The process of generating meaningful information can be understood by analyzing the relationship between metonymy and metaphor. While the metaphor is a parabolic analogy with a ‘focal point’, the metonymy rests on symbolic analogy. (1) The main difference between metonymy and metaphor has been analyzed by Roman Jakobson(1971). He distinguishes two types of sign arrangements: the combination or contexture (with the two subtypes concurrence and concatenation), as well as the selection or substitution. The constituents of a piece of information are connected to a code via an internal relationship as well as to the environment, from which the material for the generation of this information originates, via an external relation (Jakobson 1971, 243). Jacobson deduces the two constitutive principles of metonymy and metaphor from these two basic operations (Jakobson 1971, 254). Metonymy consists of a specific combination of signs on the syntagmatic axis and rests on the principle of contiguity (spatial and temporal proximity). The metaphor is a selection on the paradigmatic axis. It is produced by substituting one sign for another to which it bears a paradigmatic relation and is based on the principle of similarity. These two tropes, however, are not a categorical distinction, but poles which regulate the opening and closing of semiosis (see the articles in Dirven and Pörings 2003 as well as in Spieß and Köpcke 2015). Determining a sign element as a metonymy or a metaphor is carried out by the two interpretants as processors of the elementary semiotic system: “[…] there are always two possible interpretants (Peirce’s term) of the sign, one referring to the code and the other to the context of the message. The interpretant referring to the code is linked to it by similarity (metaphor), and the interpretant referring to the message is linked to it by contiguity (metonymy)” (Wilden 1980, 47).
Metonymy and metaphor together are necessary for generating and processing information. The metonymic combination is related other-referentially to the semiotic context from which the material for the generation of information originates, and the metaphorical selection provides for the self-reference of the semiotic code. (2) For example, the sign CHURCH, if it refers paradigmatically to a SACRED SPACE, is metaphorically linked to the religious code, and metonymically linked to the semiotic context from which the material for the generation of information originates—for example in the statement: “You should go to church again.” This sentence can be embedded in religious communication, for instance in a conversation on matters regarding one’s religious conduct of life. In this case, both the code and the information context observed from an other-referential viewpoint are determined in religious terms. However, the sentence can, for instance, also be part of educational communication. In this case, the sign CHURCH is paradigmatically related to CHURCH ATTENDANCE as an educational means in the educational code communicable/non-communicable (3). The distinction as well as the interplay between syntagmatic combination and paradigmatic selection provides an explanation for the basic polysemy of individual signs (Bartsch 2003). It is only in a particular pragmatic-semiotic context that they acquire a specific sense.
Analogies, by establishing similarity between something distinctive in comparison to some distinct other, generate and process the paradox of ‘is and is not’. This is what metaphor theories call attention to. (4) The “predicative basic structure” (Weinrich 1963, 337) of a “bold” or “living” metaphor produces a split-reference (Jakobson 1960, 371) or double reference (“suspended reference and displayed reference”) (Ricœur 1978, 261); the “dead metaphor” as an entrenched metonymy or synecdoche obscures it (Silk 1974, 27–56). This is the paradox of the metaphor: “The metaphorical ‘is’ at once signifies both ‘is not’ and ‘is like’” (Ricœur 1978, 6). In systems theory metaphors arise on the boundary between systems of meaning and their environment; they combine identity within the system with other-referentially observed similarity (Tourangeau and Sternberg 1981). (5) At these transitions, coded literalism and metaphorical surplus oscillate.
What is true for semiosis, on the whole, already applies to the complete sign as its smallest systemic unit. It is based on the two operations of metonymy and metaphor in the above-mentioned sense as well as on their synthesis (Figure 12).
According to Peirce’s sign theory, the metaphor is a type of representamen, because it is based on the principle of similarity and is self-referential. (6) The metonymy is a type of the sign object, because it is other-referential, even though it is transitively related to the respective code in the complete sign. The two tropes are both processor and process. In the sign system, the relations R1–I1 and I2–O1 are metonymic combinations that are based on a certain code (in the case of differentiated religion: on the code transcendent/immanent). However, the relation R1–I1 is self-referential, because the first interpretant transforms the metaphorical surplus by means of the system-specific code into information, which is manageable for the system. The relation I2–O1 is other-referential, because the second interpretant is environmentally sensitive and interprets the first sign object in view of other-reference. The transitions between I1 and I2 and between O1 and O2 are a metaphorical selection because they connect the level of the observing sign form with the level of the observed sign content. The relation between R2 and O2 is to be understood as a combination of metonymy and metaphor. While the sign object O2 is other-referential and therefore context-sensitive, the representamen R2 is self-referential and at the same time, it constitutes a possibility to connect to further semiosis. (7) Due to the relation between metonymy and metaphor, a change between the two can take place within the sign system (Bartsch 2003, 73–74; Goossens 1995). In the systemic process, O1 is metonymical in character but interspersed with metaphorical elements. Conversely, R2 is metaphorical in character, but transfers the other-referential metonymy into self-reference and thus also has metonymical elements. Due to the system reference of the complete sign, the relations between R1 and O2 as well as R2 and O1 are congruent.
The oscillation between metonymy and metaphor, as well as their synthesis, are the conditions for the production of semiotic information as a translation (metaphor) of the transcription (metonymy) in the interplay between closure and opening. At the same time, process (time) and structure together with the position of the elements (space) must be mediated (see Figure 12, above). In procedural terms, the metonymic transcription induces the syntagmatic closure process regarding the respective code, while the metaphorical translation is based on paradigmatic openness. In structural terms, metonymy as a sign aspect of secondness (relations) is other-referential-open, while the metaphor as a sign aspect of firstness (quality) is self-referential-closed as a result of paradigmatic selection. In the complete sign, R1, O1, and I1 are closed as a unity in the shape of the observed sign content but are open in the direction of the observing sign form. The sign system is therefore open in system-internal and epistemic terms, but it is closed in operative terms regarding the environment. (8)
The metaphorical translation of the metonymic transcription is the basal process of self-observation in a system: It draws a distinction (transcription) and observes it (translation) with respect to the synthesis of self-reference and other-reference. This process corresponds to Peirce’s understanding of metaphors. They “represent the representative character of a representamen by representing a parallelism in something else” (Peirce 1994, 2). The “parallelism in something else” consists in the metonymy of the sign object. The interrelation between metonymy and metaphor as well as the oscillation of closure and opening makes it—at least to some extent—clear how semiotic information emerges in the process of translating the transcription.
