Fairy-Tale Symbolism: An Overview

Publisher's copyright statement: This is the peer reviewed version of the following article: Vaz da Silva, F. (2017). Fairy-tale symbolism: an overview. In Oxford Research Encyclopedia of Literature.: Oxford University Press (OUP)., which has been published in final form at https://dx.doi.org/10.1093/acrefore/9780190201098.013.79. This article may be used for noncommercial purposes in accordance with the Publisher's Terms and Conditions for self-archiving.

of "incredible and impossible" matter in Märchen and myth must "possess some raison d'être." 2 This essay summarizes the main issues, clarifies some misunderstandings, and proposes a way forward in the study of fairy-tale symbolism.
One way to understand the otherness of fairy tales is to assume they are literally true to the reality of other times and places. Such was the prevailing understanding in the nineteenth century. The assumption that fairy tales are the narrative survival of customs and beliefs from other times and places has inspired two sorts of explanatory models. On the one hand, the evolutionist hypothesis stresses survival in time. As the British folklorist Andrew Lang put it, folklorists find in "proverbs and riddles, and nursery tales and superstitions … the relics of a stage of thought, which is dying out in Europe, but which still exists in many parts of the world." 3 Specifically, fairy-tale imagery-being rife with magic, cannibalistic episodes, and a general lack of distinction between animals and humans-bespeaks "an age of savage fancy." 4 On the other hand, the diffusionist persuasion stresses resilience in space. The notion that tales were invented only once and then travelled, while carrying the cultural mark of their place of origin, had a worthy exponent in Emmanuel Cosquin. This French folklorist embraced the view that fairy tales originated in India, and he repeatedly argued that fairy-tale motifs match Hindu representations. 5 Alas, it is hard to envision individuals bothering to learn something and pass it on if it means nothing to them. 6 Cosquin thought nothing of branding a chain of metamorphoses in a French text an "ultra-bizarre ending" and then, one step ahead, professing that "such an Indian ending" is a true "Made in India" mark of origin. 7 This perilous line of argument raises various issues, but focus on the main question: Why, if India's conceptions were so unique, would Indian tales have been borrowed and nurtured by people who found them "bizarre"? A related problem plagues the evolutionist model. Lang acknowledges that some tale "forms are fitter than others, survive more powerfully, and are more widely spread." 8 This notion of differential fitness supposes a process of cultural selection that would discard any meaningless contents while preserving those themes that make sense to the taletellers and their audiences. This is a sensible assumption. 9 But, crucially, it contradicts the notion that the "savage fancy" of bygone eras might survive in the modern fairy tale.
In short, the assumption that fairy tales might carry on alien cultural traits fails to explain why those traits should have survived at all. Alternatively, the premise of symbolism assumes that the bizarre elements in fairy tales are (somehow) relevant to the taletellers and their audiences. Models that address fairy-tale contents as the symbolic expression of notions that are relevant to individuals and communities have no trouble explaining why fairy tales persist in tradition. This essay concerns such models.

Fixed Symbols: Models with a Key
Although all models of symbolism agree that the contents of fairy tales are not to be taken literally, they disagree on what symbols may be. Typically, models built on nineteenthcentury assumption assume that symbols are the survivals of archaic metaphors. This assumption implies that symbols convey fossilized meanings, harking back to a primordial time, which are unintelligible at the present time. As you might predict, competing theories about the origin of the symbols offer alternative propositions about their meanings. Those propositions tend to be contentious. As Richard Dorson noted, "the celestial mythologists wrangled over the primacy of sun, storms, and stars," whereas "the psychoanalytical mythologists dispute over the symbols from the unconscious. 10 Still, the theories of Max Müller and Sigmund Freud provide valuable clues on symbolic elucidation from the perspective of survival.

A Disease of Language: F. M. Müller
In an influential 1856 essay named "Comparative Mythology," Friedrich Max Müller presented a sweeping vision of mythology and fairy tales. Before the split of the Indo-European language family into distinct branches, Müller submitted, language lost its "etymological conscience." As the original meanings of word roots were gradually forgotten, new meanings were attached to the words, "by a kind of etymological instinct," in mythical fables. 11 In other words, myths replaced forgotten etymologies. As Müller put it later, mythology comes from an "infantile disease" of language, namely, "self-forgetfulness." 12 Conversely, discovering the root of the word at the core of each myth should yield, "behind the floating clouds of the dawn of thought and language, that real nature which mythology has so long veiled and disguised." 13 For this purpose, each relevant word is to be compared to its variants in other branches of the Indo-European family. Usually the Sanskrit variant yields insight as to the root of the word (for, although "Sanskrit is not the mother of Latin and Greek … it is no doubt the eldest [sister] in so far as it has preserved its words in their most primitive state"), 14 and typically the root in question expresses solar phenomena. Given that the "divine myth becomes an heroic legend, and the heroic legend fades away into a nursery tale," solar myths are the "seeds" of the modern fairy tales. 15 Thus, Müller finds it "extremely probable" that the motif of "Red Riding Hood being swallowed by the wolf and cut out again" comes from solar mythology. 16 Candidly, Müller acknowledged "the monotonous character" of his own philological readings: "'Is everything the Dawn? Is everything the Sun?' This question I have asked myself many times before it was addressed to me by others." But his unswerving answer was, "my own researches lead me again and again to the dawn and the sun as the chief burden of the myths of the Aryan race." 17 Eventually, a parody mocking Müller's line of thought to conclude that Max Müller himself was a solar myth was well received, and the French translator added the significant title: "After All, Mr. Max Müller Never Existed." The sun of solar mythology was setting. 18 With the wane of solar mythology, an aspect of Müller's contribution lingers in undeserved obscurity. Müller recalled John Locke's point that "all words expressive of immaterial conceptions are derived by metaphor from words expressive of sensible ideas," and he added that metaphor is among "the most powerful engines" in the construction of languages. 19 Besides emphasizing the importance of metaphor in the ancient Indo-European tongue, Müller also made the point that metaphorical processes underlie myth and Märchen as well as every intellectual movement, "even in our own [age], though perhaps the least given to metaphor, poetry, and mythology." 20 Crucially, his point that myths arise whenever the awareness of metaphor is dimmed implies that myths (and Märchen) hinge on unselfconscious metaphors. This is a fundamental insight.
Because Müller saw myths and fairy tales as the survivals of ancient metaphors, he never thought of exploring the relevance of metaphoric patterns for contemporary storytelling.
Since he defined myths (and Märchen) as obscured primal metaphors, Müller assumed that the philologist must step in and clear up the "mythological misunderstanding" by reestablishing the correct etymological derivations. 21 Yet, tantalizingly, Müller was aware of the recurrence of metaphor on human thought, ancient and modern. 22 His own stated purpose of finding solar phenomena "behind the floating clouds of the dawn of thought and language" uses the solar metaphor to talk about solar metaphors. 23 Willy-nilly, Müller showed that living metaphor pervades storytelling (even the academic sort) up to his own time.

Symbols in Dreams and Folklore: S. Freud
Sigmund Freud proposed another variation on the notion that symbols are the fossilized survivals of archaic metaphors. His interest in symbols stemmed from the realization that some "'mute' dream-elements" resist the technique of dream interpretation by means of free associations. 24 Freud assumed that these elements cannot be illuminated by the dreamer's free associations because they are endowed "with a permanently fixed meaning." 25 Otherwise put, the symbolic relation between these dream-elements and the "'genuine' thing behind them" is constant. 26 On this assumption, Freud compiled a list of symbolic "stable translations," which he learned "from very different sources-from fairy tales and myths… from folklore… and from poetic and colloquial linguistic usage." 27 As Freud puzzled over why a symbol would allow people to "make use of it without knowing about it," 28 he speculated that the "ultimate meaning of the symbolic relation … is of a genetic character. Things that are symbolically connected today were probably united in prehistoric times by conceptual and linguistic identity." 29 In other words, symbols are fixed because they are the "residues" of a "hypothetical 'primal language.'" 30 And the likely reason why most of them are "sexual symbols"-which makes Freud's interpretations admittedly "very monotonous," but what can he "do about it?" 31 -is that primal words "originally had a sexual significance," but "later acquired another application." 32 Ultimately, Freud falls back on the idea of primordial metaphors to justify his understanding of symbols. Originally, he muses, "all primal words referred to sexual things but afterwards lost their original meaning through being applied to other things and activities which were compared with the sexual ones." 33 Ultimately, Freud shares with Müller a fixed reading of symbols based on the hypothesis of primal word roots and metaphorical derivation. Based on their respective definitions of the primal metaphors, each author offers a confessedly monotonous-sexual, or solar-reading of symbols. But while Müller chose to dispel the metaphors and recover the etymological roots, Freud decided to collect the contemporary survivals of primal metaphors. He set dream symbolism within the wider field of "unconscious ideation" among the folk, as found in folklore and in linguistic idioms, and proceeded to compile a list of the surviving symbolic translations. 34 Although he was a self-professed "amateur" in mythology, anthropology, philology, and folklore, 35 the corpus of fixed symbols Freud gleaned from these fields has set a template for symbolic interpretations. 36 How far this template (and its inherent view of symbolism) can illuminate fairy-tale symbolism remains a contentious issue.

Freud's Followers
Freud's hypothesis of a primal language, of which symbols are the survivals, inspired Erich Fromm's proposition that "symbolic language is a language in its own right, in fact, the only universal language the human race ever developed." Fromm adds that the understanding of symbolic language "should be taught in our high schools and colleges just as other 'foreign languages' are part of their curriculum." Thus, Fromm expands Freud's view of fixed symbols into the notion that symbolism is a conventional language. Fromm adds that the problem is "one of understanding [symbolic language] rather than of interpreting as if one dealt with an artificially manufactured secret code." 37 Yet, Fromm acknowledges internecine strife on what the meanings of symbols may be.
Each psychoanalytic school, he allows, lays exclusive claims "to the only true understanding of symbolic language," and "we lose sight of the many-sidedness of symbolic language and try to force it into the Procrustean bed of one, and only one, kind of meaning." 38 Fromm himself is a case in point. Memorably, he spars with Freud as he chooses to interpret "Little Red Riding Hood" in light of "the male-female conflict" rather than through an oedipal lens. 39 Such disagreements are fairly common. Bruno Bettelheim, in his best-selling book on the meanings of enchantment, brings back an oedipal focus to "Little Red Riding Hood." This tale, Bettelheim thinks, is about the girl's budding sexuality and her unconscious desire to get rid of grandmother (i. e., her mother) in order to be seduced by the wolf (i. e., her father). 40 This particular interpretation supposes that Little Red Riding Hood "has outgrown her oral fixation, no longer has any destructive oral desires." 41 However, fellow psychoanalyst Géza Róheim draws attention to an element of oral aggression perpetrated by the "cannibal child" in oral variants of the tale. 42 So much, then, for the presumption of understanding rather than interpreting the symbolic language of fairy tales. Nor does the plurality of Freudian interpretations foster confidence in Freud's tenet of the fixity of symbols.
Quite apart from the Freudian tenet, it is noteworthy that the possibility of cherry-picking tale variants facilitates the plurality of interpretations. As historian Robert Darnton noted, both Fromm and Bettelheim focus on the Grimm variant of "Little Red Riding Hood" and thus gloss over important motifs found in the oral variants, such as "the cannibalizing of grandmother and the strip-tease prelude to the devouring of the girl." 43 Darnton decried Fromm's "uncanny sensitivity to detail that did not exist in the original folktale," and he added: "the moral of this story should be: beware of psychoanalysts-and be careful in your use of sources." 44 This remark on the use of sources is relevant. Notably, it applies to Darnton himself. The fact that he lambasts Fromm's unfaithfulness to the original folktale suggests that Darnton thinks he knows what the urtext of "Little Red Riding Hood" is like. Indeed, he argues that an oral variant collected around 1885 represents "the tale more or less as it was told around firesides in peasant cottages during long winter evenings in eighteenth-century France." Darnton makes the case that the peasants of eighteenth-century France had lives "nasty, brutish, and short" and their folktales feature a matching worldview. 45 Quite remarkably, Darnton excises the happy end in his chosen source to suit his grim argument. 46 Although Darnton castigates Bettelheim for reading fairy tales "as if they had no history," 47 Darnton himself addresses tales as though he was no historian. (Darnton reduces the variety of folktales to the fixity of a single text, 48 declares that a nineteenth-century text represents an eighteenth-century tradition, and maims his chosen source, all of which is striking behavior for an historian.) In fact, Darnton joins Bettelheim and Fromm in the act of cherry-picking a tale variant that agrees with his predefined template for interpretation. So the contrast between the historian and the psychoanalysts is a moot point here. The relevant point is that all three authors feel free to apply a preset interpretive grid to a tale variant that suits it, and to dismiss all the variants that might disprove it.
This methodological choice raises the issue of what to do with all the unheeded tale variants. It is clear that the practice of selecting one variant over the others befits projects bent on reading meanings into tales. Conversely, abstaining from cherry-picking variants should be a condition for apprehending meanings from the tales. Understanding a fairy tale involves discovering the commonalities between all tale variants-it requires a comparative endeavor that might capture the inner workings of the tale-a principle that applies to psychoanalysts and non-psychoanalysts alike.
In short, the comparison of tale variants helps to foreground the symbolic universe of the tale. This open-ended sort of approach contrasts with the one-sidedness of psychological models that transpose the unfamiliar imagery of myth and fairy tale into (as Nadia Sels put it) "the safe register of an already familiar truth." 49 And yet, the art of metaphorically transposing whole stories-variants and all-into preset narrative templates annihilates most comparative insights. The Jungian canon is a case in point.

Jungian Interpretations
Carl Gustav Jung proposed that the structural elements of the collective unconscious, which he called archetypes, are expressed in myriad images he called symbols. The archetype itself is that which is symbolized, and each symbol captures an aspect of it. 50 These principles yield a comparative approach to fairy tales. Marie-Louise von Franz explains that, before you can understand a symbol in a fairy tale, you have to amplify it. "Amplification means enlarging through collecting a quantity of parallels." Thus you get "to know the comparative anatomy of all the symbols" in the tale, and you can proceed to "construct the context." 51 Then comes "the last essential step, which is the interpretation itself-the task of translating the amplified story into psychological language"-and von Franz is adamant that "we must use strictly psychological language. Only then do we know what the interpretation is." 52 Note that interpretation does not stem from comparing the parallel images across tale variants; rather, it springs from translating the "amplified" tale into psychological language.
This caveat addresses a fundamental point in Jungian interpretation. Jung maintains that "[c]ontents of an archetypal character are manifestations of processes in the collective unconscious," and since they refer to something unconscious "it is impossible to say what they refer to." 53 Jung adds, even "the best attempts of explanation are only more or less successful translations into another metaphorical language." The most one can do "is to dream the myth onwards and give it a modern dress." 54 In precisely this vein, von Franz acknowledges that her own translation of "the fairy tale myth into Jungian psychology" amounts to "replacing one myth by another-by our myth, the Jungian myth." 55 So the Jungian psychological "myth" becomes the template for all interpretations-again, we find meanings read into tales. Fairy tales read as so many variations on a psychological storyline permeated with Self, Shadow, Anima, and Animus. And, given Jung's notion that archetypical contents are a "twilight… which too much clarity only dispels," 56 interpreters enjoy untrammeled freedom in that "twilight." Not surprisingly, as Robert Segal remarks, Jung "is prepared to rule out all non-Jungian interpretations [of myth], but he is not prepared to rule in any one Jungian interpretation." 57 Closer to us, von Franz hints of internecine strife as she states that "many so-called Jungian attempts at interpretation," by colleagues she does not deign to "assail… by name," have strayed. 58 Here we are reminded of the conflicting interpretations by Freudian authors, and the underlying cause is the same. Invasive approaches tend to illuminate the nooks and crannies of the interpretive models, rather than the tales.
One reason why Jungians find it hard to agree on meanings is that Jung himself found symbols inscrutable. Because it is impossible to grasp unconscious contents, "every interpretation necessarily remains an 'as-if,'" Jung wrote. This is to say that "an archetypal content is first and foremost a figure of speech," and likewise all attempts at explanation are but "translations into another metaphorical language." Otherwise put, the "ultimate meaning" of symbols is inaccessible to rational inquiry. 59 Jung's tenet that the references of symbols are beyond the pale of rational inquiry is most clear in the work of Joseph Campbell. This prodigiously erudite researcher brought together, since his early inquiry on the heroic pattern of myths and fairy tales, "a host of myths and folk tales from every corner of the world" in order to "let the symbols speak for themselves." In truth, Campbell liberally used the ideas of Freud and Jung to grapple with "the grammar of the symbols." 60 But rather than following Freud's assumption that rational inquiry may transform "what is unconscious into what is conscious," 61 Campbell shared Jung's belief in unfathomable archetypes. "Such wonders simply are," Campbell writes-archetypes "are antecedent to meaning, though 'meanings' may be read into them." On this assumption, Campbell strives for "the release of the archetypical symbolic images of mythic thought from their various local matrices of culturally conditioned references and 'meaning.'" This approach amounts to addressing archetypes "as natural phenomena." 62 Indeed, Campbell reckons that myth is "an expression of the self-regulation of the psyche," comparable to "the compensatory action... of the body throwing off disease." So, insofar as archetypes have meaning, "such a 'meaning' [is] as that of a sneeze, the festering of an infected wound, or a fever." 63 In other words, the "meaning" of symbols is their biological function 64 -an absent meaning.
Fittingly, Campbell titled his "preliminary sketch" for The Masks of God (his mythological tetralogy) "The Symbol Without Meaning." This choice aligns his mythological masterplan with Jung's tenet that "the symbol is a figure by which allusion is made to an unknown." 65 In Campbell's own formulation, symbols are natural phenomena "opening back to mystery." 66 Jung's nebulous notion of archetype sets "mystery" as the gravitational core of Campbell's work on symbols.

Variation and Meanings
The foregoing discussion suggests that abstaining from translating tales into preset master narratives is a necessary condition for seeking meanings in the tales. Each folk story exists in the universe of its variants, and no single variant ever represents a tale-therefore, heeding tale variations is a necessary condition for apprehending the symbolic universe of each tale.
Yet, by the same token, taking variation in the wrong way can dispel meanings. This section revisits some influential approaches to tale variations and the study of meanings.

Tale Variation as Corruption: The Finnish School
Start with a negative case. In the late nineteenth century, the so-called historicalgeographic method-also dubbed the folkloristic method by Kaarle Krohn, its main popularizer-proposed that every tale should be tracked cartographically back to its original form. 67 Krohn and his followers assumed that folktales are authorial texts-stories created once upon a time, then memorized and orally transmitted. They presumed that, given the limitations of human memory, orally-transmitted texts inevitably get garbled. As Krohn himself put it, "If we observe the laws whose effects are recognizable in the manifold modifications of the original form of a tradition, we encounter first the influence of faulty memory." 68 This is to say that traditional variations stem from memory lapses-they are errors. As the Swedish folklorist Carl von Sydow pointed out, Krohn and his philologicallytrained collaborators transposed literary habits of thought into the study of oral traditions.
They assumed that "the original form of a folktale should be the most complete and the best, also the most logical," and they presumed that the oral variations "are products of degeneration." 69 By implication, folktales tend to meaninglessness.
Given the axiom that oral transmission fosters degeneration, the folklorist's self-assigned task is to trim off the corrupted bits in order to recover the original texts. Archer Taylor, an American expounder of the method, explains that "when the accidental variations are identified and eliminated, then and only then can the tale be studied effectively." 70 The bitter irony is that folklorists found themselves dismissing folklore. As Valdimar Hafstein put it, their goal "was to strip folk tradition away from the Ur-form, to 'recreate the original' as a 'complete creation' of an individual author." 71 This devolutionary view of tradition is unsustainable. First, it fails to account for the actual stability of tales. Krohn himself mentioned "the incredible stability of folk narrative," and Stith Thompson-who also subscribed to the notion that forgetfulness "is perhaps the most frequent cause of modifications in stories"-likewise noted the "remarkable" stability of stories "in the midst of continually shifting details." 72 The adjectives used by the two scholars are interesting. If tradition corrupts texts, then the stability of folktales looks "incredible," indeed: how could a tradition driven by memory lapses achieve a "remarkable" stability?
Second, the assumption of an error-driven tradition shuns the possibility of meaningful variations. Thus, the failure to understand variation and to address meanings are the two sides of one coin.
Alternatively, assuming that tale variations are meaningful entails that the myriad tale variants are so many equivalent permutations on enduring themes. This view accounts for the correlated facts that folktales are stable, and yet no two variants are ever alike-an insight that has been available since the inception of modern fairy-tale studies.

Tales and Variants: The Brothers Grimm
Jacob and Wilhelm Grimm, at the onset of the systematic collection and study of Märchen, explicitly held the view that folktale retellings are meaningful variations on stable patterns. The introduction to the 1812 volume of KHM clarifies that the Brothers Grimm felt free to conflate tale variants because they realized that a folktale exists in the set of its variants, and they wanted to convey a rounded view of each tale. In the introduction they note that folktales, "never fixed, transform themselves in each region, almost in each mouth," and yet "tales truly preserve the same fundamentals." Indeed, "the tales as time went by continuously created themselves anew, but for that very reason their core must be very old." The particulars are valuable, hence the Grimms often preserve in a single text "several expressions of one and the same tale owing to their pleasing and peculiar variations." 74 In the second edition they elaborate on this point: they say they often conflated complementary tale variants in one text, and kept diverging variants in the notes, so as to include everything they collected and to preserve every particularity they noticed. 75 The same idea drives Jacob Grimm, in his 1815 circular on the collection of Volkspoesie, to emphasize that "incomplete fragments are not to be scorned … all variations, repetitions, and versions of one and the same legend could each become important." 76 In short, the Brothers Grimm maintained that tradition is all about the creative retellings of capacious blueprints, which is why they chose not to select a single variant of each tale. They knew that Märchen live in myriad retellings, and no single variant ever represents a tale.

Philip Pullman, who recently retold a number of Grimm tales, captures the brothers' point as he declares, "[a] fairy tale is not a text." 77
And yet, the Grimms never set out to make sense of the themes by means of the variants.
It took a Freudian folklorist to bring together variation and meanings.

Allomotifs and Symbols: A. Dundes
Alan Dundes remarked that folktales have two defining characteristics. One is variationfor in folklore "there is no such thing as the text. There are only texts" 78 -and the other is fantasy, "collective or collectivized fantasy." 79 Much of that fantasy "is unconscious," in the Freudian sense of materials that are "repressed and disguised." Moreover, folklore "provides a socially sanctioned outlet for the expression of what cannot be articulated in the more usual, direct way." 80 It contains, in other words, "projective materials devised by a people themselves." 81 Dundes submits that it should be possible to "unlock" the "symbolic code in folktales." He does grant that Freudian readings "typically… give the appearance of being arbitrary, subjective, and unsubstantiated," 82 but he also points out that folktale variation can be called upon for the "empirical verification of purported symbolic equations." If a number of motifs can fill the same slot in a tale, then those allomotifs should be symbolically equivalent. By comparing them, "we may gain access to implicit native formulations of symbolic equivalences." This procedure, Dundes notes, can be of service to psychoanalytic theory. 83 How do allomotifs and Freudian symbols actually fit together? According to Dundes, If you find motif A and motif B on the same tale slot, you may presume they are allomotifs because "A may be used in place of B and B may be used in place of A." But you still do not know "whether A is a symbol of B or B is a symbol of A." In order to work out the symbols, you should ask if one or more allomotifs "are taboo or sensitive in nature." 84 Once you ascertain "that either A or B is a tabooed subject," it is likely "that the non-tabooed subject might be substituted for the tabooed subject rather than vice-versa." 85 For example, in different variants of one tale the protagonist may be punished "in a number of ways including throwing the hero into a snake pit, cutting off his head, or cutting off the hero's male organ." These three allomotifs suggest to Dundes, "among other things … that cutting off the hero's head is regarded as the equivalent of cutting off the hero's phallus." Dundes adds that one Ozark taleteller mentions decapitation before mixed audiences and reserves castration for men-only audiences, which confirms that decapitation is "a symbolic form of cutting off the phallus." 86 Note that Dundes reasons along Freudian lines. He does not seek the common denominator of the three allomotifs he mentions; rather, he establishes a one-way relation between an acceptable motif and a "tabooed" one, in compliance with Freud's description of symbols as the manifest elements that screen the "'genuine' thing behind them." 87 Dundes selects castration as the "sensitive" motif and beheading as its symbol, which is in agreement with Freud's assertion that decapitation symbolically represents castration. 88 In practice, Dundes never asks how the snake-pit punishment fits with the beheading and the castration allomotifs; rather, he falls back on just the data that match Freud's interpretive key. 89 He reasons in terms of Freudian symbols and (not surprisingly) tends to find such symbols.
There is no doubting the pervasiveness of the "head-phallus equation," which Dundes examines in some detail alongside other folk metaphors bearing Freudian overtones. 90 Dundes does state that he favors "relying upon folk metaphors," and he is fully aware that "metaphors are meaningful, not accidental," for "there are consistent patterns of metaphor in every culture." 91 Yet, looking for metaphoric patterns is not the same thing as seeking Freudian symbols. A Freudian symbol is the unidirectional relation between a manifest image and an unconscious element, and (as Freud noted) people will "make use of it without knowing about it." 92 Whereas a metaphor is the structural mapping between two conceptual domains, which may be bidirectional as well as conscious. Take for example the Ozark taleteller who used decapitation as a conscious euphemism for castration. 93 He actually used the metaphoric mapping between the body (with its head) and the phallus (with its glans), which is quite popular and has many entailments. This metaphor-unlike the unidirectional Freudian symbol-works both ways: a person's head connotes the phallus and, reciprocally, the phallus has a "head." All this is quite conscious. One crass Portuguese expression calls the glans of the penis "dick head," 94 and reciprocally, the risqué English expression "dickhead" likens a person's head to the phallus glans (and, by implication, the demeaned person to a "dick"). Note some entailments. Since the head maps to the phallus's glans, beheading-and all sorts of diminishment to the head, noted by Freud, such as "baldness, hair-cutting, falling out of teeth" 95 -are apt to represent castration. Reciprocally, because the penis glans maps to the head, the phallus is "a bald-headed guy," is endowed with an "eye" (hence, the epithet "cockeye"), and has a "mouth" that "spits" (hence, the expression "spitten image" to designate a child who resembles his father). 96 In short, the comparison of allomotifs yields different results according to whether one chooses to fall back on Freud's fixed translations or to follow the maze of metaphoric entailments. Freud (like Müller before him) posited that symbolism in folklore is the "residue" of metaphor in ancient language. 97 This assumption, itself a residue of nineteenthcentury thinking, has serious consequences. If symbols were indeed the survivals of bygone metaphors, then compiling a list of fossilized symbolic translations would be the way to go.
Alternatively, recognizing that metaphorical processes are ongoing-and that metaphorical mappings comprise multiple entailments, which allow perennially creative variations-clears the way for exploring the dynamic patterns of metaphor.

Symbols as Projection: B. Holbek
Yet, Dundes's call for comparing allomotifs spurred a few other folklorists to pursue the idea that "careful attention to allomotifs can often validate psychoanalytic arguments." 98 Most notably, Bengt Holbek developed Dundes's "plea for psychoanalytical semiotics" in fairy-tale studies. 99 Dundes, besides showing that folklore may be of service to psychoanalytic theory (by means of allomotific comparison), argued that "psychoanalytic theory can greatly illuminate folklore." 100 For this purpose, he proposed using "the crucial device of projection"-a psychoanalytical tool-to illuminate folk stories. 101 Holbek followed this path.
He assumes that the "symbolic" elements in fairy tales "convey feelings rather than thoughts"; and, since symbols are "vivid emotional impressions," interpretation consists in retracing all the "marvelous" fairy-tale elements back to the real-world referents of such impressions. 102 Otherwise put, "every element [in a fairy tale] may be read as pertaining to real life." 103 This leaves "no room at all for the so-called supernatural beings, the witches, fairies, dragons, ogres, etc.," since-as he stresses-"they represent aspects of real persons." 104 This perspective leaves no room for metaphorical patterns, either. When Holbek gets round to mapping allomotifs in one tale, he writes: "It makes no difference whether the queen is eating roses, onions or apples … it makes no difference whether the queen bears a wivern only or a wivern and a normal child … . It makes no difference whether the heroine's donor is an unidentified old woman or her dead mother." (my emphasis) 105 But saying "it makes no difference" is not so much about comparing allomotifs as it is about skirting the comparison. 106 Holbek's actual indifference to allomotifs reflects his choice of approaching symbols as psychological projections. Holbek surmises that the thematic axes of fairy tales express the "sensitive, even painful" problems of rural communities. 107 He takes up Dundes's idea that folklore "provides a socially sanctioned outlet for the expression of what cannot be articulated in the more usual, direct way" 108 and leads it to its logical conclusion: fairy-tale symbolism is a conscious language that screens sensitive matters. Thus, Dundes's notional project of addressing folk metaphors by means of allomotifs effectively gives way to the psychoanalytically-oriented notion of projection. In Holbek's model, fairy-tale symbols translate to the hypothetical psychological predicaments of abstract people even while metaphoric patterns recede to the background.

Metaphors and Symbols
Alternatively, exploring the dynamic patterns of metaphor is bound to encourage openended quests on fairy-tale symbolism. Maria Tatar once remarked that those who make it their business to study fairy tales "must see to it that they tell us more about the tales than about their particular school of thought." 109 With this point in mind, it is worth pursuing Dundes's interest in allomotifs beyond the remit of Freudian fixed translations.

Metaphors in Fairy Tales: M. Tatar
Freud's influence lingers in Tatar's proposition that fairy tales translate childhood fantasies into the "physical projections and representations of psychic processes." But Tatar highlights the underlying metaphoric processes. She submits that "the shift from the realistic milieu described in the tale's opening to the marvelous world of the tale proper is accompanied by a corresponding shift from the figurative meaning of words to the things that those words designate." 110 Otherwise put, in the fairy-tale world "the figurative or metaphorical dimension of language takes on literal meaning. Ideas become matter." 111 One instance of symbols "generated" by such "literalized metaphors and by visual puns," according to Tatar, is: "[t]he mother or stepmother who is like an ogress at the beginning of a tale becomes an actual witch." 112 Thus, "Hansel and Gretel" displays an "evil parent … reflected and distorted in the mirror of the fairy-tale world. The stepmother who fails to nurture the children … reemerges in the woods as … a cannibalistic fiend." 113 This depiction of fairy-tale symbols as the translation of verbal metaphors into their mirror images is reminiscent of Freud's argument that symbols translate latent dream thoughts into "visual pictures." Freud quotes an experiment by Herbert Silberer: "'I thought,' says Silberer, 'of having to revise an uneven passage in an essay.' The vision: 'I saw myself planing a piece of wood.'" 114 Overall, Tatar submits that fairy tales tell us "something about the way in which the mind draws on the double movement of language between literal meaning and figurative expression" 115 -a point that is of the essence of this discussion.

Metaphors and Symbols: C. Lévi-Strauss
Claude Lévi-Strauss also addressed Silberer's experiment. He pointed out that the transposition of a figure of speech into a literal meaning is itself a metaphorical process. "To the writer," he notes, "the work of the carpenter is an image of his own work, just as the writer's work might remind a carpenter of his own activity. A metaphor always works both ways … it is like a two-way street." 116 And so are symbols, he adds. Given a metaphor that links up different conceptual domains, symbols are the elements in each domain that are homologous to-and, thus, permutable with-elements in the other domain. It follows that signification is the product of such relations between codes-signifying, in other words, "is nothing but establishing a relation between terms." 117 Lévi-Strauss credits Müller and Freud with the merit of finding and deciphering, respectively, the "code of astronomy" and the "psycho-organic code" in myths. But he remarks that myths (and fairy tales) "always put several codes in play," none of which conveys the "better" meaning. The essence of the myth (and the fairy tale), Lévi-Strauss proposes, "is founded on the property inherent in all codes: that of being mutually convertible." 118 This property is the telltale sign of metaphoric mappings; indeed, myths reveal that "metaphor rests on the intuition of logical connections between one domain and other domains in which [the metaphor] reintegrates it." 119 This is why the elucidation of myths (and fairy tales) involves considering the "layering of codes, one on top of the other." 120 Lévi-Strauss's stresses that metaphor, in switching terms that belong to different codes, "rests on an intuition that these terms connote the same semantic field when seen from a more global perspective. The metaphor restores this semantic field." 121 Lévi-Strauss highlights the pervasiveness of metaphor as he acknowledges that his elucidation of obscure Amerindian myths matches "the very obvious analogies we make in our native tongue." Indeed, he points out, "in all languages there are more or less matching expressions that… are the emanation, in popular language, of thoughts that draw their substance from the very roots of the mind." 122 In the "metaphorical process," he suggests, the mind regresses to "the primitive apprehension of a global structure of signification." 123 Here, again, Lévi-Strauss meets Müller and Freud. But whereas his predecessors supposed that symbols are the fixed survivals of archaic metaphors, Lévi-Strauss points out that metaphorical processes are ongoing. Hence, no symbol "signifies anything by itself"; 124 rather, the signification of a myth (or a fairy tale) "is always global; it cannot be reduced to the interpretation provided by one particular code." 125 To begin making sense of the pervasiveness of metaphor, recall Müller's point "that all words expressive of immaterial conceptions are derived by metaphor from words expressive of sensible ideas." 126 The experimental finding, in contemporary cognitive science, of "the metaphorical constitution of our abstract concepts" vindicates Müller's point. 127 Indeed, the work of George Lakoff and Mark Johnson suggests that "abstract concepts are largely metaphorical" because, in essence, basic metaphors transpose "imagery from sensorimotor domains" to abstract domains. 128 Otherwise put, metaphor is required for abstract thinking. It follows that conceptual metaphors are mostly not symmetrical. While one can metaphorically map the act of grasping an object to an act of understanding (e.g., do you grasp this idea?), the converse is not true-one could not possibly use "understanding" to convey the concrete act of holding an object. In this light, recall Lévi-Strauss's suggestion that "the writer's work might remind a carpenter of his own activity" (my emphasis). 129 "Might" is the crucial term here, and it is misleading. While it may be illuminating to depict the intellectual labor of a writer in terms of the palpable craft of the carpenter, describing the act of planing a piece of wood in terms of the writer's exertions would merely befuddle the matter. The point, again, is that the thrust of metaphoric thought lies in using concrete imagery for abstract propositions.
In fact, Lévi-Strauss's entire oeuvre confirms this point. In the opening words of his tetralogy on mythical thought, Lévi-Strauss declares his aim of showing "how empirical categories… can … serve as conceptual tools to bring out abstract notions and articulate these in propositions," so as to bring out "a logic of sensory-based qualities." 130 Again and again, Lévi-Strauss recalls that in myths metaphorical thought raises concrete images to abstract propositions. 131

Fairy-Tale Symbolism
The foregoing discussion suggests that (i) fairy tales hinge on live metaphors; (ii) metaphors map relevant aspects of one conceptual domain to other conceptual domains; (iii) such mappings constitute a network of symbols, of which different tale variants choose alternative allomotifs; and (iv) metaphorical mappings often use concrete imagery to construct abstract propositions. These insights entail that (i) the meanings of fairy-tale symbols are not set once and for all; rather (ii), the values of symbols hinge on the metaphorical transpositions at play in tales; therefore, (iii) comparing the allomotifs in tale variants is crucial to determine the active metaphors; and (iv) such metaphors likely use sensorial imagery to convey abstract propositions.
When applying these insights to fairy tales, it bears heeding Vladimir Propp's warning that this genre has "a quite particular structure which is immediately felt … even though we may not be aware of it." 132 The comparative outlook of Mircea Eliade correctly identifies in the fairy-tale form the "'deaths' and 'resurrections'" pattern typical of initiatory ordeals. 133 Indeed, Propp remarked that the themes of initiation to puberty and of journeys to the realm of death account for nearly all the contents of fairy tales, which makes sense because initiatory processes are usually conceptualized "as abiding in the realm of death." 134 Propp also remarked that the "structure of the tale demands that the hero leave home at any cost," hence the narrative develops along the "route of the hero." 135 And he noted that this route often cuts across a dark forest. 136

Fairy Tale as Metaphor
Propp's insights call for two comments. The first one has to do with metaphor as a structural fairy-tale feature. Propp showed that fairy-tale adventures are journeys in space, their overarching theme being maturation into adulthood. I submit that these two features correlate, i.e., fairy tales depict spatial journeys in order to talk about maturation processes.
Differently put, fairy tales use the concreteness of spatial journeys to reason about maturation processes. This is a basic metaphorical process. Lakoff and Johnson noted that "[m]ost of our understanding of time is a metaphorical version of our understanding of motion in space." Specifically, the time-orientation metaphor "has an observer at the present who is facing toward the future, with the past behind the observer" (linguistic expressions of this metaphorical mapping include "that's all behind us now," and "we're looking ahead to the future"). Moreover, in the time's landscape metaphor "each location in the observer's path is a time," so that "time is a path on the ground the observer moves over" (linguistic expressions of this mapping include "there's going to be trouble down the road," and "we're getting close to Christmas"). 137 Fairy tales combine these two mappings of space to time as they depict characters struggling along the path of their own future.
My second comment to Propp's insights takes good note of two shortcomings in his model. First, although Propp defined the optimal span of the fairy-tale genre he never asked what the minimal condition a fairy tale must fulfill might be. 138 Second, Propp argued that the basic form of the fairy tale is about a hero who grabs magical powers in the forest, then slays a monster and liberates a princess. 139 Crucially, this androcentric model both supposes and obfuscates the princess's prior enchantment. Accordingly, I submit that the minimal fairy tale features a feminine enchantment. 140 The metaphorical expression "coming of age"-coming (as in space) of age (as in time)-arguably sums up the clockwork of fairy tales. In its minimal form, a girl walks into the woods (and comes of age).

In the Forest of Symbols
"Little Red Riding Hood" (ATU 333), a story about a girl who walks into the woods and becomes a young woman, is so deceptively simple it has ben proclaimed a children's talebut its marked sexual symbolism suggests otherwise. 141 Presently, I point out that a measure of intertextual comparison helps elucidate the symbols strewn along the path in this metaphorical depiction of maturation. 142 Virtually all the available variants (mostly from France) mention an innocent girl who strolls into the woods and meets the wolf. Charles Perrault's pointed revelation (in the moral he adds to the tale) that the wolf is a sexual predator accords with the folk saying that a girl who engaged in trysts "saw the wolf" (Littré, Dictionnaire de la langue française, s. v. "Loup"). 143 It follows that the forest path takes the girl from innocence to sexual experience. Now, the oral variants depict the path strewn with pins, needles, thorns, and other items that-as Yvonne Verdier summed up-"scratch, prick, cut." 144 The plural hints of a thorny, pricking path suggest that the girl oozes blood as she proceeds-i. e, she starts bleeding in the forest-a point that Perrault reiterates with flowers. While Perrault elides the pricking (Incidentally, this is a metaphor we still use when we call a maiden's first sexual experience "deflowering.") Overall, the tale variants hint that the girl walks the path of her puberty.
Moreover, Perrault's image of a carrier of flowers who is bound to meet a sexual predator cheekily implies that she heads to her deflowering.
Note that the wolf meets the girl at the crossroads to set her on her blood path, and again at the forest cabin to take her to bed. The illustrator Walter Crane, in his Little Red Riding Hood illustrated booklet (1875), effectively summarizes this two-tiered intervention. As the sheepskin-clad wolf meets the girl at the crossroads, his cudgel points to (is actually lodged in) a spot of flowers that are white and red like the girl's own apparel (Fig. 1).
Thus, Crane (like Perrault) both associates the girl with the flowers and hints her looming deflowering. The point is that, as Verdier pointed out, the wolf "leads the game and conducts the girl toward every step in her feminine destiny, which is realized through him." 145 Why would this be? Consider again the wolf's sheepskin. With this image, Crane certainly alludes to Perrault's insistence that "tame wolves" are "the most dangerous of all." 146 But wearing two skins is also a defining trait of the werewolf lurking behind the humanized wolf (a werewolf is, of course, a man-wolf). 147 Werewolves, deemed skin shifters, supposedly undergo cyclic shape shifting according to lunar rhythms. Indeed, shape shifting is a lunar attribute (every month the moon changes its shape as it wanes, vanishes, and waxes). 148 What is more, the moon's cycles exactly match feminine cycles. 149 Given that lunar circuits bring werewolves and cycling girls into the same orbit, werewolves befit lunar scenarios of feminine transformation and coming of age, which is presumably why this werewolf conducts this girl every step of her metaphorical path to womanhood. 150 The variant contributed by the Brothers Grimm in 1812 ("Rotkäppchen") adds its own metaphorical twist to the theme. 151 The girl is supposed to walk a straight line connoting a "straight" (obedient, chaste) disposition, which implies that straying from the path bespeaks "perdition" in the moral sense. Essentially, this variant maps the tangible act of walking along the path to the abstract realm of morals. After the wolf lures the girl into leaving the straight path, the straying girl follows her own uncharted path of flowers: "You march along as if you were going straight to school in the village, and yet it's so delightful out here in the woods! … She plunged into the woods to look for flowers." 152 Although the Grimms join Perrault in the implication that the path of flowers leads to the girl's deflowering, they avail themselves of another metaphor.
That new metaphor frames the entire story. At the outset, the mother's stark warning against straying from the path boils down to a single image: "otherwise you'll fall and break the glass." 153 This warning of moral disgrace-metaphorically, a fall from grace-makes use of the pervasive equivalence between bodies and containers, which Freud acknowledges as he notes that female genitals are often symbolized by "vessels and bottles" and other receptacles, the breaching of which signifies lost virginity. 154 In this metaphorical strain, breaking the glass connotes the shattering of the girl's integrity. A shattered wine bottle involves, of course, a red flow. Thus, the brothers convey the basic fairy-tale leitmotif: a girl bleeds in the woods.
The broken-bottle metaphor is actually transparent enough that the brothers presumably moved to contain its implication. At story's end, we find a repentant girl bearing an intact bottle. Given the mother's precise warning, and the girl's actual straying, this turn of events looks surprising. Yet, it agrees with the Grimms' well-attested zeal for expunging sexual scenes from the tales. 155 What is more, displaying an intact bottle for a prim ending is but a special use of the broken-bottle metaphor. Denying that the glass ever broke whitewashes the girl's reputation while also reiterating the figure of speech embedded in the mother's warning. Therefore, the Grimms manage to preserve the sexual symbolism and to present a morally prim tale. Such procedures should be heeded in debates about whether the Brothers Grimm were faithful to the tales they edited.

A Metaphoric Kernel: Coming of Age
The foregoing discussion of a few allomotifs in "Little Red Riding Hood" is not a proper discussion of this tale. Rather, it is meant as a quick illustration of the point that fairy tales are built on metaphorical mappings across codes, which numerous variants express in various ways. Put in a nutshell, grasping the metaphors in the variants of a tale-i.e., mapping the tale imagery to abstract propositions-is what the study of fairy-tale symbolism is about.
The above remarks on "Little Red Riding Hood" also illustrate the hypothesis that a particular metaphor-coming (as in space) of age (as in time)-is the bedrock of fairy-tale symbolism. This hypothesis requires further research that would put it to its paces. Although the study of fairy-tale symbolism is old, it is only just coming out of the woods, and a long winding path lies ahead.