, passive state of submission to the program. ¹²⁾

With the development of the recommender algorithm, the seemingly impossible is achieved: a dynamic form of communication in which theoretically any user can send and receive, yet without the need for any dangerous dialog. But how is this technical feat accomplished? 


The basic principle of recommendations is that significant dependencies exist between user- and item-centric activity. For example, a user who is interested in a historical documentary is more likely to be interested in another historical documentary […] These dependencies can be learned in a data-driven manner […] The larger the number of rated items that are available for a user, the easier it is to make robust predictions about the future behavior of the user. ¹³⁾

In the case of recommender systems the fundamental function of the apparatus is no longer to connect humans, but to bring humans and images into an ever more enthusiastic relation. The telematic dialog is turned into a dispersed sending and receiving which is brought into an exciting relation based on previous data only via the program.

As such an alignment between image and human, Flusser already characterised communication in the context of pre-telematic mass media. Thereby, it is fascinating how uncannily precise his description of communication between human and image depicts the communication dynamics found in recommendation systems: 


However, the reception of technical images does not end the communication process. Receivers are not sponges that simply absorb. On the contrary, they must react. On the outside, they must act in accordance with the technical images they have received: buy soap, go on holiday, vote for a political party. However, for the interaction between image and person under discussion here, it is crucial that receivers also react to the received image on the inside. They must feed it. A feedback loop must appear between the image and the receiver, making the images fatter and fatter. The images have feedback channels that run in the opposite direction from the distribution channels and that inform the senders about receivers’ reactions, channels like market research, demography, and political elections. This feedback enables the images to change, to become better and better, and more like the receivers want them to be; that is, the images become more and more like the receivers want them to be so that the receivers can become more and more like the images want them to be. ¹⁴⁾

A consensus between receivers and images is created. In the case of recommender systems, feedback and distribution channels are very closely and directly linked, and thus enable the creation of such a consensus between users and content much more effectively. This is amplified even more if the content to be distributed by the recommender system is user-generated content. In particular, however, it is the possibility of creating a personalised stream of content for each user that distinguishes recommender systems from the classic distribution and feedback channel systems described by Flusser in the 1980s. Through this personalization of content, a consensus between the individual and the system can be achieved much more effectively at an atomic level, without the need to create majorities within the system. Filter bubbles are an example of this. Within a platform, different users may achieve a completely different consensus with the system regarding its content. Nevertheless, the consensus does not lose its elementary meaning. Despite a wide variety of content for a wide variety of users, the underlying imperative message must remain the same: the call to further engage with the system. A deviating message would lead to a drying up of the feedback channels essential for the quality of the distribution channel, the process of alignment of receiver and image would be interrupted. In order to protect this elementary consensus between user and system, to “reduce the addressee’s criticism to a minimum”, Flusser mentions various methods, such as an “speeding up of the sequence of images” or an “inflation of images that makes freedom of choice impossible”. ¹⁵⁾

If we look at the dominant platforms of the web, such as YouTube and TikTok, we still can recognise many dialogic elements, such as comment sections, but the general evolution of the webs central platforms shows how much we are subjecting our internet communication more and more to the recommender’s logic, the human-image logic. Mark Zuckerberg’s vision of the Social Graph has not prevailed. If we look at the development from YouTube to TikTok, we see how the recommender system has become the central element of the platform. Search engines have gone out of fashion. Short videos craze enables human-image interaction at unprecedented speeds. On a platform like TikTok, there is no longer an alternative to the recommendation, the recommendation becomes an imperative to engage. In the logic of the recommender system, players become perfect functionaries who unconditionally fulfill their service to the general passive enthusiasm. On platforms dominated by the recommender system such as TikTok or Tinder, hardly anything other than hollow excitement is created collectively. Any artist who tries to inform society on TikTok or Instagram is lost. A message that is not subject to the general program of collective mental dispersion will never reach a recipient. 


The total replacement of interpersonal telematic dialogs by hyperdynamic human-image communication does not merely mean the end of collective creation, but even the concrete dismantling of existing information and complexities. The image adjusts to the person, whereby it becomes his reflection, and the person adjusts to this image, whereby he becomes a reflection of himself, whereupon the image in turn adjusts to the now distorted person. “The interaction between image and person is marked by entropy tending toward death.” ¹⁶⁾ In TikTok’s images, be it hydraulic presses destroying commodities, NPC streamers, heads in toilets, or even video collages of anything stimulating (sludge content), an unprecedentedly blunt pictorial language is revealed, a language only stimulating the passive human being via primitive stimuli. Images that are adapting to the increasingly unconscious state of their viewers and pushing it further. 


At the end of his work Kommunikologie Flusser uses an intriguing formulation, when he states how our “external situation” can be seen as a reflection of our “inner disruption,” through the recognition of which we can become aware of “our own schizophrenia”. ¹⁷⁾ Just as a movie can only be part of the communication on TikTok when it is fragmented into countless snippets, the personality of the user on TikTok is dispersed into its bare impulses. In this sense, TikTok might be the most powerful instrument of schizoanalysis to date. ¹⁸⁾ However, from a Flusserian perspective, there is no liberation in this schizophrenia, only complete subjugation, the “hellish catastrophe of autonomous totalitarianism.”¹⁹⁾ Under the enthrallment of recommender algorithms, “humanity becomes a mass that feeds the apparatus and is fed by it.” ²⁰⁾ All forms of information and complexities of society, even personalities themselves are dissolving.

To conclude…


Considering internet communication, in its myriad of different forms and in its global dimensions, the attempt made here to capture the Web’s history in its entirety may well be perceived as quite reductive. And indeed, as noted initially, Flusserian theory leaves us with serious blind spots in our understanding. For instance, any view toward the highly capitalised Internet and its development, that does not reflect such fundamental dynamics as venture capital funding or the “creator economy”, just can never be regarded as complete. And still, the interpretation chosen here does prove how much Flusserian communicology can provide for a truly macroscopic and yet compact view of internet culture and thus enable a narrative towards the Internet and its history. I believe that it is the development of such narratives that takes a vital role in the shaping of a collective awareness towards our situation, enabling a sensible utilisation of internet technology.