by Umberto Eco
1.6. The New Encyclopedic Models
Whether or not this was the unconscious model for a new ideal of encyclopedic knowledge, it must be said that the first real efforts at creating semantic representations in encyclopedic form did not get underway until the second half of the twentieth century and only after a fierce debate regarding the shortcomings of any dictionary representation.33
Clearly, although the idea of the encyclopedia as postulate and ideal model is infinite, all that could be attempted were limited and local representations, which however did not exclude the possibility of their progressive and potentially limitless enrichment.
The new encyclopedic models assumed a number of formats, among them:
(i) Matrices representing the presence or absence of traits chosen ad hoc to account for the differences among items belonging to the same semantic subset, such as chair, armchair, sofa, etc. (cf. Pottier 1965).
(ii) Contextual selection models (specifying the various meanings a given lexeme may take on in different contexts) (cf. Eco 1975, 1984a).
(iii) Models by Cases that include Agents, Objects, Instruments, Purposes (the verb “to accuse,” for example, is defined as an action in which a human Agent communicates to a human Object by means of a verbal Instrument with the Purpose of revealing to him that the action of another human Object is evil; whereas “to criticize” is explained as the action of a human Agent who by means of a verbal Instrument speaks to a human Object with the Purpose of demonstrating that the action of another human Object is open to censure; or else the verb “to kill” is analyzed as the action of a human Agent which causes a change of state, from living to dead, of an animated X—further specifying, by the use of the English verb “to assassinate,” that the X in question must be a political figure) (cf. Fillmore 1968, 1969, 1977).
(iv) Representations that take into account, in the case, for instance, of a term like “water,” the properties that determine its extension or its referent (its being H2O); labels of a quasi-dictionary variety, such as being Natural and Liquid; as well as stereotypical notions like Colorless, Transparent, Tasteless, Odorless, Thirst-Quenching (cf. Putnam 1975, 12).
(v) Representations that take into account all possible properties of a term and specify, for a chemical element for example, odor, color, natural state, atomic number, effects, history, etc. (cf. Neubauer and Petöfi 1981).
None of these proposals, however, had had recourse to network structures. It is in the field of artificial intelligence that frame-, script- or scenario-type representations appear, registering each stage of a sequence of typical events (for instance, what does “going out to a restaurant” mean: entering, sitting down at a table, ordering from the menu, eating, requesting the bill, etc.)—all models that have proved successful in the field of artificial intelligence, where, in order for a computer to understand a text and draw conclusions from it, it must first be provided with all of the competences with which (even without their being aware of it) the average human being is endowed (cf. Schank and Abelson 1977; Schank and Childers 1984).
But it is with Quillian (1968) that the notion of a semantic network, structured as a labyrinth of interconnected nodes, first appears. To simplify things, all we have to do is take another look at Figure 1.17. Any node can be taken as the point of departure or type of a series of other nodes (tokens) that define it (let’s say the point of departure is dog and that this node is defined by its links with animal, quadruped, able to bark, faithful, etc.). Each of the defining terms may in its turn become the type of another series of tokens. For instance, animal could be exemplified by dog, but also by cat, and would include quadruped but also biped; or, if a node cat were to be identified, it would be defined by a number of nodes it shared with the definition of dog, such as animal and quadruped, but it would also refer to nodes like feline, which it shares with tiger, and so on.
A network model implies the definition of every concept (represented by a term) through its interconnection with the universe of all the concepts that interpret it, each of them ready to become the concept interpreted by all the others.
If we were to expand the network of linked nodes ad infinitum, from a concept assumed as type it would be possible to retrace, from the center to the outermost periphery, the entire universe of the other concepts, each of which may in its turn become the center, thereby generating infinite peripheries.
Such a model is also susceptible of a two-dimensional graphic configuration when we examine a local portion of it (and in a computer simulation, in which the number of tokens chosen is limited, it is possible to give it a describable structure). But it is not in fact possible to represent it in all its complexity. It would have to be shown as a kind of polydimensional network, endowed with topological properties, in which the paths become longer or shorter, and every term gains in proximity with the others, by way of shortcuts and immediate contacts, while remaining at the same time linked to all the others according to historically mutable relationships.
It has been said that, if we assume a maximal notion of competence about the world, the meaning of a term would then consist of all the true propositions in which it has appeared or could appear. In fact, this would presuppose the ideal model of the encyclopedia. But in scientific practice and the way in which, in our daily lives, we try to make sense of sentences, we do not make a global appeal to the encyclopedia for every sentence, and it is the content that selects the local zones of competence that must be activated. Two flexible criteria may be assumed: (i) information is potentially part of the average encyclopedic competence if it can be supposed to be sufficiently shared by a collectivity (which may also be a “regional” collectivity—in this sense the definition of neutrino would form part only of the regional competence of a community of nuclear physicists—see the concept of Specialized Encyclopedia discussed below in section 1.9); (ii) the format of the network to be activated is prescribed by the contexts and the circumstances of the proposition (accordingly, if someone uses the word torus in speaking of topology a network is constituted which is concerned with mathematical objects, and all concepts regarding the fields of architecture, anatomy, and botany are excluded).
While in an ideal encyclopedia there are no differences between necessary and contingent properties, it must be admitted that, within a specific culture, certain properties appear to be more resistant to negation than others, on account of the fact that they are more salient: it could feasibly be denied, for instance, in the light of a new system of classification, that a sheep is ovine, or again this particular trait might not be deemed necessary to the understanding of the term sheep in the sentence: “the sheep was bleating in the field.” There can be no doubt, however, that it is hard to deny that a sheep is an animal—and the characteristic also remains implicit for the comprehension of the example we just cited. It has also been observed (Violi 1997: sect. 2.2.2.3) that some traits seem to be more resistant than others, and that these uncancelable traits are not only categorical labels such as ANIMAL or PHYSICAL OBJECT. In the life of semiosis we realize that we are also reluctant to cancel some “factual” properties that appear more salient and characteristic than others.
To explain why certain properties appear more resistant than others, Violi (1997: sect. 7.2) distinguishes between essential and typical properties: it is essential that a cat be an animal; it is typical that it meows. The second property can be canceled, but not the first. But if this were to be the case we would be back again to the same old difference between dictionary and encyclopedic properties. Violi (1997: sect. 7.3.1.3) instead considers properties that are functional and certainly encyclopedic in nature to be similarly uncancelable: hence it is difficult to say of something that it is a box and at the same time deny that it can contain objects (if it couldn’t it would be a fake box).
Often, however, in order to construct and presuppose a local portion of encyclopedia needed for the comprehension of a determined context, we must resort to simplified local represen
tations that set aside many properties that are otherwise (in other contexts) resistant.
In Eco (1984b: sect. 2.3.4) I gave the example of a dialogue between a wife and her husband at midnight in a suburban home. The wife looks out the window and says with a preoccupied air, “Honey, there’s a man in the garden.” The husband takes a look and says, “No, honey, that’s not a man.” The husband’s reaction certainly violates a pragmatic rule because it provides less information than the situation calls for, since denying the presence of a man could on the one hand suggest that what is there is a child or a cat, while on the other hand it could also lead his wife to imagine something more dangerous (why not an invader from outer space?).
In this context, when she is afraid there may be a man there, the wife surely does not assign to the term the properties of rationality, bipedality, or the ability to laugh—all properties that in that context are narcotized (cf. Eco 1979a: ch. 5) and considered irrelevant, but instead those of a living being, capable of movement and aggression and therefore potentially—at night and in someone else’s garden—dangerous. Because it is also part and parcel of the infinite encyclopedic properties of man to be prone to take up a life of crime (don’t we all know that homo homini lupus, man is a wolf to men?) The husband ought then to adjust his iteration on the basis of a local encyclopedic representation, as in Figure 1.18, one that he conjecturally considers shared (given the circumstances) by his wife.
Figure 1.18
If the husband wishes to calm his wife down he must either exclude immediately the property of mobility (by saying, for example, that what she saw was the shadow of a tree) or deny any suggestion of properties suggesting dangerousness (in which case he might say that it wasn’t a man but a stray dog).
The ad hoc construction of a local portion of encyclopedia, that organizes only the properties pertinent to the context, is the only strategy that will allow the husband to interact in a reasonable way with his concerned wife.
1.7. The “Ontologies”
In the most recent research on artificial intelligence and the cognitive sciences, the notion of semantic networks has given rise to a theory of ontologies. Despite the inappropriate use of a concept like “ontology,” which has quite a different meaning in philosophy, the term is used in this context to refer to the categorical organization of a portion of universe that may take the form of any kind of classificatory tree or semantic network. In this sense, the husband in the example just considered could (without being aware of it, like Molière’s Monsieur Jourdain who spoke in prose without being aware of it) have constructed an ontology.
According to one recent definition, an ontology is “a specification of a representational vocabulary for a shared domain of discourse—definitions of classes, relations, functions, and other objects” (Gruber 1993: 199). The definition is very broad and can be adapted both to a complex semantic network and to a mere classification. In fact, in much of the literature on ontologies, the starting point is the model of the Arbor Porphyriana, used to exemplify the most common semantic relationship, that is, that of subsumption (see, for instance, Sowa 1991, 2000). If this seems disappointing, it occurs because the producers of ontologies are responding as a rule to practical needs (even supplying a business firm with a satisfactory organization of its data and products) and sometimes a tree structure can serve that purpose.
There are ontologies in the form “part-of,” in which, for example, the meaning of car is analyzed, representing its various components and functions (see, for instance, Barsalou 1992: 30). From the theoretical point of view, they do not go much further than the representations already present in various versions of encyclopedia semantics—except that a particular representational structure is devised to give instructions to a computer. Other times graphs are designed in which each node has not just one single superordinate but allowance is made for multiple heredities and a node may derive properties from each single superordinate node or from all of them.
There is some discussion as to whether the ontologies should be adequatist, that is, maximal, or reductionalist, that is, referred to a single universe of discourse. It is usually conceded that the domain of an ontology should not be complete but simply cover the area of interest that produces it. In the vast literature on the subject the ontologies are sometimes no more than ingenuous diagrams designed to illustrate perfectly intuitive links and differences, classifications in the most traditional sense of the word—like those used in the natural sciences since Linnaeus—or mere shorthand notes or mnemonic devices. Even apart from the nonchalance with which the word “ontology” is used to indicate (and to sell) representations so dissimilar in scope and purpose, the variety of the models suggests that, if they really reflect states and structures of the mind, this would mean that our brain articulates its competence through different data-organization models depending on the problem to be resolved or committed to memory.
The aspects that make ontologies interesting are highlighted by Smith (2003): (i) they do not pretend to be representations of the world but of our modes of conceptualization in given domains—sometimes representing also commonsense knowledge; (ii) therefore an ontology has nothing to do with questions of ontological realism and is a purely pragmatic undertaking; (iii) the entities present in an ontology possess only the properties represented in that structure (we would add that the others are narcotized because they are irrelevant to the specific universe of discourse). As Smith remarks, it is as if Hamlet, whose hair is not mentioned in Shakespeare’s tragedy, was neither bald nor nonbald, but instead was a man without properties as far as his hair was concerned.
In this sense, an ontology, however clumsy and ingenuous it may be, is the local representation of a portion of encyclopedic knowledge relevant for the purposes of a given universe of discourse.
1.8. Ontologies and Semiosic Creativity
It appears that, if we are to understand a text or the meaning of a word, we need an underlying ontology, as is shown in the example illustrated in Figure 1.18. In the same way, it seems obvious that if the encyclopedia, as Leibniz already opined, is a body subject to a constant process of renewal and expansion, many expressions produced in the context of a given culture can contribute toward changing the current encyclopedia. The contentions of Copernicus, for instance, and later those of Galileo and Kepler noticeably modified the encyclopedia of the modern world (which from that moment on did not stop citing the theories of Ptolemy but placed an asterisk in front of them to show they were mistaken).
But, alongside these cases of scientific innovation, or transformation of common sense, there are cases of artistic creativity in which a new text requires—if it is to be understood in all its innovative aspects—that our encyclopedia be modified.
1.8.1. Metaphor as a Tool for Producing New Ontologies
In this historical rereading of the adventures of the encyclopedia we must once again return to Aristotle, to consider an aspect of his thought that has apparently nothing to do with the history of definitions, either dictionary or encyclopedic. We are talking about his theory of metaphor.
What makes Aristotle’s theory of metaphor interesting for us today is not simply the fact that it is the first rigorous discussion of this trope, but above all the fact that this first theorization of metaphor does not consider it as a mere ornament of discourse but assigns it a cognitive function.34
The key suggestion in the Poetics is to be found in 1459a 8, where the author declares that metaphor is the best of all the tropes because understanding metaphor means “knowing how to recognize similarity” or “the related concept.” The verb he uses is theorein, which means to perceive, to investigate, to compare, to judge. It is then clearly a verbum cognoscendi. Aristotle gives examples of banal metaphors, such as those from genus to species (there lies my ship) or from species to genus (Verily ten thousand noble deeds hath Odysseus wrought), but he already lists metaphors that are more interesting poetically when he speaks of the metaphor from species
to species (with blade of bronze drew away the life). As for the metaphor by analogy he appears to be listing expressions that are already quite solidly codified such as the shield of Dionysus (god of wine) for the cup and the cup of Mars (god of war) for the shield or the evening as the old age of the day. But he identifies an effective and original poetic expression in sowing the god-created light, said of the sun, perhaps by Pindar, and he likewise appreciates a quasi-riddle like a man I saw who on another man had glued the bronze by aid of fire, said of the suction cup or cupping glass. These are cases in which the poetic invention leads us to investigate the similarity, suggested, but not immediately evident.
The relevant passages in the third book of the Rhetoric are far more in number. What arouses wonder (to thaumaston) is pleasing; metaphor manifests itself (phainesthai) when we examine (skopein) a possible correspondence or analogy. The talent for metaphor is not something that can be learned from others, and therefore it is not a matter of mere imitation but of invention. The examples he gives of analogy are not in the least banal, as in the famous example (1405a) in which pirates refer to themselves as “purveyors.” The rhetorical move is persuasive because it insinuates that the plunderer and the merchant share a characteristic in common, since both of them facilitate the transfer of goods from a source to the consumer. The identification of the characteristic they share (in addition to being brazen) is daring, because other discordant characteristics, such as the opposition between a peaceful means and a violent one, are narcotized, but it is undeniably ingenious and provokes surprise, encouraging us to reconsider the role of the pirate in the economy of the Mediterranean.