A Systems Theory of Religion

The Contingency Formula “God”

If everything determined in the medium of meaning is accompanied by endless referential surpluses, how is communication even possible? How can it have any hope of a successful outcome, whether in the form of consensus or of dissent? Negotiating meaning is thus the initial problem we face with respect to the function of religion. Since no other medium is available, however, we need meaning both to exclude and to connect to meaning. This might be said to ruin the concept of “meaning,” but it nonetheless very precisely indicates the problem, which calls for contingency formulas.

The paradox involved—that connection can only be implemented by exclusion—can be transmitted in multiple forms. We had also asked how it is possible to transfer indeterminability into determinability, to transfer infinite burdens of information into finite ones. This (already functional) problem can be rendered more precisely by substituting the “how” for a “who or what.” I shall call this specification, which sees a narrower version of the problem as its solution, a contingency formula.

From a functional perspective and described externally, one can re-resolve/re-dismantle the unity of such a formula. The paradoxical formulation that the problem appears as its solution, that the different is the same, has already been adopted for this. The distinction defining the concept of function thus disappears. In the internal domain of the system that is shifted by coding to contingency and reflexivity, the contingency formula vreplaces the reference to function. That is where the system’s self-descriptions find their grounding. Although the system is a difference and, seen as an operation, is the reproduction of this difference, it can be made accessible to itself as a unity. It can produce a relationship to its own hypostatization, yet only such that the paradoxes of reentry and of coding are replaced by an identity that is held onto afterwards. An external observer can see how that works—and that it cannot work differently. Because the observer has no responsibility for how the system operates, he may still ask how such contingency formulas function. With the greater freedom of a second-order observer, he can tap into a comparative perspective not applicable in the compared systems. Yet it is only appropriate for him to account for the difference in perspectives, bearing in mind that the system-internal observations are at the mercy of the system-internal contingency formula, like a blind spot without which observations and self-observations in the (observed) system would not be possible. If the second-order observer imagines a correctible error here, a defect in consciousness, or an ideology, then he would be making an error himself. He would himself be using a distinction (such as truth/ideology) that impedes commensurate access to the object.

Contingency formulas are found in all function systems that are opened by their codes to contingency, reflection, and to paradoxical/tautological possibilities of self-observation. They are distinguished depending on which of their operational types needs to have a capacity for connection. Thus, in the economic system, the contingency formula shortage is used for making sure that the economy can count on an operationally independent constancy of sums (although that is not accurate, either for goods or work or money).¹ The artificiality of the principle is apparent in the difficulty of implementing effective monitoring of the money supply. Its indispensability is revealed in the rule that no one can spend his money more than once. A corresponding principle of “common good” is to be found in politics,² or at least that was true in the medieval and early modern traditions, which assumed that people could still distinguish between public and private interests (according to their nature). Since this principle is less and less evident, “legitimacy” is the only thing that can still be specified as a political contingency formula, in that it refers to continuously acceptable values. These nonetheless leave open the question of how value conflicts will be decided.³ It therefore becomes a matter of legitimized opportunism (though saying that would expose the meaning formula legitimacy as a paradox). In the educational system, the contingency formula has to provide learning objectives, whether in the form of an educational canon tied to content, or in the form of learning to learn.⁴ In the process, it must be ruled out or go unsaid that other things are also learned at school, such as getting used to dullness or how in life it is more important to be deceitful than diligent. Operations of science depend on the truth code functioning restrictively, meaning that refuting an assumption of knowledge consolidates truth’s chances. Yet the contrary experience that an increase in knowledge is tantamount to an (overproportional) increase in new uncertainties [Nichtwissen] is also abundantly available.⁵ In the legal system, something analogous is achieved with the contingency formula justice, which prescribes the consistency of decisions with the double directive to treat like cases alike and different ones differently. These provide the observational formula of like/unlike that makes it meaningful to look for comparative standpoints in the legal system, while leaving open how these can be found.⁶

These examples make it clear (and clearer than a terminological definition could) what a contingency formula does and how it works. Its aim is to suppress other possibilities also given. The other side of this form may be reproduced as accompanying knowledge, but it can no longer claim that it records the system’s meaning. Official communication is oriented along the guidelines of contingency formulas, thereby operating on the secure side of intended acceptance. Since the nineteenth century, the concept of value has been used to signal the consensual capacity of preferences. But we also know that professional knowledge cannot make do with that alone. Rather, completion occasionally imposes itself by transgressing boundaries, by glancing at the other side. In a normal situation, there may be little promise of success, since we can only operate in systems within the framework of contingency formulas. Communication itself favors assumptions in which one has to be involved if one wants to get around to anything. But none of these answers makes the slightest claim to satisfying the questions. With concepts of value, only what is wished for is beyond question—fewer shortages, more justice, and so on. But such formulations already open up horizons that these concepts themselves can no longer integrate.

The major religions have, it might be pointed out, experimented with contingency formulas for religion, without achieving a unanimous result. The most successful attempts, those of Buddhism and the monotheistic religions, appear to be founded on a common element: a redemptive outlook. They hold out the prospect of access to transcendence as a corrective for suffering from distinctions. They also hold out the prospect of sublating every distinction into something beyond all distinctions. That is the form in which the difference of immanence and transcendence is presented. The programs that become necessary appear as conditions of redemption. That in turn requires that one builds a temporal perspective into religion. The temporalization of the problem is associated with a devaluation of the world as it is found here and now, a devaluation of the hicmundus [this world] and thus the exploitability of chances for plausibility that are already given in the social-structural (situative, role-based, institutional) differentiation of religiously defined communication.

This much can be assumed as a common starting point, and that is a good deal. It suffices for providing an image of how the contingency formula “God” works. In the following, I shall limit my analysis to this contingency formula, even though (and because) it is probably easier to find access to Buddhism from constructivist starting positions.⁷ At this point, we shall also not be dealing with polytheistic religions, in the case of which it is too tempting to define gods by their difference from other gods and to rule out questions going beyond that. In addition, their specifications were based too much on what they provided for certain formations of society (providing clues for differentiation, whether for genealogies of nobility, or for roles and competencies not yet differentiated into particular function systems).

A characteristic of religions formulated with reference to God is that they are also familiar with other holy figures that populate the religious cosmos and release the high god himself from unreasonable demands of definition: lesser gods, angels, spirits, saints, ancestors who live on spiritually, and so forth. At the same time, they serve as mediators and intervening instances when one does not dare (or see a way) to contact God directly. (Obviously, this reflects the social structures of an aristocratic society under the rule of a king.) The special position of the high god is thus distinguished by reason of his alone realizing the complete characteristics of transcendence, in particular, the characteristic of limitlessness, of ubiquity, and of universal presence even in the domain of immanence; he alone realizes the unity of the difference of transcendence and immanence. Only the high god can serve as the concluding formula of the religious cosmos. Yet that is precisely why it is difficult to know how to behave toward him. The problem of the contingency formula is therefore conveyed as a need for supplements, auxiliary institutions, and finally professional assistance. Supplements, though, are always (as is known from Derrida’s analyses) moments of essence (that need supplementing). They are indications of a concealed paradox.

Contingency can best be grasped in personal behavior, because, in referring to a person, one assumes that the person might also behave differently. That, however, leads to the difficulty of thinking the highest transcendence as a person who would have other possible ways to behave—and who excludes them. More than all other religions, then, strong monotheistic religions are striking in the boldness with which they specify a contingency formula. Hence, they are also striking in the demands they make on individual religious faith. And they are just as striking in the burden of their dogmatic apparatuses (weighed down by the need to secure plausibility), in the strict way they distinguish between the faithful and nonbelievers, and (lastly) in their dependence on organization. All this is anchored in (and presupposes) the transferring of all transcendence (“there is no transcendence beyond transcendence”) into an existential proposition. Transcendence exists as a person, it is the one God. And whoever does not believe in him is excluded. The “other side” of this semantic boldness is its correlative in the social structure: the possibility of being “excommunicated” and the (more or less) grave social consequences that ensue.

As long as gods were honored as house gods in ancestor veneration, it was not difficult to understand them as (invisible) persons and hence as observers of the living. If, however, the reference to those formerly alive breaks down, special reasons are needed to continue the idea of being observed by (invisible) persons and assigning it ultimately to a universal God. Personalizing ideas about God would have been a difficult, virtually counterintuitive process, especially if the idea of a transcendent power was supposed to be maintained and extended simultaneously.⁸ The main risk of this semantics, the way it endangers itself, probably consist in compelling us to think of the good and bad effects—the attractive and repulsive (essential) traits of the holy—as the intention of a person. The Greeks had helped themselves out with terms such as “immortal” and “ageless.” In this framework, they assigned their gods freedoms to decide, preferences, and dispositions to conflict. In the process, personality remained bound to a majority of gods, each of whom in their sphere of action personalized themselves (as it were). But there was too much contingency there for making greater demands of religion. The case that a single, transcendent, all-responsible God could ultimately be thought of as a person (albeit one without a name) must have corresponded to a need that was not apparent at first glance. It is presupposed that there was such a need in a social-structural development at the same time that activated social differentiations and individualizations, and that was only able to comprehend unity through the idea of an observer. For personality is nothing but a cipher for observing and being observed.

Yet why does God have to be given the quality of an observer? Why can he not simply exist? The best answer to this question comes from reflecting that a listing of all things that exist (P1, P2, . . . Pn) would never lead to a notion of God. Rather, it would always result in assuming that there are additional things. And this would also apply if one assumes (along with Deleuze) that each listing produces two series, one that moves forward and another backward—and that their unity can only be understood as a paradox.⁹ God must then exist on another level, in another qualifying situation, and the notion of the observer provides at least one possible interpretation of this position. For it does not exclude—in fact, it virtually demands—that predicates like existence and thing (in the sense of res) also apply for an observer.

If the observer is understood not only as a form of indication that distinguishes (and that, in the case of God, distinguishes everyone and everything in its individuality) but also as a person, then it can also be made plausible for the observer to set conditions of liking/dislike. These then lead in the context of society to a difference of inclusion and exclusion that spreads across all other distinctions. Norms are always in part conditioned by what they exclude. And to the degree that they are formulated more generally, they can ultimately only be recognized by their exclusionary effects. None of that can be changed by “justification.” By the form that god-based religions choose for their differentiation, they must (from a sociological standpoint) be dependent upon socio-structural conditions. They give weight to exclusion and, in the Middle Ages (for instance), they might even have forced an emperor to his knees. Conditions for inclusion/exclusion, however, are subject to a deep social transition,¹⁰ and they can thus endanger the embedding (of the “outbedding”) of such religions. They do so without the possibility of being intercepted on the semantic level by exchanging divine attributes (for instance) or giving up “father” symbolism (which nowadays seems somewhat anachronistic). The question thus becomes whether god-based religions today have to rely on the coincidence of individual belief decisions or on the persistence of an irrefutable error, and whether their semantics (themselves) have deeper motivations and social affinities, moving in the same direction as the function of religion. If one pursues this question from a sociological (i.e., religion-external) perspective of observation, one can examine more precisely how the contingency formula “God” actually functions (and it “functions” not in the sense of a trivial machine, certainly, but in solving the referential problem of religion: the transformation of the indeterminable into the determinable).

III

I shall not concern myself here with “proofs of God’s existence.” By means of such proofs, people try to deny God the contingency more customary of worldly things. This is done precisely because the world, seen from God’s position, is contingent. Yet nowadays, hardly anyone still disputes that these proofs have a circular structure that must be interrupted (or made asymmetrical) one way or another, perhaps most clearly by the dogma of God’s self-revelation.¹¹ Concluding from the beauty and wellorderedness of the world that it has an intelligent cause means offsetting any awkwardness as an error. We could also assume the contingency of all meaningful world descriptions, thus turning the argument around. In that case, God’s existence is not proven by the contingency of the world; rather, the contingency of the world is proven by God’s existence. In each case, the circularity is resolved into an asymmetrical relation of God and the world, which cannot be treated as reversible. And precisely that is the structure enabling us to acquire more information and that we designate as a contingency formula.

We can thus assume that an asymmetry is involved, and that the process must be hidden. The really important question is how to determine the specifics of the form making this operation possible. The notion of an asymmetry between God and world is already helpful because it prevents the development of monotheism from stagnating into a sterile idea of perfection. An absolutely perfect God is a God who cannot add anything to himself. But can such a God make a difference?¹² This problem can be metaphysically resolved, following Plato, by a theology of emanation.¹³ “The concept of Self-Sufficing Perfection, by a bold logical inversion, was—without losing any of its original implications—converted into the concept of Self-Transcending Fecundity.”¹⁴ This alternative can be elaborated with a semantics of abundance, outflow, and giving. Yet it does not make it either necessary or comprehensible to think of God as a person.

The Christian tradition combined the assumption of an observer God with an ontological metaphysics, with a description of the world assuming the distinction of being/non-being—and ordering all other distinctions according to that one. If this world is understood as divinely created, it contains a normative moment. It is understood as a requirement of order. What is should not be what it is not. Hence, a human being is not an animal, a man is not a woman, and a Christian is not a pagan. Whenever there are breakthroughs within these ontological exclusions, they are understood as “miracles” and are then fed back into religion. The world of being is divinely created by God and what contradicts its organization also confirms something: that it is created by an all-powerful God whose power is not lost into the world but is maintained as a possibility of also deciding differently. In this way, metaphysics and religion corroborate each other, doing so in the form of a circle that closes the thought system.

The starting situations that have led to this special development need not be addressed here. A polytheistic cosmos was able to apply the person-form as a distinguishing principle. For the Hebraic tradition, the ability to make covenants had been important and also presupposed a divine personality. There is a rich, controversial literature on the relationship of tribal societies to notions of a high god and on the analogies and formal parallels to political rule. For the problem here of the contingency formula “God,” what matters is not the semantic genealogy of this evolutionary form—“pre-adaptive advances” naturally have to be assumed—but rather the function of constructing transcendence as a person. What might be meant by “person” cannot be determined by an analogy between God and humanity—neither in one direction as anthropomorphism nor in the other in which God creates humanity “in His image.” These kinds of interpretations are available, based on analogies of content that do not appear (at least nowadays) to have any apparent plausible basis. This schema of analogies, which is (as always) charged with differences, is replaced with the hypothesis referred to above: God is defined as a person because that establishes him as an observer. The form of asymmetry one is asking about is thus the form of the operation of observing.

What we have been asking about is contained in the idea of observing. The operation of observing is an asymmetrical (i.e., irreversible) operation that takes place exclusively in the observer but implies for him a distinction between observer and observed. It consists of a distinguishing indication, in which the applied distinction is not identical with the implication that the observer distinguishes himself from the observed. In other words, he has to distinguish himself in order to be able to make distinctions. This can be seen, in the case of God, as a reason why he creates the world.

Additionally, the idea of observing is so general that when applied to humanity it includes experience and action, cognitive and willed activity. The (internally experienced) distinction between experience and action depends, as Gotthard Günther has shown, on the distinction of other- and self-reference and thereby (indirectly) on the distinction of environment and system.¹⁵ In the case of experience, the system counters the experience of external determination with internal binarizations—such as the schema of true/untrue or the schema of liking/disliking. By contrast, in the case of action, the system produces a difference that is found in the environment—such as the difference of the achieved purpose in contrast to what otherwise would obtain. Yet since one cannot assume that God distinguishes between other- and self-reference, the distinction as derived will not be appropriate for him. It does not rule out understanding God as an observer. In reference to God, however, the idea of observation does not need the distinction of intellect/will. The related controversies of the Middle Ages are now only of historical interest. The unity of observation as intellect and will tells us primarily that divine will is not bound to insights previously gained through knowledge. We do not need to argue, as in those medieval controversies, that omniscience prevails over omnipotence of the will (or vice versa). They are identical.¹⁶

In particular, there is a special status of this observer God that is correlated with the transcendence value of religion’s code. God does not need a “blind spot.” He can realize every distinguishing schema simultaneously as a difference and a unity of the distinguished. That includes the distinction between being distinguished and not being distinguished. And since this applies to each (and all) of his observations, his manner of observation—if one wished to try and observe it—can only be regarded as paradoxical. We shall return to this issue later on. For now, let us merely note that observing the transcendental observer proceeds precisely by contrast to all world-immanent observation, thus supplementing it by what it is missing. The God/human difference lies in the question of whether the unity of the operation of observing can observe itself in the operation (transcendence) or not (immanence). It could also be said that the transcendent person is self-transparent, and the immanent person is not. To that extent, the contingency formula “God,” understood as an observer, is aligned with the necessity of coding religion and with religion’s special way of reflecting contingency.

If God is seen as distinct from the “world” (which is unavoidable if he is said to observe it), the world can be differentiated from God, and this distinction determines them. Yet at the same time, if one assumes that this observer God has complete knowledge of the world, he cannot disturb its order, since possible disturbances (miracles, for instance) are always foreseen in God’s mode of observing things. God and the world are in a relationship of harmony—a particularly important idea in the natural philosophy of the Renaissance. On the other hand, the distinction between God and the world is still meaningful. It enables (as is often pointed out) a desacralization of nature. As nature, the world is assigned for human use.¹⁷ The God/world distinction further makes it possible to devalue the world in the sense of hic mundus, thereby societally differentiating religious concerns from subjective preoccupations with money, power, status, or one’s sex life. This then leads relatively easily to calls for an anti-worldly religious (monk-like) lifestyle. And, finally, if it suffices to define the world by its difference from God, one can leave all other characteristics wide open, accepting uncertainty, hypotheticals, and so forth.¹⁸ Even if modern, polycontextural society should require the world to be conveyed as an unmarked state, and thus as unobservable, religion might still maintain that it could itself distinguish the world, specifically from God.

All this is formulated so abstractly that neither a cosmology nor a morality is required. For human beings, it seems obvious to refer God’s observation to humans (or primarily to them) and to be pleased with the illusion that nature has been established for humanity’s sake. And (some say) toilets would not be able to flush without divine assistance. But the dogma of omniscience and omnipotence goes far beyond that, becoming explicitly visible for example in the doctrine of the creatio continua: divine observation is a condition for preserving the world from moment to moment in all its details.

Even as logos, as word or as text, the observing God is beyond all distinctions, especially all temporal distinctions. That is what guarantees the presence of being, independently of the merely momentary concurrence of all topical [i.e., individual] human experience, for ontological metaphysics. It also underwrites a life after death, that most far-reaching of theological generalizations. It is the basis for all moral conditionings of human lifestyle, condensing them into a unity (without having to rely on principles, moral laws, or the like). Time is thereby understood as a duration, and an immortality of the soul is deduced from the unthinkability of a beginning (without a “before”) and an end (without an “after”).¹⁹ Derrida’s critique of the metaphysics of presence and its logocentrism is still indebted to this idea; in fact, it cannot be formulated on its own but only as a rejection of the idea.²⁰

Despite existing independently of distinction, God must therefore be transferred to the good, beautiful, and existentially stronger side whenever humans make distinctions. Obviously, this is not only true of such distinctions as being/nonbeing, existence/appearance, good/bad but also of more remote considerations concerning God’s image in art.²¹ If one wished to proceed in the other direction, using the negative side of the distinction as a symbol for the distinction, one would come close to a symbolism of death, negating all distinctions. This reflection shows that the (interest-laden) figure of life after death integrates two logical impossibilities in a single paradox, indicating the unity of every distinction either from its negative or positive side—that is, either as death or God.

One does not have to rely on employing an ontological mode of observation, observation with the help of distinguishing being from nonbeing. For doing that would only lead back to the question: who is the observer observing with this very schema of distinction, dividing up the world in precisely this way and not another? And what is his blind spot? Perhaps the impossibility of comprehending time? In any case, God observes without being dependent on the distinction of being and nonbeing, and as a result, there is no “excluded third possibility” for him—and thus no logic, either. Only thus can God be thought of as the creator—further abstracted here to “observer”—of being out of nonbeing, as Nicholas of Cusa holds.²² To theologians who still think ontologically, the asymmetry of God and world gives way to the possibility of thinking of God as total meaning, as perfection of being, as ens universalissimum. This asymmetry has to be replaced by (an utterly faded, formal) notion of the unity of difference, for example, one based on formulations for difference such as ens finitum/infinitum or ens creatum/increatum.²³ The theory of (operative) observation is substituted here for this ontological weak point, which can only be filled by an empty formula.

In place of determining the minimal components of distinctions and how they predetermine interpretations of the world, a more exciting question emerges: How can the observer God be observed? Whoever can observe God as an observer (and not simply as a holy, untouchable object) obtains a final guarantee of meaning with regard to an infallible constructor. This constructor created the world, sees what happens, and will not change his construction—even though he could. As is reported, he found “it was good.” And that is underscored by many dogmas. He bound himself to a covenant. He sees humanity as the crown of creation. And he loves us.

Yet as soon as one accepts that God observes everything (“nothing escapes him”) and that he therefore has to distinguish himself from everything, he cannot be observed in (or with) the world. One cannot determine by worldly means whether God exists or not. Proofs of God’s existence come to contradict what they set out to demonstrate. One must therefore repeat the question: how can a human being observe the observer God? Or, put more sociologically, how does one deal with differences of opinion that can be anticipated when observations of the unobservable are made? With this question about the possibilities of second-order observation, one comes across solutions that, on the one hand, make use of privileged positions and, on the other, reflect their own imperfection—setting up a rich semantic field for theological discussion.

That man is ignorant of some things—an untaught unknowing, as it were—even religions that are not self-reflective concede.²⁴ Such a situation is then resolved through an ambivalence of attraction and repulsion, seeking and avoiding, and asking and fearing. This state of affairs can be differentiated according to opportunities or situations, or it can be dealt with by having God transcendentally loved and immanently feared. Furthermore, there is that solution enshrined in myth where persons are punished for their arrogance and for claiming they know better than others what God actually wants. Such is the fate of the angel who is punished by falling (from heaven) and who cannot regret having fallen, since his ultimate goal was to observe God. And there is the solution of paradox, confused speech, silence, of “I know it because I don’t know it,” of speculative fusion (“God’s eye is my eye”)—the mystical solution of Nicolas of Cusa.

Theologians are only rarely insightful enough to interpret the fate of the fallen angel as a paradox of love rather than a divinely conceded freedom to do evil (an interpretation that should require explanation).²⁵ For that angel who loves God the most cannot be satisfied with limiting himself to discovering this paradox. Nor can he shrug his shoulders and declare: “Ultimately, He must know; it is not my affair.”²⁶ Satan’s love becomes an existential paradox, an expulsion due to love. The issue is not freedom for the devil; rather, his activism of seduction is something theologians invented to build their moral code into creation (at the expense of their theology). But such an invention is not really so inspired.²⁷ If it were to address the cognitive problem directly, a theology that reflected on the observation of observing could more readily make the transition from the impossibility of consciousness to a consciousness of impossibility.²⁸ Yet who is going to see the advantage in that?

Regardless of how deeply they might reflect, theologians find themselves, as professional observers of God and/or interpreters of his texts, subjected to questions and expected to provide answers. They have to communicate within the religion system, imparting what they know, which more or less compels them to make a fatal reinterpretation of the contingency formula as a selection criterion: “Yes to this, no to that—that is how God wants it.” There is no third way [tertium non datur]. But in contrast to the devil—who does not himself partake of the fruit forbidden in the Bible (!)—how can one avoid the arrogance of knowing better?

One of the possibilities [two more follow below] is to appeal to divine self-revelation. Because it is so central, revelation is one of the most difficult concepts in religious studies. It has also been theologically controversial up to the present day. While we have no desire to intervene in the internal theological discussions, a distinction should be made here: revelation is not only about mantics, divination, interpretation of signs, or formal analogies between the visible and the invisible. It is therefore not about getting an answer (no matter how that answer is encoded) to questions posed to God based on some important situation or another (such as “Should I marry this person?” or “Should we go to war?”). Divination is and always has been observing the lines of a surface in the hope of being able to discover something deep and concealed.²⁹ Even if it is accepted that the gods make decisions each day about human destinies, the problem remains of how (and by means of what omens) we can figure out those decisions. The thought of God’s self-revelation breaks off there—which is the source of its evolutionary achievement and historical volatility. A special idea of revelation is therefore only meaningful if it includes a divine initiative and an autological moment. That rules out posing the question of truth (in a scientific sense).³⁰ In terms of form, the revelation reveals itself, meaning that there is no other cognitive access to it besides assuming that the revelation is a revelation. In terms of content, the issue is the relationship of observing (noted above), of establishing a possibility of observing the observer God. And in terms of history, the revelation changes history so that it no longer suffices to understand religion as memory. Something has commenced that is new in history.³¹

The troubles start specifically whenever one thinks about divine self-revelation, with its well-measured instructions on observation, as communication. In particular, the trouble is caused by the distinction between information and utterance [Mitteilung]. It does not make much sense to ask why revelation, as information, is being communicated. If the issue is religious truth, there are no additional reasons or motives needed for it to be communicated. Moreover, if revelation is communication, the continuous relationship of observing also has to be understood as an event, a transitory happening that cannot be replicated. That leads then to the paradox of substantiating presence by means of absence, for which the form of text presents itself. That in turn assumes the revelation has been completed textually and historically. It would be difficult to cope with if God (with all his possibilities) were always announcing new things, if he were commenting on the ecological state of society, or if he suddenly judged sexual behavior differently than one thought. (The visionary phenomena of the Late Middle Ages were in any case not treated as instruments for proclaiming dogmas, whatever might be seen as their cause and distinguishing effect.) The dogmatics of revelation, as in the doctrine of the Sinaitic revelation, are always legends developed after the fact. They respond to a historically given need for legitimation in already established religions.³² In the Jewish Talmud, a way out of this dilemma was developed using the written/oral distinction to guide revelation (according to God’s will) in both ways: the text is given in writing for oral (and thus revisable and open) interpretations. As such, it is a particularly impressive form of a resolved paradox.³³

A second possibility is to include the first observer of God’s observing, the devil, and then to distinguish oneself from him either by demonization and/or defamation, or by reconstructing the relationship of God and devil with the moral coding of good/evil. Regardless of what one doesn’t know to say about God, one can issue warnings about the devil’s seductions and denounce all those vices that would lead sinners into his realm. In that case, one is using negative communication, which is easier from the standpoint of communicative technique or rhetoric. One likewise has fewer problems identifying bad taste—even when what is good taste cannot be defined.

The third (and probably most demanding) possibility is a mystical solution: the gazes coincide and are collapsed. No criteria are needed here. Being seen can be seen directly. All distinctions in existence are sublated, even if only for the moment. Such certainty cannot be distinguished and therefore cannot be surpassed. But it cannot therefore be communicated. Getting into communicatio with God (and here the old idea of producing commonality is appropriate) impacts communication between human beings. One can discuss this as one does other matters. Yet this experience cannot be conveyed to anyone who has not had this experience (and does anyone really know if he has?). One might go a step further and claim there can only be mysticism if religious experience is transferred to the level of second-order observation, and if the issue of distinguishing criteria is problematized, and if ways are sought to avoid the issue as something unsuitable. When embedded in culture like this, mysticism derives its powers of persuasion from the temporary character of all other answers, both dogmatic and professional—all of which are clearly inadequate. But that does not mean that mysticism would be able to replace dogmatics in communication about religion.

In short, the complicated structure of second-order observation is important for elaborating the contingency formula “God.” On the one hand, that formula unifies the code of immanence/transcendence, absorbing contingency in this capacity. On the other hand, it is a selection criterion, an almost complete formula for religious programmatics and indicating what is right and wrong in the relationship of immanence to transcendence. None of that could be achieved by object propositions, by indicating attributes of the thing called “God.” The doctrine of attributes could only specify things that are unobservable. In addition, it isn’t necessary to construe God himself as an observer until there is actually a context of second-order observation. One doesn’t have to know what he is. Rather, one has to know how he renders judgment—in order to adapt one’s life to his love.

If we now return to the operation of communication that reproduces society and religion within society, then the contingency formula “God” is revealed as something that closes itself off from communication. It dismisses any attempt to be analyzed. Its “holiness” consists in refusing all access to its mystery, in punishing every scrutinizing or differentiating communication by recognizing that what one sought was not what one found. Yet communication is the only form in which society is capable of realizing itself. There is no other possibility for realizing meaning in society. When it is necessary, then, one has to communicate incommunicability.

To explain this state of affairs, there are a variety of major ideas that depend for their plausibility on the framing conditions of society. We can distinguish three forms (that overlap a great deal), correlating roughly to the transition from tribal (or segmentary) to stratified societies and from them to a functionally differentiated one. Three models—“mystery,” “paradox,” and “external (functional) analysis”—present themselves.

In early notions about sacred objects, techniques, and relationships, the communication problem was solved by the figure of a mystery equipped with sanctions. It’s an idea that can apply to sacred matters of all kinds. These matters are treated as external givens and their limited accessibility is part of their essence. If one foolishly tries to penetrate mysteries, break taboos, or say too much, one has to count on there being supernatural sanctions. One had to accept that numinous powers were interested in their own sacrosanctity. The problem was thus externalized. In communication, one can only request that such conditions be adhered to, and this is underwritten by a type of collective liability. The punishment not only applied to the perpetrator but also to his kin. The sacred powers were not really geared toward individual guilt. If they operated according to moral criteria at all, they did so rather imprecisely. Society was left to take precautions so that nothing will happen, and to discipline or expel its members appropriately.

Such a notion clearly becomes problematic when monotheistic religions emerge that assume a high god or even a single god whose expectations are oriented toward morality. We have already discussed the problems with coding that become apparent in such frameworks.³⁴ Here, the concern is that such a god can no longer keep his identity, or how he judges people, a mystery. In such circumstances, mystery was still an acceptable thesis, but it had to be modified. God’s will was now seen as unfathomable. Although he makes laws and reveals himself, he also burdened human beings to such a degree with sin that they ultimately did not know if they were satisfying his demands or whether they would be granted the mercy necessary to make up for their deficits.

The core point of incommunicability now takes on the form of a paradox. Final insights can only be communicated as paradoxes. These are specially tailored for second-order observers, that is, those who attempt to observe God’s observing, particularly the devil and theologians. For curiositas remained prohibited in religious matters and was handled the old way, as a prohibition on intruding into mysteries [deemed sacred]. It was still regarded in the seventeenth century as “pride and wantonness of Knowledge, because it looketh after high things that are above us and after hidden things that are denied us.”³⁵ But attempting to penetrate the mystery was now no longer punished with thunder and lightning, with poor harvests and illness—but with the banality of the formulations that emerge for it. Mysteries are sought to no avail, or (if a starker assessment is wished for) they are “a source of error.”³⁶ God is now clearly attuned to human individuality. He punishes people by rendering their efforts useless. And, as a loving God, he makes them learn. And in the specific religious practices of the seventeenth century, it was indirectly suggested (and understood by theologians) that man could only find salvation in himself and not in curiosity about things that cannot (or should not) be known.³⁷

In this phase of development, the impact of writing (and later, of printing) was clearly evident. Writing becomes a reprieve for the “obscenity of questioning.”³⁸ And yet, texts answered questions that had not even been asked. When questions and answers could not be coordinated, the starting situation for communicative prohibitions and controls was now different. And it was probably not a coincidence that the rhetoric of paradox also experienced its highpoint with the late Renaissance introduction of printing. Texts can treat author and reader as absent, thereby choosing a representation in the form of paradox that leaves what can be said about it to the person participating in the communication.

Paradoxical communication—whether oral (as in Buddhism) or written (as in reports of mystical experiences)—claims its own kind of authenticity (indeed rationality) by destroying communications that can be continued by saying “yes” or a “no” to what is communicated. In this form, it documents something “beyond yes and no.”³⁹ It puts its own honesty on display, for whenever one communicates true knowledge, one could always be making a mistake or even end up in a lie. Paradoxical communication is in this respect invincible, and it symbolizes the highest form of knowledge since it does not expose itself to refutation but represents itself as unconditional knowledge. Nonetheless, it is a mode of communication conveyed by faith assumptions that are themselves open to refutation. If it is acceptable to society, one arrives at a point in which the forms mystery and paradox are just as effective as before, but only for those who believe something here is closing itself off from communication.

Society reacted to these conditions by permitting ways of dealing with religion that did not have to be believed, that really could not be believed. This began with legal regulation of state religions and prohibitions on religion, then extends to the open, legal establishment of religious freedom.⁴⁰ Once placed in the hands of the law, there could be regulation of what could be said or what could be punished as blasphemy or as violating other people’s religious feelings. Leibniz’s solution to the problem of theodicy is developed in a similar fashion: God is restricted to a remainder function. He produces worlds by assuring the compossibility of this best of all possible worlds and by excluding whatever is incompatible with it.⁴¹ Yet such an approach did not work for long. With Kant and Hegel’s newly formulated dialectics, it became another issue to be addressed as part of the world: the problem of unstable antitheses.

Starting in the second half of the eighteenth century, there was a new way of discussing religious topics, and it qualifies as scientific. Viewed from the outside, religion appears to be culture, and it can thus be compared historically and regionally.⁴² Theology itself was historicized—taking part in the same process.⁴³ In the empirical research on religion that emerged later, the status in society of faith in God could be analyzed either statistically or functionally. Such an external description usually dealt more with externally obtained data, such as the results of polling research or a content analysis of texts.⁴⁴ In principle, however, nothing stands in the way of doing what has just been engaged in: treating questions of dogma (all the way to their most subtle esoteric constructions) as constructions with respect to their function and the historical and societal conditions of their plausibility. A scientist, for example, could arrive at an understanding that it is no longer advisable to speak about God in the form of paradoxical observation because it might weaken the willingness to believe.⁴⁵

When functional differentiation had been accepted, the mystery or paradox of God could be resolved in a polycontextural manner. It was assigned to different communicative contexts depending on the system to which communication seeks to relate. There was a great deal of nonreligious communication on religion that can be treated as irrelevant in the framework of the religion system. The operative closing of function systems and the universalization of their specific functional competence made possible a great diversity of combinations in intensity and indifference. Only when the social-structural “logic” of this arrangement is understood can continuity and discontinuity be seen in relation to the forms with which earlier societies had exposed the contingency formula “God” to communication—while at the same time protecting it against communication.

With the institution of a single personal God, a problem already touched on many times took on dramatic significance: the relationship of religion to morality. When religions are compared, this kind of overlap in religion and morality is seen to be more the exception than the rule.⁴⁶ In the animistic religions of tribal cultures, it is almost out of the question for numinous powers to involve themselves in the moral affairs of humanity.⁴⁷ Rather, what has to be regulated normatively in society has to be protected against magic, against single-case interventions from the outside. Structural interrelations between religion and social life are not mediated by moral directives [Gebote] but by the schema of pure/impure, to name but one example.⁴⁸ That way, contingencies can be reduced. In addition, external threats emanating from magical powers are a reason to react internally by moralizing the directives for handling such contact.⁴⁹ Whoever angers the powers in the other world not only endangers himself but society in general. Yet that indicates the precise lack of morality, which should mean the same thing in both the transcendent and immanent realms. Polytheistic societies also avoid too close an association between religion and morality. The gods and goddesses with their capabilities and interests represent only differences of standpoint. If there is a higher god at all (as is typical in the African religious sphere), it becomes difficult to assign him moral preferences. In this respect, his essence remains ambivalent.⁵⁰ Nonetheless, toward the end of the development under discussion, in the eighteenth century, one encountered a God who was largely inactive. He could not let his moral preferences be known but had ordered the world with his invisible hand so well that it could run on its own. Morality, in this case, became a societal institution which (if nothing else) also evaluated religions with respect to their civilizing qualities.⁵¹ The transcendence of the religious code represented by a personal God, who at the same time is seen as epitomizing the good, is therefore a special constellation faced with related problems that have to be avoided by a supplemental semantics.

There are a number of (nearly) compelling reasons to burden the one God, the observer God, with morality. If he already observes and sees everything but also wants all that he sees, then how could he behave neutrally in a matter as important as morality? The God of the Hebrews, unique as a universal god, learned this lesson in a paradigmatic way,⁵² and one may suppose that the idea of a god having a covenant with his people supported such a development. If the observers of God are imagining an observer God who himself can be observed (at least in part), then it is hard to avoid infecting this God with morality. And that means that there is a problem of how the codings of immanent/transcendent and good/bad relate to each other.

Yet in the beginning and well into modernity, morality implied only that God had arranged for order and dependability in the realm of free human activity. Despite this freedom, there was no sphere of natural creation incapable of being ordered. Seen thus, it is self-evident that establishing moral order among human beings—and thus morality itself—is good and not bad. By using its positive quality, the difference of good and bad is attributed quite naïvely to creation. And using this kind of natural moral awareness, one could extrapolate back to what God preferred to see.⁵³

It isn’t surprising that the necessity of distinction has a moral valuation in the case of older societies. What might be surprising, though, is the perverse idea of priests that God would enjoy being repeatedly confronted with the moral defects of his creation in the form of confessions.⁵⁴ If confession is then enforced organizationally and intended to have a socializing effect, doesn’t theology have to take on itself in order to make it comprehensible that permission to sin comes from God?

The resolution of this figure did not begin until the seventeenth century. On the one hand, the history of sin was seen as a success story, triggered by the felix culpa of the Fall,⁵⁵ and man might now see himself as capable of moral improvement. On the other hand, God fell into a trap set by himself: if he established the world according to moral criteria, then wasn’t the world morally optimal? There could be no moral reason to create a world that is morally defective. God was acting on moral necessity. And if his knowledge were thought to be defective, he could be recalculated out of the world. We thus arrive at Leibniz’s “theology.”⁵⁶

There was another problem as well: the old religious symbols had to go without authoritative interpretation and explicit reference. In practicing public communication, both “papists” and “enthusiastic” Puritans could make themselves suspect.⁵⁷ Consequently, it could no longer be safely assumed that observing God directly or institutionally would help one figure out what he expects. Or there is the question on internal enlightenment posed by Hobbes, Locke, and Leibniz: how can one distinguish between divine inspiration and the devil’s insinuations without engaging one’s own (human) reason? It was thereby suggested that the natural capacity for moral and aesthetic feelings should be valued more. It was also assumed that one might rely on this capacity because God had given us such inclinations. But it should be enough to observe humanity on the path of civilized progress; as a result, the eighteenth century developed a semantics of civilization and culture that spoke to this issue.

The issue was no longer that man was observing how God observes him. Nor was the issue “theological”—even if religion was not being questioned as a core certainty or as a social institution or accomplishment of “culture.” If reason (reflection) had to be entered into the equation, then it really meant that human beings had to be assumed to function as nontrivial machines.⁵⁸ And only then was it deemed worthwhile to observe their observing, since one could no longer manage only with divine knowledge, principles, and laws.

In the eighteenth century, the unity of religion and morality fell apart completely as the result of the Enlightenment and with the knowledge that a historical and regional majority of “cultures” had their own particular religions. Morality was reformulated on an anthropological basis, using the concept of moral sentiments in particular. At least in “educated strata,” it became morally unbearable to despise someone, such as a Jew or Muslim, because they had a different religion. But in so doing, one struck the criteria from the hands of one’s own God, who would like to see himself being worshipped.

These considerations all lead to the question: what are the effects of such changes on societal communication? One would need an elaborate sociological theory of moral communication to see the problems involved.⁵⁹ A few keywords will have to suffice for the time being. Moral communication regulates the division of respect [Achtung] and disdain [Mißachtung] toward [both] those present and those absent. This distinction of present/absent already causes problems, since respect is more readily shown for those present, and disdain more often for those who are not. This problem and the similar one of a certain temporal consistency are solved by conditioning the “deserving” of respect and disdain. As a consequence, moral communication can refer back to itself recursively. These conditions (or rules) can assume a normative form but can also determine the criteria by which accomplishments earn recognition (accomplishments that cannot be demanded “beyond the call of duty”). The problem of morality then lies in the invariance and certainty of such programs. On the same level of generalization, a universal moral code of good/bad is formed (or, if internal attitudes are included, a code of good/evil). Yet that code does not create the necessary certainty since it actually reflects contingency, leaving open which modes of conduct relate to which moral values. It seems to be this uncertainty that has triggered demands for a transcendent guarantee of moral criteria.⁶⁰ What appears and is practiced as morality is a supplement of its coding, and the need for reasonable explanations is another supplement.

The attempt to solve the problem of moral rules by externalization, however, returns the problem to the source. It brings us to the question of why God has created or permitted the difference of good and bad, and why he does not use the possibilities available to him to help the good triumph instead of the bad. Only in the modern era do people speak of “theodicy,” but it is a problem as old as identifying God with morality.⁶¹ One can imagine the most diverse solutions. An opponent of God might be projected into a transcendent realm, creating an ambivalent situation until the final victory of the good. Or the observability of the observer God could be acknowledged then restricted. God lets himself be observed only to a limited extent (in giving humanity its freedom, for instance).⁶² In these cases, significant temporary conditions are tolerable because one presumes there is immortality. Hence, the final victory of good or (in the other case) the Last Judgment will bring clarity to these conditions, albeit in the manner already proclaimed (but nonetheless surprising to stubborn moralists). Whoever had relied too greatly on morality will think he has been deceived, and an indirect path via sin will prove necessary for achieving redemption.⁶³ Yet redemption can only be achieved using the dialectic of moral values applied beforehand. It is only necessary to take this zigzag path to it because the problem of theodicy had already been raised.

Further solutions come into view when one dares to reconstruct how divine calculations are made. For a solution in which only the good achieve happiness also has disadvantages that God would clearly wish to avoid. In such cases, those deemed good could not at all know if they are good for the sake of what is good or for the sake of the happiness associated with it.⁶⁴ In order to avoid such calculations, the distinctions of good/bad and happiness/sorrow have to be uncoupled. Although some good people have to be asked to suffer, they are at least able to know that their suffering is observed and approved of by God. They may then cling all the more to their goodness instead of its consequences. Here as elsewhere, theodicy also means having more variety within a more reliable order.

Another difficulty is when the moral rules not only have to be confirmed religiously but (precisely for that reason) when the binary code of morality also has to be copied into religion. If one wagers on a solution that makes use of temporality, then two final repositories have to be available: These are heaven and hell, the realm of God and that of the devil. Such logic is difficult to dispense with, no matter how hard people have been working at it since the eighteenth-century Enlightenment.⁶⁵ For how should a God be imagined who himself is “the good,” who has made a clear moral choice, but who seems not to have made any provisions for whether our actions are good or bad?

While the questions dealt with up to now have received in-depth theological attention, a different problem has been perceived, mainly by external observers. The identification of morality with religion and the reidentification of religion with morality have infected religion as well with the “polemogenous” structure of moral communication.⁶⁶ For morality is easy to argue about, particularly when the criteria for respect and disdain are not fully insulated from their relationship to social structures. Moreover, there are always friends and enemies, dependents and those more highly placed, as well as ways to anticipate the effects of communication. To further moralize questions of faith is to throw holy oil on the fire. The effect, especially in the domain of religions that link religious and moral codes, is that millions of people have been killed. The paradox of coding has thus found a fully practical way of resolving itself: one tortures and murders in the name of the “good God” because other ways of persuading people have failed, and that is the only way the difference of respect and disdain can ultimately be sanctioned. For it is no longer acceptable to say that such things do not matter.

These troubles got worse in the Middle Ages owing to the legal and organizational powers of the Catholic Church, and once again in early modernity with the wars of religion, which were in part politically motivated. It got so bad that society developed immune responses, with assistance from the law. But in particular, the form of societal differentiation was altered and along with it, the context in which religion and morality were utilized. In the transition to a primarily functional differentiation, the religious positing of the world lost its significance for the whole society (which does not, however, permit us to draw conclusions about the intensity of religious communication). In addition, it became clear that morality was no longer appropriate as a form of societal integration. With stratified societies using either religion or morality to express unity, despite differences in hierarchy,⁶⁷ something else ultimately became clear: under a regime of functional differentiation, none of the codes of function systems could identify their positive/negative values with those of morality.⁶⁸ Moral communication was certainly reproduced on a massive scale and looked for its themes in the domain of function systems.⁶⁹ Yet only the code of morality could still be applied universally, owing to its formalism and strict bivalence.

In contrast, moral programs are hardly able to achieve consensus any more. Unlike positive law, and influenced as they are by the mass media, they are not provided with normal guarantees that they will be implemented. As a result, societal integration has to be left up to the reciprocal limitations on the movements of function systems. Even (academic) ethics has lost control over morality. Instead of getting involved in problems and consequences of moral communication, it accepts the function of preparing legal regulations discursively. At this point, one might say, morality is something that is only relevant in everyday communication (including press and television) and really only in the most pathological cases.

What has been just as dramatic are the changes in the domain of communications that use the mode of second-order observation. In the meantime, all function systems have switched to this mode:⁷⁰ science with help of publication, the economy through its market orientation, politics via public opinion, and family or other close relationships on the basis of personal intimacy. Teachers are required to observe how they are being observed by children, yet no one expects them for that reason to think of themselves as gods. Jurists find their rules of positive law in decisions that make perceptible how law observes itself in a given situation. Hence, it is no longer striking that religion too makes use of this mode of second-order observation with the help of its professional experts. Here the contingency formula “God” serves as a functional equivalent for what elsewhere is realized by the marketplace or public opinion, by a look from one’s sexual partner, or by the notion that children can be trained—that is, by the reflexive processing of the system using system-specific observations of system-specific observations. If one can still speak of a discourse common to all of society, then it is up to the intellectuals—who deal with such matters according to the (pleasantly unpleasant) formula of critique—to describe how other intellectuals describe what it is they describe.

In retrospect, it appears as if religions with an observer God might have prepared this shift from all-important, broad-based communication to the mode of second-order observation. To some extent, “God” has been used to try out and rehearse what was later supposed to become a universal social way of dealing with greater complexity. And if relevant function-specific forms can be found for this state of affairs (market prices and media themes, children and rules for representation in scientific publications or grounds for legal decisions), then it is no longer necessary to unify the totality of meaning in a religious-moral cosmology. This problem may recur in the future, but religion seeks to have a ready-made answer for it wherever it becomes virulent due to internal or external maladjustments of the societal system.

The observer God had offered a security of orientation that was nearly unequalled. If that idea of him is given up, “orientation” becomes a problem (and a buzzword too). For he had homogenized existence, making it appear to be a continuum of rationality. He had guaranteed that everything that exists could be known (if not always by human beings). Not knowing was thus an anthropological (if not humanistic) idea; it was not a metaphysical one. In other words, we did not have to count on non-knowledge [Nichtwissen] being a condition for the possibility of knowledge or on efforts to know being able to result in still more ignorance. The limits on attaining knowledge were marked by the stop signs of mystery and prohibitions on curiosity.

Further, the observer God had provided distinctions (or in any case those which are most important) with a preferential side, a side on which actual existence, perfection, or nature could be found. And that had made it possible to see this side as crucial to the meaning of the distinction itself. One could consequently see man as a basis for the man/woman distinction. The city (or the political) became a basis for the polis/oikos distinction. The oral/written distinction replaced writing as a merely technical externalization. The soul (undying)/body (dying) distinction privileged eternal life. The concept/metaphor distinction assumed that the distinction itself was conceptual. The good was a basis for the good/bad distinction; the true for the true/untrue distinction; being for the being/nonbeing distinction. Everywhere in “old European” thought, one finds this structure of a hierarchized opposition, of a hierarchy that outdid itself. Logic had closed itself off with this (logically unreliable) double application of the preferred side as the meaning of the distinction itself—until Gödel came on the scene. In this kind of resolved paradox, the world could be read as something God wanted.

The philosophy of deconstruction takes aim at this decisive point of onto-theological metaphysics. It is the metaphysics (if it truly is one) of the regrounding upon difference; it is the metaphysics of spirit that can no longer exclude what it is not, the metaphysics of paradox. “The motif of homogeneity, the theological motif par excellence is what must be destroyed.”⁷¹ Even this “must” can clearly be deconstructed again. (That is always part of reflecting on it.) It becomes set on the question of how it would be if it were not so. It repeats (though from a greater distance) the “writing” of theology. This should not be understood as a denunciation—or only in the sense that the denunciation is repeating (albeit in the sense of différance) what it is denouncing. The denunciation lives “de-reconstructively” from what it is rejecting.

Do we thus know a great deal more when we know this? What we gain certainly does not consist in a type of knowing better which—knowing being to be on one’s side—can be easily deconstructed. But what we gain is a larger structural abundance of forms available for observation, and thereby an expansion of the possibilities for communication.

Besides this type of hierarchy, which makes use of asymmetrical oppositions, there is the more renowned hierarchy of essences that resolves the observer paradox by distinguishing between levels. God also created angels between himself and man, giving them a different form of cognitive access to the world. Angels are capable of perceiving ideas. In their minds, there is a different world, an ideal one, a purely spiritual one that man can use to measure defects in his own state of knowledge.⁷² In turn, this differentiation of levels is capable of accounting for the asymmetry in oppositions that, on their good side, direct our gaze upward.

Today, all these things are supposed to have come to an end: history, humanity, metaphysics, art, the book, and God. But perhaps we need only learn the difference it makes when we describe something this way.