From the Tree to the Labyrinth: Historical Studies on the Sign and Interpretation

by Umberto Eco

  At the very end of this citation Dionysius continues with an apparent contradiction: he observes that “there is nothing which lacks its own share of beauty” (ibid., p. 150), given that Scripture states that God saw everything He had made, “and behold, it was very good” (Gen. 1:31). But what we have here is a bow in the direction of that pancalistic sensibility that will pervade the entire Middle Ages. The problem is rather that at this point Dionysius introduces the idea, which will return with some frequency throughout his corpus, of naming through dissimilar similarity or inappropriate dissimilarity (see, for example, chapter 2 of The Celestial Hierarchy, trans. Luibheid, p. 138), whereby the divinity is sometimes given a lowly name: “Sometimes the images are of the lowliest kind, such as sweet-smelling ointment and corner stone, Sometimes the imagery is even derived from animals so that God is described as a lion or a panther, a leopard or a charging bear. Add to this what seems the lowliest and most incongruous of all, for the experts in things divine gave him the form of a worm” (The Celestial Hierarchy, trans. Luibheid, p. 152).31

  Concerning this point, it has frequently been understood that for Dionysius the name that best expresses the inexpressibility of the divine nature is based on an inverse analogy, according to which what is emphasized are not the similar but the opposed properties. Some occultist interpretations of these passages speak of an image of God reflected as it were on the surface of the terrestrial sea in inverted symmetry (and this would be the sense in the famous passage from Paul’s First Epistle to the Corinthians 13:12, where he says that we see now “through a glass darkly”). If this were indeed the case we might expect a theory of inverse analogy, which would go a long way toward confirming the idea of a symbolic naming that obscures in order to spur the intelligence to seek further—and we would therefore be quite close to the idea of metaphor as cognitive process. And this could be tied in with a strong suggestion from Aristotle (see Rhetoric 1405a): “since opposites are in the same class, you do what I have suggested if you say that a man who begs ‘prays,’ and a man who prays ‘begs’; for praying and begging are both varieties of asking.” And it would be quite a challenge to require a semiotics of metaphor to account for a process by which two things are substituted for each other based, not on the properties they have in common, but on the maximum tension between opposite properties (like calling the sea solid, God malevolent, the gaze of the Medusa benevolent, and so on). To tell the truth, none of the examples given by Dionysius constitutes a case of dissimilar similarity (in the above sense), but at most of audacious similarity, linking the divine and the human on the basis of “unseemly” resemblances, but resemblances nonetheless.

  The most extreme case of dissimilarity is cited in Letter Nine, which examines a passage from Psalm 78 in which God appears to get drunk. Since the image of a divinity shamelessly intoxicated is unacceptable, Dionysius engages in a prodigious example of exegetic subtlety only to conclude as follows:

  In our terminology, inebriation has the pejorative meaning of an immoderate fullness, being out of one’s mind and wits. It has a better meaning when applied to God, and this inebriation must be understood as nothing other than the measureless superabundance of good things which are in him as Cause. As for being out of one’s mind and wits, which follows drunkenness, in God’s case it must be taken to mean that incomprehensible superabundance of God by virtue of which his capacity to understand transcends any understanding or any state of being understood. He is beyond being itself. Quite simply, as “drunk,” God stands outside of all good things, being the superfullness of all these things. He surpasses all that is measureless and his abode is above and beyond all that exists (Letter Nine, trans. Luibheid, p. 287).

  A memorable example of an author clutching at allegorical straws, whereas all the Psalmist is doing is describing the wrath of God: “Then the Lord awaked as one out of sleep, and like a mighty man that shouteth by reason of wine” (Ps. 78:65). Describing God as wrathful is in keeping with the Bible’s normal process of anthropomorphization; and what we have here is actually a simile: God awakens so full of wrath as to appear to be drunk. A powerful image, that truly puts before our eyes, as Aristotle has it, the wrath of God, but which Dionysius, with his lack of interest in the mechanism of metaphor, does not see, his attention being concentrated on the subtler exercises of allegory. So that, as Augustine points out, seeing that the literal sense appears repugnant, we look in factis for a spiritual sense.

  The real problem is that Dionysius does not make a clear distinction between metaphor and allegory and tends to lump both together in the category of the symbolic. The difference between metaphor and allegory has already been made abundantly clear. What constitutes a symbol, compared with these two rhetorical techniques, is still an open question at this point in time and will remain so for centuries (see Eco 1984a: ch. 4): an image in the form of a luminous glowing mandala may be thought of as a symbol in a number of cultures, without its being either an allegory or a metaphor. After all, maybe the best way to grasp Dionysius’s hallucinated semiotics is to reconsider Goethe’s famous distinction:

  There is a great difference, whether the poet seeks the particular for the sake of the general or sees the general in the particular. From the former procedure there ensues allegory, in which the particular serves only as illustration, as example of the general. The latter procedure, however, is genuinely the nature of poetry; it expresses something particular, without thinking of the general or pointing to it. Allegory transforms the phenomenon into a concept, the concept into an image, but in such a way that the concept always remains bounded in the image, and is entirely to be kept and held in it, and to be expressed by it.

  Symbolism … transforms the phenomenon into idea, the idea into an image, and in such a way that the idea remains always infinitely active and unapproachable in the image, and even if expressed in all languages, still would remain inexpressible. (Goethe, Maxims and Reflections, Nos. 279, 1112, 1113)

  Now, we might expect Dionysius to consider allegories as didactic procedures (or procedures calculated to conceal the truth from the eyes of the profane) and symbols as epiphanies that make secret knowledge evident. The truth is that all the examples of symbolic theology provided by Dionysius have nothing whatsoever to do with a modern theory of symbols, nor do they propose an alternative. Let us consider a few examples.

  In chapter 2 of The Celestial Hierarchy (trans. Luibheid, p. 148), Dionysius affirms that the Scriptures use poetic forms to represent formless celestial intelligences. It is unclear whether by poetic forms he means allegories (in verbis) or metaphors. And in the passage previously cited in which he speaks of God being named through the lowliest creatures, such as the bear and the lion, the example Dionysius has in mind is clearly Hosea 5:12–14, where God, still angry with Israel, says that he will be unto Ephraim as a moth, and to the house of Judah as rottenness, and unto Ephraim as a lion, and as a young lion to the house of Judah. The moth and the young lion are not “symbols” of the divinity. The Bible does not say that God is a moth or a lion, but that in a certain circumstance He will behave as His children are used to seeing the moth and the lion behave. These are perfectly comprehensible similes or metaphors (in verbis naturally) to which the prophets have accustomed us. Thomas Aquinas would have said that what the biblical author intended to say literally was that God, at the height of his wrath, was not about to give his erring children any respite.

  Similarly, when in Letter Nine (trans. Luibheid, pp. 286–287) Dionysius speaks of those “occult and audacious enigmas” in which the Scriptures compare divine things to dew or honey, he is still thinking of Hosea 14:5, where the Psalmist says, “I will be as the dew unto Israel,” or Psalm 19:9–10, where he affirms that “the judgments of the Lord are … sweeter also than honey and the honeycomb.” This time God is not angry but most loving, and the metaphor makes this clear. In no sense, however, is honey a symbol of God.

  It should be obvious that these metaphors are comprehensible, b
ecause the traditional attributes of honey are its pleasant taste and sweetness, of the moth its annoying persistence, of dew its beneficial fertilizing qualities. When Dionysius is afraid that his audience may not be familiar with all the properties of the metaphorical vehicle, he lists them, as any self-respecting encyclopedist of the early centuries A.D. would have done. In The Celestial Hierarchy, for example, speaking of the symbolic presentation of fire, he points out that the Scriptures give us examples of flaming chariot wheels, fiery animals, men radiating fire, braziers of red-hot coals, rivers of flame, and he observes: “And indeed it seems to me that this imagery of fire best expresses the way in which the intelligent beings of heaven are like the Deity” (p. 183), and he proceeds to list a series of properties traditionally associated with fire. Fire passes through all things without mixing with them, it cannot be grasped but it seizes everything, it lies hidden until it finds the proper kindling, it transforms things, it vivifies them with its heat, it shuns adulteration, it tends upward, it penetrates, it moves by itself and makes other things move, it embraces everything but nothing can contain it, it is efficient, powerful, and when ignored it appears to be dead, but it springs unexpectedly to life when stirred, it flings itself upward and cannot be checked, and so on. With such an encyclopedia it is easy to produce not just metaphors but whole allegories based upon fire. Fire is not an obscure symbol that names without naming, that alludes without revealing: when intimately known in its very nature, as Dionysius shows that he knows it, it puts before our eyes the supernatural realities of which it is a metaphor or an allegory, and it does so effortlessly.

  The same can be said of light, and of the sun as the source of light, to which Dionysius devotes a number of fine pages in the Divine Names (trans. Luibheid, p. 74), pages that will inspire many medieval theorists of the aesthetics of light (see Eco 1956, 1987).

  The pages of the Divine Names in which Dionysius says that God can be called Good, Beauty, or Being belong to a different register. In this case he is not talking about earthly entities, animals, objects, natural phenomena capable of becoming images, or metaphors of divine things. Here he is talking about what the Scholastics will call the transcendental properties of Being. The problem is that we, knowing the moth from experience, can compare it to God, but we are able to say that something is good or beautiful only insofar as we are able to see that certain things in our experience participate in a reflected fashion in the properties of the divinity. “For we recognize the difference in intelligible beings between qualities that are shared and the objects which share them. We call ‘beautiful’ that which has a share in beauty, and we give the name of ‘beauty’ to that ingredient which is the cause of beauty in everything. But the ‘beautiful’ which is beyond individual being is called ‘beauty’ because of that beauty bestowed by it on all things, each in accordance with what it is” (The Divine Names, trans. Luibheid, p. 76).32

  Likewise, what is suprasubstantially Good and Beauty is “that which truly is and which gives being to everything else” (The Divine Names, trans. Luibheid, p. 98). “Every being and all the ages derive their existence from the Preexistent. All eternity and time are from Him. The Preexistent is the source and is the cause of all eternity, of time and of every kind of being” (The Divine Names, V, 5, trans. Luibheid, pp. 98–99).

  What we have here is a leap. Here the trajectory is no longer upward (from the moth to God) but downward, from God to whatever is good and beautiful. The divine names belong strictly speaking to the divinity, and only at a subordinate level to things. This subordination, however, is not of a metaphorical order, but of a metaphysical one. If the properties of the moth are similar to those of God, it is because of a defect of our imagination. This is the only way can imagine the implacability of God’s wrath (which is obviously something quite different). The simile is couched in verbis, and the verba are clearly inadequate to express an object so sublime. Therefore the metaphor from low to high appears capable of making us know, by putting the thing before our eyes; but it makes us know in an extremely pallid fashion what is by definition unknowable. The properties of beautiful things on the other hand are what they are because they participate in the beauty of the divinity. The similitude is not in verbis but in re. The sharing of transcendental properties by creatures is always a pallid sharing, but it is not a pallor of the imagination (or of language); instead the pallor is ontological.

  This is tantamount to saying that in the symbolic theology of Dionysius there is no room for a coherent theory of metaphor, and so be it. But this position implies a fine cognitive dilemma. In fact we have it on faith that God is Goodness and Beauty, but in what precise way He suprasubstantially possesses these properties we do not know. Or rather, either we know it by illumination or arcane knowledge, or we must imagine it in a pallid fashion taking the properties of things as our point of departure. A problem of which Thomas Aquinas (who is not a cultivator of any hidden or mysteriosophic science of the divinity) is fully aware when, from these very same pages of Dionysius, he derives the idea of knowledge by analogy: somehow or other, “prout possumus,” to the best of our abilities, we must elevate ourselves from earthly things to knowledge of the First Cause (Expositio Sancti Thomae V, 3, n. 668). Are we justified in saying, then, that such knowledge is merely metaphorical?

  3.7. The Analogia Entis

  Rosier-Catach (1997: 167–173) cites a number of cases in which the canonical example of prata rident serves to highlight the difference between metaphor and translatio in divinis. Boethius (De Trinitate IV, 1, 5, 21) had already remarked that when predications had to do with God, the things predicated are thereby modified. Gilbert of Poitiers (Dialogus Everardi et Ratii) will say, apropos of the ten categories of Aristotle (praedicamenta), that “si quis ad divinam verterit praedicationem, cuncta [praedicamenta] mutantur” (“if one proceeds to the predication of divine things, all the [categories] change”). Theodoric of Chartres follows the dictum of Dionysius, according to which a substantial predicate does not mean that God is a substance, but that he is beyond all substance.

  So, in the case of predication in divinis, it is not the thing that is predicated, only the name. At the same time, the idea makes headway that, despite this difference, the predicate “quodam modo innuit nobis substantiam” (“in a certain way suggests the substance to us”). As a result, what we have is not an unbridgeable divarication, and predication by pure negation, but instead some form of connotation.

  To what extent the difference between univocal predication and predication in divinis posed an insurmountable problem is confirmed by the Regulae theologicae of Alain of Lille, in which a distinction is made (somewhat obscurely) between: (i) the transfer of the name and the thing, as in “linea est longa,” where length, which is the property of the body, is said of the line that distinguishes it and makes it possible to call it “long”; (ii) the transfer of the thing, as in “seges est laeta,” where the thing (in this case laetitia or gladness) is attributed to the subject, the cornfield; (iii) the transfer of the name alone, as in “monachus est albus”—in which the white monk is not himself white (we are talking about a white-robed Cistercian). But this is precisely the way we say “Deus est iustus.”

  Later in the text, however, Alain admits that God is called just “a causa quia efficit iustum” (“rightly, since he brings about justice”). Here we are close to the position taken by Dionysius, for whom Goodness and Beauty really are divine properties and may be applied to earthly things only insofar as, through participation, they cause something very similar in them. In that case it is not simply a question of transferring the name: indeed it would not even be a metaphor (judging by the above-cited classification).

  Since this is not the place to venture into the boundless territory of the discussions on the analogia entis (pointing out the frequently subtle differences between one author and another, right down to the Second Scholasticism of the Counter-Reformation, from Cajetanus to Suarez, we will simply attempt to see what
were the basic models for univocal and equivocal discourse that inspired the whole of Scholastic debate. And the fundamental model is always the one derived from Aristotle (see Owens 1951) and from Boethius’s commentary.

  The discourse on equivocity is already present in the Metaphysics, where Aristotle discusses how being can be “said in many ways.” After saying that there is a science that considers being in and of itself, when we might have expected his first tentative definition of the object of this science, he repeats as the only possible definition what had appeared in his first book (992b 18) only as a parenthetical observation: “being is said in many ways” (“to de on leghetai men pollachos”)—according to multiple meanings (1003a 33).

  In fact, Aristotle reduces these many ways to four. Being is said: (i) as accidental being (this is the being predicated by the copula, whereby we say that a man is white or standing); (ii) as true—it may be true or false that the man is white, or that man is an animal; (iii) as potentiality and act, whereby, while it may not be true that this healthy man is ill right now, he could become ill, and (as we might say today) we can think of a possible world in which it is true that this man is ill; and finally (iv) as ens per se or as substance. However we speak of being, we say it “with reference to a single principle” (1003b 5–6), that is, to the substances: “The first meaning of being is the essence that signifies the substance (semainei ten ousian)” (1028a 14–15).

  Is this saying in a number of ways an equivocal way of saying? Aristotle is unclear on this point. In the Categories (1, 1a) he says that we have homonymy or equivocity when entities that require a different definition have a single name in common. The classical example is zoon, used both for an animal and a painting, a homonymy that exists in Greek. It should be said that medieval thinkers, who did not know Greek, failed to grasp this homonymy, thinking that the word animal was used both for the animal and the image of the animal, and that Aristotle gave a broader meaning to equivocity than they did. See, for example, Thomas Aquinas: “Philosophus largo modo accipit aequivoca, secundum quod includant in se analoga” (“The philosopher takes equivocal terms in a broad sense, so they include analogous terms”) (Summa Theologiae I, 13, 10 ad 4).