Now, if I go back and review the entire gist of my philosophical reflections, I realize that I always placed them under the sign of the limit.
— Umberto Eco, “Weak Thought versus the Limits of Interpretation”
I. THE DREAM OF REASON
In a short but very interesting essay dating back to the end of the 1970s, Ernst Gombrich investigated the iconography of “reason” during the French Revolution.1 Gombrich closes his article, almost incidentally, mentioning what is probably the most famous illustration of the Caprichos series by Francisco Goya, El sueño de la razón produce monstruos. He points out that in Spanish this title may mean “the sleep of reason produces monsters” (as it is usually interpreted) but also “the dream of reason produces monsters.” Gombrich alludes to the ambiguous and embarrassed attitude of twentieth-century culture towards reason and rationality (two concepts I will overlap here). From one side, reason is considered a fundamental human faculty, bringing progress and tolerance; from the other side, if treated as an end in itself or applied without control, it becomes an accomplice to or instrument of mass tragedies. This attitude is quite widespread, even in those who refuse to seek refuge in irrationality. Max Weber says that a capitalist society closes itself in the “iron cage” of rationality.2 And Max Horkheimer and Theodor Adorno, who by no means can be suspected of being tempted by irrationality, open Dialectic of Enlightenment with an apocalyptic and unappealable verdict: “the wholly enlightened earth is radiant with triumphal calamity.”3
Gombrich’s words remind me that there is an expression I have always associated with Umberto Eco’s works, quite instinctively: this expression is “reasonableness.” And I now realize how often, and in what key positions, this word emerges in his writings. The aim of this essay is to understand what meaning Eco gives to the word “reasonableness” and where, in his theory, we can find this reasonableness. I only want to explore the possibility of a very specific point of view on Eco’s work and to observe links among elements that are already connected in other ways.
II. REASON AND REASONABLENESS
Oversimplifying, in philosophical language the term “reason” indicates either a general human faculty that leads us in our investigations and research or one of the specific ways in which this general faculty acts (for instance, reason as a discursive procedure). The term “reasonableness” usually has a connotation of limit; it indicates doubts about the infallibility of reason; or it restricts an argument’s validity to certain circumstances or fields. Eco never gives a definition of “reasonableness” or “reasonable” but he explains many times what he intends by “reason” as opposed to “reasonableness.” He does it very clearly in 1980, reviewing an Italian book titled The Crisis of Reason.4
According to Eco, if we consider reason as the ability to produce abstractions and to speak through abstractions, we cannot say it is in crisis. In this case reason is a “natural knowledge, characteristic of man, opposed on the one hand to mere instinctive reactions, and on the other to intuitive knowledge (such as mystical illuminations, faith, subjective experiences not communicable through language, and so on).”5 The kind of reason that is in crisis is the one that can be identified with the intuition of the absolute or with Platonism. In the first case reason is “a special faculty of knowing the Absolute by direct view; it is the self-knowledge of the idealistic ego; it is the intuition of the prime principles which both the cosmos and the human mind obey, and even the divine mind.”6 In the second case it is a system of universal principles that precedes human abstractive capacity; at most man may recognize these principles, “perhaps with difficulty and after long reflection.”7 So, what is in crisis is a classical or strong notion of reason; the kind of “reason” or “rationality” that may have contributed to disastrous events, as Gombrich knew. “The problem is not to kill reason, but to render bad reasons harmless, and to dissociate the notion of reason from that of truth.”8
We now have a clear idea of what Eco considers a “bad” reason or a not reasonable reason. But what I am looking for is what he means by “reasonableness.” As I have already said, there is not a general definition of “reasonableness” in Eco’s work; nevertheless, it is possible to trace the contours of this concept and to understand how it is implied by many of his writings. I think we can recognize the idea of “reasonableness” in at least three important aspects of Eco’s theory. I will briefly discuss the first two aspects because they have already been treated many times, and I will focus more on the third one.
III. FIRST REASONABLE ASPECT: A MINIMAL DEFINITION OF REALISM
The first reasonable feature I find in Eco’s thought is a general attitude based on a “minimal definition” of realism. It could seem strange to say this of a person who has been accused of extreme idealism for his theoretical positions,9 or of nihilism for his narrative works (above all for The Name of the Rose).10 But, as I will explain, it is very hard to think about Eco’s ideas in these terms and, besides, realism is one of the main pillars of the work of Charles Sanders Peirce, an author who plays a relevant role in Eco’s system.
What is a “minimal definition” of realism? “According [to it] a realist is someone who believes that things go a certain way—even though we may not know which way and may never succeed in knowing.”11 What we can do is produce interpretations of the world; we have to revise them, to set them against the resistance of the world, to never consider them definitive.
Though we may accept that the descriptions we give of the World (or of a text as World) are always prospective, this ought not to prevent our readings from attempting to keep pace with the world, at least from a certain point of view, without ever pretending that the said readings, even when they seem on the whole to be “good,” are to be considered definitive. And, since in this series of interpretations we stick to the parameters offered by the facts to be interpreted (even if only locally and from a certain perspective), we must presume that the readings we give of the facts may be accurate or they may be completely off the wall.12
It is in Kant and the Platypus, at the end of the 1990s, that Eco has more extensively treated this point.13 But this does not mean that he arrived at this position only then or that we cannot find some traces of these ideas before. In fact during the 1970s and 1980s semiotics was still a newborn discipline and it had to find its space and specificity. From this point of view, the problem of the relationship between our representation of the world and the world itself was not pertinent; it was more interesting to investigate how this representation of the world (or of a possible world) was organized and used. In other words, in the classical triangle expression–content–referent semiotics was interested by the expression‒content axis, while it left to other disciplines the investigation of the content‒referent axis. It was probably a wise choice, but some themes, such as the semiotic status of visual signs or, later, the cognitive basis of semantics, continued to pose the problem of the relationship between semiotic systems and the world. And when we come to it—that is, to an ontological problem—we cannot refuse to make an hypothesis, even if it is open-ended.
From the end of the 1960s to the early 1980s semiotics was deeply involved in the so-called “debate on iconism.”14 Oversimplifying, the question at stake was whether visual signs were conventional like words or were in some way “natural” or, better, “motivated.” Eco was one of the main actors in this debate and was accused of idealism (for instance, by Thomas Maldonado) because he had underlined how our mental schemas are shaped by cultural influences. But his position was very far from idealism and was essentially equivalent to that expressed in Kant and the Platypus. Underlining cultural influences on perception does not mean to deny that we perceive a world outside and that this world influences perception in its turn. Eco simply focused his attention on a different aspect of the problem. In the 1960s and 1970s a “specular,” Aristotelian conception of knowledge was still strong and perhaps dominant: mental schemas were considered to be perfectly in alignment with external reality and it was thus necessary to lay particular emphasis on the conventional aspects of vision. In the 1990s this necessity no longer existed and the attention shifted to the motivated origin of the schemas and on how they are modified through continual contact with reality. In other words, it shifted to the dialectic between an interpretation that tries adequately to represent the world and a world that stimulates us but cannot be completely grasped.
In the opposition between an Aristotelian realism and the arbitrary dogma that could lead to extreme idealism, Eco proposes a reasonable settlement on a minimal realism. Reasonableness is a theoretical style that is inclined to a synthetic approach and that renounces pure and strong concepts, preferring more complex and hybrid ones, such as, in Kant and the Platypus, hypoicons—iconic signs that result from mixing cultural factors and the pressure of reality.
IV. SECOND REASONABLE ASPECT: A LIMITED KNOWLEDGE OF THE (LIMITED) ENCYCLOPEDIA
The second aspect in which I recognize reasonableness in Eco’s thought is the passage from a semantic theory based on the model of the dictionary to the idea of encyclopedia, which is discussed in depth in this volume by Patrizia Violi. Dictionary semantics is founded on semantic primitives and on genus-species definitions. Encyclopedic semantics is founded on a highly interconnected network of units that can be used, according to the specific situation, as expressions, sememes (meanings to be defined), or semes (meanings used to interpret other meanings).
The relevant point for my discussion is that we can never get the whole encyclopedia (and thus the world it represents). There are two reasons for this incapacity. First, the global encyclopedia is so complex that we can only select partial and local structures of it, choosing a specific point of view. If we demanded something more, we would have to go beyond the capacity of every known discipline and technology: it would be like asking to “catch” the whole of Borges’s Aleph. The second reason for this incapacity is that the description of the encyclopedia would be part of the encyclopedia itself. It is a sort of cultural version of Heisenberg’s uncertainty principle: to describe the encyclopedia means to alter it and to make it something different from what we were describing.
In the theory of encyclopedic semantics, reasonableness consists of a double limit: a limit to the possibility of an absolute knowledge; and the consequent need to always limit our observations to a specific field.
V. THIRD REASONABLE ASPECT: THE INDETERMINACY OF INTERPRETATION
I would like to focus more on the third aspect of reasonableness. It concerns the indeterminacy of interpretation. In September 1986 Eco gave a talk at a symposium of semioticians and immunologists, organized to investigate if the immune system could be considered as a communication system among lymphoid cells. In his talk Eco compares two kinds of process:
Let us consider two different cases: (i) I push a button and a bell tolls; (ii) I say rose and somebody answers “you mean red flower.”
By pushing the button I implement a process that cannot but end with the tolling of the bell, while uttering rose I implement a process that can end with these (or other) responses: “you mean the past tense of to rise,” “you are quoting a word used by Gertrude Stein,” “And so what?” “I am not interested in what you are saying,” or “buru buru aba boom!” (in Macomba dialect: “what kind of damned language are you speaking?”).
The first phenomenon is based upon a stimulus-response mechanism, the second one requires a comparison between the received expression and a given sign-system, plus the decision of interpreting the expression.15
The stimulus-response process is dyadic: two elements, A and B, are copresent and one immediately brings us to the other; it is a causal sequence. The second process (which is the real semiotic process) is triadic and it implies three elements: A, B, and C. One of A or B has to be absent; if B is absent, A is used as a sign of B thanks to the third element C, which is a code or the interpretive process based on a code.
The dyadic relationship between A and B is necessary; once we recognize it, we know for certain what the response to A will be (it will be B). In other words, A causes B without any mediation, whereas in the interpretive process there is an unpredictable and potentially infinite series of C. The space of interpretation is called by Eco the “C space,” and he considers it a “space of choice and of supposed indeterminacy.”16
The opposition between a dyadic and a triadic model stimulates two reflections. The first one concerns the evolution of semiotic theory and Eco’s contribution to it. The distinction between these two processes is a direct consequence of the passage from a conception of sign as equivalence (smoke = gas produced by combustion) to one of sign as inference (if “smoke” is in contexts x or y then it means “gas produced by combustion,” while if it is in contexts z or k it means “the activity of a person ingesting gas produced by the combustion of tobacco”). This idea was already present in the 1970s but was fully developed by Eco in the 1980s.17
The second reflection needs a longer discussion and is strictly related to the idea of reasonableness. Talking about the C space as a “space of choice” and of “indeterminacy” might be intended as a theoretical surrender, or at least as a withdrawal. Interpretation would become a sort of black box into which semiotic inquiry is not allowed. We can know the input but we ignore how it is processed and, consequently, what the output is. Or, at least, it could be intended as an example of what cognitive sciences call the homunculus argument (or homunculus fallacy): an infinite regress where a phenomenon is accounted for in terms of the same phenomenon that has to be explained. In this case semiotic theory has to explain how the interpretation process occurs. To do this, it proposes rules and mechanisms of interpretation; but to explain how rules are chosen and applied it would need to suppose an interpretation activity that is internal to the interpretation process, thus falling into an infinite regress.
I do not think this can be said of Eco’s theory. Since the late 1950s, when it was an aesthetic theory that was at stake, Eco has always declared his impatience with theories that use the argument of “ineffability”;18 he has always preferred to outline disciplinary boundaries or to declare temporary impossibility of going further in the research. In the next pages I will try to support this point of view and to explain how this “indeterminacy” has to be conceived, and how reasonable this conception is.
From Eco’s point of view, the C space is not a black box. He has meticulously described the interpretive mechanisms that regulate the process of semiosis. He has done it in the first chapter of his Italian book Lector in fabula, where the fundamentals of an instructional semantics are outlined;19 he has come back to this theme in the second and third chapter of Semiotics and the Philosophy of Language (where he also gives a precise account of how metaphors work); he has given a really detailed description of the interpretive process in one of the essays of The Limits of Interpretation, where he pretends to report the amusing and instructive dialogue between a cognitive scientist and the Charles Sanders Personal, a computer equipped with an interpretive software.20
In these works Eco has given, as far as possible, an account of the interpretive process, and the aim of my discussion is also to understand what should be intended by “as far as possible.” But, certainly, it cannot be said that interpretation has been taken for granted and not deeply investigated. Eco gives at least two answers to a very central question: “How does interpretation work?” The first answer is offered by instructional semantics, as I have anticipated. The second is Peirce’s interpretant chain: this is a mechanism that is more difficult to handle and it may be more easily suspected of some kind of unsatisfactory theoretical indeterminacy.
An instructional semantics manages to represent contexts and circumstances, or, in other words, incorporates pragmatics into semantics. It may explain how we can disambiguate meaning and choose the right contextual and circumstantial selections, if we have the right encyclopedia. For instance, it lets us explain how we understand the difference between the sentences “should we take Junior back to the zoo?” and “should we take the lion back to the zoo?” From a syntactical point of view they are equivalent, as they are on the lexical level, except for the difference between “Junior,” which is probably the name of a young boy, and “lion,” which is the common name for an animal that usually lives in the savannah (and we presume he does not like living in a cage very much). On the basis of this difference (the presence of “Junior” or “lion” in the sentence) we can activate some aspects of the meaning of “zoo” and narcotize some others: for Junior the zoo is the place of extraordinary experiences, for the lion it is a place of detention. But even at this initial level we may have other possibilities, for Junior’s parents could be fervent animal-rights activists and they could take Junior to the zoo to show him what a horrible place it is. Even if I have just admitted some degree of flexibility, all these interpretive choices and the contexts that activate them should already be recorded in the encyclopedia (or they can be recorded as a result of an inferential activity or of an explication).
But this first-level disambiguation is only a part of the interpretation process. Further connotations will be activated, together with common or intertextual frames, which could, for instance, suggest how the story will continue (we could expect a final fight between the lion and the zoo custodian, or a clever trick devised by the latter); and each passage brings us towards other interpretations. This is exactly what is intended by the expression “interpretant chain.” The point, as Peirce knew, is that this chain is potentially infinite and this is, at the same time, its strength and its weakness. It is a strength because this unlimited and unpredictable succession of interpretants explains why meaning can be so rich, and stratified, and why multiple interpretations of the same text are possible. But it may also be a theoretical weakness because, if the interpretant chain is really unlimited and occurs in an encyclopedic space where everything is interconnected, we could suspect that it is a fascinating but useless hypothesis, a night in which all cows are black.
It is for this reason that both Peirce and Eco emphasize many times that, even if potentially infinite, in the actual interpretation process the interpretant chain stops when it arrives at a certain point or, more precisely, when a habit has been established that is fit for the present situation. But a theoretical problem remains unsolved: how does this process go ahead, how is the connection between an interpretant and the next one established?
The answer is easy to find: instructional semantics and the interpretant chain are not two different explanations, but, rather, complementary ones. First of all, it has to be considered that even if the cultural unities forming an encyclopedia are (potentially) all interconnected, it does not mean these connections have to be equiprobable. Instructional semantics shows how, when we come to a fork in the interpretive process, contextual and circumstantial information make us choose one road or the other; and if more than one choice is legitimate, it helps us understand the implications of the choice we made. In terms of the interpretant chain, an instructional semantics can contribute to explaining how the process of semiosis goes ahead and which directions it could take at any moment. From this point of view, the process of semiosis, far from being undetermined and uncontrollable, has a regularity.
As Patrizia Violi has pointed out in this volume, even if it is explained in terms of the interpretant chain, the interpretation process can be guided by a certain pragmatic “laziness” and tends to follow existing paths. At any step there are many (but not infinite, at least from a synchronic point of view) possible alternatives, and they could even be contradictory because an encyclopedia is so complex that it can contain both if p then q and if p then not q. This is not a problem, even when it occurs not in the same encyclopedia, but in the same text. Let’s think about aesthetic texts, where we are hit by a wave of emotions, connections, ideas, a complex and even confused (but still able to be analyzed) meaning. Aesthetic experience is—happily—contradictory, and contradiction is often what makes a text interesting. When we analyze a text we usually try to trace coherent interpretive lines and, after that, to connect them with other compatible lines or to compare them with contradictory ones. But the real interpretive process does not occur in this aseptic and simplified way, and exposes the reader (or, in general, the user) to the wholeness of these stimuli and possibilities, making him or her follow contradictory interpretations in parallel or repeatedly jump from one to the other so that contradictory lines can overlap or clash at any moment.
Anyway, if we do not consider idiosyncratic interpretations (we leave them to psychoanalysis) or radical inventions (we will come back to them soon), these interpretive paths should already be present in the encyclopedia we are using; or, at least, there should already be the parts that will form them, probably arranged in a way that suggests or fosters their connection. So, as a matter of principle, these interpretive paths should be predictable. This does not mean that we can know in advance the result of an interpretive process, but we can at least foresee the interpretive paths that a certain encyclopedia allows and that a text stimulates. Which direction will be taken at a certain textual fork cannot be said (it depends on too many factors, and part of them are idiosyncratic, so not of a semiotic pertinence) but we can say: (i) that at that point there will be a textual fork, (ii) which are the possible alternatives it offers, and perhaps (iii) which of them are more probable, ceteris paribus.
Thus the indeterminacy Eco talks about concerns our certainty about the result of the interpretive process, not our understanding of its mechanism, nor our capacity to approximately forecast possible results within a certain encyclopedia. It is thus an indeterminacy that is completely compatible with a clear and sound theoretical frame. An analogy will clarify this point. Geneticists know they have done their duty when they describe DNA structure and the mechanisms of meiosis; they do not feel guilty that they cannot exactly forecast the genetic sequence of a baby, because they know chromosomal crossover will always shuffle the cards. On the contrary, they think this is a good thing because it lets us have genetic variability.
In conclusion, to pass from a dyadic to a triadic process is not to renounce any precise and detailed theoretical model to explain interpretation. A dyadic relationship implies that the same stimulus will always have the same response. A “C” space, in contrast, explains why the interpretive process may bring us to different results (the indeterminacy). But this does not mean that the C space is made of uncontrollable and unknown random processes. Moreover, the difference between dyadic and triadic processes may exist only at a macroscopic level. The C space could be decomposed into smaller dyadic mechanisms. As Eco writes: “Perhaps some day science will demonstrate that C space is only a figment like ether, presupposed in order to fill up an ‘empty’ interval where deterministic phenomena—that escape our present knowledge—take place. But until that moment we have to deal with C space.”21 This deterministic project is simply impossible for at least two good reasons: (i) the C space, where contextual selections and other kinds of inference occur, is the space of the encyclopedia, and when it comes to single interpretations this encyclopedia is different from person to person; (ii) the amount of information and factors to be considered would be, once again, beyond the possibility of the human mind or machines. If these two problems are solved, then we will be able to forecast not the possible but the real result of an interpretation, and we will have a discipline very similar to the psychohistory described by Isaac Asimov in his science fiction Foundation saga. As for now, we have to accept the idea of indeterminacy and that it is compatible with a sound and respectable (and reasonable) theory of meaning.
VI. WHAT WE ARE NOT, WHAT WE DO NOT SAY22
I do not think we could use the attribute “postmodern” to label Eco’s semiotic theory but—forgive my pun—it is certainly a postindustrial one. What does this mean? As we know, the twentieth-century industrial society spirit lies in Taylorism, according to which there is always one best way to do a thing.23 In contrast, we can say that Eco renounces the comfortable idea of “one best way,” of a unique interpretation, and he describes processes that may lead to many possible results. But—this is a very important point—not to every result. According to Eco, a semiotic theory should above all be able to indicate those interpretations that we can certainly exclude.
This idea is extensively discussed in the essay “Weak Thought versus the Limits of Interpretation,” which I have already cited. This essay deepens two aspects of Eco’s theory. The first one is the moderate realism that I have already treated in section III. Eco reminds the reader how he has never maintained that only interpretations, and not facts, exist. Facts exist; the problem is that, if we intend them as Peirce’s dynamic object, we can never grasp them as such and we can only know them through interpretations.
Since the idea of a sequence of interpretations is conceivable and makes sense only if we admit that there is something to interpret, wouldn’t it make sense too to come to grips with that something?
This is the problem I came back to definitively in Kant and the Platypus, but even prior to that, it was hard to credit me with the idea that there are no facts, only interpretations, considering I had written a book entitled—no less!—The Limits of Interpretation. And the problem had already come up previously in some of my earlier writings.24
This passage brings us to the second main theme of Eco’s essay: not all interpretations are legitimate. Eco had already treated this point in Lector in fabula, where he distinguished the interpretation (that takes into account a dialectic between the model reader and the author strategy) from the use of a text (where the text is considered as a mere stimulus for imagination). “In the face of the affirmation that there are no facts only interpretations, I have always maintained that every fact is the occasion for diverse and conflicting interpretations, but one curious thing about facts is that they resist interpretations they do not legitimate or support. In other words, though it may be difficult to decide whether one interpretation is better than another, we can always recognize untenable interpretations.”25 Recognizing and marking some interpretations as untenable or not valid means to expel them from a well-ordered debate where texts and arguments are considered for their legitimate meaning(s) and not as pretexts. Thus the question of the limits of interpretation is directly connected with the possibility of a fruitful discussion in human society. This point had already been treated in the book review of the early 1980s I cited above, where Eco traces the contours of a “rational” or “reasonable” way to go beyond the classic idea of reason he had just criticized. Reason is not only a faculty we use to investigate the world but also the criterion that should inform our interaction with others in public life. Eco’s idea is that a critical reasonableness should be based on the sharing of rules; these rules would make possible a common and public discussion; they do not always have to be the same and they vary in their strength according to the situation.
Rationality is exercised through the very fact that we are expressing propositions regarding the world, and even before making sure that these propositions are “true,” we have to make sure that others can understand them. So we have to work out some rules for common speech, logical rules which are also linguistic rules. Which is not to assert that when we speak we have to say always and only one thing, without ambiguity or multiple meanings. On the contrary, it is rather rational and reasonable to recognize that there exist also discourses (in dreams, in poetry, in the expression of desires and passions) that mean several things at once, contradictory among themselves. But precisely because it is fortunately obvious that our speech is also open and has multiple meanings, every so often, and in certain matters, we have to work out agreed norms of speech, for specific situations where we all decide to adopt the same criteria for using words and for linking them in propositions which can then be debated. Can I reasonably assert that human beings love food? Yes, even if there are dyspeptics, ascetics, and anorexics. We must simply agree and establish that, in this area of problems, statistical evidence can be held reasonable.26
Before discussing this point more deeply, I would like to underline how this is a very important issue for our society. Digital media have multiplied information and sources but not all of them are fully reliable. Of course it has always been so, but the number of unreliable sources has probably increased both in absolute terms (for mere quantitative reasons) and in terms of percentage: the access to the public sphere is now easier thanks to blogs and social media and that means a greater freedom of expression for every citizen, but it also implies that the access to the public debate is no longer regulated by professionals such as journalists or politicians. Even if they may be considered as interested and biased actors, it is true that as gatekeepers they could also filter untrue news or arguments unfit for a decent debate. Thus digital media are an extraordinary opportunity but they also make verification efforts more necessary than ever. Here I do not mean a verification of facts (fact checking still ought to be done by journalists, even if they may be helped by citizens) but a verification of interpretations of facts, in order to free public debate from narrations based on stereotypes, subreptitious arguments, and uses of other people’s texts. In this situation the aberrant decoding is no more an effect of asymmetrical encyclopedic competence or a way to deconstruct power mechanisms, but it is used to create, sometimes on purpose, the opportunity for violent discussions in the public sphere.
VII. REASONABLE ANALOGIES BETWEEN UMBERTO ECO AND JOHN RAWLS
Eco is not the only one to have used the notion of reasonableness in recent decades. In political philosophy, and on similar themes, this term has been widely used by John Rawls.27 There are interesting analogies and differences in the ways Eco and Rawls talk about reasonableness.
Rawls aims to build a system that could support a democratic, free, and pluralistic society. This system is based on three pillars: (i) citizens as free and equal persons; (ii) the idea of a well-ordered society, that is, “a society effectively regulated by a public conception of justice”; and (iii) the idea of “a society as a fair system of social cooperation over time from one generation to the next.”28 Linguists and semioticians could not help comparing the idea of social cooperation with the cooperative principle stated by Paul Grice.29 Just as Grice founded his conversational maxims on this cooperative principle, so Rawls’s social cooperation “is guided by publicly recognized rules and procedures which those cooperating accept as appropriate to regulate their conduct.”30
The sharing of rules for a civilized public interaction is a common point for Rawls and Eco, but this is not the only analogy. Both authors underline the importance of reasonableness for the success of this cooperation. Reasonableness is one of the most central concepts in Rawls’s theory.
As applied to the simplest case, namely to persons engaged in cooperation and situated as equals in relevant respects (or symmetrically, for short), reasonable persons are ready to propose, or to acknowledge when proposed by others, the principles needed to specify what can be seen by all as fair terms of cooperation. Reasonable persons also understand that they are to honor these principles, even at the expense of their own interests as circumstances may require, provided others likewise may be expected to honor them. It is unreasonable not to be ready to propose such principles, or not to honor fair terms of cooperation that others may reasonably be expected to accept; it is worse than unreasonable if one merely seems, or pretends, to propose or honor them but is ready to violate them to one’s advantage as the occasion permits.
Yet while it is unreasonable, it is not, in general, not rational. For it may be that some have a superior political power or are placed in more fortunate circumstances; and though these conditions are irrelevant, let us assume, in distinguishing between the persons in question as equals, it may be rational for those so placed to take advantage of their situation.31
“Rationality” is a separate but complementary faculty and applies to single agents and to their ends and interests, and to the ways to reach them.32 “Reasonableness” is the acceptation of shared rules that may, sometimes, be against our interest, but in the long run they assure the fairness and the order of the society we live in. There is another passage in which this aspect emerges very clearly:
The reasonable is viewed as a basic intuitive moral idea; it may be applied to persons, their decisions and actions, as well as to principles and standards, to comprehensive doctrines and to much else. . . . What constraints, though, are reasonable? We say: those that arise from situating citizens’ representatives symmetrically when they are represented solely as free and equal, and not as belonging to this or that social class, or as possessing these or those native endowments, or this or that (comprehensive) conception of the good. While this conjecture may have an initial plausibility, only its detailed elaboration can show how far it is sound.33
Both Rawls and Eco share the idea that reasonableness includes some sort of limit or constraint. For Eco, as we have seen, a reasonable approach sets various limits to the philosophical investigation and to the certainty of its result, or, in the case of public discussion, it pushes us to negotiate our meanings with other people. For Rawls, reasonableness is the necessary self-limitation of our potential convenience.
In addition, both authors think that reasonableness is not only a preliminary condition for a fair public arena but also the criterion that should orient the public discussions occurring in it. Eco stated that we need shared rules to negotiate meanings and to link them in propositions to be debated. Statistical evidence is one of these rules but not the only one. “Is statistical evidence valid in establishing what is the ‘right’ meaning of the Iliad, or whether Bo Derek is more desirable than Sigourney Weaver? No, the rules change.”34 For his part, Rawls claims an agreement between the principles of reasoning and how to evaluate empirical data.
we allow the parties the general beliefs and forms of reasoning found in common sense, and the methods and conclusions of science, when not controversial. . . .
The second kind of political values—the values of public reason—fall under the guidelines for public inquiry and for the steps to be taken to ensure that inquiry is free and public as well as informed and reasonable. They include not only the appropriate use of the fundamental concepts of judgment, inference, and evidence, but also the virtues of reasonableness and fair-mindedness as shown in the adherence to the criteria and procedures of commonsense knowledge and to the methods and conclusions of science when not controversial. These values reflect an ideal of citizenship: our willingness to settle the fundamental political matters in ways that others as free and equal can acknowledge are reasonable and rational.35
VIII. HOW TO INTEND AND MANAGE REASONABLENESS?
For the sake of clarity, it is time to separate two different aspects of the notion of reasonableness. At a first level reasonableness is an attribute of Eco’s theory: it consists in the consciousness of the limit (the limit of what we can study; the limit of our forecasting capability) and in a knowledge style that is more inclined to compose thesis and antithesis than to bring an idea to its extreme consequences. At a second level reasonableness is a concept, an explicative instrument within Eco’s theory, and perhaps it should be further discussed.
Is reasonableness a new “black box” in interpretation theory? No, it is, rather, the public debate’s equivalent of the indeterminacy we have found in the single interpretive process (the “C space”).
In the face of the already given, we proceed by conjecture, and we do our best to get others to accept our conjectures. That is to say, we publicly compare our conjectures with what others know of the already given. It may be that this attitude does not define a “strong” thought, in the sense in which the various tribunals of Reason and Faith (more closely related than may appear at first blush) see themselves as strong. But it certainly defines a thought that continually runs up against the “forces” that oppose it. And since racing improves the breed, a conjectural thought, while it may not be strong, may not be weak either, but it will be well-tempered.36
The reasonableness of our conjectures is the first requisite to make them accepted and discussed by others. Even if we admit that this implies continual negotiations, it does not mean that the theory of meaning also has to be negotiated every time. It is the meaning about which we have to agree that is unstable; the mechanisms through which we discuss this meaning or, better, the models that we use to describe these negotiations have a certain stability; this stability fails only when the theory has to be modified, not each time we apply it. The Charles Sanders Personal software would probably say the problem is which of the possible and predictable results of a certain encyclopedia the interpretation process will bring us to; in our case, which of the possible and predictable results of the public negotiation of meaning and which of the reasonable conjectures in discussion will prevail. It is not how the system will work once we have chosen the encyclopedia, nor, perhaps, how the encyclopedia has to be chosen.
Thus semiotics can describe these processes; it cannot forecast their results but it can at least say which ones should not be possible because they would be untenable interpretations. The following points should explain how to intend, manage, or study reasonableness in the interpretation process, be it public or not.
(1) The basic case is the judgment about the legitimacy of a certain interpretation within a given encyclopedia. How can we recognize a reasonable interpretation? A good method would be to verify if that interpretation is the result of an interpretant chain whose existence is, at least partially, already documented in that culture (we do not have to decide if the interpretation is “true” but only if it is legitimate). This would mean an encyclopedia is condemned to repeat itself for ever. And what if an interpretation is not already present in the system? In that case it would be doomed to rejection and then oblivion (unless it reemerges later) or it could become stronger and stronger and, in the end, it would change the system. This leads us to the case of invention.
(2) If we have an invention,37 what we can do is try to understand how this invention originated within that encyclopedia. An example could be the iconologist investigating the formation of a new type, for instance Erwin Panofsky explaining why, between the sixteenth and the seventeenth century, we can find some representations of Judith holding a plate with the head of Holofernes on it (while the plate should be an attribute of Salome).38 Would this mean we can only produce an ex-post explanation? But we could also operate ex ante, forecasting an invention. I do not mean to exactly forecast when and how this invention will take place, but, rather to forecast the possible combinations of already existing elements. Following my example, we could find out that at a certain point in the fourteenth century some representations of the head of Saint John the Baptist on a plate began to appear alone, that is, without Salome; and the existence of this isolated and autonomous theme (the head of a man on a plate) made it easier to be absorbed by another type (that of Judith, which already included the head of a man). To explain an invention, we would need rules of cultural innovation, that is, rules describing ways in which a given culture tends to recombine elements to produce new cultural units. These rules will more probably be local (intracultural) rather than general, and I think they should be the main object of a semiotics of culture.
(3) A different case is the legitimacy of the interpretations a culture gives of the interpretations produced by another culture; for instance: what do Muslims mean by the term “jihad”? This case can be treated as a problem of translation, and it is certainly, again, a matter of the semiotics of culture.
(4) Finally, if the problem is how to face, judge, and govern the meeting and the clash between interpretations coming from different encyclopedias (a central issue in multicultural societies), that is definitely an extrasemiotic problem and it is more likely an issue for political philosophy. But I do think that the semiotic approach traced in the first three points of this list could help to solve these political problems.
IX. SAD AND REASONABLE AND FINAL
So far I have discussed reasonableness as an attribute of and a concept in Umberto Eco’s theory. But in the case of Eco it is quite impossible to isolate his theoretical activity from his narrative work. I will not even try a systematic investigation of reasonableness in his literary production, but I would like to hint at a comparison that I have always found very interesting and instructive.
As you all know, The Name of the Rose ends with the fire in the library and then in the whole abbey. All the books are destroyed. Facing this devastation, an overcome Guglielmo confesses his distrust regarding the possibility of finding order in the world. Even his cogent reasoning and his broad culture did not enable him to grasp the world; the solution of the case, too, was a matter of chance. The novel leaves us here, with an unarmed and powerless Guglielmo, no longer able to say anything; it is as if he could only wait for his end, which arrives many years later during the Black Death.
Confronting itself with the defeat of rationality (Guglielmo) against the elusive reality, reasonableness looks for a handhold, a clasp. It is in human nature to do this. Society hopes to find it in the principle of cooperation, in the confidence in shared values that make public debate and civil life possible. The individual, as a rational being, tries to grab the “marble veining” that crosses reality, tries to confront these “resistance lines” of the world to verify or modify his or her knowledge. But individuals do not only need gnoseological handholds; our attitude, our reasonableness forces us to look for a meaning, a base, in each aspect of our existence. The contrast has always struck me between Guglielmo, who will still live for a long time but gives in to the defeat of reason, and Casaubon, who—as far as we know and he knows—is waiting for his death and calmly seizes on the memory of a sensation (walking barefoot, with calloused heels, among Langhe vineyards) or on the contemplation of the hill; thinking “It’s so beautiful.”39
It is a semiotic rule not to ask for confirmation from the author, but I hope this can be considered as a legitimate interpretation.
PIERO POLIDORO
LUMSA UNIVERSITY, ROME
DECEMBER 2015
NOTES
I wish to thank Patrizia Violi and Stefano Semplici for their kind suggestions in the preparatory phase of this work.
1. Ernst Gombrich, “The Dream of Reason: Symbolism in the French Revolution,” British Journal of Eighteenth Century Studies 2, no. 3 (1979): 187–205.
2. Max Weber, The Protestant Ethic and the Spirit of Capitalism, trans. Talcott Parsons (New York: Charles Scribner’s Sons, 1958), 181.
3. Max Horkheimer and Theodor W. Adorno, Dialectic of Enlightenment, trans. Edmund Jephcott (Stanford: Stanford University Press, 2002), 1.
4. Umberto Eco, “On the Crisis of the Crisis of Reason,” trans. William Weaver, in Eco, Travels in Hyperreality (New York: Harcourt Brace Jovanovich, 1986). The article was originally published in the Italian journal Alfabeta on January 9, 1980. The reviewed book was Aldo Gargani, ed., La crisi della ragione (Turin: Einaudi, 1979).
5. Ibid., 127.
6. Ibid., 128.
7. Ibid.
8. Ibid., 126. On the opposition between reason and reasonableness see also Umberto Eco, “Antiporphyry,” in Gianni Vattimo and Pier Aldo Rovatti, eds., Weak Thought, trans. Peter Carravetta (Albany: State University of New York Press, 2012), 75–100.
9. The accusation of extreme idealism occurred above all during the so-called “debate on iconism,” which I briefly discuss later in this paragraph.
10. Umberto Eco, The Name of the Rose, trans. William Weaver (New York: Harcourt Brace Jovanovich, 1983). Nihilism as a consequence of nominalism is one of the main accusations some reviewers made of Eco after the publication of The Name of the Rose. Some of these critics have later been published in Renato Giovannoli, ed., Saggi su Il nome della rosa (Essays on The Name of the Rose) (Milan: Bompiani, 1985). See also Guido Sommavilla, “L’allegro nominalismo nichilistico di Umberto Eco” (The merry nihilist nominalism of Umberto Eco), La Civiltà Cattolica 3156, no. 132 (September 1981): 502–6.
11. Umberto Eco, “Weak Thought versus the Limits of Interpretation,” in From the Tree to the Labyrinth, trans. Anthony Oldcorn (Cambridge, MA: Harvard University Press, 2014), 576.
12. Ibid.
13. Umberto Eco, Kant and the Platypus: Essays on Language and Cognition, trans. Alastair McEwen (New York and San Diego: Harcourt Brace & Co., 2000).
14. On the debate on iconism and Eco’s role in it, see Eco, Kant and the Platypus, ch. 6. See also Piero Polidoro, “Umberto Eco and the Problem of Iconism,” Semiotica 206, nos. 1–4 (2015): 129–60; and Polidoro, Umberto Eco e il dibattito sull’iconismo (Umberto Eco and the debate on iconism) (Rome: Aracne, 2012).
15. Umberto Eco, “On Semiotics and Immunology,” in Eli E. Sercarz et al., eds., The Semiotics of Cellular Communication in the Immune System (Berlin: Springer, 1988), 8.
16. Ibid., 9.
17. Umberto Eco, Semiotics and the Philosophy of Language (Bloomington: Indiana University Press, 1984), ch. 1.
18. An example is the polemic against the aesthetic theory of Benedetto Croce. See Umberto Eco, The Open Work, trans. Anna Cancogni (Cambridge, MA: Harvard University Press, 1989), 25–26.
19. Umberto Eco, Lector in fabula: La cooperazione interpretativa nei testi narrativi (Milan: Bompiani 1979). Part of the Italian book has been translated into English in Umberto Eco, The Role of the Reader (Bloomington: Indiana University Press, 1979). Instructional semantics has also been extensively treated in Eco, Semiotics and the Philosophy of Language, ch. 1.
20. Umberto Eco, The Limits of Interpretation (Bloomington: Indiana University Press, 1990), 265–82.
21. Eco, “On Semiotics and Immunology,” 9–10.
22. This title is a reference to the famous verses of Eugenio Montale’s poem “Non chiederci la parola” (Do not ask us the word): “Codesto solo oggi possiamo dirti, / Ciò che non siamo, ciò che non vogliamo” (“We can only tell you today, / What we are not, what we do not want”).
23. Actually, Taylor wrote about “one best method” (not “one best way”) to do things; see Frederick W. Taylor, The Principles of Scientific Management (New York: Harper & Brothers, 1919), 25.
24. Eco, “Weak Thought versus the Limits of Interpretation,” 567.
25. Ibid.
26. Eco, “On the Crisis of the Crisis of Reason,” 129–30.
27. John Rawls, A Theory of Justice (Cambridge, MA: Harvard University Press, 1971); Rawls, The Law of Peoples with “The Idea of Public Reason Revisited” (Cambridge, MA: Harvard University Press, 1999); Rawls, Justice as Fairness: A Restatement (Cambridge, MA: Harvard University Press, 2001); Rawls, Political Liberalism, expanded edition (New York: Columbia University Press, 2005).
28. Rawls, Justice as Fairness, 5.
29. H. Paul Grice, “Logic and Conversation,” in P. Cole and J. L. Morgan, eds., Syntax and Semantics—Speech Acts (New York: Academic Press, 1975), 41–58.
30. Rawls, Justice as Fairness, 6.
31. Ibid., 6–7.
32. See the definition of “rational” given in Rawls, Political Liberalism, 51: “The rational applies to how these ends and interests are adopted and affirmed, as well as to how they are given priority. It also applies to the choice of means, in which case it is guided by such familiar principles as: to adopt the most effective means to end, or to select the more probably alternative, other things equal. However, rational agents are not limited to means-end reasoning, as they may balance finite ends by their significance for their plan of life as a whole, and by how well these ends cohere with and complement one another. Nor are rational agents as such solely self-interested: that is, their interests are not always interests in benefits to themselves. Every interest is an interest of a self (agent), but not every interest is in benefits to the self that has it. Indeed, rational agents may have all kinds of affections for persons and attachments to communities and places, including love of country and of nature; and they may select and order their ends in various ways.”
33. Rawls, Justice as Fairness, 82.
34. Eco, “On the Crisis of the Crisis of Reason,” 130.
35. Rawls, Justice as Fairness, 89–92.
36. Eco, “Weak Thought versus the Limits of Interpretation,” 586.
37. See Umberto Eco, A Theory of Semiotics (Bloomington: Indiana University Press, 1976), ch. 3.2 and 3.6.7–3.6.8.
38. Erwin Panofsky, Studies in Iconology (Oxford: Oxford University Press, 1939).
39. Umberto Eco, Foucault’s Pendulum, trans. William Weaver (New York: Harcourt Brace Jovanovich, 1989).