(Print in German: Pompe, Hedwig; Scholz, Leander (ed.): Archivprozesse. Die Kommunikation der Aufbewahrung. Köln 2002, p. 297-315; print (Engl.): in: Configurations (in preparation). - website creation date: 16. 5. 02, update: 16. 5. 02, expiration date: 16. 02. 2005, 127 KB, url: www.uni-paderborn.de /~winkler/modell_e.html, language: German, © H. Winkler 2001 home- (keys: media, theory, discourse, technology, writing, cultural continuity, monument, memory, language, architecture)).
Discourses, Schemata, Technology, Monuments:
Outline for a Theory of Cultural Continuity
Translated by Geoffrey Winthrop-Young and Michael Wutz
1 Introduction: Acts versus Deposits—Two Media-Theoretical Paradigms
Few phenomena in media studies have been given as much sustained attention as writing, the various types of material depositing,  and media technologies, which remain a problematic central concern for all media theories. In the following, I would like to introduce a model that attempts to solve various problems within the fields of media and cultural studies in a systematic way. The model is not new; it resurfaces in widely disparate theories, and I have in fact argued for it in some of my own writings in the past.
What is new is that I now introduce it as a model in a compressed and abstracted form, and as a key for the understanding of certain problems that would otherwise appear different, puzzling, or remain altogether invisible. The model is, at first glance, so simple as to appear almost trivial. I will proceed, first, by introducing the background of the investigation and then the model itself. In a series of additional steps, I will consider both the analytical reach and certain limitations of the model, eventually attempting to arrive at some sort of summary. I will only be able to demonstrate the plausibility and limitations of this approach in layered form: by playing through a set of media problems which seem to have little in common, and by playing through different media that appear to be of different conceptual orders. The main contribution of this particular approach, as I see it, is that it is able to relate these heterogeneous questions at all. Its “abstractness” creates a platform for media- and theory-based comparisons and a kind of switchboard that makes it possible for me to give much of my own research a kind of organizational center.  (Besides, my model is good against smallpox, diphtheria, and bad weather.)
The starting point for the model is the question as to how discourses organize their continuity. Basically, media can be considered from two perspectives: either, they are understood as a fluid discourse, as a link among actions. Such an approach focuses on communicative acts and, since these acts are tied to human actors, necessarily also on humans, the carriers of these communicative actions. As a result, this approach has been labeled “anthropological media theory.”
By contrast, other approaches center on writing, on technology or other forms of material depositing. They derive their legitimation from the controversial question whether media—as part of a larger socio-technological environment—can indeed still be adequately grasped as a “means” (e. g. of communication) from the perspective of “the human,” in terms of functional purpose or consciousness. If the development of technology is seen—at least in part—as an autonomous process that extends the blind evolution of nature, then we can do no more than trace the consequences of this evolution for social formations and the positioning of the individual. These theories represent the enlightened mainstream of media theory since the eighties and have, in the wake of Foucault, been labeled “discourse analysis,” or “techno-centered” by their opponents.
Both approaches have their defenders who argue vigorously and almost in as polarized a fashion as I have just sketched them. Naturally, there are many attempts at mediation. Beginning with the example of a single medium,  or under the iridescent banner of a “media culture,”  some approaches recognize polarization as a problem. Nevertheless, up to this point, polarization could not be done away with. Paradoxically, both approaches are undoubtedly right. Leaving aside their historical and philosophical premises, we are dealing with the radicalization of two perspectives that could be united merely through a theoretical effort on the terrain of a valid theory of media. To develop a sketch in that direction is the first goal of the model introduced here.
Theoretically, as noted above, we are dealing with the question as to how discourses establish continuity.  As chains of discrete communicative events, discourses, one might think, are in constant danger of disruption or abrupt changes in direction. Several media theories, indeed, among them such prominent ones as that of Luhmann, understand discourses as chains of discrete events and from the perspective of “connectivity.” 
The observation, however, indicates that discourses are astonishingly continuous and rather resistant to changes. Below the surface of harried innovation they resist de facto innovations with considerable inertia.  In my judgment, the central puzzle in the functioning of medial discourses is not “connectivity” or unforeseeable “articulation,”  but this capacity for inertia.
The nature of our inquiry, therefore, relates to what I like to call an economy of discourses that combines the unforeseeable chains of acts of utterance with moments of inertia. Discourses organize their changeablility, and we would fail in our inquiry were we to ignore such real changes and irruptions. At the same time, however, discourses also organize their “weight,” as it were, with which they offer resistance to such changes. To date, we have no model that mediates between these two. My argument is that such a model would be a variation of the question about “technology-centered” and “anthropological” media theories.
2 Monuments and Repetitions
A particularly suggestive approach to describe such mechanisms of continuation was developed by Jan Assmann.  Through the example of Ancient Egypt, he demonstrates the existence of—and here I introduce another binary structure— basically, two polar cultural techniques that are capable of stabilizing and continuing discourses: monuments and repetitions. In the case of Ancient Egypt, Assmann observes, two modes of life were juxtaposed to each other: on the one hand, hieroglyphic writing and the architectonic funereal monuments, built from stone and with the assumption of, quite literally, eternal duration; on the other hand, the more transient living quarters built from clay, changeable cursive writing, and daily routines which (analogous to the rhythms of the Nile) were seen in terms of a cyclical structure.
In more general form, this model has its origins in the research into orality: while writing cultures invest in material deposits and juxtapose the monumental duration of the writing medium to a transient temporality, oral cultures are vested in repetition and ritual. From a contemporary perspective, this is a technique of cyclical rejuvenation, which, as Friedrich Nietzsche put it, literally burns memory into humans.
What is irritating in the work of Assmann, as it is in the scholarship of orality, is that both techniques are juxtaposed to one another and are put into the service of cultural continuation without insisting on or demonstrating a systematic connection between the two. This is all the more puzzling given that the theory of writing maintains that the monumentality of writing can substitute the mechanisms of oral repetition. As soon as a culture adopts the technique of writing, it devalues ritual repetition and, to a certain degree, relieves human memory from the burden of having to provide continuity. If the model of repetition, however, can be replaced by the monumental one, such a replacement points—beyond a functional parallel—to a structural similarity or a systematic relationship.
3 The Relationship between Monument and Repetition
In the following I’d like to focus on this relationship. Initially, monuments and repetitions are far apart. Once it is erected, a monument wants to persist. It plays off its material solidity and persistence against the change of daily routines. The cultural significance of the Cheops pyramid may have changed profoundly—migrating from the realm of ritual to that of tourism—but it has occupied one and the same piece of property for the last 4,700 years. Certain daily practices are, therefore, marked out in advance. Just as the architecture of a city predetermines and stabilizes the paths of its inhabitants, so daily routines surround the monuments and seek their orientation through them.
Repetitions, by contrast, are in much greater jeopardy. Frequently, they can achieve continuity only by securing the identity  of repetitive acts through repressive means: each tradition has its guardians, priests and authorities, and if Egypt managed to maintain its hieroglyphs in unaltered form for thousands of years, it could do so only through an extremely repressive scribal culture that did not tolerate deviations and adaptations to their historical moment.  Aside from repression, the stability of repetitive cycles can be explained (and is already observable in the animal world) through habit, the proclivity for repetition and schemata, as well as the economy that comes with such repetition. Initially, therefore, repetition and monument fall apart.
At a second glance, however, things begin to get interesting. We note that the monument (paradoxically) harbors within itself an element of repetition, and that repetition (again, paradoxically) harbors within itself an element of monumentality. The material persistence of the monument initiates a series of encounters with that monument. Over centuries, a written text can be read by tens of thousands of readers who take it in hand and integrate it into their lives; select readers may read it repeatedly. Its material durability asserts itself, above all, by bringing about a certain type of repetition that creates a kind of center of gravity for that repetition; this center of gravity forces the repetitive act to, in fact, return in cyclical fashion to a describable point. Seen from a practice-based point of view, the monument operates as a machine that produces this particularly stable type of repetition.
Conversely, as I noted above, repetition too contains an aspect of monumentality. Repetition can take place only if the two acts of repetition are conjoined through an instance which in itself has a monumental (or quasi-monumental) quality. In the case of oral societies, this is the human memory which—while requiring cyclical rejuvenation—is capable of storing the to-be-repeated pattern in the interval between two acts of repetition. What becomes evident is the possibility of linking repetition and monument—little as they seem to have in common—in a combined and more abstract relationship.
4 The Model
I will return to that question in a moment. Prior to doing so, however, I want to introduce the basic model which underlies our ensuing reflections like a system of coordinates.
Monuments originate in an initial act of inscription. In the case of the pyramids, that act is the (rather complex) process of construction; in the case of a written text it is the act of depositing undertaken by an author in combination with the widely ramified material and organizational processes of the publishing industry that turn authorial manuscripts into marketable print products. If a book is to persist, the initial act of depositing has to be complemented by additional instances and agencies such as distribution networks, libraries, the lack of natural catastrophes, air raids, etc. On this first level, therefore, act and monument are linked through a process of inscription.
Once a monument has been erected, it has an effect on a culture’s daily practices. In the simplest case, a book is read, a pyramid is marveled at; their deposit “is dissolved” into daily practices by determining or shaping them. The monument unfolds its effects precisely because it does not remain in isolation, but, rather, writes itself back into daily practice.
As a model, we can observe the intertwining of two movements:
Or, more precisely:
Daily practices and monuments/deposits are linked in a cyclical movement. And since daily practices don’t have any priority in this cycle, one could also formulate the above as follows:
^ \ ^
/ \ /
As well, the model should illustrate the monument’s material persistence, that is, the possibility that daily practices may return to the same monument:
\ ^ \
\ / \ /
Deposit —> material persistence, ‘tradition’
This rather simple model, it seems to me, is considerably far reaching, which is why I expend quite some energy making it strong within the field of media theory. It is capable of interrelating in a systematic way questions arising in media theory, cultural theory, semiotics, the theory of technology, psychoanalysis and in several important subdiscourses.  As well, as I mentioned earlier, the model opens up questions that would otherwise remain invisible.
Obviously, it provides a solution to the dispute, which, as I described earlier, currently characterizes media theory. Whether I grant autonomous status to a given technology and examine its effects on social processes, or whether I insist that technology has its roots in social and communicative acts and practices, merely serves to indicate which phase of the cycle I am primarily interested in. In either case we are dealing with the one-sided treatment of a comprehensive process, which, fundamentally, encompasses inscription and rescription, that is, the transition from practices to deposits and the second transition from deposits to practices. 
5 At the Macrolevel: Technology and Language
If we are now to project this model onto various configurations (with the possible consequence that it might appear less simple), we need to add a significant modification, for the model is by no means limited to a single text, as I have demonstrated it so far. Likewise, technology per se can be understood as a “deposit” on a social level. At every point in its history, single technologies merge into a landscape of technology: whatever we comprehend as the present state of technology is the result of past practices and, at the same time, the point of departure for future practices. At this level of abstraction, certainly, the cycle of inscription, depositing, and re-inscription into practices is precisely the same as on the microlevel, which links an individual technology with the macrolevel of technology.  The same is certainly true of the world of texts, the social library, and so on.
In particular, however—and that thought is anything but trivial—that mechanism applies to language. The semantic system of a language, the system of conventionalized meanings we experience as a stable lexicon, did not fall out of the blue, but (as can be glimpsed from numerous linguistic theories  ) is the result of billions of speech acts and single texts that, in the manner of a collective work of art, have given form to a language. In concrete terms, this means that language too must be described in terms of a dialectic between linguistic practices and material deposits—deposits whose material location is dispersed among the heads of millions of language users.  Language can, thus, be framed in terms of a technology which, on the social level, intertwines acts of inscription/depositing with speech practices. This already took place prior to the development of writing, which illustrates the technological character of language all the more.
Almost imperceptibly, we have significantly enlarged the notion of technology: while much of media theory, in particular, narrowly focuses on hardware,  and on writing as a comparatively compact and materialized object of investigation, the model proposed here urges a more complex understanding of technology—an understanding that intertwines material depositing with practices and that comprehends practices themselves systemically from the point of view of their technicity. Paralleling some contemporary theories of technology,  we ought to return to the ancient notion of technè, which has always encompassed both of these elements.
6 Conventions and Schemata
These reflections about technique and language have to be extrapolated. If language functions as a social technology that intertwines linguistic practice and the language system and that subordinates the apparatus of signification at any one time to the linguistic events of the past (acts of speech, utterances), we have found a model that describes, in rather precise fashion, not just linguistic events, but conventions, in a generalized sense. Conventions are congealed practices: sedimentations, deposits of, actually, fluid acts and events that accrue and accumulate, and eventually transmute into a structure.
Were one to inquire into the concrete discourse-economical mechanism that brings about conventions,  one would, quite likely, first point to repetition. Conventions are grounded in repetitions, and they trigger entire chains of future repetitions. Deposited as a system, however, they become an agglomerate, and hence monumental.
Through the notion of convention, we open up an entire universe of theoretical problems that can now be linked to the model described here: first off, theories of schemata, which have become important in the analysis of visual media: all the approaches ranging from Gestalt psychology to the theory of stereotypes and from iconography in fine arts to the notion of aesthetic form center, at bottom, around what semiotics would subsume under the notion of a code.
As difficult as it has proven to formulate a semiotic theory in relation to visual media, it is, simultaneously, undeniable that—in the field of technical images, in particular—repetition and scheme formation play a dominant role when it comes to media socialization and media competency and to shaping the structure of expectations with which recipients approach concrete products. Schemata and stereotypes are deposits that profoundly impact the structure of visual discourse, even if the field of film studies has a rather critical perspective on such stereotypes and schemata. Stereotypes are a kind of hidden skeleton embedded in technical images and—at least in terms of their structure and function—very similar to the conventionalized schemata residing in language.
Last but not least, the notion of convention makes it possible to relate systems of action, as they are examined by sociology and the social sciences, to the sketched-out model. The realm of silent practices as well is dominated by the same logic of singular act and scheme, repetition and conventionalization. By demonstrating the regularity of acts, sociology is concerned with the kind of deposit which I have been talking about. Above all, the model presented here operates as a switchboard because it puts the general notion of convention at its center and because it, at the same time, precisely defines that notion as a deposit suspended in the dialectic between singular act, repetition and depositing.
At this juncture, it may be prudent to correct the impression of excessive overestimation and point out some specific limits of our model. I certainly do not believe that what I have presented here is a kind of universal key or the emc² of media scholarship; its theoretical problems are all too evident.
These problems suggest themselves already on the very level of the model’s formulation. Is it really possible to combine the pyramids and the conventional system of language under the notion of a deposit? Are we talking about the same type of deposit, given that pyramids persist in a material-monumental way, while the semantic system of language with its discourses lumbers forward, subject to constant change? To insist on the notion of “deposit” means—notwithstanding such clear distinctions—to point out the fact that in both cases we have to envision a material storage device side by side with interaction. A second, more serious question is, in what sense can we speak of a “cycle” if this cycle combines chains of different acts, which is to say, it does not simply return to its point of origin?  And finally, isn't it an extremely conservative model which emphasizes historical continuities, without being able to reflect on the ruptures and radical changes which are at the center of postmodern debates? The list of theoretical problems could easily be expanded. Therefore, let us return to the sunny side of my model and its possible achievements.
8 Subjects as Depositing Sites
Let me draw attention to an important shift which the preceding arguments have produced, possibly without being noticed. While I started by pointing to the material depositing in texts or technologies, the type of depositing encountered in language—and, even more, in conventions—is wholly different. In such cases, the material depositing site is not an environment of objects, but, on the contrary, the subject. More specifically, it is, on the one hand, the individual memory/body memory in which the linguistic system and the system of conventions are located. On the other hand, it is a collective memory, which, through its distribution into individual memories, constitutes itself as a phenomenon of social redundancy.
This shift from objects to subjects as the site of inscription, irritating as it may be, is not simply deficient. While subjects as carriers of practices were, initially, systematically juxtaposed to all forms of “material and object-like” deposits, they are now themselves understood as belonging to the side of objects assuming a passive role. Does that not—at least, from one perspective—correspond to current conditions? Poststructuralism, above all, has shown us that we are the objects of our media socialization, the objects of social inscription and the unconscious-involuntary carriers of linguistic and extralinguistic conventions which we execute without our prior approval, and which we transmit without being able to control them.
In view of the central question of my essay, which focuses on those cultural practices that secure a continuation of discourses, I would argue that subjects can indeed be found in both positions: in the subject position as a carrier of acts that result in deposits—deposits which, in turn, become the origin for renewed practices; and—functionally parallel to such deposits—as carriers of conventional, congealed structures that counterbalance fluid discourses as an instance of resistance, inertia and restraint. That this is an extremely dramatic dimension of cultural continuation becomes apparent when we consider that, following the collapse of the Third Reich in 1945, it was a great deal easier to “purify” the holdings of German libraries than to cleanse what goes on inside German heads. Humans themselves, in that sense, are “monumental,” and as astonishingly adaptable as they are, they also resist, with leaden heaviness, projects of change even when they are emancipatory.
Naturally, we will also have to distinguish between the depositing into material storage devices and those into human memories. Ideally, material storage devices are supposed to preserve their contents faithfully. Human memories, on the other hand, tend to select, reconfigure and forget their contents; and we know from memory theory that this is the real achievement of human memory. A sober and quantitative reflection indicates that we have to forget the large majority of the infinite perceptions we make on any given day, simply because of the limited human processing capacity and because an unstructured accumulation of perceptions is impossible. Forgetting, in that sense, is not a defect, but an absolutely necessary form of protection.
What is more, we can assume that this forgetting leaves its traces. Even though memory theory offers surprisingly few models on that score, Freud’s notion of the “miracle block” already takes note of the fact that the concrete act of perception—while being submerged in the act of forgetting—changes the perceiving subject with each perception. Forgetting appears to be a machine that transforms the infinite space of singular perceptions into subject structures; or, to put it more precisely: it transforms these perceptions into those structures of expectation with which the subject encounters new perceptions. Forgetting, therefore, is always a “forgetting into the structure” of subjects,  and such forgetting can easily be related to Freud’s notion of “condensation,” as formulated in The Interpretation of Dreams. 
10 Collective Condensation, Medial Condensation
There appear to be wholly comparable mechanisms on the collective level. If the system of language originates in the speech acts of the past, which have given form to the semantic system through a gigantic process of accumulation, language in its entirety must be seen as a product of ‘condensation.’  What is of relevance here is the quantitative proportion: Billions of speech acts register as deposits in linguistic structures, whose virtue is that they are so compact as to fit into puny human skulls. Given our limited mental resources, this is an astonishingly compact and economical form of representation and a brilliant compromise.
Perhaps this is the most admirable aspect of language: As a social technology it transforms speech acts into compressed semantic-mental structures. And this conversion, the mechanism for the production of structures, is at the center of what I elaborated above as a generalized model. The necessary dialectic between act and deposit, discourse and structure, is centered in the notion of condensation.
Naturally, this mechanism does not solely apply to language. It is evident that the stereotypes and theories of schemata in the visual media, to which I referred above, follow similar mechanisms: stereotypes and schemata assume their structure—even more visible than the units of language—in the progression of discourses; a long chain of western movies has given shape to the genre, and the structure of expectation with which recipients encounter it. Prior experience, as I said earlier, condenses into media competency, which, in turn, shapes a system of socio-symbolic topoi shared by both producers and receivers. It is time to retire the notion that visual media can be accessed without prerequisites.
Technology and architecture can be seen as products of condensation as well—this time, outside of human heads. The practices and knowledge of the past have accrued into the respective “state of the art.” Paralleling language, technology is a compressed structure containing the practices of the past and anticipating practices of the future. The same quantitative proportion applies here as well: Trivializing Hegel, one could say that the practices of the past—that is, the technological practices of the past—have been “sublated in,” or displaced from, the collective Kunstwerk of technology.
This is relevant even on the level of the single product: As a socio-symbolic technology, the feature film—leaving all other medial differences aside—conceivably surpassed literary fiction only because it shows ‘[a] higher [level of] condensation.’ Deposited onto one material carrier, which can easily be consumed in 90 minutes, is the work on which entire divisions of industrial specialists have been working for years, and they have done so with the help of an advanced technology in which is condensed, in singular form, societal work and technological know how. The novelist’s solitary act of writing, by contrast (supported solely by the collective Kunstwerk language), appears to be ‘technologically under-equipped.’ 
11 Recapitulation: Monuments and Repetition
Revisiting the question of the relationship between monuments and repetitions, one connection should be sufficiently clear. Monuments can replace repetition because they themselves are social engines aimed at initiating repetition. Discourses manage to secure their continuity by establishing agencies of inertia that persist side by side and in tension with them.
At the same time, we do well to distinguish between the various types of agencies. Type 1 would be the pyramids that combine persistence with—in an ideal case—unchanging duration thanks to their material durability. Type 2 can be represented by the human memory and the system of language: both are products of condensation and both exist and unfold within discourse; at the same time, their inertia and relative immobility provide a counterbalance to the tendency of discourse toward abrupt changes. All actual utterances and events must be seen with a view toward this agency of inertia.  Paradoxically, therefore, type 2 embodies the historical-plastic monument that is mutable in itself.
In that context, technology has a dual face: on the level of the single technological artifact it, no doubt, belongs to the first type of simple material inertia; in the social realm, operating as a social technology and analogous to language, it belongs to the second type. Both types are characterized by a model of condensation; and the point of each individual artifact seems to be that it freezes a certain level of condensation in a material stasis. Both types as well force practices into cycles of repetition. That was the reason for abandoning the original notion, argued by Jan Assmann, to see monuments and repetition in terms of a polarity.
Finally, what insights have we gained with this speculation? First and fundamentally, medial acts have to be referred to medial deposits, and medial deposits, in turn, have to be referred to medial acts. Only this dialectic will allow us to show how media bring about cultural continuity. As I have pointed out, repetitions are no exception and have to be seen in the context of the interplay between a pattern and its reenactment, a moment of action and a moment of persistence. Deposited in material fashion, and hence “monumental,” the pattern awaits its reactivation and renewal.
Second, human subjects are not located exclusively on the side of actions. Given that human memory has to be conceptualized as a site of inscription, and, in a more general sense, subjects as the carriers of an unconscious socio-semantic structure, the subjects themselves, despite their undeniable mobility, represent a source of cultural persistence.
Third, next to the relatively stable, material monumentality of individual artifacts we have to posit a second type that achieves monumentality by way of accumulation and condensation. Towns, technologies and languages may serve as an example of what, because of their constant morphing, can only be described in terms of condensation. Subsequently, the notion of convention can only be apprehended by way of conventionalization, which, in turn, has to be seen in the context of chains of repetitive acts.
The notion of condensation is the core of the model and its real theoretical gain. What characterizes condensation is that it combines a quantitative with a qualitative aspect. When the incalculable range of linguistic utterances turns into a linguistic structure, acts are transformed into structures in ways that might recall Frederick Engels’ dialectics of quantity and quality.  This enables us to connect our insights to quantitative-economic models. The mechanisms of circulation and distribution, which come quite naturally to economic analyses, are still hardly investigated in media theory. Split into empirical and theoretical analyses, quantities are left to superficial statistics, while theoretical models concentrating on the circulation of signs are quite rare, and an ‘economy of discourse’ is, at best, a desideratum.
Referring to the concept of ‘condensation,’ technological reproduction, to use one of the most prominent concepts of media theory as an example, would be conceived of as a certain type of repetition. Technological reproduction generates structure (and redundancy) and—this is its monumental aspect—achieves cultural continuation. It would be the task of an ‘economy of discourses’ to facilitate more precise synchronic and diachronic descriptions of such mechanisms.
Finally, the model presented here may help to correct certain systematic distortions of contemporary theorizing. It seems obvious that current media theories suffer from a flagrant ‘forgetfulness of language’ and almost completely evade issues of language and code. My explanation is that semiotics, formerly a hopeful candidate in the theory debates, has fallen into the abyss that separates anthropological from technology-centered approaches. With their grounding in action theory, the former focus on individual acts, but in doing so they forget that acts cannot be conceptualized independently of repetitions, that is, independently of a system of habits and conventions. This makes it necessary to reflect upon the tension between acts and ”monumental” code.
Those centered on technology, in turn, view code as a media-theoretical residue from the humanist-anthropological era because it is still tied to human carriers. The category of “meaning” is judged to be bloated and fuzzy, opposed to any materialist description of discursive processes, and hence negligible; especially since the actual history of media itself appears to have moved from “natural languages” to hardware-intensive visual media and, most recently, to the “pure” sphere of mathematical algorithms.
In light of what has been proposed here, however, this view is an illusion. If a code is obviously still at work in the case of visual media (and a crux of visual media is precisely that they systematically obscure this fact  ), and if, second, technology itself has to be conceptualized as a code, that is, as a condensed social deposit that is capable of determining subsequent practices, then it appears that the same question has to be asked about the computer. If we are to take this model seriously, we will have to pinpoint the exact instance in the computer where the arbitrariness of the code reappears. I have attempted something along these lines by suggesting that one should approach the project of formal languages and formalization from this point of view. 
Rather than declaring code to be obsolete, it may be necessary to describe the production of meaning itself as a social technology, that is, in materialist and discourse-economical categories. I have very roughly suggested as much by pointing to the transformation of discourse into structure; in such a way it may be possible to avoid both the allergy against ‘meaning’ and the truncation of the concept of technology by way of eliminating language.
Media theory seems to depend on the elaboration of models that go beyond what is evident in media. These models are necessarily abstract, simply because only abstract models are able to cross the established, sizeable boundaries between different media that inevitably resist media-theoretical comparisons. As abstract models they are by necessity wrong. They are bound to miss precisely those mechanisms that are particularly characteristic of individual media and that have to be part and parcel of any single-medium analysis. But focusing on individual media (as well as one so-called “intermediality”) cannot spare us the labor of theoretical effort.
To develop a general notion of convention and conventionalization, an idea of cultural continuation, an idea of how monuments and repetitions work together, and to grasp what material persistence has in common with other types of continuation—all this appears to be necessary for any comparison between media. Once we assume that the various levels of acts, institutions, the symbolic and the technological—which are without doubt the four basic registers for any elaboration of media—do not simply exist on their own, we have to ask on what level they are mediated. This is precisely the question to which the model of the dialectical change of discourse into structure presented here wants to provide a first, tentative answer.
 Translators’ note: In consultation with the author we are using deposit/depositing to translate Niederlegung (derived from the verb niederlegen which, according to context, can mean ‘to lay down,’ ‘to put down,’ ‘to deposit’, ‘to record’).
 To give myself such methodological self-clarification was the occasion for writing the present essay. My book Docuverse contains most of the observations articulated here; there, they are located within the project of the book which attempts to formulate (an, as far as possible, immanent) critique of the present computer discourse (Docuverse: Zur Medientheorie der Computer. München: Boer 1997; see http://www.uni-paderborn.de/~winkler. html for a structure of the book and the full text of the first chapter in German).
 Relating to TV, Raymond Williams discusses the question in terms of a general theory of technology (Television. Technology and Cultural Form. [1973/75]. London: Routledge, 1992: 9-31, “The Technology and the Society”).
 For example, Claus Pias et. al. (eds.) Kursbuch Medienkultur. Die maßgeblichen Theorien von Brecht bis Baudrillard. (Stuttgart, 1999), 8.
 In the following I use various terms of discourse: Underlying all of them is, (1), the common conception of discourse as the totality of all acts of utterance both oral and written: “Discourse is [all of] a person’s realized linguistic utterances based on his or her language competency in the process of linguistic communication” (adapted from Duden. Fremdwörterbuch. Mannheim 1974, p. 182). (2) More generally, “discourse” frequently designates the totality of symbolic practices, as when visual discourse is juxtaposed to linguistic discourse. (3) In the work of Foucault, the term discourse encompasses utterances as well as practices, for example, the construction of prisons and the formation of the body through torture or drill. At the same time, Foucault’s term of discourse designates a specific epistemological process; this epistemological process is claimed by the discourse-analytical approaches.
 See Niklas Luhmann. Social Systems. Trans. John Bednarz, Jr., with Dirk Baecker. Foreword by Eva Knodt. Stanford, CA: Stanford Univ. Press, 1995, 36.
 Horkheimer and Adorno have notably drawn attention to this inertia. Rather shocked by their exposure to American mass culture, they spoke of the “constant sameness [which] governs the relationship to the past” (Dialectic of Enlightenment. Trans. John Cumming. New York: Herder and Herder, 1972, 134).
 Within the current media debate, this term has gained considerable currency.
 Jan Assmann “Stein und Zeit. Das “monumentale” Gedächtnis der altägyptischen Kultur,” in Tonio Hölscher (ed.). Kultur und Gedächtnis (Frankfurt: Suhrkamp, 1988), 87-114. See also Jan Assmann, Stein und Zeit. Mensch und Gesellschaft im alten Ägypten (München: Fink, 1991).
 A rather precarious identity, as Derrida demonstrated in his well-known debate with Searle. The debate advanced in three steps: 1. Jacques Derrida, “Eight. Signature Event Context” , Glyph,1977 (1); 2. John R. Searle, “Nine. Reiterating the Differences. A Replay to Derrida,” Glyph, 1977 (1); 3. Jacques Derrida, “Nine. Limited Inc. a b c....,” Glyph, 1977 (2). I discussed the debate and the 'certain self-identity' of the repetitive acts in Docuverse, pp. 281.
 This stability is valid only for hieroglyphs, not for cursive writing. For that reason, it is owing not only to repression, but always already also to the material co-presence of written documents from the past. See also Jan Assmann, “Ancient Egypt and the Materiality of the Sign,” Materialities of Communication. Ed. Hans Ulrich Gumbrecht and K. Ludwig Pfeiffer. Trans. William Whobrey. Stanford: Stanford Univ. Press, 1994: 15-31.
 In the following, I will only briefly touch upon some of these discourses, esp. semiotics and psychoanalysis. I will explore these connections in a more extensive fashion in a future essay.
 I have developed this argument in detail in my essay, “Die prekäre Rolle der Technik. Technikzentrierte versus ‘anthropologische’ Mediengeschichtsschreibung,” in Claus Pias (ed.). Medien. Dreizehn Vorträge zur Medienkultur (Weimar 1999: 221-40); see http://www.uni-paderborn.de/~winkler/henne.html for the text.
 It is baffling that the dialectic between speech and language is not a central concern of linguistic theory. Instead, and in abbreviation of Saussurean categories, synchrony and diachrony are juxtaposed in abstract and static fashion. Nevertheless, time and again one can find isolated approaches that are close to my arguments: “Speech always implies both an established system and an evolution; at every moment it is an existing institution and a product of the past.” “It [language] is a storehouse filled by the members of a given community through their active use of speaking, a grammatical system that has a potential existence in each brain” (Ferdinand de Saussure, Course in General Linguistics. Ed. Charles Bally and Albert Sechehaye. Trans., with an introduction and notes by Wade Baskin. New York: McGraw-Hill, 1996: 8, 13f.)
Similarly, in research into orality: “[T]he meaning of each word is ratified in a succession of concrete situations... all of which combine to particularize both its specific denotation and its accepted connotative usage. This process of direct semantic ratification, of course, operates cumulatively” (Jack Goody and Ian Watt, “The Consequences of Literacy,” in: Literature in Traditional Societies. Ed. Jack Goody. Cambridge: Cambridge UP, 1968: 29).
In Michel Foucault: “What civilizations and peoples leave us as the monuments of their thought is not so much their texts as their vocabularies... the discursivity of their language. ‘The language of a people gives us its vocabulary, and its vocabulary is a sufficiently faithful and authoritative record of all the knowledge of that people’” (The Order of Things: An Archeology of the Human Sciences. New York: Vintage, 1994: 87; Foucault is here quoting Diderot.) And finally in Marshall McLuhan’s more general view of media and technology: “The classic curse of Midas, his power of translating all he touched into gold, is in some degree the character of any medium, including language.... All technology has the Midas touch.... Language, like currency, acts as a store of perception and as a transmitter of the perceptions and experience of one person or of one generation to another” (Understanding Media. The Extensions of Man. New York: Signet, 1966: 130-31).
 At one point, this was not unjustified, especially when it was a matter of countering philology’s forgetfulness of technology.
 See, e.g., Carl Mitcham, Thinking through Technology: The Path between Engineering and Philosophy. (Chicago & London: Univ. of Chicago Press, 1994); Merrit Toe Sith and Leo Marx, eds., Does Technology Drive History? The Dilemma of Technological Determinism. Cambridge: Cambridge Univ. Press, 1994; David Rothenberg, Hand’s End: Technology and the Limits of Nature. Berkeley: Univ. of California Press, 1995.
 “Discourse economy” refers more to a work scheme than to an already existing, fully elaborated scholarly approach: based on the model of classical political economy, which investigates the production of commodities as well as the circulation and accumulation of capital, discourse-economical research would have to clarify how also in the realm of signs and symbolic exchange quantitative processes generate structures.
The notion of repetition contains
the entire problem: it combines the idea of linear progression (as it is
presumed by the notion of an act) with the idea of a cyclical return. Both ideas, initially, contradict one another.
Repetition, however, is inconceivable without this contradiction.
Even more: it can easily be seen as the model or concept for this
contradiction. Repetition, as I said
earlier, contains a moment of identity or similarity; otherwise, it could
not be recognized as such in the whirl of events.
At the same time, it also contains a moment of difference in that
it always combines self-contained/heterogeneous events. Instead of speaking
of a cycle, therefore, one could speak of a spiral (if one is to remain
in the problematic sphere of geometrical illustration). A spiral moves forward in linear fashion along
one of its axes (the moment of difference); at the same time, it also describes
a cyclical motion (the moment of identity). Naturally—what complicates the situation even
more—the interplay of both moments can proceed in different constellations.
 See Winkler, Docuverse, 143ff.
 The most striking version of this thought was formulated by Christian Metz, whose book offers a psychoanalytic-semiotic theory of the cinema: "It is indeed a characteristic of language—and another aspect of the 'problem of the word'’—that it has this constant but never fully realised tendency to encapsulate a kind of complete (but concentrated, compressed) 'argument' in every word: a tendency which is also intrinsically condensatory. Even the most ordinary word, lamp for instance, is the meeting-point for several 'ideas' ... each of which, if it were unravelled, or decondensed, would require a whole sentence." "Past condensations meet in each word of the language.... [T]his is to define the lexicon itself as the product of an enormous condensation" (The Imaginary Signifier. Trans. Celia Briton, Annwyl Williams, Ben Brewster and Alfred Guzzetti. Bloomington: Indiana Univ. Press, 1982: 225, 239).
 Comparison and thesis are, admittedly, rather crude, but I do not consider them to be out of place. Whoever judges media-historical transitions to be in need of interpretation will have to clarify why feature films obviously achieve a higher level of signification. Popular explanations along the line of, “movies are more successful because they are entertaining and easy to consume,” are insufficient. What I have provisionally called “level of signification” would also have to be elaborated within the framework an of economy of discourse.
Practices 1 Practices 2
\ ^ \
\ / \
Monument—> material persistence
\ ^ \
\ / \
 See Frederick Engels, Dialectics of Nature. Karl Marx and Frederick Engels, Collected Works. London: Lawrence and Wishart, 19xx. Vol. 25: 356-361.
 In the context of the debates on realism, the problem of the ‘invisible code’ is discussed under the heading of the ‘illusion of transparency.’
 See Hartmut Winkler: “Ueber Rekursion: Eine Überlegung zu Programmierbarkeit, Wiederholung, Verdichtung und Schema,” at http://www.uni-paderborn.de/~winkler/rekursio.html