Traditionally fairytales were analyzed by their plot; however, this approach omits tone, mood and character of the storytelling. Using Factor Analysis and Natural Language Processing we have identified five factors describing the mood of fairy tales.
Folklorists classified folktales by recurring pattern plots to organize and analyze them. The first such classification was done by the Finnish folklorist Antti Aarne (Aarne, 1910) for Scandinavian folktales and later enlarged by the American folklorist Stith Thompson (Thompson, 1928) to the international scale. Nowadays, Aarne-Thompson tale type index is regarded as one of the “most valuable tools in the professional folklorist's arsenal of aids for analysis” (Dundes, 1997). Similarly Russian folk tales were analyzed by Vladimir Propp in Morphology of the Folk Tale (Propp, 1928). However, while Aarne-Thompson classification scheme depends solely on narrative and a folktale can’t be both, an animal tale and a tale of the fantastic, but only one of them, Propp’s classification allows any combination of motifs and characters much in the same way that a plant could be described by a combination of leafs, trunks and roots.
Nevertheless, Propp’s work itself was criticized by Claude Lévi-Strauss for removing all verbal considerations from the analysis (even though the folktale's form is almost always oral) and all considerations of tone, mood, character, and, anything that differentiates one fairy tale from another (Levi-Strauss, 1976). Despite this disadvantage, these two systems are still popular and in use (Tatar, 2003).
The purpose of this explanatory research is to find factors describing the moods of fairytales in a similar way Big Five personality factors, defined by Louis Leon Thurstone, describe human behavior:
Sixty adjectives that are in common use or describing people ... were given to each 1300 raters. Each rater was asked to think of a person whom he know well and to underline every adjective that he might may us in a conversational description of that person ... the ... correlation ... coefficients for the sixty personality traits were then analyzed by means of multiple factor methods and we found that five factors are sufficient to account for the coefficients (Thurstone, 1934).
To see the actual results try some examples at the classification page.