The "Motif Compass" Method by Olga Freidenberg and Tove Jansson's "Grey Duchesse":Sinister Knitters and Embroiders
This article revisits the theoretical approach proposed by the Russian scholar Olga Freidenberg (1926/1987) and suggests that Freidenberg's analytical method, called "motif compass," can be a valuable tool for modern narrative investigations. Although it is relatively unknown and is not currently used in mainstream narratology, the motif compass method offers numerous advantages in studying "narrative DNA" (Ofek et al. 166) and can potentially open new avenues in comparing otherwise unrelated narratives. In order to demonstrate how the aforementioned method operates, the article presents a conceptual diagram and a motif analysis of a story by the Finnish-Swedish writer Tove Jansson titled "The Grey Duchesse." Freidenberg's method, as applied to Jansson's story, helps uncover how unrelated narratives and narrative elements may be considered genetically similar at a conceptual level, provided they stem from the same conceptual network comprised of similar universal images, motifs, and beliefs. The article shows that such genetic similarities are often manifest in various forms across genre and language boundaries.
Olga Freidenberg, Tove Jansson, narrative DNA, narratology, data mining, machine learning
[End Page 83]
Memory is the seamstress, and a capricious one at that. Memory runs her needle in and out, up and down, hither and thither. We know not what comes next, or what follows after.—Virginia Woolf, Orlando
This article revisits the theoretical approach proposed by Olga Freidenberg, a brilliant but often overlooked Russian scholar (see Perlina, Olga Freidenberg's Works). I argue that the method that Freidenberg called "motif compass" ("Metodologiya" 126) can be a valuable tool for modern narrative investigations. Although it is relatively unknown and is not currently used in mainstream narratology, the motif compass method offers numerous advantages in studying "narrative DNA" (Ofek et al. 166) and can potentially open new avenues in comparing otherwise unrelated narratives. This is especially important in the context of new interdisciplinary approaches to narratology and cognitive science (see Sommer). While showcasing the advantages that the motif compass method can offer narratology, this article briefly discusses Freidenberg's scholarly inheritance. Despite the fact that Freidenberg is less known in the West in comparison to her contemporaries, such as Mikhail Bakhtin, Lev Vygotsky, and Vladimir Propp, her legacy is as significant as theirs, and is as worthy of examination.
In order to demonstrate how the aforementioned method may be used, the article analyzes a story by the Finnish-Swedish writer Tove Jansson (1914–2001) titled "The Grey Duchesse" (published 2014). The choice of the story is not arbitrary—it appears to be ideal for exemplifying the proposed method. Its plot can be easily dissembled into "nuclear" conceptual elements in order to illustrate how genetically similar narratives, stemming from the same universal images, motifs, beliefs, and concepts, emerged in various forms across genre and language boundaries.
This article consists of four parts: (1) a brief introduction to Freidenberg's scholarly legacy and discussion of her motif compass method in its connection to modern narrative research and cognitive science; (2) the conceptual diagram and the analysis of "The Grey Duchesse" with the help of the motif compass method; (3) manifestations of "The Grey Duchesse" conceptual diagram in unrelated narratives; and (4) conclusions.
Olga Michailovna Freidenberg
Olga Michailovna Freidenberg, born in 1890 in Odessa, died in 1955 in Leningrad as a marginalized academic. She was Jewish, and she was also the first woman to receive a PhD in Classical Philology from the Soviet University—both facts are important because they took a heavy toll on her academic career and life, and to a large extent contributed to her subsequent obscurity. Freidenberg possessed unparalleled expertise [End Page 84] in ancient folklore, history of culture, and historical aesthetics. Due to her enormous erudition, her ability to detect analogous patterns in a wide range of cultural phenomena was remarkable (Perlina, Olga Freidenberg's Works 4–5). In the West, Freidenberg is mostly known as Boris Pasternak's cousin and correspondent, and her scholarly contributions are mistakenly thought to be focused on the history of ancient literature and the origins of Greek tragedy. Her role as a prominent and brilliant scholar, whose theoretical significance is equal to Propp's, is less known, especially in the context of narrative studies. As opposed to Propp's Morphology of the Folktale, which is widely used in the study of narratives, Freidenberg's works are virtually unknown outside of Russia. As for the universality of Freidenberg's ideas and approaches, there are very few discussions on how they fit into modern narratological research, and even those are rather indirect and not dealing with Freidenberg's methods themselves (see, e.g., Vrabie). The titanic efforts of Nina Perlina, Nina Braginskaya, and Kevin Moss to bring Freidenberg's legacy to the attention of the Western academic community have still not yielded the desired results (see Braginskaya, "Analiz"; Braginskaya, "From the Marginals"; Perlina, "From Historical Semantics"; Perlina, "Olga Freidenberg on Myth"; Perlina, Olga Freidenberg's Works; and Freidenberg, Image and Concept).
According to Braginskaya, Freidenberg's works are still insufficiently known in the West: "Her interpretations are being used to a certain degree as historical data" ("Analiz" 116). Only a very small portion of Freidenberg's work is translated into English; as a first step, Freidenberg's book Image and Concept was fully translated into English by Moss (with a foreword by V. V. Ivanov). The rest of her articles, which are scattered across various Russian publications, as well as her fundamental book Poetics of a Literary Plot and Genre, still await their turn. Despite Freidenberg's relative obscurity in the West, her methods, which were innovative for her time, are still valuable today.
According to Ivanov, Freidenberg can be justly considered our contemporary because of the wealth of materials she used, and because she was a "forerunner of Claude Lévi-Strauss and Mikhail Bakhtin" (Freidenberg, Image and Concept x–xi). Ivanov calls her "a scholar interested in the archaic layer of consciousness as revealed in human psyche, myths, language and cinema" who was "also concerned with archetypal proto-image," and maintains that she developed ideas that were close to the results of recent investigations in cognitive science (xi). Ivanov also suggests that it is imperative to revisit Freidenberg's insight on how images develop into concepts, because it can be reinterpreted in the light of new developments in neuropsychology (xi).
In Image and Concept, Freidenberg concentrates on "reinterpretation of the old form in a new conceptual system of thought" and shows how the mythological image reproduced itself later in various forms, different on the surface but "united by the identity of their mythological semantics" (Image and Concept 17). In the forties and fifties, Freidenberg further developed her idea of the transition from internal (content) to external (form) in another aspect: as a shift in cognitive function from image to concept. She called her approach "genetic" as opposed to "evolutionary" (the Formalists' preferred term). Although there is no doubt about the value of Freidenberg's works for studying narratives, one of the hurdles in popularizing her ideas is that they often [End Page 85] seem too dense. As is often the case with those scientists who are ahead of their time, Freidenberg explored uncharted territory and "had to invent a new metalanguage to describe the interaction between psychology, language, and phenomenology" (16). It should be emphasized that, using Braginskaya's words, Freidenberg was a "scholar in isolation" ("From the Marginals" 65), and that her method of accounting for plot similarities was paleontological—that is, it searched for "retentive archetypal patterns" (66) rather than for influence, imitation, adaptation, borrowing, etc., although she never denied such aspects when relevant and appropriate (69).
It should be noted that Propp's and Freidenberg's approaches are not mutually exclusive—they are complementary. While Propp deals with functions, Freidenberg's method targets deeper semantic/conceptual structures across texts; these approaches can be applied simultaneously or consecutively to the same data to reveal different levels of abstraction. Whereas Propp's method, which always enjoyed popularity, currently witnesses a new wave of interest within a broad range of disciplines dealing with narratives, Freidenberg's method still awaits its rediscovery.
The recent attempts to integrate Propp's Morphology of the Folktale with computational models of narratives clearly demonstrate that, in addition to Propp's functions (i.e., the "syntax" or surface structures of the plot), another level of abstraction is needed at a conceptual level (roughly corresponding to a semantic level or deep structure). In this respect, the DNA metaphor introduced by Ofek et al. seems especially useful because it captures an important property of the narrative data across texts, which often exhibit innate similarities despite external differences. This property is well known to the scholars of folklore (see Thompson's Motif Index ["AT"] and the Aarne-Thompson-Uther Tale Type Catalog ["ATU"], the most comprehensive classification and bibliography of international folk tales). AT/ATU tale types recently started being studied extensively in computerized modeling and data mining. For example, Matthew L. Jockers has developed data mining software that discovered six basic plots of fiction—the same idea that was famously presented by Kurt Vonnegut as the "simple shapes of stories" (cited in Fusco). The existence of innate similarities in unrelated plots is a well-known fact across disciplines, but there is still no adequate theory explaining and accounting for it, despite some attempts in the field of folklore studies. The most popular tool—Propp's model—brings inevitable challenges precisely because it deals with functions. Thus, instead of finding genetically similar narratives or a "narrative genome" (Ofek et al. 166), motif analysis only reveals external similarities or helps grouping narratives by types. For example, Ofek et al., using sequence mining methods from bioinformatics and applying the idea that genetic variation ("mutation") also occurs in storytelling, were able to build a machine learning algorithm capable of predicting the ATU tale type (e.g., magic tale, novella, or joke) of a text based on its motif sequence (170). By "motif" the authors understand "abstracted, generic content tags which summarize segments of the plot" (167). The authors have found out that multi-motif narratives, unlike one-motif ones, were not always predictive of their tale types, hence the caveat that "tale type is a bicomponential concept, having to satisfy both a formal and a topical constraint" (171). In other words, tale types (the formal aspect) are less predictive of genre groupings than [End Page 86] motifs, which are "topical" and have the ability to concatenate and "to generate new narrative types" (171).
Along the same lines, Finlayson, who was interested in an analogical mapping algorithm to find the semantic and structural similarities in folktales, arrived at the conclusion that it was necessary to study descriptions at a higher semantic level: to take an event timeline as described in folktales, and "abstract [it] to the next higher level: structures such as Villainy, Struggle-Victory, and Reward" (Finlayson 3; emphasis original).
It is my contention that the motif compass method proposed by Freidenberg can be successfully used to circumvent some of the above challenges because it targets concepts. First, it can explain plot similarities on a "topical level" (by focusing on the interrelationships of concepts underlying them, because concepts are connected to each other through a web of associations). Second, it can be used for data mining, artificial intelligence, and machine learning (via detailed annotated conceptual diagrams similar to the one discussed in this article).
The motif compass method draws on comparative folklore, literature, and historical data, and therefore it can be used for establishing associative rules between the various concepts, elucidating patterns of their co-occurrence, and building conceptual diagrams and mappings with historical and cultural validity and universality. The latter consideration is especially important because it has been noted that there exists a "fundamental nature of the mental processes involved in reading, as well as [a] generic nature of the narrative frames and schemata which are activated in the reception process" (Sommer 93). Moreover, according to Phelan, "If readers need conceptual schema to construct interpretations, authors also need conceptual schema to construct structural wholes" (49). In my view, then, conceptual diagrams can illustrate the above conceptual schemata and also be highly predictable and stable across a variety of narratives.
The idea of a motif compass first appeared in Freidenberg's article "Metodologiya Odnogo Motiva" in 1926, where she demonstrates how seemingly unrelated plots may be genetically related on a conceptual level (i.e., "topicality"), and argues that it is important to compare various plots by focusing only on their conceptual features, while completely severing those plots from their authors' intentions, ideologies, and biographies. As an example, Freidenberg isolated a concept of [VILLAINY] in "La Devoción de la Cruz," a play by Pedro Calderón (1600–81), a dramatist of the Spanish Golden Age. Freidenberg unravels the play's genetic connections to other narratives that were built on adjacent, overlapping, or juxtaposed concepts, such as [DEVIL], [DEMON], [SAINT], [RIGHTEOUS], and others (recall Finlayson's "higher level of abstraction" mentioned above).
In the discussion that follows, Freidenberg draws numerous literary and folklore parallels to demonstrate how different plots are genetically connected through a web of related underlying concepts. She calls this method "motif compass," and convincingly proves that it can reliably and robustly uncover genetically related plots, which on the surface do not seem similar at all.
To exemplify this underlying cognitive process, Freidenberg dissects the poem "Daemon" by the Russian poet Mikhail Lermontov (1814–41) and shows that the [End Page 87] concepts of [EVIL] and [VIRTUOUSNESS] are semantically connected because one is an antithesis of another. As such, they are related to a plethora of concepts, such as [RIGHTEOUS ROBBER], [VILLAIN TURNED HERO/SAINT], and [REPENTANT SINNER], to name just a few. While the concepts may have varying degrees of connectedness (or fuzziness—to be discussed further), what is essential is that their connection is not accidental but stereotypical and predictable (and, as a result, it does not surprise the reader). Furthermore, such connections are highly clichéd and reproducible in both literature and folklore. In other words, manifestations of conceptual links are to be found on a plot level as a "gist" or scheme, as occurs, for example, in such plots as the medieval "Robert the Devil," the legend of Robin Hood, and Schiller's play The Robberss (1781).
Later, Freidenberg demonstrated that one widespread concept of the [RIGHTEOUS ROBBER] was also robustly related to the concept of the [DOUBLE] through a similar conceptual link with [DUALITY]. This idea was further developed and elaborated in her opus magnum Poetika Sjuzheta i Zhanra, where she discussed the semantics of a double in the historical perspective. Freidenberg applied this method with the same invariable result to other conceptual networks, for example [SLEEP], [IMPRISONMENT], and [DEATH], while calling her own approach "unlawful from the point of view of orthodox literary theory" ("Metodologiya" 126). In fact, the method was so innovative that it freely crossed the boundaries of genre and chronology, dwelling only on the human cognitive ability to build a web of free associations, which were not accidental but highly predictable. It is important to keep in mind that Freidenberg used this method not to identify genetically similar plots but rather to uncover the underlying concepts and their related conceptual networks in a wide variety of narratives. Thus, her goal was to exemplify the cognitive/mental processes per se.
For example, she argues that Calderón developed his play based on the conceptual associations (i.e., cognitive "networks") that had already existed before him. The same goes for Lermontov, who used preexisting materials and imagery to develop the classical idea of a Daemon, which was new to him. Hence, both Calderón and Lermontov drew on preexisting conceptual relationships, and that is precisely why, on a conceptual level, their works exhibit such a high degree of isomorphism, although on a formal or ideological level they seem unrelated.
According to Freidenberg, a given concept, along with its semantic connections, already exists in a collective consciousness in various forms prior to the individual author's realization ("Metodologiya"123). The author's role, therefore, is merely to arrange the concepts, and to connect them with other concepts and their corresponding imagery, which are never accidental. Although such arrangement may be individual, the conceptual and imagery inventory is preexisting, and the most incipient concepts (such as [LIFE] or [DEATH]) are universal.
The motif compass, according to Freidenberg, leads readers toward recognition of other conceptual materials familiar to them, in the process of retrieval of the related narratives from their memory. Freidenberg has chosen Calderón's play to exemplify her method. However, for its further application, the choice of narrative material does not matter. Because the method is paleontological and targets deeper [End Page 88] conceptual structures, the data for analysis may vary, and it is highly justified to apply the method to any folklore, fiction, or nonliterary narrative materials.
In the subsequent analysis, dwelling on Freidenberg's theoretical framework, I use the motif compass method to uncover the incipient conceptual level of one specific plot: "The Grey Duchesse" by Jansson. I build a conceptual diagram of the story and demonstrate how this plot is related to other adjacent, overlapping, or juxtaposed concepts and images found in unrelated but genetically close narratives.
The choice of the story for the analysis—"The Grey Duchesse"—is not arbitrary. First, to the best of my knowledge, an analysis of its universal and symbolic features has not yet been conducted. Second, the story is very short and is thus ideal for exemplifying a motif compass method. The story's plot is based on a conceptual association of [THREAD], [FATE], and [VISION], which is stable, predictable, and can be found in world literature and folklore. It should be noted that in this article I am applying Freidenberg's method to one specific set of conceptual associations, which was selected arbitrarily based on my own prior knowledge of their existing conceptual interrelationships in world folklore and ethnography.
In what follows, each literary example is investigated from the conceptual perspective only, which does not preclude it from having a myriad of other dimensions. It is important to reiterate that the analysis suggested here is not mutually exclusive with any other way of looking at the same text. From the methodological perspective, motif compass simply targets a set of similar concepts and motifs "artificially" extracted from the large corpus of texts. Beyond doubt, other methods of analysis could and would identify other dimensions (such as additional dominant or secondary motifs, different conceptual interrelationships, or important historical, ideological, or biographical information) of "The Grey Duchesse" or any other text presented here. In other words, any text/fragment presented here can figure in multiple motif analyses and additional motifs/conceptual associations or interpretations.
The goal here is to demonstrate the validity of the motif compass method for establishing similarities at the conceptual level between unrelated texts. If applied, the motif compass method should be seen as an additional tool to other techniques employed in traditional literary criticism/narrative analysis.
"The Grey Duchesse" by Tove Jansson
Following Freidenberg, I treat "The Grey Duchesse" by Tove Jansson as a plot/narrative scheme only, leaving aside its author's intentions and biography. The story is about a seamstress, Marta, who has a gift to see when those around her are soon to face death. She cannot help having these visions, and feels compelled to share them with the person who will die soon. This, naturally, does not make her popular, and eventually she has to leave her native village to settle in the city. Because she is a very talented seamstress able to create beautiful patterns and combinations of colors, she soon becomes a leading embroiderer in the most prestigious fashion house. Once she receives an order to decorate a dress for a client, and chooses a grey duchesse for decoration (hence the title). The client is not happy about the color, but Marta says [End Page 89] that it does not matter because the dress will bring misfortune in any case, and the client will die very soon. Soon Marta's sinister prophecy is fulfilled and the woman dies in a car accident, leaving Marta feeling very tired. In the end, Marta loses her prophetic gift, but with it goes her talent as an embroiderer.
The story is very simple, short, and told in a realistic manner, leaving the reader with a vague impression of something ordinary yet eerie. Despite its apparent simplicity, the story imprints in one's memory: the underlying concept of the inescapability of one's destiny, along with this concept's accompanying imagery, make the story very powerful.
In order to research the DNA of this story and find genetically close narratives, following Freidenberg, I use the motif compass method. To exemplify how it operates, I first reduce the story to its narrative scheme:
[A PERSON WITH A PROPHETIC GIFT TO FORETELL DEATH] AND [A PERSON APT IN EMBROIDERY]
Second, I decompose the resulting narrative scheme into two nuclear conceptual elements:
[DEATH] AND [EMBROIDERY]
As a next step, I extract three main conceptual components from the reduced narrative scheme by semantically connecting embroidery with thread, and then by connecting thread with life.
I argue that the story contains a powerful cue to the preexisting conceptual materials about thread, fate, and life, which are omnipresent in world folklore, beliefs, and literature. The validity of conceptually connecting thread with life is strongly justified based on the ethnographic and linguistic evidence. First, the imagery of a temporary and precarious life as literally being attached to a thread separating one from death permeates figurative language (in such expressions as "thread of life," "life hanging on a thread," "hang by a thread," etc.). Second, the association of life with thread seems to be universal, as it can be found across various cultures. Thread is one of the primordial human symbols; literature devoted to its symbolism is copious and well known. Because it is impossible to survey it here, it suffices to mention that, according to Eliade, who studied symbolic representations extensively, in many countries the thread of life "symbolises human destiny" (114). Thread, and weaving as an activity related to thread, is also a universal symbolic representation of the umbilical cord, which is literally connected to birth (life) (see, e.g., Cooper).
The resulting conceptual diagram of the studied story is as follows:
[EMBROIDERY] ➔ [THREAD] AND [THREAD] ➔ [[LIFE] AND [DEATH]]
Because Marta's gift (prophecy) is uncanny, and because it triggers fear as it is related to death, additional semantic elements are added to the conceptual network: [SINISTER]; [FEAR]; [MISFORTUNE]. [End Page 90]
Also, because embroidery is semantically related to any activity that manipulates yarn and thread, I add to the conceptual network the following elements: [KNITTING]; [SPINNING]; [WEAVING]; [YARN]; [WOOL]; [NEEDLE].
Likewise, [WHEEL] is added because spinning of the yarn is related conceptually to the wheel.
Prophecy, like future telling, is related to destiny and fate; hence, the additional conceptual element [DESTINY/FATE] is added, and because one cannot know one's own destiny, [MYSTERY] is added.
There is also an association between fate and vision: being able to foresee fate requires special vision, and hence [VISION] enters the diagram. The duchesse color is grey, which is also added as an additional element because of its symbolic meaning.
The resulting extended diagram connects all the above conceptual elements with [THREAD] as its central concept, as in Figure 1.
To reiterate, the above diagram of semantic connections was built with the help of Freidenberg's motif compass method and following the model of the semantic priming paradigm, which is one of the most widely used paradigms in psychology to investigate the organization of a cognitive network (cf. McNamara). I deliberately connect many of the concepts to each other and predict that such conceptual connections will robustly emerge in one form or another; that is, sets of concepts may exhibit a predictable pattern of co-occurrence. Further analysis demonstrates that, indeed, most of the elements included in the above diagram have stable and recurrent [End Page 91] conceptual associations, which reproduce themselves reliably in various literary and nonliterary contexts. I bring ethnographic and historic evidence to bear in support of the view that some of the above conceptual associations are primeval and not accidental. Although such ethnographic evidence is not a narrative per se, it does indicate that, in the historical perspective, there existed beliefs and narratives connected to it. Hence, following Freidenberg, I maintain that ethnographic facts are part of the common system and stock of beliefs and imagery. Additional parallels can be provided using the AT Motif Index, which is the worldwide repository of narrative materials pertaining to world folklore (with the caveat that their variants or "mutations" may exist elsewhere).
To answer the question of whether the above conceptual links can be considered motifs per se, we need more data and critical reflection. In my view, any of the above conceptual combinations may have the potential to develop into full-fledged motifs either on their own (as shown below) or through adjacent concepts. The concepts presented here could be semantically connected with other concepts because boundaries of any given concept are always fuzzy: "Natural language concepts have vague boundaries and fuzzy edges" (Lakoff, cited in Aitchinson 401). This phenomenon, well studied in linguistics, is known as the "fuzzy meaning" assumption (46). With respect to the interrelationships of separate motifs, it has been observed that there also exists a certain degree of overlap and fuzziness: in literature and folklore, universal motifs have the ability to "merge" or "branch out" (Segre 193) so that any motif (X) may appear on its own, but also as "merged" (i.e., X+Y, or concatenated with elements of Y and/or Z).
The above conceptual network diagram suggests that Marta is a personification of fate, whose attributes often include yarn and thread. The cultural association of fate and thread is recurrent universally: consider, for example, the Greek and Roman personifications of fate, Moiras and Parcas, who determined human destinies, and in particular the span of a person's life and his allotment of misery and suffering; also Slavic equivalents with the same attributes named Dolia, Sretcha, Rojenice, Sujenice, and Sudice; the Scandinavian Norns, often depicted as young women spinning yarn; the Lithuanian Laima and Werpeja; and the Latvian Kārta, also spinners of the thread of life (see Afanasiev). Hence, "The Grey Duchesse" builds on the primeval conceptual association of [FATE] with [THREAD], which is universal, stable, and recurrent. In the modern form, its manifestations can be found in various beliefs about the protective and beneficial properties of a thread/string; for example, a protective talisman (red string) is worn around the wrist in the modern Jewish and Israeli tradition, and "the string symbolically functions as a boundary marker between life and death" (Teman 32). Yet another modern manifestation is the "Martenitsa," a bicolor thread talisman omnipresent in the Balkans, which appears in March and symbolizes a birth of spring and new beginning (see Golant).
It is impossible to know one's destiny, and that is why the association of [FATE] with [FEAR] is similarly stable. Fate is rarely benevolent, as exemplified by Marta's clairvoyance. It is related to her ability to foresee, but by the same token her special vision imbues her with a remarkable ability to combine colors in her embroidery. It is not accidental, therefore, that when Marta loses her ability to predict death, she [End Page 92] also loses her unusual ability to see and combine colors (color symbolism is yet another level of analysis, which is outside of the scope of this article).
To reiterate, the analysis presented here only intends to capture a specific conceptual structure and to compare it with its other manifestations. By no means does it imply that this story has no other dimensions. For example, according to Larissa Naiditch, specialist in Nordic linguistics, it would be interesting to trace the Scandinavian mythological underpinnings in "The Grey Duchesse," and to study its connection with the mythological Norns, as well as to investigate further the "Prophecy of Death/Marked by Death" motif in light of the following linguistic evidence: Old Icelandic feigr ("person marked by death") etymologically relates to the Middle English fey ("fated to die"), Old English fǣġe ("doomed to die," "timid"), Old Saxon fēgi, Dutch veeg ("doomed, near death"), Old High German feigi ("appointed for death," "ungodly"), and Icelandic feigur ("doomed to die") (Larissa Naiditch, personal communication, Jerusalem, April 2019).
To recapitulate, in this section, following Freidenberg, I used the motif compass method, which led me from the reduced narrative scheme [DEATH] AND [EMBROIDERY] to the extended conceptual diagram. In the following sections, I demonstrate that the motif compass reliably and predictably yields similar conceptual links between the elements of the above diagram in various unrelated narratives.
Manifestations of the "Grey Duchesse" Conceptual Diagram in Unrelated Narratives
One of the most famous examples of the association of [THREAD] and [FEAR] is the case of Madame Defarge—one of the world's most sinister knitters and the antagonist of Charles Dickens's novel A Tale of Two Cities (1859). In the novel, bloodthirsty Madame Defarge constantly knits the names of the aristocrats whom she plans to send to the guillotine: she encodes the names in her knitting patterns, passes them to executioners, points at the condemned with her sharp needle from her perch near the [End Page 93] guillotine, and is generally connected with executions. Madame Defarge knits and causes fear and death.
Hence, the underlying conceptual links are:
[KNITTING] ➔ [FEAR] AND [DEATH]
Madame Defarge's image is very powerful; it imprints indelibly into a reader's memory. Similarly to Marta from Jansson's story, through the same association of knitting with thread, Madame Defarge is a personification of Fate, and is connected to death. Does this powerful image seem a unique product of a writer's imagination? Yes and no, if we follow the motif compass method: the association between Madame Defarge's sinister knitting and death is much deeper, and has to do with the association on a conceptual level.
The historical facts are in favor of this explanation. Madame Defarge was a personification of a collective fear that filled the public during the French Revolution. "The Knitters of the Guillotine" (les Tricoteuses de la Guillotine) existed in reality—they were working-class women who attended the executions while habitually keeping their hands busy with knitting. They were a common sight accompanying guillotines, were surrounded by fear, and eventually became personages of a mythical scale and even became known as "the furies of the guillotine. [. . .] The tricoteuse [. . .] evokes feelings of violence, hate, death, and blood [. . .] her fetish, the needle, is really a weapon" (Godineau xviii). The tricoteuses' behaviour was allegedly vicious, and it was believed that they were ubiquitous on the Paris streets, harassing suspected aristocrats. The popular imagery of the French Revolution was permeated by the sinister knitters; for example, in 1862, the London magazine The Dark Blue published a likely fictional account of a tricoteuse:
The instinct of the fury of the Revolution detected an aristocrat in the quiet lady who [. . .] was walking timidly along the terror-laden streets. The tricoteuse confronted, and stopped abruptly opposite the poor lady, who shrank before the fierce and baleful eyes out of which glared hatred and suspicion [. . .] "I promise you that I suspect you. You are no true patriot, I see . . ." The nostrils of the woman of the guillotine were dilated as if they scented blood.(Buck; emphasis original)
A sinister knitter of the French Revolution can also be found in Baroness Orczy's novel The Scarlet Pimpernel (1905). According to Godineau, "the tricoteuse [. . .] evokes feelings of violence, hate, death, and blood [. . .] allying the most delicate tenderness with the most extreme violence" (xviii; emphasis original).
Although a knitting woman is mundane, that is, neutral, the symbolism of her image is not: the irrational mass fear of knitting women which emerged during the French Revolution is a manifestation of the conceptual linkage in the above conceptual diagram:
[THREAD] ➔ [[DEATH] AND [FATE]] [End Page 94]
[End Page 95]
Indeed, the historical evidence points towards the existence of the urban myths and legends surrounding the "Knitters of the Guillotine," and it is assumed by Godineu that the negative reaction to the fact that women implicated themselves in politics, a typically masculine sphere, was mostly due to their crossing the boundary of the private sphere where they belonged socially (xxxiii–ix, 27, 276). Although such a feminist interpretation is valid and adequate on the ideological level, it does not explain the choice of the imagery, its persistence, and the scale of mass fear caused by it—that is, it does not explain the patterns of its co-occurrence (i.e., knitting and death) on a conceptual level.
This is where Freidenberg's motif compass method is invaluable, because it demonstrates that the stable association of knitting with death is highly predictable. Then it is safe to assume that, in the popular imagination, Madame Defargue and other Knitters of the Guillotine were perceived as sinister because of their knitting and its conceptual association with fate and death. The question for further investigation is: does the above belong to the domain of fiction or to that of historical narrative?
Two notable and well-known symbolic representations of a thread as related to fate in literature (through its conceptual association with knitting and embroidery) are found in Joseph Conrad's Heart of Darkness (1899) and in Nathaniel Hawthorne's The Scarlet Letter (1850).
In Conrad's Heart of Darkness, two women knitting black wool in the Company's head offices only appear briefly. But they are some of the "last faces" the protagonist sees (Mclntire 271), and the women, "at once comforting and sinister," are "reminiscent of the three fates of Greek myth who weave and unweave destinies regardless of individual wishes" (271). Despite their transitional place in the plot, they represent a very powerful image, which imprints in the protagonist's (and the reader's) memory, precisely because they personify Fate.
Hence, the underlying conceptual scheme of their image is the same as above:
[[WOOL] AND [KNITTING]] AND [[FATE] AND [DEATH]]
Likewise, in Hawthorne's The Scarlet Letter, thread and embroidery are not neutral and mundane activities, but are laden with symbolism. Embroidery in the novel manifests itself in an embroidered scarlet letter, which is connected to the inescapable destiny through the same conceptual association:
[[NEEDLEWORK] AND [EMBROIDERY]] AND [[THREAD] AND [FATE]]
It should be noted that, although it was demonstrated above that the conceptual association of [THREAD] with [FATE] was stable, its connotation was not always negative. In Hawthorne's novel, the symbolism of thread undergoes the following permutation: what starts as a symbol of the adulteress later becomes a symbol of righteousness (Roy 298). Such a change of valence is not surprising, and is in line with Freidenberg's theory, which demonstrates that a concept's connotations may bifurcate or shift (recall her [RIGHTEOUS ROBBER]). Freidenberg's prediction is that, once an image and its related concepts receive a stable cultural representation, their [End Page 96] symbolic association becomes robust, but with one caveat: the valence may change depending on the context.
Embroidery and weaving, as associated with fate, can also be found in Salman Rushdie's novel Shame (1983) and in Heinrich Heine's poem "The Silesian Weavers" (1844).2 In Shame, one of the protagonists, Rani Harappa, "embroidered shawl after shawl [. . .] persuading the villagers that she was composing the tapestry of their fate" (157). Although this motif is not central in the novel, the image is very powerful, and rests in the reader's memory: Rani's eighteen shawls, completed over six years, tell forbidden stories about atrocities and crime, and are condemned to be locked in a chest so as not to be seen by anyone. Here again, Rushdie's choice of the image and its artistic development may be individual, but the image itself belongs to the world image inventory, and is universal.
In "The Silesian Weavers" by Heinrich Heine, weaving is not only a mundane activity, but also has a metaphorical meaning pertaining to the studied conceptual network. Taking into account the stability of the conceptual association of weaving with fate, the poem's refrain and the powerful imagery it conjures—"We're weaving, we're weaving!"—is not accidental, and has a symbolic significance:
They sit at the spinning wheel, snarling cheerless:Germany, we weave your funeral shroud,A threefold curse be within it endowed—We're weaving, we're weaving!
While the historical interpretation of this poem is valid (it refers to the Vormärz movement and the Silesian weavers, who in 1844 started an uprising against exploitation and wage decreases), if we follow the motif compass method, the powerful image is not simply a product of a unique constellation of a poet's imagination and real events, but is also rooted, through its universal imagery, in the deeper web of conceptual associations discussed above.
It was demonstrated above that weaving and its paraphernalia (yarn, spindle, etc.) is not a neutral, ordinary activity; in literature, it often manifests some symbolic associations and has variable connotations. The universality and depth of the thread symbolism are such that it frequently appears in emotionally charged contexts.
Thus, for example, knitting in nineteenth-century literature and discourse is not only mundane but also emblematic. According to Alison Lurie,
In nineteenth-century literature, it's often true that good women knit and bad women crochet or do fancy work. In Jane Austen's "Emma," the long-suffering good girl, Jane Fairfax, is a dedicated knitter, as is her aunt, Miss Bates. [. . .] On the other hand, Thackeray's anti-heroine Becky Sharp, in "Vanity Fair," practices fine netting in order to show off her long white fingers and captivate Josiah Sedley. [. . .] In Tolstoy's "Anna Karenina," Levin's loyal and lovable wife, Kitty, knits while she is in labour with her first child. Anna herself crochets nervously and automatically [reflecting her status as an adulteress].(Lurie) [End Page 97]
The [MYSTERY] element of the conceptual diagram, as it was predicted, also emerges elsewhere in its association with knitting; for example, Agatha Christie's famous amateur sleuth, Miss Marple, is often depicted with a pair of knitting needles. In view of the aforementioned discussion, the conceptual connection of [KNITTING] with [MYSTERY] in Christie's Miss Marple is not accidental either: in this context, knitting it is not a neutral mundane activity, but has a deeper symbolic meaning.
One interesting case of the conceptual association of [KNITTING] with [FATE] can be found in twentieth-century "patriotic knitting" narratives. As Lurie maintains, "the tradition of patriotic knitting for the military continued for years, both in fiction and in real life." Apart from its practical side, knitting in these narratives is surrounded by a high degree of uncertainty and legends. It was believed, for example, that wartime spies "used knitting to encode messages, [and] the message was a form of steganography, a way to hide a message physically" (Zarelli). Yet, as plausible as it may seem, Zarelli maintains that a 1918 Pearson's Magazine article's report that Germans were knitting whole sweaters to send messages was "perhaps an exaggeration." Along similar lines, Zarrelli further maintains:
As with many things spy-related, getting the proof and exact details on code knitting can be tricky; much of the time, knitters used needles and yarn as a cover to spy on their enemies without attracting suspicion. Knitting hidden codes was less common.
The Pearson's account of code knitting seems a bit convoluted, but the rumors were not pure fantasy. Because women were encouraged to knit socks, hats, and balaclavas for soldiers during many conflicts, including the American Civil War, and the World Wars, knitting and textile work was a common sight—and one that could be easily used to the spy's advantage.(Zarelli)
Knitting, though by itself a mundane activity, thus receives a symbolic meaning and becomes related to the concept of fate when placed in a patriotic context. In twentieth-century patriotic knitting narratives, the conceptual association was as predictable and robust as it was two centuries before in the context of the French Revolution. Hence, it was not accidental that it also triggered similar emotions of fear and suspicion.
Although some of the accounts of spy-knitters seem credible, some of them do not, which, in view of the above conceptual associations, is not surprising, and may warrant additional investigation. As has been demonstrated throughout this article, once a conceptual link is established culturally and historically, it becomes stable and recurrent, and permeates the popular imagery. Subsequently, concepts become further intertwined, and their resulting connections are reinforced through a variety of vehicles ("ethnographic" manifestations, narratives, etc.). The images underlying specific concepts are stored in, and retrieved from, the collective or implicit individual memory. This is in line with Freidenberg, who maintained that searching for an origin of any specific motif or symbol would be futile, because nobody was the first to "invent" it or "use" it. The very fact of its recurrence points to its universality. [End Page 98]
The number of literary and historical parallels to the above may be increased, and it is important to emphasize that the examples in this article are used only to illustrate the predictability of co-occurrence of conceptual patterns and the validity of the motif compass method (rather than amassing data and finding further parallels). Development of a data mining algorithm based on conceptual diagrams similar to the one discussed in this article might be a task for the future.
To summarize this section, one may say that any given manifestation of a set of conceptual links is just a tip of the iceberg. The relevant concepts are intertwined semantically and culturally, and are loaded with symbolic meaning. Once they appear in narratives, they often play a pivotal role in plot development or make indelible imprints on recipients, invariably evoking in them strong emotions and some kind of recognition.
In this article, I revisited the motif compass method by Olga Freidenberg, demonstrating that it can be a valuable tool in narrative investigation and that it can be useful for studying narrative DNA. I argued that Freidenberg's scholarly legacy is momentous, and that it needs to gain more recognition, validity, and currency, especially in modern narratology. Following V. V. Ivanov, I maintained that Freidenberg's innovative approaches, as well as her ability to bring together evidence from a broad range of data, might have deep theoretical implications for the cognitive sciences [End Page 99] and for narratology. Similar analysis has been done before, for example by Valeria Eremina, who studied the "Temporary Death/Co-Death" motif on a wide range of ethnographic and literary data ranging from funeral customs and rituals to literary sources such as Sigrid Undset's trilogy Kristin Lavransdatter (Eremina 581–621).
In presenting Freidenberg's concepts, I deliberately omitted such traditional terminology as "archaic" and "archetype," which is used by her other interpreters (e.g., Braginskaya, Perlina, and Moss). Such omission was due to this article's goal—to present Freidenberg's method and its interpretation within the framework of contemporary approaches to narrative.
Using Freidenberg's method, I built a conceptual diagram connecting related concepts that emerge in unrelated narratives. In its turn, the diagram helped unveil patterns of co-occurrence of the conceptual elements, and demonstrated that they had stable, predictable, and universal associations across a variety of narratives. Hence, the presented method of the motif compass is reliable and can be used for data mining and machine learning, provided that comparable extended conceptual diagrams are built and annotated in advance.
Such annotation ideally should be carried out by trained folklorists, who have a proven capacity to recognise recurrent motifs in their various disguises. Because any human activity yields narratives, narratology is essentially an interdisciplinary field (see Sommer), and folkloristics, with its well-documented ability to identify "mutations" in various narratives, is essential for studying narratives in various forms. [End Page 100]
Olga Levitski is an independent scholar affiliated with The Propp Centre for Humanities-Based Research in the Sphere of Traditional Culture. Olga has obtained a graduate degree from the University of St. Petersburg, Russia, specializing in comparative folklore under Professor V. I. Eremina. Olga also holds an MA in Theoretical Linguistics from York University in Toronto, Canada, and is a member of the International Society for Folk Narrative Research.
Note on the transliteration of Russian names: There is a variation in transliterations of some of the Russian names cited in this article—for example, Braginskaia and Braginskaya; Freidenberg and Freudenberg; Ol'ga and Olga. For consistency purposes, I use the same transliteration for all instances throughout the article.
Note on translations of Russian texts: Unless otherwise noted, translations of original Russian texts are my own.
1. I am grateful to Hebrew University Professor Larissa Naiditch, who pointed my attention to Aitchinson's book.
2. This example was kindly provided by Professor Larissa Naiditch, Hebrew University of Jerusalem (personal communication).