4 / THE CONSEQUENCES FOR A CONCEPT OF RATIONALITY

At the end of chapter 1 I argued that an adequate theory of rationality can only be developed on the basis of a comprehensive analysis of processes of understanding. Such an analysis is only comprehensive if it includes the understanding of nonlinguistic communicative actions that can account for the untranslatability of nonlinguistic articulations. In conformity with one use of the concept of a “medium” in everyday language, I then assumed that some processes of nonlinguistic understanding are implemented with the help of means that are analogous to the medium of language, which makes verbal communication possible. Because no stable concept of medium can be found in the contemporary theory landscape—despite the fact that the media concept plays a central role in the most influential social theories of our day—in chapter 2 I set out to develop the main features of a general media theory. This should in particular show itself to be up to the task of making nonlinguistic processes of understanding, as they paradigmatically appear in aesthetic communication, accessible to theoretical analysis within the framework of an interpretationist theory. If the intuition is correct that an adequate concept of rationality must be developed on the basis of an adequate concept of understanding, then it is now time to examine what benefits the media-theoretic amelioration of the concept of understanding has brought to the theory of rationality. In other words, the issue is to examine whether this book rightly bears the title Media of Reason. This chapter is thus devoted to addressing the question of the degree to which our view of rationality can be refined by the proposed media-theoretic means and the consequences that this might have for the further development of the project of the Enlightenment.

With a view to the mentioned issue, two possible benefits of the media-theoretic amelioration of the concept of understanding are to be distinguished, which I will attempt to balance in succession. For one, explicative benefits may result from the fact that a concept of understanding that is broadened in this way allows us to apply our view of rationality more comprehensively and with greater differentiation and greater homogeneity, so that we can, for example, clearly show, with such a concept, the degree to which aesthetic communication instantiates rational competencies and what the differences from the established forms for instantiating reason consist in. From the perspective of the study of the explicative benefits, I hope to be able to show that the common critique of rationality is attractive because it is based on a view of rationality that centered on—historically plausible but contingent—philosophical explications of rationality that only take account of linguistic competencies. For another, however, one can also expect a normative benefit from a theory of rationality ameliorated by media theory. From the perspective of the study of the normative benefit, I would especially like to examine whether we ought to understand reason in the Humean tradition rather as a bundle of such competencies with the help of which we can best achieve our respective contingent preferences, or whether, following a Habermasian impulse, general norms can be obtained from the analysis of understanding—now with a complete variety of processes of understanding at our disposal—whose normative power is not dependent on specific interests, that is, on particular norms that would be relevant for the conception of Enlightenment processes.

4.1. THE EXPLICATIVE BENEFITS

If we inquire into the explicative benefits of the reflections on media theory for a theory of rationality, then it is obvious that we must address the question of what reason is. As we saw in the introduction, there are a series of well-established answers to this question. Admittedly, particularly using these theories in the attempt to understand what artistic action is, we encountered numerous problems, precisely the problems that motivated us to develop the media theory. However, from the fact that the established rationality theories do not provide any particularly favorable instruments for analyzing artistic actions, it by no means automatically follows that these conceptions of rationality are to be fundamentally modified. In order to determine the explicative benefit of media-theoretic reflections we should thus examine whether medial action can be subsumed under the existing concepts of rationality. We must thus examine whether, starting with a well-contoured understanding of rationality, we can also understand artistic action and nonlinguistic thinking as instantiations of already sufficiently specified rational abilities. Only if we show that this is not possible can we justifiably assume that our view of rationality itself must be differentiated in order to provide a place for abilities, the analysis of which has not yet been able to contribute to a characterization of rational competencies, although the same set of entities can be characterized with the help of these abilities as can be characterized with the classical definition of reasonableness. In the framework of these reflections, the concern is thus not to grant one ability or another a place among the range of exemplary rational competencies; rather, we must show that they have a rightful claim to this status.

4.1.1. Media as Mere Instruments of Reason?

However, if, to begin with, one pursues the first path, then a charitable reader of the media-theoretic reflections might come to the conclusion that they may well yield something interesting about media, art, and nonlinguistic communication, but that media in principle are nothing more than instruments that can be used by actors with artistic, expressive, or general communicative intentions. If that is the case, however, then the connection between media and reason appears to be purely external. Then, in this perspective, media come into purview, under the assumption of existing intentions, only as possible instruments for implementing such intentions. They contribute nothing to defining reason, which reflects on the adequacy of these means. Even if this objection is based on a specific understanding of reason, one could maintain the thesis that each of the concepts of reason that was explained in chapter 1 in principle would be able to integrate the media-theoretic reflections so that the respective definition of reason of each of them would remain untouched. In the face of the list of paradigmatic rational competencies schematically summarized in the findings of the reconstruction of the established concepts of rationality,1 we must be able to say the following: If rationality essentially

1. is a culturally conveyed competence of actors to orient themselves with the help of transindividual but context-varying standards in action situations (Kambartel), then even in the context of aesthetic phenomena, or more generally in contexts of nonlinguistic communication, this competence may arise and develop;

2. is to be understood as a set of normative demands, which, if followed, ought to promote the production of forms of life that find general agreement, that are generally perceived to be acceptable, and that present a suitable basis for justified developments (Mittelstraß), then we can assume that (some, perhaps even many) people will tend to view the expressive and communicative possibilities that depend on the social existence of media as a characteristic of generally acceptable conditions of life;

3. is the competence of communicative actors to subject their communicative actions to universal validity claims and thus to fulfill conditions required to criticize and, hence, understand them (Habermas), then, under the assumption that communicative actors pursue communicative objectives, it is certainly possible that they make use of nonlinguistic means in doing so; here, however, the adequacy of these means in the case of expressive action is only subject to limited intersubjective examination;

4. is the competence of actors to optimize their actions and beliefs relative to their well-understood individual interests (Gosepath), then media can be means to implementing such optimizations;

5. is understood as a set of normative demands that are directed to interpreters who want to describe and understand the behavior of beings as action by imputing desires and intentions to them (Davidson), then such interpreters might feel compelled to impute to these beings expressive intentions that they implement by means of media;

6. is understood as a competence of systems to reduce the complexity of their environments under conditions of self-maintenance and to assume a perspective that relates the selectivity of the systems to the improbability of their existence (Luhmann), then we can easily imagine that they accomplish this by the use of medial means—whatever that means.

If we go through the individual points, it quickly becomes clear that all the concepts are able to integrate the media-theoretic reflections insofar as they can view (nonlinguistic) media as means for implementing communicative intentions as well as other desires. From this perspective—which completely concurs with the intuition of the charitable reader—media appear to be instruments, and indeed instruments like dough hooks or toothbrushes. They are a part of what Heidegger called equipment (Zeug). If one wants to evade this kind of assimilation (Vereinnahmung), then it must be shown that the main point of media theory has escaped the allegedly charitable reader, and, as a result, he deals with a sterilized media concept that virtually guarantees the subsumption of media under an antecedent rationality concept. If I am correct, this subsumption is possible because the established concepts of reason assume a specific relationship between reason and language, which, for its part, has consequences for the relationship between reason and media. For what would happen if, in the face of the above finding, we were to say: now media behave exactly like language. Media and language are means for fulfilling (communicative) ends. I think that the proponents of conceptions of reason would attempt to show the validity of the view that the relationship between language and reason is fundamentally different from the relationship between media and reason insofar as reason is something that simply cannot exist independently of linguistic competencies. The connection between reason and language is accordingly constitutive rather than instrumental.

However, precisely because, according to this view, reason exists independently from nonlinguistic media, the concepts can place media under the umbrella of reason in the way sketched out above; here media do not play a part in constituting reason. However, as I attempt to show in what follows, this is only plausible if one falls back behind the media concept that I have developed. Because the established theories believe that they can solely entrust language with constituting reason, they are able to apprehend the relationship between reason and media as merely instrumental. If we would like to oppose the view that media have a mere instrumental status, then we must first of all understand what makes the constitutive relationship between language and reason so plausible and why it also appears to be clear that nonlinguistic media cannot play a role in constituting reason.

With a view to the relationship between language and reason, in essence the four following theoretical motifs can be offered as arguments for the constitutive relationship. First, language is thought to make possible the individuation of (higher) intentional states, i.e., of contentful mental states; this comes to expression, inter alia, in the fact that having a thought is identified with the inner articulation of a sentence. However, because only beings with thoughts can be rational, it is at the same time clear that only linguistic beings can be rational. Second, language is apprehended as making the development of a self-relationship possible, so that only one who masters a language can develop the ability for self-reflection and with it the ability for self-criticism. However, because self-criticism is a condition for the process of weighing alternatives, which is specific to rational beings, rational beings have to be linguistic beings. Third, because only language is considered to allow the formulation of rules, and the ability to orient one’s actions on principles, i.e., on rules, is a basic dimension of rational behavior, one would have to deny nonlinguistic beings a fundamental aspect of rationality. Finally, fourth, only language is entrusted with the ability to make something explicit that our behavior is implicitly based on. By making something explicit, however, one places it in the realm of reasons; here it is subject to the light of critique. And, of course, only beings that operate within the realm of reasons are considered rational beings.

Even if these four motifs are connected with each other in diverse ways and are weighted differently in the different conceptions of rationality, they nevertheless depict the core of those beliefs that form the background for the view that only linguistic beings can be rational beings. Because it is unclear what it might mean for a thought to have content if this content cannot be linguistically articulated, and because, beyond this, it is unclear how nonlinguistic media might make a self-relationship possible, how they could explain the formulation of rules or the elucidation of implicit norms, it appears at the same time to be evident that the relationship between nonlinguistic media and reason is external, in a word, instrumental. If one would like to show that this assessment is wrong, then one must show that and how nonlinguistic media can assume roles that are comparable with the role of language with a view to the competencies mentioned above. In short, one must answer the question of whether it is also sensible to say that nonlinguistic media have the ability to be constitutive conditions for reason.

4.1.2. Media as Conditions of Reason

In what follows I would like to proceed by discussing the four mentioned motifs relatively concisely. I will look into whether it is possible to imagine that their basic relations can also be realized by means of nonlinguistic media. While the reflections of chapter 3 were marked by the attempt to buttress an intuition, now I will not mince words in treating the ideas developed there, but will attempt to examine them as forthrightly as possible.

4.1.2.1. Content

From an interpretationist framework, to speak of the content of thoughts is to speak of theoretical entities of interpretation. Content is what we assume if we understand beings by rationalizing their behavior. Here we impute to them states that have an identity within a holistic network of propositional attitudes, which, for its part, is structured inferentially. According to this basically Fregean perspective, however, content then has the form of a sentence of necessity, because sentences are the smallest unities that can play inferential roles, i.e., the role of reasons.2 However, with Brandom, we must then say that musical expressions or works possess no content, because they cannot be put into the form of sentences, and thus they cannot be understood within the framework of the inferentialist theory of content. For “understanding in this favored sense is a grasp of reasons, mastery of properties of theoretical and practical inference.”3 If one accepts this view, then it follows (according to interpretationist premises) that those mental states that are causes of musical action cannot have content. If, in contrast, on the basis of the reflections of the previous chapter, one wants to claim that the B-intentional states that I postulated there certainly have content, then this content can certainly not be due to the fact that these states can play the role of premises or of conclusions. If B-intentional states are supposed to measure up to linguistically individuated thoughts in the sense of having content, then an interpretive relation is required that ascribes them content without thereby imputing to them the status of reasons.

In the previous chapter I proposed4 that nonlinguistic expressions, in the context of the social interactions that they are developed in, acquire content by virtue of the fact that, in a first step, through the formation of expectations, they are correlated with the regular reactions of the recipients of such expressions, and, in a second step, they are joined to those reactions that the producers of such expressions experience for themselves or anticipate as reactions of potential recipients. With a view to the process in the course of which expressions and reactions are joined to one another, we need not assume that the constancy of the reactions to types of expressions, for its part, is supported by the recipients’ reason-guided behavior. We can thus assume that the content of nonlinguistic expressions is ascribed by a form of social interpretation that is itself not a linguistic interpretation. However, if interpretation is the only way in which a mental state or an expression can acquire content, and there are nonlinguistic interpretations, then there are beings that fulfill one prerequisite for being rational beings—namely, of having thoughts with content—without necessarily thereby being linguistic beings. With B-intentional states we have not yet entered the realm of reasons, yet we have left the realm of nature insofar as we are concerned with socially conferred content.

A fundamental pragmatic prerequisite for the ability to socially confer content is that media—regardless of whether these are nonlinguistic or linguistic media—are not instantiated by types of expressions that themselves already have content, independently of social interpretations. In other words, it requires types of expressions that are semantically intrinsically empty, i.e., that have no intrinsic content. Here it should hold:

(IC) A form of behavior B has intrinsic content if and only if recipients of B react to B as they would react to a natural sign.5

Given a certain biological endowment in a recipient R, R cannot receive a form of behavior that has intrinsic content B without reacting to B as if that which B stands for is present; in the case of natural signs, this consists in a cause C that brings about B. Here B is caused by C and only C. In contrast, only those expressions are suited to be types of medial expressions that bring about states or events that do not play the role of natural signs, because these would indicate—in a way that authenticates natural law—the states of a being that brought them about.6

The media-theoretic analysis thus draws our attention to the fact that the social dealings with types of activities that are semantically intrinsically empty create the foundation for our ability to develop something that is called mind; for the operation of mind consists in dealing with (inhibited) types of expressions that obtain content in the context of social interaction. Here, with nonlinguistic media—as with languages—the differentiation of content is dependent on the differentiation of the medial constellations.7 In the development of this foundation of mind, nonlinguistic media thus mark an important step, which is further specified by language. In this way, from a media-theoretic perspective, we at the same time gain a clearer view of the role of language; here, an advantage of this perspective is that it allows language to be described in a theoretically homogeneous vocabulary as a specific medium—a medium, namely, that is subordinate to normative restrictions that allow it to become a medium for giving and asking for reasons. If the creation of content justifies the intimate relationship between language and reason, then we must account for the fact that nonlinguistic means also make the individuation of content possible if we replace the mere instrumental relationship between reason and media with a constitutive one. If, beyond that, it can be shown to be plausible that the content that can be individuated with the help of nonlinguistic media cannot necessarily be articulated with the help of language—if then this content is connected to certain forms of nonlinguistic articulation, which, for their part, cannot be translated—then we can claim that this content cannot be reduced to what is able to be linguistically articulated. For nonlinguistic medial expressions this is fulfilled in particular by the fact that the experiencing of the medial expression remains linked to those physical attributes that the implementation of a medial expression exhibits. At the same time, the ability to substitute, paraphrase, and translate linguistic expressions shows that the medium of language has, to a significant extent, been emancipated from the sensory elements that constitute one of the causes of the untranslatability of nonlinguistic medial expressions.

4.1.2.2. Self-Relationship

An answer to the question of whether a sphere that is rightfully denoted a sphere of linguistically independent self-interpretation can be characterized by media-theoretic means will have to be positioned in reference to two views: on the one hand, to the interpretation that quite self-assuredly tends, for example, to understand the painting of children as a form of expression by which children develop a relationship to themselves; on the other hand, to the interpretation that becoming well versed in a medial practice means only that one possesses a disposition to react to experiences by producing certain pictures. While the first position interprets the children’s pictures as a nonlinguistic form of self-interpretation, the second one views those pictures as mere symptoms; and while the first position has to postulate something like a self-relationship, which at any rate is not necessarily generated with a linguistically realized self-interpretation, the second position assumes that the symptomatically understood pictures can only be observed as material of an interpretation from the perspective of an interpreter that possesses language.

Within the framework of the four-phase model, I have attempted to show that it is plausible that medial constellations obtain content in the phase of medial communication by virtue of the fact that individuals who are familiar with a medial practice react to the medial performances of children in an exemplary way. These reactions succeed reactions to affect-steered expressive behavior in which reference persons mirror the states of children in a process of marked mimetic communication. If medial behavior is joined to the expectation of a nonlinguistic response behavior, and the response behavior covaries with the compositional identity of the medial expressions, then a child that is socialized in a medial practice learns to order medial expressions in line with the proprioceptive perception of the act of carrying out the expression and in line with the effects of the social response behavior on itself. In this process, types of medial expressions become the means for referring to types of experiences.

Already at an elementary level, nonlinguistic medial constellations can thus assume a form in which, by virtue of the effect of the medial means on themselves, producers of medial constellations can envision experiences. Because these means do not have any intrinsic meaning, like the mimetic expressions of affects, constellations can obtain meaning in line with their receptive effects. Because the medial production of medial constellations is a self-controlled form of articulation, and the production can be controlled by the producer in line with its effect on the producer, for their producers medial constellations are instantiations of a self-relationship. Here, with medial means, they refer to an experience that is adequate or inadequate relative to the state in which they find themselves. Then, however, one can say that the medial constellations structure the access to those states, relative to which, on the basis of their effects, they are adequate or inadequate.

4.1.2.3. Rules

With a view to the problem of whether the aspect of rational action that makes it possible to follow maxims or principles, that is, to follow rules, can be reconstructed by media-theoretic means, we must, first of all, distinguish between two levels. On the one hand, it is clear that in order for an action to be defined as a rational action, it is not required that the action have a rule-following character. For rational actions, we do not necessarily expect the content of rules to establish the reason for the action. In order to understand an activity as a rational action, what we need are reasons (in the framework of assuming minimal rationality) that we can ascribe coherently. On the other hand, the ability to follow rules or to act on the basis of principles is a specific characteristic and prominent aspect of rational action.8 If language is the only medium for formulating principles, then we must admit that nonlinguistic media cannot contribute to constituting this aspect of rationality. However, I have attempted to show that we can plausibly postulate a special form of explicit rules with the help of the concept of demonstrative rules that is characterized by the fact that it articulates rightness nonlinguistically and medially.9 However, because the normative component is articulated within the parameters of this concept, with the help of an imperativist linguistic form, the possibility that explicit rules can exist in a language-independent form appears to be dependent on our ability to demonstrate how the imperativist component can also be nonlinguistically implemented. In attempting to show this, we are confronted with the following task: if we want to reconstruct the concept of rule following solely using nonlinguistic instruments, then we must ensure that both the part of a rule that characterizes the right or wrong behavior as well as the part that articulates the imperative character that is expressed by rules can be implemented nonlinguistically.

In connection with the conception of demonstrative rules, we, after all, find ourselves in an interesting theoretical position, namely, that we can implement these two components by different mechanisms, and it is not necessary to burden one concept alone—for example, the concept of a sanctioning practice—with the task of specifying the content and conveying the imperative.10 The strategy sketched out here is thus a strategy of decoupling the individuation of the content of rules and the implementation of the imperative modus. And because the concept of demonstrative rules allows us to entrust the individuation of the content of nonlinguistic rules to a deictic practice, a sanctioning practice could limit itself to conveying the imperative. For this, we must only assume that, for example, a teacher, who can deictically refer to and bring about what is right, negatively sanctions a student if and only if the student, for example, makes a mistake in the production of medial elements.

Admittedly, it can be objected that the deictic practice is overtaxed as a content-individuating mechanism insofar as, in deictic situations, it is not at all clear what is being pointed to or what is demonstrated as right. In opposition to a deictic individuation mechanism, one could thus formulate an objection along the line of arguments on which Føllesdal11 bases his objections to Davidson’s causal reference theory: just as, in the triangulation situation, there is a lack of clarity about the cause of the similar reactions among students and teachers, so with the deictic individuation mechanism, there is a lack of clarity about what the sanctioner is really pointing at. Is the nonlinguistic vocal instructor concerned about the positioning of the lips, the effects that the singing has on animals in the surroundings, or the pitch? Arguments of this type can perhaps not be dispelled in principle, but they can be largely rebutted if sufficiently strong restrictions are introduced for deictically implemented teaching practices that are carried out without the help of language. Acceptable deictic practices—that is, practices that can be accepted as mechanisms for individuating the content of nonlinguistic rules—must then exhibit the following characteristics:

1. The right performance must be repeated frequently.

2. The repetition of the right performance, which is held stable, occurs in different contexts in which the deixis does not refer to attributes that are varied.

3. The right performance is marked by nonlinguistic means, normally by exaggeration.

While this last criterion admits a potential attribute of deictic action that is deeply rooted in the prelinguistic communication between infants and their reference persons,12 the first and second criteria prescribe characteristics for deictic actions whose meaning can be made plausible in the conditionality analysis or other neuronal models of learning: the repetition functions to establish a type, and varying the environmental conditions functions to validate the type ex negativo.

In light of the consequences—which are difficult to take into purview—I do not want to commit myself;13 yet a strategy may begin to show itself that makes rule-following behavior possible, even independently of language. If this analysis allows medial competences to be described as competencies of rule following that are not joined to a linguistic form of explicit rules, then, as a character of rational action, the competence to act according to rules or principles is not linked to language either.

4.1.2.4. Explicitness and Reasons

If we explain a particularity of rationality in a Brandomian perspective so that reason consists of the ability to make explicit the inherent norms of rightness in social practices and thus to make them potential objects of critique, then it is not possible to see how elucidative rationality could have an analogue in nonlinguistic thinking. For from the perspective of elucidative rationality, medial practices may indeed be based on standards of rightness, but these practices do not make things explicit, and, when they do, they require language. In short, I do not see how we can understand nonlinguistic communicative practices to be practices that make their normative basis explicit. Nonlinguistic medial practices instantiate rightness; however, without language they are not able to form rightness into rules or principles, and thus to explain it. If we view the rightness that is specific to a specific medial practice, then we can say that what is considered right in this practice, what is accepted as right, can be made explicit in a deictic practice. What is right can be demonstrated at any time, and the function of the demonstrative rules consists precisely in providing what is demonstrated with an exemplary character. Here the concept of demonstrative rules14 allows us to accommodate the fact that it is possible that what is right (in a given social context) cannot be made explicit linguistically at all. But it is important to see that, with this form of explicitness, it is only possible to affirm that which is considered right. In the context of the mere deictic reference to what is right, critique is not carried out by critiquing explicit norms, but by bringing about other paradigmatic forms that are endowed with the status of rightness by becoming media of communication and thus being passed down. In these contexts, “critique” is carried out more by replacing what is valid than by disputing the justification of what is valid.

On the other hand, one could maintain that, with the help of nonlinguistic media, it is possible particularly to make explicit what the content of a medial constellation is. We could then say that medial constellations, for example, make the structures of experiences explicit. This, however, would imply that something would be made explicit by medial means that exists independently of its being articulated. And this does not appear to me to be the case. For van Gogh’s Peasant Shoes does not make anything explicit that someone with a practiced eye could discover about peasant’s shoes or—after having seen the picture—that can be perceived about such shoes with hindsight. The picture shares rather a B-intentional state or an experience that is reliant for its individuation on the individuation of the picture.

What about the possibility of understanding reasons as something that can be implemented by nonlinguistic medial means? If we view B-intentional states as medial thoughts, that is, as states that have content and an identity for the being that thinks these thoughts, then there is no reason that B-intentional states cannot assume the role of reasons. This assessment can also be supported by the fact that B-intentional states are among those mental states that, by virtue of their content, can also play the role of causes of action. On the other hand, this attribute of the mental cause of an action depends on the existence of a rationalizing interpretation of the activity that actualizes the action, in which the cause is regarded as a reason. I have proposed (H2)15 as a schema for rationalizing artistic action; and in the first premise of (H2)—fully analogous to the common rationalizations of action in which beliefs or desires are ascribed—a medial intentional state is ascribed. In addition, the medial intentional state would also fulfill the counterfactual criterion for indentifying the right causes of an action: if P had not found herself in state Ib, P would not have produced the work of art. Notwithstanding, the medial intentional state has a different status than, for example, the desire to express Ib. Such desires or beliefs are irreducible components of explanations of action, and Ib only plays the role of a reason in contexts that incorporate desires and beliefs. The fact that Ib plays the role of a reason in (H2) is dependent on the fact that Ib is embedded in the context of desires and beliefs. Because Ib is neither a belief nor a desire, Ib requires this context in order to be a reason. In short, medial intentional states can function as reasons, but we can only ascribe them this function in the context of the classical intentional vocabulary.

4.1.2.5. Summary

The attempt to describe competencies characteristic of rational action as competencies that can be carried out by medial means has yielded the following view:

1. Mental (B-intentional) states that are individuated by media have content, and they have this content for the beings that have these states. But, unlike propositional states, here it frequently is not possible to express this content in any but one—and only one—medial constellation.

2. Media can be understood as the conditions for the development of a self-relation, because, with the help of medial-affecting experiences, beings can draw on inner states.

3. Media are products of social practices that can be understood as practices that are guided by rules without needing to assume that what is right is characterized linguistically. But medial practices only instantiate this rightness; in a narrower sense, they are not means with which rules—as with the help of language—can be formulated. Admittedly, the correctness of this evaluation is posited on the implausibility of a form for articulating rules that is completely supported by deictic means.

4. Strictly speaking, media do not make anything explicit. But they allow an access to structures of perceptions and experiences that we must assume are partially structured independently of media and language insofar as the acquisition of media and language competencies already presuppose the ability to identify medial expression types.

5. Medial intentional states can take on the role of reasons insofar as their content can be causes for medial action. But it is only from the perspective of an interpreter capable of speech that they claim to take on the role of reasons.

4.1.3. Another Perspective

The questions of the preceding sections are concerned with appropriating the diagnosis that rational competencies are based on linguistic capabilities. With this perspective, I have tried to clarify the degree to which these competencies can be understood as being based on medial competencies. However, it is specific to such questioning that the reason-constitutive aspects of media are described according to standards of paradigmatic linguistic capabilities. One can, however, also question whether media generate a specific reason-constitutive achievement that is not brought adequately into view from the perspective of linguistically authored competencies or that is not brought into view from that perspective at all. In order to make this plausible, it must be demonstrated that the concept of reason, which media are supposed to make an independent contribution to constituting, within the parameters of this procedure, is not just broadened precisely to include those aspects that guarantee that the media make a constitutive contribution. In other words, one must ensure that the concept of reason has an inner homogeneity such that it is not so malleable that it merely embodies all those capabilities that one places in a constitutive relationship to it. If this threat is not precluded, then arbitrary constituting relations can be designed so that the concept of reason would only reflect what one is presently characterizing as its constituting basis. If one thus wants to show that media make an independent contribution to defining reason, then this contribution must be included in a conception of reason in such a way that it can be defined independently from these specific contributions. In the introduction to this book I provided a diagnosis—namely, that the theory of rationality is converging toward a theory of understanding as the fundamental level of explication—and I stated that the core rational competencies are those that allow us to understand something. If media are thus supposed to be able to contribute to defining reason, then it must be possible to illustrate this contribution with the concepts of understanding.

Before I attempt to show the specific contribution that nonlinguistic media make to the competence of understanding, in a short excursus I would like to show that some of Kant’s thoughts can be read as a suspenseful attempt to characterize a broadened concept of understanding (Verstehen) that moves beyond the view that the process of understanding is carried out solely by conceptual-linguistic means. In doing this, I would like to show, within the framework of what is possible here, that theoretical motifs of the Kantian aesthetic can be reformulated by media-theoretic means within the framework of the interactionist paradigm, while avoiding some conceptual ambivalences and counterintuitive presuppositions.

4.1.3.1. Kant’s Theory of Aesthetic Experience as a Theory of Nonconceptual Understanding

Very little is there to reveal things to us . . .

I. Bachmann16

In his Critique of Judgment, Kant, among other things, sought to show that aesthetic experience is based on specific competencies that cannot be reduced to theoretical or practical competencies for dealing with objects of aesthetic experience. In contrast, as Höffe vividly describes, “in the aesthetic relationship to the world . . . [there is] a distinct form of rationality,” which “can neither be traced back to objective cognition nor to morality, nor to both.”17 In the context of the reflections here, it is, of course, of particular interest that Kant sees a particularity of this form of rationality to lie in the fact that we develop interpretive thought in aesthetic experiences, which have content that systematically evades conceptual articulation. In what follows I would like to suggest an interpretation of Kant that aims to uncover theoretical motifs that are suited to serve as links for the theoretical formulation of nonconceptual rational competencies.

As is well known, Kant conceives of the particularity of aesthetic experience such that in aesthetic experience sensibility and spontaneity interact with one another in such a way that the representations (Vorstellungen) that we form of a sensory-perceived phenomenon are brought together, under the direction of the reflective judgment, with conceptual productivity such that an integrated representation emerges without the perceived phenomenon thereby being subsumed under a concept. In contrast to the determinant judgment, which subsumes the particular that is provided by the senses under the general, a rule or principle, that is provided by the understanding (Verstand), the reflective judgment searches for the general under which the particular that is provided through intuition could fall. As reflective judgment, in this it proceeds—according to the subjective transcendental principle—to view the particular, in accord with its form, as purposive for our cognitive faculty.18 If we now examine the representation of a particular from the perspective of the subject that has the representation, then we come up against its aesthetic constitution insofar as it is subject to a specific evaluation by taste—an evaluation in which the purposiveness of the object for our cognitive faculty is judged. If objects of sensory experience manage to bring imagination and understanding into a harmonious relationship—without this having been intended—then this relationship generates a feeling of pleasure that Kant traces back to the purposiveness of the object of the representation for the reflective judgment, which is not dependent on its conceptual determination.19

In this perspective, an aesthetic experience does not emerge as an exotic special kind of experience, but as a fundamental kind of experience; on the basis of it, the conditions of the elementary individuation of experience can be analyzed without the need for a concept of the object of experience. The analysis of aesthetic judgment thus also aims at an explication of experience, which does not allow the conceptual structuring of intuitions. For even if, in contrast to the common experience, we do not end up, within the framework of aesthetic experiences, with a conclusive conceptual structuring of intuitions, the process in which the sensibility and spontaneity interact with one another is not in vain. From the perspective characteristic of judgment toward subjective purposiveness of the form of the object for our cognitive faculty, the object provided in the experience is indeed, on the one hand, apprehended independently of concepts; on the other hand, however, this is carried out precisely by virtue of the fact that the intuition and the concepts come together in cognition generally (einer Erkenntnis überhaupt).20

1. The concept of cognition generally, which Brigitte Scheer places at the center of her interpretation of Kant,21 here has the function of characterizing a synthesizing activity of mind, which emerges from the free play of imagination and understanding in the face of an aesthetically experienced phenomena and “need not yet be cut short by the stipulation of a concept.”22 Representations of objects of aesthetic experience do not have an identity due to the fact that they are (identifying and) identified by means of the classifying power of concepts, but due to the fact that they exist in a (challenging) relation to our cumulative possibilities to make phenomena intelligible. I suggest that we interpret the expression “cognition generally”23 as a characterization of a kind of intentional state that indeed stands in an interpretive relation to an object of experience, but neither in order to identify the object nor to subsume it under concepts. If we would like to more precisely understand how such states are theoretically designed, then we must above all more precisely understand the unique place of the conceptual in aesthetic experiences. If we call to mind Kant’s thoughts about the role of concepts in the context of aesthetic experience, then, however, the following ambivalent view emerges:

a. Concepts are necessary conditions for referring to objects of experience at all; they are, as Kant showed in his Critique of Pure Reason, necessary conditions for the knowledge-constitutive synthesis.

b. Insofar as the aesthetic experience is oriented toward the sensory appearance of things, their conceptual determination plays no role: aesthetic judgments are not knowledge judgments.24

c. Because aesthetic experience puts into motion the free play of the cognitive faculties, which understanding, with its concepts, necessarily participates in, concepts are a necessary condition for aesthetic experiences, even if having an aesthetic experience does not entail the conceptual determination of an object.

d. In principle, the structure of what emerges with the free play of the cognitive faculties, namely, cognition generally and aesthetic ideas, evades conceptual articulation.

e. The ability to judge an object of experience aesthetically is not a conceptual power, but a power of feeling (Gefühlsvermögen).25

If I am correct, the ambivalences of these motifs can be resolved if we limit the role of concepts so that concepts, on the one hand, play a necessary, though not a sufficient, role for the possibility of aesthetic experiences; on the other hand, however, especially for determining the content of an aesthetic experience, they are insufficient. Because Kant does not want to understand aesthetic experiences as experiences that simply have not yet led to a synthesis that identifies the object of the experiences—which would thus merely be an early stage of a process that, for contingent reasons, is incomplete—a solution must be found in which conceptual aspects are among the necessary framework conditions of aesthetic experience, but wherein they are not part of what defines those experiences. To do this, I will attempt to understand Kant’s reflections such that his concern in identifying “cognition generally” is with a nonconceptual form of intelligibility, a form of comprehensibility (Einsichtigkeit) that sensory objects can have for us because they confront us as if they were made for us and our cognitive apparatus. However, while Kant joins this relation to the point that, from the perspective of understanding, we must view this “fitting” in the case of natural objects as coincidental, and we thus cannot expect it,26 and our pleasure comes to indicate that we fit into nature, I would like to orient myself on a less perplexing view of the comprehensibility of sensory objects. I would like to claim that it is necessary to reformulate Kant’s view, namely, that the fit of nature to our cognitive faculty can be made plausible not with causal concepts, but with teleological ones, in such a way that human artifacts are the exemplar for teleological structure, and we only derivatively describe nature with teleological concepts. To speak with Dennett, when faced with nature, we assume the design stance, which we initially rehearse with reference to human products. However, if, in inversion to Kant’s weighting, we place the primacy of the artistically beautiful before the naturally beautiful, and analyze aesthetic phenomena at a fundamental level as communicative phenomena, then the nonconceptual form of intelligibility can be shown to be more plausible. While the first view certainly stands in contrast to Kant’s ideas, a series of links can be found in Kant’s text that supports an emphasis on the basic communicative structure of aesthetic experiences.

2. In the analytic of the beautiful, in which Kant is concerned with the analysis of aesthetic judgment, the communicative function that such judgments can take on plays an important role with a view to determining the validity claim (quantity) of the judgment of taste. Kant analyzes aesthetic judgments not as merely subjective evaluative-expressive acts that have no claim to general validity, but as evaluations that allow those who are making the judgments to lay a claim to a transindividual dimension of aesthetic experience. The basis for this potential “general validity” (Gemeingültigkeit) of aesthetic judgments is that the different recipients have the same mental apparatus, the game states (Spielzustände) of which are supraindividual. Kant can thus expect a “universal capability of communication of the mental state in the given representation,”27 which is not brought into being by concepts. In comparison to Kant’s general strategy, namely, to show that nothing can stake a claim to being a principle of reason that cannot be followed by all—thus that is not in a basic sense transindividually intelligible—the reference to the ability to communicate pleasure, which begins from a representation in reflection, seems rather weak. For there appears to be no particular reason—that can be explained and thus transindividually accepted—why an aesthetic judgment can be demanded of everyone; there appears rather to be an empirical cause, namely, that we share a certain mental endowment. Further, however, beyond the fact that a representation has led to a pleasure originating from reflection, the content that is communicated with the help of judgments of taste is not further determined.28

In contrast, if we do not view the communicative function of aesthetic judgments but of works of art, another element is introduced that Kant calls Geist and that could also be called Esprit. We expect that works of art originate from the “faculty of presenting aesthetical Ideas.” And an “aesthetic Idea”29 is a “representation of the imagination which occasions much thought, without, however, any definite thought, i.e., any concept capable of being adequate to it; it consequently cannot be completely encompassed and made intelligible by language.”30

However, the intersubjective principle from which the presentation of aesthetic ideas can take on a communicative function remains unclear. Kant does not even view the faculty of expressing aesthetic ideas as a competence that is intersubjectively acquired in order to portray something with the help of aesthetic (and not “logical”) attributes, but as a talent, which he understands as a natural gift.31 That the talent is apprehended as a natural gift, for its part, has the function of reconciling the artifact character—which is connected to precedent rules—that every work of art possesses with the view that a work of art should not merely be viewed as resulting from the arbitrary following of rules that can be conceptually explained: “But since at the same time a product can never be called Art without some precedent rule, Nature . . . must give the rule to art, i.e., beautiful Art is only possible as a product of Genius.”32 Herewith, however, the freedom from which the genius creates, in contrast to the systemization otherwise found in Kant, is precisely not a freedom that results from mind’s connection to comprehensible principles, but from the “innate mental disposition (ingenium) through which Nature gives the rule to Art.”33 In order to avoid these difficulties, a procedure is required with the help of which the particularities of the aesthetic experiences and articulations can, in harmony with the Kantian intuitions,34 be connected with recourse to transindividual structures, which nonetheless do not simply have the form of explainable rules. In short, we must identify structures that make possible an intersubjective practice that is open to all. If, as I believe, nonlinguistic media can play this role without what is articulated with their help being subsumed under the conceptual, then it should be possible, in a clearer way, to maintain Kant’s intuition that the aesthetic attitude is based on a separate form of rationality. A necessary condition for this, however, is precisely that aesthetic practices are described as communicative practices.

3. If we attempt to accommodate this last-mentioned condition, then we must free Kant’s aesthetic—entirely in the sense of Hegel—from the primacy of natural beauty and maintain without qualification that natural beauty is derivative from the beauty of artifacts.35 For Kant, the concept of nature especially plays a foundational role because, with its help, Kant can grant aesthetic phenomena an ambivalent position between a separate purposive freedom, which protects the phenomena from being subsumed under purposes, and a formal purposiveness for our cognitive faculty. Using the concept of purposiveness, Kant articulates the possibility of structuring nature, which he views as a subjective prerequisite for the possibility of experience. The faculty of placing sensory appearances under the perspective of formal purposes is the faculty of structuring sensory appearances in accord with an assumption that makes it plausible that the sensory appearance is made for us. The fact that we experience this as “beautiful” positions us in a specific way toward nature. But this assumption would, of course, lose all quasi-religious connotations if the phenomena were made for us because they were made by other people. If, along with Hegel, we view art as a form by which we develop a relationship to ourselves and link the possibility of such a relationship to the requirement that others stand in an interpretive relationship to our articulations, then a concept of beauty that leans on Kant’s views can be explained with recourse to the success of communication that does not owe this to explicit rules, which are, because of this, potentially generally valid rules. In this sense we might be able to call an artifact “beautiful” if we manage, with its help, to communicate experience. According to this view, beauty is an attribute that an experience of the intelligibility of a perception possesses; here the medial structuring of an object succeeds such that, in this, we generate an integrated episode of experience. If we assume that cognition generally, in the sense I have explained above, is made possible by the fact that the sensory object of experience is nonconceptually structured, and we further assume, in line with interactionist requirements, that the means for this structuring are not inborn but are socially acquired, then the “general validity” of judgments of taste can be made understandable in a simple way.

Of course, one could claim, against this reconstruction, that this is irreconcilable with Kant’s definition of beauty as “the form of the purposiveness of an object, so far as this is perceived in it without any representation of a purpose,”36 because beauty in the proposed reconstruction would be viewed from the perspective of communicative ends. But because of the formal purposiveness relative to our cognitive structure, the reconstruction of the formal purposiveness using concepts of communicative intelligibility remains sufficiently connected to the Kantian definition. It can thus still serve as a reconstruction insofar as the intelligibility of nonconceptual communicative artifacts is projected onto artifacts similarly to the way the subjective principle of formal purposiveness is projected onto nature. An assumption serves as a subjective condition for the possibility of judging such objects aesthetically; namely, it is necessary to view those experienceable things that other people generate as having a form that is lent dimensions by the structure of our capacity for having experiences. If one now understands formal purposiveness as resulting from the fact that the objects of experience display a medial structure with which we are familiar, at least in principle, then those tensions can be avoided that characterize Kant’s description of the role of rules in the context of aesthetic experiences. For, on the one hand, aesthetic judgment is “a special faculty for judging of things according to a rule, but not according to concepts.”37 On the other hand, “every art presupposes rules by means of which in the first instance a product, if it is to be called artistic, is represented as possible.”38 Here, however, the work of art should not leave a trace “of the rule having been before the eyes of the artist and having fettered his mental powers.”39 Kant appears to want to bring these different motifs together in the following argument:

a. There are rules that artists must follow if they want to create works of art at all, because the character of works of art as artifacts depends on it.

b. Nothing is a work of art that merely applies rules.

c. Because conceptual rules are general, and by following them only applications like (b) can be produced, the rules named in (a) cannot be rules that are able to be fully conceptually determined.

d. “Therefore beautiful art cannot itself devise the rule according to which it can bring about its product,” and “Nature in the subject must . . . give the rule to Art.”40

If I am correct, (a)–(c) can also be integrated into conclusion (d′).

d′. Thus: the necessary rules that artists follow are demonstrative rules as laid out in (R1).41 They only prescribe how the material from which the work of art is formed is produced, but not how it is to be used in order to produce a work of art.

In my view, conclusion (d′) would have the great advantage of being able to situate artistic production in an intersubjective social context, and thus it would be connected better than (d) with Kant’s basic intuition that links the possibility of rationality with the condition for the existence of practices that are in principle open to all. If these practices are analyzed as communicative, then we can say that aesthetic communication is centered around the communication of experiences, and indeed in the sense that experiences are communicated in the form of medial-structured objects of perception. And Kant himself points out that the structuring competencies, that is, those capabilities that allow us to position a perception in the light of the products of imagination, are socially transmitted. Kant writes:

The skill that men have in communicating their thoughts requires also a relation between the Imagination and the Understanding in order to associate intuitions with concepts, and concepts again with those concepts, which then combine in a cognition. But in that case the agreement of the two mental powers is according to law, under the constraint of definite concepts. Only where the Imagination in its freedom awakens the Understanding, and is put by it into regular play without the aid of concepts, does the representation communicate itself not as a thought but as an internal feeling of a purposive state of the mind.

Taste is then the faculty of judging a priori of the communicability of feelings that are bound up with a given representation (without the mediation of a concept).42

In the Critique of Judgment, with the concept of “cognition generally” and the concept of the “aesthetic idea,” Kant introduced two concepts that can be interpreted as products or objects of a nonconceptual understanding. I think Kant’s observation that successful works of art present specific challenges for our structuring capabilities is correct and would like to suggest that a particularity of aesthetic experience consists precisely in the fact that what the subject of an aesthetic experience provides for structuring the object is not merely conceptual (in the narrower sense), but precisely is an attempt to reconstruct the phenomenon medially. From the perspective of these considerations, we would try to explain the preestablished harmony that is assumed by aesthetic judgment to exist between the forms of artifacts and the structure of our cognitive faculties to result from the fact that we share medial competencies with the producers of aesthetic phenomena. The core of aesthetic experience would thus lie in the correlation of an experienced phenomenon with those structuring medial-interpretive correlates that—because of our familiarity with the medium in which the work of art has a compositional identity—have a history of generating experiences. As a consequence of these reflections, aesthetic experience is thus described as a kind of experience that integrates experienced phenomena at a fundamental level into the realm of things toward which we can behave interpretively.

In the Critique of Judgment Kant centered attention on a dimension of aesthetic experience that must evade every merely experience-centered aesthetic, and he attempted to show that aesthetic experience possesses its own rationality. However, because Kant must define all the particularities of this rationality in relation to conceptual competencies, and the analysis of the communicative functions of aesthetic experiences is not available to him in a systematic way for determining the specifics of rationality, his determinations turn out, on the one hand, to be strangely formal, and, on the other hand, to be oddly negative. From a media-theoretic perspective, one can concur with Kant that aesthetic experiences are not exclusively conceptually structured; that the production and interpretation of works of art do not or cannot follow conceptually explicit rules; that the content of works of art cannot be conceptually identified; and, finally, that aesthetic experiences are not limited to an appreciation of sensory impulses, but are characterized by a certain form of reflection. However, insofar as one can say with media-theoretic means that this reflection results from a medial structuring of experiences, and the competencies for such acts of structuring that are acquired in communicative processes form the core of nonconceptual reason, a perspective emerges that allows the difficulties of the Kantian track to be avoided. This would at least indicate how a philosophy of media based on interactionist basic beliefs could reconstruct those competencies that have thus far largely found no place in the philosophy of the linguistic turn.

4.1.3.2. The Rationality of Medial Understanding

The examination of Kant has at least shown that my attempt to uncover a special form of mental activity in aesthetic forms of experience, which is subject to distinct principles, is a project that is not without its historical precursors. In the following I will attempt to establish that this kind of mental activity is a form of understanding, and that the analysis of this can contribute to the explication of rational competencies. Here, those correlations in which medial constellations (K) either appear as interpreting structures or that are themselves the object of an interpretation by medial structures must form the basis of such an analysis. Systematically then, the following correlations can appear:

(C1) P interprets X with the help of K.

(C2) P interprets K with the help of K′.

In order to be able to analyze both cases separately, apart from the fact that X is not a medial constellation, there should be no restrictions about what X can represent.

1. In examining the correlation (C1), we are examining the relations that form the basis for the fact that a medial constellation (K), from the perspective of a person (P), can assume a form in which P develops an interpretation of something (X); and the one standard case (to which I will limit myself here) may be that P taps a sensory experience. In this perspective K is a means for articulating the attributes of X in a form that is determined in some way by X.

Let us assume a very simple example: In the sky we can see very deep-hanging, dark clouds. It is humid. The body of P feels heavy, etc. If we assume that there is some form in which P has a sensory awareness of this, then a medial constellation could be a form in which P structures and identifies aspects of the experience of the situation by using relations between medial elements to articulate the attributes of the experience of the situation.43 In contrast to the case of a linguistic description of the experience, however, the generated constellation K would itself constitute something that P initially taps in a sensory reception. However, P is at the same time connected to K in another way, for as the producer of K, the medial structure of K is transparent to P. In this case, understanding would constitute a correlation between two experiences; here the medially generated experience would be connected with the capability to generate this by medial means.

Admittedly, in this analysis we have assumed that P already understands K, for only under this assumption can K become a means to understanding X for P. Of course, this assumption is met, in particular, if P is the producer of K. However, if this assumption is not met, then we must analyze the case (C2). For if K is not transparent to P from the perspective of the producer, then K must be understood medially, even by an interpreter. In other words, K must be interpreted with the help of K′.

2. So, if we must analyze what it means to understand a medial constellation (K) by the medial means (K′), then two levels are to be differentiated; for one level of understanding is related to the reconstruction of the identity of a medial constellation, and another is related to the content. Here the level of understanding, as a specific form of reference to the identity of a medial constellation, can be analyzed in the context of two different cases:

a. The simpler case is one in which the interpreter P perceives a medial constellation K, which consists of media elements that P is practiced in bringing about. This is the case, for example, if P hears a song that she herself could sing. An interpreter like this will structure the perception of the song such that she organizes the flow of acoustic perceptions, breaking them up into a sequence of activities that actualize those media elements whose constellation is specific to K. P does not simply perceive K ; rather, she comprehends the structure of K by an inhibited performance of K′. Insofar as P in this way manages a performative reconstruction of the perception of K, the perception does not simply befall P; rather, P refers to the compositional identity of K by performing something internally that has the same compositional identity as K. At this level, to interpret K simply means to identify K against the background of medial alternatives by producing K′, and K′ is compositionally identical to K.

b. A more complex case exists if an interpreter is confronted with a sensory-structured artifact K*, which has a structure that—in part or fully—evades a medially structured replication (Mitvollzug) by P. Let us assume that P knows or assumes that K* is a communicative artifact and that it was created with an expressive intention. Then how could the medial aspects of the understanding be explained with respect to the apprehension of the identity of K*? I would like to suggest that we analyze this more complex case as follows: P attempts to structure K* so that K* is apprehended like a constellation whose compositional identity exists in a hypothetical medium. P thus attempts to organize the sensory-perceptible attributes so that they can be described as medial attributes. Here, P, so to say, integrates K* into a network of distinctions that consists of systematic alternatives to the attributes that K* depicts. In this way, P places K* within the scope of possibilities of a hypothetical medium and determines the compositional identity of K* relative to this scope of possibilities. Here P at the same time drafts patterns for the comprehension (Nachvollzug) that can be described as sequences for realizing medial elements.

Of course, it seems questionable that an interpreter of K* could design a hypothetical medium without the help of language. However, while this possibility should not be precluded, the construction of the hypothetical medium can be described as the attempt to test forms of practical comprehension. This, however, raises the question of whether every practical comprehension ought to be considered a form for understanding the identity of K*. Obviously this cannot be the case; understanding is a concept that implies success, so it should be possible to provide criteria for successful medial understanding. While quantifying the normative benefits, I will attempt to show what such criteria might look like.

Having now gained an idea of understanding by employing concepts that refer to the compositional identity of a medial constellation, what it means to understand its content is still to be explained. As I have shown in chapter 3,44 the content of a medial constellation is initially dependent on the exemplary reactions of (parental) recipients that covary with their medial structure. This dependency of the content on compositional identity continues in developed medial practices. The content, however, becomes independent from the exemplary reactions of others to the degree that the initially external reception is internalized. However, then, by specifying (M8) and (M9)45 more precisely, it can be said:

(M12) The content of a medial constellation is the experience that accompanies the comprehension of its compositional identity.46

If the content for both producers and recipients is formed by reference to the compositional identity of a medial constellation, then one can say that, with the help of medial practices, processes of understanding can be organized. However, this means that the medial competencies that these processes are based on must be counted among the competencies of understanding; as such, they form the core of the rational competencies. Beyond that, through these reflections, the portrayal of the communicative function of medial expressions takes on clearer contours. For in the context of interpersonal communication, media obviously allow contents to be transmitted in more independent forms; it is through their dependency on transindividual individuating patterns that these are able to be communicated.

In the last section in which the explicative benefits of a media-theoretic amelioration of our understanding of rationality is quantified, I would like to clearly show that existing medial practices always at least partially realize the rationality potential of medial understanding.

4.1.3.3. The Inherent Rationality of Media Practices

In the preceding chapter I have defended the thesis that nonlinguistic media play a foundational role in the process of the development of higher intentional states. In doing so, in an explicative perspective I have clearly shown that media make a necessary contribution to the realization of those structures that exemplify a minimal rationality and, thus, are constitutive for our mind. In the following, I would like to clearly show that medial practices can be described as if they are oriented on certain principles. I will thus attempt to show that our dealing with media, particularly with artistic media, exhibits an inner systematization that can be described as resulting from the social implemented fulfillment of conditions for constituting media.

If we consider a community in which certain medial practices have been established, then we can say that, in this community, certain (demonstrative) rules hold that subject the behavioral scope of medial-interacting members to certain media-constitutive restrictions; we can at the same time say that it is precisely these restrictions that the actors, first of all, owe to the medially structured possible actions. The freedom to perform this or another medial constellation, to choose this or another action alternative, presupposes that the members of the community can move amidst a scope of alternatives whose existence is owed to the observance of restrictions. To that extent, we can say:

(M13) All successful medial practices realize the principle of productive constraint.

But (M13) is not sufficient precisely insofar as it sets out from an implicit counterfactual assumption, namely, the assumption that beings can subject their behavior to constraints before they have acquired medial competencies. If we consider the case of a novice who is socialized in a medial practice, it appears that the “negative freedom” that the novice would have to constrict can only be ascribed from the perspective of an observer that already moves within the scope of alternatives that is due precisely to the adherence to restrictions. For only from the perspective of beings with media competence can forms of behavior be interpreted as behavioral possibilities that, for beings that do not yet have these competencies, do not represent possibilities they could choose from. (M13) thus already presupposes that the participants in a successful medial practice move within a scope of alternatives that is dependent on media-constitutive restrictions. However, as long as a being only has A-intentional states at its disposal, it does not have the possibility to constrain itself, because it does not have possibilities that are possibilities for it. But this comes down to the following:

(M14) Media are fundamentally social mechanisms for constituting basal rationality, insofar as only those beings can be rational that have action alternatives that are alternatives for them.

Besides these elementary implications of medial practices that are constitutive for action and mind, quasi-economic principles play an interesting role. So, a medium in the sense of (M2)47 plays the role of mediating between a few medial attributes—often organized in simple grids—and a sphere of complexity that can be built on the basis of the medial elements. Insofar as the possible choices are organized in a grid with relatively few attributes, from the perspective of an actor, the possible choices are more clearly arranged, and they are easier to use in developing arbitrarily complex medial constellations. With a view to the communicative function of the media, this means that because one can learn to bring about simple medial elements, and any medial constellation can be analyzed in media elements, medial constellations can in principle be both reconstructed and reproduced. At the same time, any medial practice characterizes spheres of various degrees of plasticity by devoting stronger resistance to changes in the set of media elements than to changes at the level of the production of medial constellations.

Besides these limitations, which develop in complexity, it can be maintained that every factual implementation of a medial practice looks as if the participants in the practice orient themselves on the principles for securing communicative success. I have already mentioned one property that is relevant in this respect:48 medial elements are semantically intrinsically empty; consequently, the constitution of medial constellations is not bothered by primary forms of semanticity. Other properties of medial practice can be interpreted as culturally developed successors to that marking that we got to know as a property of parental response behavior in the phase of affectual communication.49 Examples of such properties are forms for contextually characterizing medial expressions, as, for example, they are realized with the help of frames and stages. Further quasi principles for securing communicative success aim at something that could be called the economy of attention; for, as a rule, medial constellations are placed in a context that ought to make it clear to the senses what is a part of the medial performance and what is not. In short, as a rule, medial performances take on their contours against a sensorily uninteresting (redundant) background.

These rather cursory remarks should merely show it is plausible that the relations of understanding that were reconstructed in the preceding section were not only carried over from a theoretical perspective to the medial constellations, but that structures can be found in the existing medial practices, whose meaning (Sinn) is tapped with a view to the individuation and transference of intelligible objects of experiences.

4.2. THE NORMATIVE RETURNS

By balancing the explicative returns (or benefits) for the view of rationality ameliorated by media-theoretic reflections, the plausibility of the following view should be shown. Nonlinguistic media allow correlative structures to be actualized with which it is possible to individuate contentful mental states, but with which experiences can also be structured and communicated and with which we can develop a self-relationship at a basic level. The reflections in connection with the determinations of the specifics of processes of medial understanding have led us to a point where we could see a particularity of medial processes of understanding to consist in the following: a medial interpreter taps an experience by attempting to reconstruct the compositional identity of the medial constellation by medially structuring the medial constellation that is to be understood, if necessary by means of a hypothetical medium. Admittedly, here it has remained an open question whether every comprehension of the structure of a medial expression can also be considered an apprehending comprehension. In other words, what the conditions for the success of adequate medial interpretations consist in has remained an open question. In a last step I would thus like to propose a principle that can serve as a standard for a good interpretation and that, in the best-case scenario, ought to result in a nonlinguistic analogue to Davidson’s principle of charity. Here, however, from the outset we must be aware of a difficulty: a principle like this will not be able to have recourse to the type of normative relations (e.g., inferential ones) that exist between ascribed propositional contents in linguistic interpretations.

4.2.1. A Principle of Medial Understanding

In order to construe a situation of radical medial interpretation, let us assume the case of an interpreter of a sensory-perceptible artifact K, who does not have the capability to routinely identify the medial structure of K. In this situation, should every correlation between K and the scope of possibilities of a hypothetical medium be considered a case of understanding? Or can criteria be provided with the help of which a boundary can be drawn between correlations based on understanding that exist between the sensory attributes of K and hypothetical scopes of possibilities and some other correlations?

If we assume that K had a set of perceptible attributes E1, E2, . . . , En, then we could judge the correlation that K has to a hypothetical medium according to how many of the Ei can be viewed as attributes of medial elements. Here it is, of course, not clear from the outset which of the Ei are relevant attributes of the object of experience. An interpreter can indeed prescind from any attributes that cannot be perceived in the context of the presentation of the object of experience—attributes like the weight of a drawing, the crystal structure of an applied pigment, etc.—but among the contextually accessible attributes, all perceptible attributes are potentially relevant. However, because it is not clear which of the perceptible attributes are in fact relevant, a medial interpreter could be prompted to consider as many sensory-perceptible attributes in the scope of possibilities of her hypothetical medium as possible. She thus must adopt the following principle:

(1) Maximize the number of those sensory-perceptible attributes that you apprehend as medial attributes in the medial structuring of an object that you want to understand as a communicative artifact.

This maximization principle, however, does not integrate the attributes that are taken into account. In the worst case, a hypothetical medium would be put together namely from a number of medial elements identical to the number of attributes taken into account. The interpreter could thus postulate a medial element for the implementation of each medial attribute, which would contradict the idea that the structure of a medial constellation must be able to be tapped on the basis of a manageable number of medial elements. A medial interpreter is thus prompted to replace the simple maximization principle with a more mature principle that combines the maximization demand with a minimization demand. This principle would have the following form:

(2) If you structure a sensory-perceptible phenomenon with the help of a hypothetical medium, then keep the number of medial elements whose combination implements the maximum perceptible attributes of the phenomenon as small as possible.

This principle prompts the interpreter of a nonlinguistic expression to apprehend the maximum amount of sensory-perceptible complexity of the expression as a product of a minimum number of different medial action types. I believe that this principle has a counterpart for the producer of medial constellations. If one considers that not a few artists in the process of creating a work (K) produce a number of preliminary studies and sketches, then one can ask why, in a series of sketches, S1, S2, . . . , K, one passes over from a sketch Sn to Sn+1, and in particular what, in the view of the artist, distinguishes K from all the sketches. Of course, one could reconstruct this process by assuming, within the framework of an instrumentalist understanding, that Sn+1 is simply preferred to Sn because Sn+1 better expresses what the artist wants to express. But this reconstruction assumes that the artist has some access to Si that is independent from what she wants to express.

In my view, one can understand the reflections that lead to K being granted a special status compared to the sketches as an answer to questions like the following: Does the medial constellation depict something that is identified as a unity in perception? Are all of the applied elements necessary in order to give a medial constellation the character of an integrated unity that can be experienced? Does the medial constellation obtain an “independence” such that the integration of the elements prevents their disintegration? If these questions actually play a role in the decision about which medial constellation is preferred in comparison to the sketches, then one can understand the question about the necessity of the applied elements to mean that an artist anticipates the minimization demand directed to the interpreter. In the same sense, the question of the unity in the perception and the question of the “independence” can be understood as anticipating the pressure, which arises from the minimization demand, to reconstruct the work from the smallest possible number of densely integrated elements. Because the maximization demand prompts the recipients to pay attention to any potentially relevant attribute of the work, artists must expect that the recipients will include all sensory-perceptible attributes of a work in the medial structuring.

Simply following the mentioned principle cannot secure a good interpretation. For numerous interpretations are always possible, and the suggested principle is only selective if these interpretations significantly differ in their breadth and their level of integration. In this respect, it is no different than the principle of charity, for Davidson also expects indeterminate interpretations. Admittedly, with the help of this principle, correlations of understanding can be distinguished insofar as correlations are subject to restrictions that prescribe that interpretive correlations orient themselves on the conditions for individuating medial constellations insofar as differentiating an object that can be experienced is linked with perceiving it as a unity. In other words, the restrictions articulate conditions for successfully conveying experiences with the help of medially structured objects of experiences, and, with them, the conditions for the transindividual individuation of B-intentional states so that the discussion of medial processes of understanding in any case appears to be justified. Under the assumption that the core rational competencies can be explained in reference to concepts related to competencies of understanding, this in turn means that medial processes of understanding can be drawn on to determine rational competencies. But then it must also be possible, aligned with the model of that catalogue,50 which explains rational competencies as prerequisites for the ability to understand and for the de facto understanding at different levels—to characterize basal, habitual, and optional rationality in the context of medial competencies:

1. In the context of medial action, basal rationality legitimizes a manner of speaking about beings whose activities can in part be described as if they use sensory-perceptible states of affairs in the world or processes for communicative purposes.

This implies especially that basally rational beings

a. have a typified repertoire of semantically intrinsically empty behavioral alternatives that they use to produce medial constellations;

b. teach the production of medial elements;

c. frequently react to the expression of medial constellations by expressing medial constellations; and

d. frequently use types of medial constellations so that these covary with the circumstances of the expression.

If these criteria are not fulfilled, then the behavior of these beings loses the status of action. However, this is not because we cannot ascribe reasons for the expression, but because we cannot ascribe a medium that allows the individuation of higher intentional states and legitimizes the assumption that these beings choose their activities against the background of alternatives that are alternatives for them.

2. Habitual rationality can only be assigned to those beings that fulfill the criteria of basal rationality and participate in a social practice of interpretation. In communicative practices, such beings frequently act in compliance with

a. principles of the economy of attention; and

b. principles of the economy of individuation.

3. At the level of medial practices, optional rationality is also a second-order rationality. It presumes a medium, language, with which beings can draw on their medial practices. Here, forms for realizing medial practices or medial strategies themselves become the content of intentional states, and indeed in the perspective of the systematic formulation and opening up of forms of articulation. With respect to the ability for medial articulation (on the part of the producers) and the ability for medial interpretation (on the part of the recipients), the maximization of optional rationality underlies education processes and the development of an artistic or aesthetic attitude.

The criteria of optional rationality are normative insofar as the realization of optional medial rationality prescribes the further development of articulatory possibilities and interpretive capabilities. Norms of medial rationality demand that their addressees

a. maximize the articulation possibilities by

α. making full systematic use of the scope of possibilities of the media; or

β. developing new medial strategies; or

γ. developing new media; and

b. maximize the interpretive capabilities by

α. attempting to maximize the originality of the medial constellation in their interpretations;

β. developing new strategies of medial comprehension;

γ. tapping new forms of the medial organization of possibilities of experience,

in order to systematically expand the means that we have at our disposal for the processes of (self-)understanding.

4.2.2. Media of the Enlightenment

In the definition of (A1), I described the Enlightenment as a process in which we develop and learn to understand our abilities to understand.51 In the preceding reflections I have attempted to plausibly show that an adequate and comprehensive understanding of our capabilities to understand must also include forms in which nonlinguistic media play a key role. For nonlinguistic media are both means for articulating and means for developing articulations; they are means for developing a self-relationship as well as means with which we can individuate contentful intentional states. In short, they are means for nonlinguistic thinking.

In many places, but especially in chapter 3, the development of my reflections is influenced by the impulse to understand the possibilities that nonlinguistic means offer us as sui generis possibilities. That means that these determinations often occur independently of language or after setting aside language. However, it is important to see that this impulse is indebted to the theoretical interest in, if possible, bringing everything into view that is possible in media without language, and it does not attempt to play down the significance of language. I hope, to the contrary, that my reflections have contributed to better understanding the prominent role of language. For if language can be understood as a medium that shares part of the phylogenetic and ontogenetic history of origins with nonlinguistic media, then, in a theoretically homogeneous vocabulary, it can be clearly shown which normative relations a nonlinguistic medium must be embedded in so that it can take on the role of a language. If the justification of medial expressions is tied to normative criteria, which, for their part, are connected to intersubjectively accessible truth-conditions, then a nonlinguistic medium is transformed into a language.52 And because linguistic utterances can themselves be counted among the truth conditions for linguistic utterances, a metalinguistic reference is possible in language for which there are only limited functional equivalents in nonlinguistic media.

However, none of this changes anything about the finding that nonlinguistic media have a specific capacity available especially to make experience intersubjectively communicable.53 In nonlinguistic medial expressions, the possibility of arousing our sensibility is connected with the possibility of structurally tapping this sensory dimension of the reception of medial articulations and, through this structuring, of making it intersubjectively accessible. Because medial articulations have a compositional structure, in the course of a comprehension they convey their sensory attributes, but also indications regarding their development (Erschließung); such a reconstruction can orient itself on the compositional identity of a medial expression against the background of a regularly structured scope of possibilities. This specific combination of an intelligible structure and sensorily “dense” perceptible attributes forms the basis on which medial interaction processes can be viewed as processes of understanding, and it forms the basis for the fact that a principle for judging interpretations can be characterized, but without the following of this principle ensuring interpretive success. Here, however, interpreting medial expressions is related to interpreting linguistic utterances even if interpreters of linguistic utterances, because of the normative relations between sentences, are subject to considerably more restrictions than the interpreters of nonlinguistic medial expressions.

If the discussion of medial interpretations can be shown to be plausible in this way, then we can understand the development of medial articulation competencies and interpretation competencies to be genuine components of the Enlightenment process. We can, in a basic manner, claim that this process is more complex than its critics would have us believe. For as paradigmatic cases of developed aesthetic communication, in this view, works of art play neither the role of a compensatory or correcting mechanism that relieves us of conceptual work nor the role of enigmatic expressions whose rationality must be unlocked with aesthetic commentary or critique.54 The proposed concept should rather reveal that one can take seriously the thought that aesthetic communication is a full-fledged dimension of the development of rationality. To the degree that the development of medial and linguistic competencies in the context of educational processes is the development of competencies of understanding, and the articulated self-understanding is a necessary condition for knowing who we are and what we want, to this degree, the development of these competencies is a value that does not stand in need of an external justification. Because justifications of what is required, for their part, presuppose a self-understanding, those practices in which we transmit and develop the preconditions for developing self-understanding constitute unconditional values; for we are what we are for us only within the framework of linguistic and medial self-interpretations.

Media, understood in the manner developed here, are media of reason in a twofold sense: they are media that lead us to reason, and they are something in which our reason is realized.