Project Report
Introduction
Folk fairy tales, like folklore in general, differ from literature in that they lack a specific author and often reflect the beliefs, customs, and cultural practices of the community in which they were created.1 Because of their simultaneous ability to reflect a culture and perpetuate its values, the study of the messages that fairy tales contain lends itself well to an examination of gender roles. Fairy tales have been both criticized and praised for their depictions of young women and girls. Is the princess a passive object awaiting the salvation of her prince, or an intelligent and kind role model who teaches young girls important life lessons?2
These types of questions have become particularly interesting to many writers and scholars not only because of the way in which fairy tales mirror society, but also because of the increasing prevalence of fairy tales in current culture. Because of this, it is often well-known tales that are viewed as culturally relevant and thus become the topic of research and discussion. Feminist scholars are frequently interested in popular tales because they are more likely to affect the acculturation of children.3
One way to explore the gender implications of folk fairy tales is to examine the relationship between gender and agency. The allocation of power can be analyzed through the type and amount of speech associated with different male and female characters. It is through speech that characters curse, bestow blessings, perform magic, and solve riddles. The ability to perform a vocalization, whether it be conversing with magical creatures or singing to enchant a prince, often affects not only a character’s outcome within the narrative, but also the reader’s perception of the character. Conversely, the inclusion of silence may also confer certain connotations and affect the course of the narrative.
The relationship between speech and gender in fairy tales has been researched in tales from the German, French, and, Italian traditions with mixed results. In her research on tales from the Grimm corpus, Ruth Bottigheimer found a strong correlation between speech and gender.4 In the tales she studied, both positive male characters and negative female characters speak frequently, while positive female characters are often silent. This trend is not found in comparable tales from the French and English traditions.5 In addition to the unequal distribution of speech, the allotment of verbs forms a hierarchy in which more active verbs, such as to “ask,” are associated with men, while more passive verbs, such as to “answer,” are commonly associated with women. Bottigheimer attributes the speech patterns in the Grimm corpus to the valorization of female silence and passivity found in German society during the period in which the tales were being collected.6 The tales from the Grimm corpus were collected and assembled by the Grimm Brothers in the early nineteenth century. The Russian tales analyzed as part of my research were collected roughly fifty years later. Alexander Afanas′ev, who is commonly regarded as the Russian counterpart to the Brothers Grimm, released his first collection of Russian fairy tales in 1861.7
The primary goal of the research discussed in this paper is to determine the connections between speech and gender in Russian fairy tales and to examine how these connections may differ from those found in other fairy-tale traditions. Some of the questions my research seeks to answer revolve around not only the speech patterns of men and women, but also the speech patterns of positive and negative characters. For example, do witches and other negative female characters speak more often and more forcefully than their male counterparts? Do “good” women—whether heroines or fairy godmothers—speak words of wisdom or are they mostly silent? I am also interested in the relationship between gender and speech in certain tale typologies: Do positive female characters speak differently in “Wicked Stepmother” tales than they do in “Wise Maiden” tales?
Answering these questions involves examining not only the volume of speech, whether measured by the number of speech acts or length of speeches, but also the relative use of direct and indirect speech and the specific verbs of speaking associated with different character types. In order to retrieve this data from a large number of tales, this project has employed a variety of computing tools to collect and manage information. In this sense, the research conducted falls under the umbrella of digital humanities, as the project employed computing tools in order to answer humanities-based research questions.8 The tales that I have studied have been encoded using XML, which supports the formal association of specific textual moments with narrative functions.9 The speeches within each text have been tagged and the speaker has been associated with his or her speech. When a verb of speech occurs in conjunction with a speech act, it is also associated with that act and thus becomes linked with the speaker as well. Additionally, the research employs a number of ancillary XML-based technologies, including XSLT, which is used to extract, analyze, and display the information tagged in the original documents.
Methodology
All of the tales examined as part of this research are from the Russian tradition of folk tales. No previous research regarding the relationship between speech and gender in Russian tales has ever been done, and there is very little research available regarding the overall gender connotations and implications of the Russian tradition. Several factors, some textual and some historical, make Russian fairy tales a fitting area of research in terms of exploring the gender associations of fairy tales. The Afanas′ev collection originated not much later than the Grimm collection, making it a relevant point of comparison chronologically. In addition, several tale typologies found in the Russian tradition feature strong gender stereotypes. For example, both “Wise Maiden” and “Bad Wife” tales assume strong intrinsic connections between female speech and female character.
“Wise Maiden” tales are a type not discussed in research on the Grimm Corpus, and initially the nature of these tales seems to diverge from the principle that “good” women in tales are silent. In these tales the heroine assumes a less passive role than in other fairy tales, and manifests her wisdom through prescience, cleverness, and riddles. Bad Wife tales, on the other hand, would initially seem to support the preexisting hypothesis that “negative” women speak frequently and with an active voice: bad wives may be nags, scolds, or tyrants. The Afanas′ev collection also features a large number of “Wicked Stepmother” tales. The tales in this genre lend themselves well to a study of gender and power, since they are driven by positive and negative female characters locked in opposition to one another. One of the most famous “Wicked Stepmother” tales, “Cinderella,” has been highlighted as an example of positive females being relatively silent in the Grimm corpus.
Until now, the research done on the connections among speech, power, and gender in fairy tales has been limited to a small sample of tales. Ruth Bottigheimer’s study on the use of speech in the Grimm corpus is limited to fifteen tales, which she categorizes according to genre, with six of them falling under the category of “popular” tales, including such titles as “Cinderella” and “Snow White.”10 In addition to the limited number of tales, only the five most commonly occurring verbs were studied: speak (sprechen), say (sagen), ask (fragen), answer (antworten), and cry out (rufen).11
In this paper I will be analyzing the use of speech in a larger sample of thirty-seven tales12 and I will be examining seventy-nine verbs of speech that appeared more than once throughout this sample of tales. In order to collect the data from a large number of texts effectively, it was necessary to employ a systematic approach to data collection, which enabled objectivity and a greater degree of accuracy. It would be impossible to record thousands of individual speech acts and determine statistical correlations without computing tools. The computing methods used to collect and process the information relevant to this research required the creation of digital texts that could be encoded to allow further analysis.
Rather than transcribe the tales manually, the textual sources for all of the tales researched through this project were obtained from The Fundamental Digital Library of Russian Literature and Folklore (FEB).13 The FEB website functions as a full-text digital resource that accumulates information on Russian folklore and other types of literary and non-literary texts. This source was selected because of its academic nature and accurate representation of materials. All texts in the digital library include the original structure, pagination, orthography, punctuation, and graphics of the source editions.
Instead of tallying instances of speech and verb use manually and entering the data into a spreadsheet, I used XML, eXtensible Markup Language, to encode relevant information within the texts. XML allows for the use of descriptive encoding, which ascribes specific and unique meaning to certain aspects and portions of a text. The XML-based approach, which extracts the analytical data directly from the prose text, minimizes the opportunity for error by obviating the need to enter the same information in multiple places. The data collected in this project is hierarchical in nature, which fits perfectly with the tree-based structure of XML. In addition, with XML the same file is being used to create a readable digitized version of a tale, to extract and organize the information about the speech in the tale, and to generate reports about the entire collection of tales. By using XML, any changes that are made, whether they involve adding a new tale or generating a new report, can be updated simultaneously in all views.
The hierarchy created to organize and represent the texts revolves around the connection between a speech act and the speaker and verb associated with it. By connecting speeches and verbs with a speaker, speeches and verbs also become linked with the gender and character attributes of a speaker. The basic model used to encode speech acts can be seen in the following example:
Наутро жена ]]>говорит]]>: ]]>«Поезжай, старик, проведай-ка дочь — что напряла она в ночь?»]]>14
The next morning the wife ]]>said]]>: ]]>«Go, old man, visit your daughter, see how much she has spun during the night?»]]>
This example includes both elements and attributes, two types of markup used to categorize information in the XML hierarchy. Elements contain a start tag, an end tag, and content, which is what occurs between the tags. The tags are bound by angle brackets and the end tag contains a forward slash after the initial bracket. The name of an element is called a generic identifier. In the preceding example, the generic identifiers of the two elements are, vb and speech.15 Attributes contain additional information about the elements with which they are associated. The attribute of the ]]> element is called @infinitive. This attribute contains the lexical infinitive of the verb being tagged by the ]]> element.
This model of markup allows for connections to be made among speeches, verbs, and speakers. Although the verb is tagged separately from the speech, the content of the @infinitive attribute matches that of the @verb attribute in the ]]> element, thus linking verbs with specific speeches and characters. The characters for each tale are listed in a character list at the beginning of the XML document, and each ]]> element in that list contains a @gender attribute (m, f, or mx for mixed gender groups) and a unique @id attribute that distinguishes the character from all others in the tale. The @speaker attribute within a ]]> element contains the same value as the @id attribute for that person in the character list, which establishes a formal association (amenable to computational processing) among the speech, the speaker, and the speaker’s gender.
Other attributes that can be included in the ]]> element are: @number, @role, and @value. The @number attribute specifies whether the character in question is singular (sg) or plural (pl). For example, a character such as “Ivan” would be singular, whereas “geese” would be plural. The value of the @role attribute can be one of seven character types set forth by Vladimir Propp: father, princess, hero, donor, helper, villain, or dispatcher.16 The role of a specific character is determined by his or her function within the narrative. The @value attribute specifies whether a character is positive or negative. Whether a character is determined to be “good,” “bad,” “neutral,” or “other” depends on his or her role within the text. Because of the connection among character attributes, speeches, and verbs, many queries can be executed based on the qualities of a given character.
Not all of the speeches within the tales occur in a straightforward form and the preceding model alone does not adequately express the more complex vocalizations within the texts. A series of additional attributes contained within the ]]> element serve to describe further variances within the text. For example, the @type attribute categorizes instances of indirect speech, as in the following:
Вот девушка хлопочет у печи, а сама горько ]]>плачет]]>]]>.17
While the girl is busy at the stove, she ]]>cries]]>]]> bitterly.
In this example, crying is an instance of indirect speech. Indirect speech is reported speech; unlike direct speech, it does not include the vocalized content in the form of a first person quotation. In this case, since the speech does not have any textual content, the indirect speech is tagged with an empty speech element following the verb. Some instances of indirect speech contain content which is demarcated by a start and end tag. This system of encoding direct and indirect speeches is important for analysis, since the presence of direct speech implies a more active voice while the allocation of indirect speech can be seen as a form of narrative silencing.18
The tagging of the content within both direct and indirect speeches means that not only can the number of speeches be counted, but the length of a speech can be calculated as well.19 In addition, the speeches themselves can be pulled from the text. In each of these cases, because of the way they tagged, the speeches being queried can be chosen based on a number of characteristics. These can include the attributes of the speech itself, such as whether or not a speech is direct or indirect, as well as properties associated with the speaker and therefore indirectly (from the perspective of formal markup) with the speech, such as the character’s gender or role.
The @sn attribute allows such queries to return the correct count and content of speeches even if some quotations include content that is not part of the speech being quoted. For archeographic reasons I reproduced the texts exactly as they appeared in the source. However, the punctuation conventions employed in that source often failed to distinguish speech from descriptions of speech acts. For example:
]]>«В избушке]]>, — ]]>подумал]]>, — ]]>лучше оставить дочь»]]>.20
]]>«The little hut ]]>, — ]]>he thought]]>, — ]]>is the best place to leave my daughter»]]>
The preceding excerpt includes “подумал” ("he thought") within the quotation marks, even though it is not part of the quoted text. The quote is tagged as two separate speech acts and the two portions are linked together by a shared speech number @sn attribute. This markup strategy uses the shared @sn attribute value to formalize the fact that the two instances of speech are parts of a single speech act. The quotation marks in the original source erroneously suggests that “подумал” ("he thought") was part of the speech. The model above ensures that when the contents of a speech are retrieved for analysis, descriptive words and phrases are not also returned. Even though the speeches are tagged separately, when speech frequency is calculated the @sn attribute allows the speeches to be calculated as one speech act.
The @vl attribute, which is contained within a ]]> element, has a value of negative (ng) and is used to describe instances of silence. Negative speech can take two slightly different forms, the first of which is a direct reference to a character’s silence using a non-negated verb: “Она молчала” ("She was silent"). The second implies silence by referring to a specific lack of speech with a negated verb: “Она не говорила” ("She did not speak"). Silence, the opposite of speech, can be viewed as a lack of agency and a removal of power. Given the prominent way in which silence shapes the portrayal of gender within the Grimm corpus21, the study of silence within the Afanas′ev collection is necessary in order to draw a more complete comparative analysis.
Of the five attributes that can be included in a ]]> element, only the @speaker attribute is mandatory. All vocalizations are performed by a character within the text, but the other attributes describe unique textual situations that are not true of all speeches. For example, not every speech has a verb connected with it:
Рыбка приплыла к берегу: ]]>«Что тебе, старик, надо?»]]>
The fish swan to the shore: ]]>”What do you need old man?” ]]>
In this example, there is no verb of speech associated with the fish’s question. Because of this, the Relax NG schema that is attached to the XML files specifies that the only attribute required in a ]]> element is a @speaker attribute. The schema further dictates that the value of the @speaker attribute must match the value of an @id attribute associated with one of the characters. In this way, the schema, which contains a blue print for the model of markup that has been employed, eliminates the potential for certain types of error and inconsistency. The schema also prevents multiple or incorrect values from being entered for the other attributes in the XML.
The only two attributes found in a ]]> element that can contain multiple values are the @speaker and @verb. Multiple characters performing the same utterance occurs in several places throughout the tales:
Приехали вместе в лес, отыскали ель, ]]>крикнули]]>: ]]>«Дверцы, дверцы, отворитеся!»]]>22
They arrived at the forest together, found the fir tree, and ]]>shouted]]>: ]]> “Little door, little door, open!” ]]>
In this example, two unique characters, the father and older brother, are shouting the same speech in unison. Although this is a single speech, it is two concurrent instances of identical male speech by different speakers, and for certain analytical purposes it is appropriate to count the speech twice. If the two speakers were of opposite genders, the example would be understood as containing one male speech and one female speech.
The inclusion of multiple values in the @verb attribute is slightly more complex, and can take two different forms. In the first, two different verbs refer to the same speech act:
]]>« Дочка, дочка!]]> — ]]> говорила ]]> мать. — ]]>Мы пойдем на работу, принесем тебе булочку, сошьем платьице, купим платочек; будь умна, береги братца, не ходи со двора»]]>. Старшие ушли, а дочка забыла, что ей ]]>приказывали]]>…23
]]>«Daughter! daughter!]]> — ]]>said]]> the mother. — ]]>We are going to work, we shall bring you back a bun, sew you a dress, and buy you a kerchief. Be careful, watch over your little brother, and do not leave the yard.»]]>. The parents left, and the girl forgot what they had ]]>ordered ]]>…
In this example, both the verb “say” and the verb “order” are associated with the mother’s speech, but she is nonetheless a single character who utters this speech only once. For this reason it would be incorrect to tag the example as if it contained two speeches by the mother, one associated with each verb. It would also be incorrect to associate only one of the two verbs with the speech, since each verb has its own connotations with respect to agency and neither is more relevant than the other in the terms of my analysis.
The second way in which multiple verbs can connect back to the same speech involves instances of the same verb referring back to a single instance of speech multiple times. For example:
Девушка пошла за водой, сидит у колодца и ]]>плачет]]>]]>; рыбка выплыла наверх и ]]>спрашивает]]> ее: ]]>«Об чем ты, красная девица, ]]>плачешь]]>…]]>24
The girl went to fetch water, sits at the well and]]>cries]]>]]>; the fish swam up to the surface and ]]>asks ]]> her: ]]>«Why, pretty girl, are you ]]>crying]]>…]]>
Here, the same verb “cry” is associated with the same speech act two separate times in the narrative. Since the speech act itself, which in this case is indirect, occurs only once, it should not be tagged as if it occurred twice. However, in this case the verbs in question do not have different connotations and the verb to “cry” is not associated with the girl twice because the girl was crying on two separate occasions. Rather, it is associated with her once and the fish references the girl’s indirect speech act. Because of this, the verb is only included once as a value for the @verb attribute. However, each instance of the verb is still tagged as a ]]> element.
In addition to its attributes, a ]]> element may also contain a nested ]]> element. This often occurs when a character mentions another character's vocalization:
]]>«Что ты, дитятко, ]]>плачешь]]>]]>?»]]>25
]]>«Why are you ]]>crying]]>]]> my child?"]]>
Unlike the fish’s referral to the little girl’s act of crying in the previous example from tale 292, here the mother is mentioning a vocalization that is not referrenced earlier in the tale. This is the first and only reference to this instance of crying. The creation of an embedded ]]> element captures the complexity of this type of speech. Because of the markup, all of the text spoken by the mother is categorized as such, including her reference to the fool's vocalization. Additionally, the fool's act of crying is tagged as an indirect speech in which the fool is the speaker.
The purpose of this complex network of elements and attributes is to render a thorough, consistent, and descriptive structure that allows connections to be made among all of the aspects relevant to the analysis. This structure is made accessible by a combination of XML-related technologies. This research has primarily utilized XPath to traverse the XML hierarchy and XSLT to gather, analyze, and display data. The following snippet of code is an example of the way in which these technologies have been used to answer the research questions presented earlier.
]]>
The preceding XSLT expression creates a variable that stores information that can be used to calculate the number of speeches in a tale. The first line of code specifies that the name of the variable being created is speeches. The second line of code contains an XPath expression and dictates that the variable speeches represents all of the ]]> elements that contain a @sn attribute that is not equal to a preceding @sn attribute. This ensures that the ]]> elements that are connected by the same @sn attribute are counted as one speech, as opposed to two separate vocalizations. This line of code finds all of the speeches within a tale and additional qualifiers can be added to search for speeches that are associated with a specific gender or type of character.
Findings
The frequency count of both direct and indirect speech for all 37 tales examined shows that men speak more frequently than women. Out of a total number of 1210 speeches present in the tales, 480 (39.67%) were spoken by females, 713 (58.93%) were spoken by males and 17 (1.40%) by characters of mixed gender, that is, groups of people whose gender is not specified or most likely mixed, such as “the townsfolk.”
Of the 1210 speeches in the corpus, 332 were indirect, comprising 27.44 % of the total. 194 indirect speeches had a male speaker, and 133 had a female speaker. Male characters perform more indirect utterances overall, and the percentage of male and female speeches that are indirect is almost the same: 27.21% of male speeches and 27.71% percent of female speeches were categorized as indirect.
There are 79 verbs of speech that appear more than once within the tales. The 7 most frequently occurring verbs are: говорить (speak), сказать (tell), спрашивать (ask), отвечать (answer), думать (think), рассказать (recount), and велеть (order). The following chart shows how frequently each of these verbs is associated with the speech acts of male and female characters.
| Verbs | Female Count | Female Pct | Male Count | Male Pct | Total |
| Говорить | 68 | 36.36% | 119 | 63.64% | 187 |
| Сказать | 50 | 61.73% | 31 | 38.27% | 81 |
| Спрашивать | 20 | 36.36% | 35 | 63.64% | 55 |
| Отвечать | 20 | 45.45% | 24 | 54.55% | 44 |
| Думать | 6 | 17.65% | 28 | 82.35 % | 34 |
| Рассказать | 7 | 33.33% | 14 | 66.67% | 21 |
| Велеть | 9 | 29.03% | 22 | 70.97% | 31 |
The distributions of “спрашивать” (ask) and “отвечать” (answer) appears to diverge from Bottigheimer's observations regarding the Grimm corpus, in which women asked questions infrequently, but answered questions more often than men. “Велеть” (order) and “думать” (think) are the verbs that display the most uneven distribution between genders in the table. Some verbs are associated only with men or women, but none of these verbs appear more than 4 times in the sample.
The amount of speech in each of the 37 tales was also examined with respect to tale typology. “Wicked Stepmother” tales were characterized by the greatest use of speech by female characters in the corpus. Even though each of these 7 tales included at least one male character, the character that spoke the most often was always female. The single highest instance of female speech in all 37 tales occurred in the Russian counterpart to “Cinderella,” “Василиса прекрасная.” (“Vasilisa the Beautiful”)26 In this tale the heroine Vasilisa spoke 26 times, which amounts to 32.50% of the speech in the tale. This directly diverges from the speech allocation Bottigheimer found in the Grimm version of “Cinderella,” in which the heroine only receives 18% of the speeches while the prince, who speaks the most frequently, receives 31%.27 In three of the 7 “Wicked Stepmother” tales examined, the heroine speaks the most frequently.28 In an additional 3 tales the character who speaks the most is the villainous stepmother,29 and in 1 of the 7 tales it is the fairy godmother character, Baba Yaga.30
Conversely, “Bad Wife” and “Wise Maiden” tales were not characterized by frequent occurrences of female speech. This is particularly curious because the titles of most of these tales suggests that a female will play a predominant role in the narrative. In all but 1 of the 6 “Bad Wife” tales examined31 a male character speaks more frequently than the bad wife. In 4 of these 5 tales it is the husband, a positive male character, who speaks most frequently.32 In the other tale the young man who assists the husband speaks the most frequently.33 Similarly to “Bad Wife” tales, the 5 “Witch” tales studied feature a female antagonist central to the plot.34 In none of these tales does the witch character speak the most frequently, and in 4 of 5 the positive male character has the highest speech frequency.35 As previously noted, these patterns deviate from the speech hierarchy Bottigheimer observed in the tales from the Grimm corpus, since in those tales, it is the witches and other negative female characters that speak frequently and actively.
In 7 of the 8 “Wise Maiden” tales a male character speaks more frequently than any single female character.36 In several of the tales the frequency of female speech is negligible compared to that of male speech. Given the titles and plot lines of these tales, the infrequency of female speech is surprising. Each of the 8 titles refers directly to the wise maiden and 4 of the titles reference the heroine by her name.37 In "Мудрая девица и семь разбойников" (“The Wise Maiden and the Seven Robbers”)38 the wise maiden speaks significantly less often than her father, neighbor, and the male villains. The wise maiden only performs 7 of the 106 speech acts present in the tale. In 6 of the 8 tales, the total count of male speeches is greater than the total count of female speeches.39 Even in the tales that center around a seemingly active female, such as “Марья Моревна” ("Maria Morevna"),40 the heroine speaks less frequently than a male character. In “Марья Моревна” ("Maria Morevna") the warrior queen heroine, Марья (Maria), speaks only 12 times, while the male character that rescues her, Иван (Ivan), speaks 38 times.
Interestingly, the only “Wise Maiden” tale that features a female character that speaks more frequently than any single male character is “Василиса поповна” (“Vasilisa the Priest’s daughter”).41 In this tale, the heroine cross-dresses and presents herself as a male throughout the course of the narrative. This suggests that in addition to being associated with male behavior, such as drinking vodka and shooting, Vasilisa the Priest’s daughter is associated with the speech patterns commonly linked with males in “Wise Maiden” tales.
Further Research
While the initial findings seem promising, especially the results of the verb analysis and the speech patterns found in specific tale types, a larger sample of tales needs to be studied in order to have results that are more indicative of the corpus. Further research will focus on analyzing “Wicked Stepmother,” “Bad Wife,” and “Wise Maiden” tales. These tales all feature female characters that have distinctive roles and whose speech patterns initially seem to differ across typologies. Thus far the “Wicked Stepmother” tales have included both negative and positive female characters who speak frequently. Conversely, the negative female characters in “Bad Wife” tales and the positive females in “Wise Maiden” tales speak relatively little. Is this because they appear relatively late in the narrative, or are they present but silent? Does the frequency of their speech change over the course of the tale?
Further research will also examine the amount of speech as opposed to its frequency. Do principal characters with a small number of speeches have longer speeches, and therefore a greater verbal presence in the tale than can be discerned from counting speech verbs or speech acts? The concepts that emerged from the Grimm corpus, that positive female speech is infrequent and unobtrusive while negative female speech is frequent and active, may, for example, be mirrored in the length of speeches made by characters in the Russian tradition, even if not in the count of speech frequency.
Two of the longest tales already researched, “Соль” (“Salt”) and “Василиса прекрасная” (“Vasilisa the Beautiful)”42 feature narratives that revolve around a central protagonist. In “Соль” (“Salt”), the hero is the young son of a merchant who must go out into the world to acquire wealth and a bride. In “Василиса прекрасная” ("Vasilisa the Beautiful”) the heroine must undergo trials and tests in order to escape the cruelty of her stepmother and obtain the love of a King. The tales feature many lexical and narrative similarities, one of which is the frequent amount of speech performed by each tale's protagonist, even though they are of opposite genders. Both the hero in “Соль” (“Salt”) and the heroine in “Василиса прекрасная” ("Vasilisa the Beautiful”) speak more often than any of the other characters in their given tale. A comparison of these tales might suggest that speech patterns reflect the function of a character within a tale. Future research will attempt to answer the question: “Is speech frequency in Russian tales dictated by character roles more than by gender?” One potential way to explore this question is to measure when and how often characters appear in a tale.
Footnotes
1. For an alternative perspective on the creation of folklore, see: Bottigheimer, Ruth B. Fairy Tales a New History. Albany, NY: Excelsior Ed., 2009. Print.
2. For more information on a general feminist critique of popular fairy tales, see: Lieberman, Maria R. "Some Day My Prince Will Come: Female Acculturation through the Fairy Tale." College English Vol. 34, No. 3 (Dec., 1972), pp. 383-395.
3. Lieberman, Maria R. "Some Day My Prince Will Come: Female Acculturation through the Fairy Tale" p. 383-384
4. Bottigheimer, Ruth B. “Silenced Women in the Grimms' Tales: the ‘Fit’ Between Fairy Tales and Their Historical Context.” Ruth Bottigheimer, ed Fairy Tales and Society: Illusion, Allusion, and Paradigm. Philadelphia: University of Pennsylvania. 1986.
5. Bottigheimer, Ruth B. “Silenced Women in the Grimms' Tales: the ‘Fit’ Between Fairy Tales and Their Historical Context.” p. 125
6. Bottigheimer, Ruth B. “Silenced Women in the Grimms' Tales: the ‘Fit’ Between Fairy Tales and Their Historical Context.” pp. 115-119
7. Afanas′ev, Aleksandr Nikolaevich, Norbert Guterman, Alexander Alexeieff, and Roman Jakobson. Russian Fairy Tales. New York: Random House, 2006. Print.
8. For examples of Digital Humanities research projects see: http://dh.obdurodon.org/projects.html
9. For more information on XML see: http://www.w3.org/TR/xml/
10. Bottigheimer, Ruth B. Grimms’ Bad Girls and Bold Boys. New Haven: Yale University, 1987. pp. 177-192
11. Bottigheimer, "Silenced Women in the Grimms' Tales: the ‘Fit’ Between Fairy Tales and Their Historical Context." pp. 127-126
12. Afanas′ev, tales: 75, 96, 97, 98, 101, 102, 104, 113, 159, 169, 232, 233, 236, 242, 256, 267, 277, 292, 316, 326, 327, 328, 335, 345, 365, 366, 368, 369, 403, 404, 433, 434, 436, 437, 444, and 445. These tales were chosen both because of their typologies and because of their popularity in Russian culture.
13. The Fundamental Digital Library of Russian Literature and Folklore: http://feb-web.ru/indexen.htm
15. For the sake of clarity, all elements will by marked by angled brackets and all attributes will be marked with an atmark.
16. Propp, Vladimir. Morphology of the Folktale. Austin: University of Texas, 1968. Print.
18. Bottigheimer, "Silenced Women in the Grimms' Tales: the ‘Fit’ Between Fairy Tales and Their Historical Context." pp. 125-127
19. The length of a speech can be measured in several ways: word count, letter count, or syllable count. Syllable counting can be inferred from Russian orthography because except for a very small number of lexical exceptions, in Russian every vowel letter is syllabic and no non-vowel letter is syllabic, which means that a count of vowel letters can serve as a surrogate for a count of syllables. Given the inherently oral nature of the transmission of fairy tales, the syllable count may be a more useful measure of a character’s verbal presence than word or letter count. Syllable counts and word count are mentioned in the section of this paper entitled "Further Research."
21. Bottigheimer, "Silenced Women in the Grimms' Tales: the ‘Fit’ Between Fairy Tales and Their Historical Context."
27. Bottigheimer, Grimms’ Bad Girls and Bold Boys. p. 59 The percentages refer to the 1857 version of "Cinderella."
28. Afanas′ev, tales 104, 101, 292
29. Afanas′ev, tales 96, 97, 98
30. Afanas′ev, tale 102. Baba Yaga is a complex character in the Russian folk and fairy-tale tradition, corresponding in some ways to the wicked witch of western tales and in other ways serving as a magic helper, or Proppian donor. See: Johns, Andreas. Baba Yaga: the Ambiguous Mother and Witch of the Russian Folktale. New York: Peter Lang, 2004. Print.
31. Afanas′ev, tales 75, 256, 433, 434, 436, 437
32. Afanas′ev, tales 256, 433, 434, 436
34. Afanas′ev, tales 328, 365, 366, 368, 369
35. Afanas′ev, tales 328, 365, 366, 368
36. Afanas′ev, tales 159, 169, 232, 233, 236, 316, 327, 345
37. Afanas′ev, tales 159, 169, 236, 316
39. Afanas′ev, tales 159, 169, 233, 236, 327, 345
42. Afanas′ev, tales 242 and 104 respectively
.