The classic work on forms of the folktale.
This book is the classic work on forms of the European folktale.
- Preface to the Second Edition
- Introduction to the Second Edition
- Introduction to the First Edition
- Author's Foreword
- I. On the History of the Problem
- II. The Method and Material
- III. The Functions of Dramatis Personae
- IV. Assimilations: Cases of the Double Morphological Meaning of a Single Function
- V. Some Other Elements of the Tale
- A. Auxiliary Elements for the Interconnection of Functions
- B. Auxiliary Elements in Trebling
- C. Motivations
- VI. The Distribution of Functions Among Dramatis Personae
- VII. Ways in Which New Characters Are Introduced into the Course of Action
- VIII. On the Attributes of Dramatis Personae and their Significance
- IX. The Tale as a Whole
- A. The Ways in Which Stories Are Combined
- B. An Example of Analysis of a Tale
- C. The Problem of Classification
- D. On the Relationship of Particular Forms of Structure to the General Pattern
- E. The Problem of Composition and Theme, and of Themes and Variants
- F. Conclusion
Since the appearance of the English translation of Vladímir Propp's Morphology of the Folktale in 1958, there has been an ever increasing interest in attempting structural analyses of various folklore genres. In view of the enormous impact Propp's study has had on folklorists, linguists, anthropologists, and literary critics, one can only regret that there was a thirty-year time lag between Propp's completion of the Morphology in 1928 and the time that most European and American scholars read it.
The stimulating effect of Propp's seminal ideas is indicated in part by the number of studies it has inspired (Lévi-Strauss 1960, Dundes 1962a, 1964b, Bremond 1964, Greimas 1966b:172-221). To be sure, some of the studies are critical (cf. Taylor 1964), but from the criticism has come even more insight (e.g., Fischer 1963:288-289). Even though the flurry of activity initiated by the publication of Propp's Morphology has really barely begun, some preliminary comments may be made.
First of all, there seem to be at least two distinct types of structural analysis in folklore. One is the type of which Propp's Morphology is the exemplar par excellence. In this type, the structure or formal organization of a folkloristic text is described following the chronological order of the linear sequence of elements in the text as reported from an informant. Thus if a tale consists of elements A to Z, the structure of the tale is delineated in terms of this same sequence. Following Lévi-Strauss (1964: 312), this linear sequential structural analysis we might term "syntagmatic" structural analysis, borrowing from the notion of syntax in the study of language (cf. Greimas 1966a:404). The other type of structural analysis in folklore seeks to describe the pattern (usually based upon an a priori binary principle of opposition) which allegedly underlies the folkloristic text. This pattern is not the same as the sequential structure at all. Rather the elements are taken out of the "given" order and are regrouped in one or more analytic schema. Patterns or organization in this second type of structural analysis might be termed "paradigmatic" (cf. Sebag 1968:75), borrowing from the notion of paradigms in the study of language.
The champion of paradigmatic structural analysis is Claude Lévi-Strauss and it should be noted that he presented a paradigmatic model as early as 1955, that is, well before the English translation of Propp's work. The hypothetical paradigmatic matrix is typically one in which polar oppositions such as life/death, male/female are mediated. Lévi-Strauss is certainly aware of the distinction between Propp's syntagmatic structure and his paradigmatic structure. In fact, Lévi-Strauss's position is essentially that linear sequential structure is but apparent or manifest content, whereas the paradigmatic or schematic structure is the more important latent content. Thus the task of the structural analyst, according to Lévi-Strauss, is to see past or through the superficial linear structure to the "correct" or true underlying paradigmatic pattern of organization Lévi-Strauss 1955: 482, 1958:18; 1964:313). Although some of the differences between syntagmatic and paradigmatic analyses have been pointed out (cf. Waugh 1966:161), most folklorists are not aware of them and they wrongly lump both Propp and Lévi-Strauss together in the same category. (Propp himself attempted to comment on Lévi-Strauss's extended critique of the Morphology but this exchange is available only in the 1966 Italian translation of Propp's work to which Lévi-Strauss's 1960 critique and Propp's rejoinder are appended.) Generally speaking, the syntagmatic approach tends to be both empirical and inductive, and its resultant analyses can be replicated. In contrast, paradigmatic analyses are speculative and deductive, and they are not as easily replicated. (For examples of paradigmatic analyses, see the studies by Greimas, Leach, Sebag, and Köngäs and Maranda.)
One of the most important differences in emphasis between the syntagmatic and paradigmatic brands of structural analysis has been the concern or lack of concern with context. Propp's syntagmatic approach has unfortunately dealt with the structure of text alone, just as literary folklorists generally have considered the text in isolation from its social and cultural context (cf. Dundes 1964c). In this sense, pure formalistic structural analysis is probably every bit as sterile as motif-hunting and word-counting. In contrast, Lévi-Strauss has bravely attempted to relate the paradigm(s) he "finds" in myth to the world at large, that is, to other aspects of culture such as cosmology and world view. It is in this light that Lévi-Strauss's approach has helped lead to the new notion of myth (and other forms of folklore) as models. (Note that Malinowski's basically diachronic conception of myth as charter [set back in primeval time] has had to be updated to include a more synchronic conception of myth as model. The intellectual shift from "myth as charter" to "myth as model" is surely one significant consequence of synchronic structural analysis.) However, the emphasis upon context is rather one of application of the results of structural analysis than one inherent in the paradigmatic approach. The problem is that Propp made no attempt to relate his extraordinary morphology to Russian (or Indo-European) culture as a whole. Clearly, structural analysis is not an end in itself ! Rather it is a beginning, not an end. It is a powerful technique of descriptive ethnography inasmuch as it lays bare the essential form of the folkloristic text. But the form must ultimately be related to the culture or cultures in which it is found. In this sense, Propp's study is only a first step, albeit a giant one. For example, does not the fact that Propp's last function is a wedding indicate that Russian fairy-tale structure has something to do with marriage? Is the fairy tale a model, a model of fantasy to be sure, in which one begins with an old nuclear family (cf. Propp's typical initial situation "The members of a family are enumerated" or Function 1, "One of the members of a family is absent from home") and ends finally with the formation of a new family (Function 31, "The hero is married and ascends the throne")? Whether this is so or not, there is certainly no reason in theory why the syntagmatic structure of folktales cannot be meaningfully related to other aspects of a culture (such as social structure).
Many other fruitful areas of investigation are opened up by Propp's study. To what extent is Propp's Morphology an analysis of Russian fairy tales (as opposed to the fairy tales of other cultures)? Many, if not all, of the tales are Aarne-Thompson tale types and thus Propp's analysis is clearly not limited to Russian Materials. On the other hand, Propp's Morphology provides a useful point of departure for studies attempting to identify oicotypes. Von Sydow's notion of oicotype (1948:243) meaning a recurrent, predictable cultural or local variant must be amended in view of Propp's work to include oicotypes of structure as well as of content. Thus in addition to local penchants for specific content (motifs) within stable cross-cultural frames (such as Aarne-Thompson tale types), there may be culturally favored structural patterns (motifemic sequences) as well (cf. Dundes 1962b, 1964b:99-100).
Some of the other questions arising from Propp's work include: to what extent is Propp's analysis applicable to forms of the folktale other than the fairy tale? The English title Morphology of the Folktale is misleading. Propp limits his analysis to only one kind of folktale, that is, to fairy tales or Aarne-Thompson tale types 300-749. What about the other Aarne-Thompson folktale types? If, for example, Von Sydow is correct in grouping Aarne-Thompson tale types 850-879 under what he calls chimerates (the major portion of which are Aarne-Thompson types 300-749), then presumably Propp's analysis should also apply to this group of tales (cf. Von Sydow 1948:70). There is also the question of whether Propp's analysis might be applicable to non-Indo-European folktales. Attempts to study African tales (Paulme) and American Indian tales (Dundes 1964b) suggest that parts of Propp's Morphology may be cross-culturally valid.
Another question concerns the extent to which Propp's analysis applies to forms of folk narrative other than the folktale. For example, what is the relationship of Propp's Morphology to the structure of epic? (In this connection, it is noteworthy that the last portion of the Odyssey is strikingly similar to Propp's functions 23-31.) To what extent does Propp's analysis apply to genres of folklore other than those of folk narrative? It would appear that the structure of folk dances and games may be illuminated by Propp's analysis (Dundes 1964a). And what of the structure of nonfolkloristic materials? If there is a pattern in a culture, it is by no means necessary that it be limited to only one aspect of that culture. Quite the contrary. Culture patterns normally manifest themselves in a variety of cultural materials. Propp's analysis should be useful in analyzing the structure of literary forms (such as novels and plays), comic strips, motion-picture and television plots, and the like. In understanding the interrelationship between folklore and literature, and between folklore and the mass media, the emphasis has hitherto been principally upon content. Propp's Morphology suggests that there can be structural borrowings as well as content borrowings.
Propp's Morphology may also have important implications for studies of thinking and learning processes. To what extent is the structure of the fairy tale related to the structure of the ideal success story in a culture? (This also asks whether actual behavior is critically influenced by the type of fairy-tale structure found in a given culture.) And how precisely is fairy-tale structure learned? Does the child unconsciously extrapolate fairy-tale structure from hearing many individual fairy tales? Do children become familiar enough with the general nature of fairy-tale morphology to object to or question a deviation from it by a storyteller? (This kind of question may be investigated by field and laboratory experiments. For example, part of an actual or fictitious (=nontraditional) fairy tale containing the first several functions of Propp's analysis could be presented to a child who would be asked to "finish" the story. His completion could be checked against the rest of Propp's functions. Or a tale could be told with a section left out, e.g., the donor sequence, functions 12-14, and the child asked to fill in the missing portion. Such tests might also be of value in studies of child psychology. Presumably, the kinds of choices made by a child might be related to his personality. For example, does a little boy select a female donor figure to aid him against a male villain? Does a little girl select a male donor figure to assist her against her wicked stepmother?) In any case, while there have been many studies of language learning, there have been very few dealing with the acquisition of folklore. Certainly children "learn" riddle structure almost as soon as they learn specific riddles. Propp's Morphology thus provides an invaluable tool for the investigation of the acquisition of folklore.
Finally, Propp's scheme could also be used to generate new tales. In fact, Propp's Morphology has been programmed for a computer (Dundes 1965). Such techniques might be of interest to those seeking new species of literature based on folk form and content, or to those seeking to show the traditional nature and limited number of the combinations of narrative motifs actually found in oral tradition as opposed to the total number of theoretically possible combinations. In addition, analysis of the "rules" by which tales or portions (Propp's moves) of tales are generated or transformed is clearly another research prospect made possible by Propp's pioneering study.
There can be no doubt that Propp's analysis is a landmark in the study of folklore. Despite the fact that there is no mention of it in the standard treatises on the folktale, Propp's Morphology will in all probability be regarded by future generations as one of the major theoretical breakthroughs in the field of folklore in the twentieth century.
Scholarly literature concerning the tale is not especially rich. Apart from the fact that few works are being published, bibliographical sources present the following picture: mostly texts themselves are published; there are quite a number of works concerning particular problems; there are no general works on the tale. Such works as do exist are of an informational rather than an investigatory nature. Yet it is precisely questions of a general character which, more than all others, awaken interest. Their resolution is the aim of scholarship. Professor M. Speránskij characterizes the existing situation in the following way: "Without dwelling on conclusions already reached, scientific anthropology continues its investigations, considering the material already collected as still insufficient for a generalized doctrine. Science, therefore, once again sets about the task of collecting material and evaluating it in the interests of future generations. But what general conclusions will be made and when they can be made is still unknown."
What is the reason for this helplessness, and why has the study of the tale found itself up a blind alley?
Speránskij places the blame on an insufficiency of material. But ten years have elapsed since the above lines were written. During this period the major three-volume work of Bolte and Polívka, Anmerkungen zu den Kinder- and Hausmdrchen der Brilder Grimm, has been completed. In this study, each tale is presented with its variants from the entire world. The last volume ends with a bibliography which lists sources, i.e., all collections of tales and other materials which contain tales that are known to the authors. This listing consists of about 1200 titles. It is true that among these materials there are those which are incidental and insignificant. But there are also major collections, such as the Thousand and One Nights, or the 400 texts of the Afanás'ev collection. But that is not all. An enormous amount of tale material has not yet been published, and, in part, not even described. It is in private hands or stored in the archives of various institutions. Specialists do have access to some of these collections. The Folktale Commission of the Geographic Society, in its Research Survey for the Year 1926, registers 531 tales as being available to its members. The preceding survey cites approximately three times as many examples. Thanks to this, the material of Bolte and Polivka can, in certain instances, be augmented. If this is so, then just how great is the number of tales that we have at our disposal in general? And moreover, how many researchers are there who have fully covered even the printed sources?
It is impossible, under these circumstances, to say that "the material already collected is still insufficient." What matters is not the amount of material, but the methods of investigation. At a time when the physical and mathematical sciences possess well-ordered classification, a unified terminology adopted by special conferences, and a methodology improved upon by the transmission from teachers to students, we have nothing comparable. The diversity and the picturesque multiformity of tale material make a clear, accurate organization and solution of problems possible only with great difficulty. Let us examine the manner in which the study of the tale has been carried out and the difficulties which confront us. The present essay does not have the aim of systematically recounting the history of the study of the tale. It is impossible to do so in a brief, introductory chapter; nor is this necessary, since this history has already been treated more than once 4 We shall try only to elucidate critically several attempts at the solution of some basic problems in the study of the tale and at the same time introduce them to the reader.
It is scarcely possible to doubt that phenomena and objects around us can be studied from the aspect of their composition and structure, or from the aspect of those processes and changes to which they are subject, or from the aspect of their origins. Nor is it necessary to prove that one can speak about the origin of any phenomenon only after that phenomenon has been described.
Meanwhile the study of the tale has been pursued for the most part only genetically, and, to a great extent, without attempts at preliminary, systematic description. We shall not speak at present about the historical study of the tale, but shall speak only about the description of it, for to discuss genetics, without special elucidation of the problem of description as it is usually treated, is completely useless. Before throwing light upon the question of the tale's origin, one must first answer the question as to what the tale itself represents.
Since the tale is exceptionally diverse, and evidently cannot be studied at once in its full extent, the material must be divided into sections, i.e., it must be classified. Correct classification is one of the first steps in a scientific description. The accuracy of all further study depends upon the accuracy of classification. But although classification serves as the foundation of all investigation, it must itself be the result of certain preliminary study. What we see, however, is precisely the reverse: the majority of researchers begin with classification, imposing it upon the material from without and not extracting it from the material itself. As we shall see further, the classifiers also frequently violate the simplest rules of division. Here we find one of the reasons for the "blind alley" of which Speránskij speaks. Let us consider a few examples.
The most common division is a division into tales with fantastic content, tales of everyday life, and animal tales. At first glance everything appears to be correct. But involuntarily the question arises, "Don't tales about animals sometimes contain elements of the fantastic to a very high degree?" And conversely, "Don't animals actually play a large role in fantastic tales?" Is it possible to consider such an indicator as sufficiently precise? Afanás'ev, for instance, places the tale about the fisherman and the fish among animal tales. Is he correct or not? If not, then why not? Later on we shall see that the tale ascribes with great ease identical actions to persons, objects, and animals. This rule is mainly true for so-called fairy tales, but it is also encountered 1n tales in general. One of the best-known examples in this regard is the tale about the sharing of the harvest ("I, Misa, get the heads of the grain; you get the roots"). In Russia, the one deceived is the bear; in the West, the devil. Consequently, this tale, upon introduction of a Western variant, suddenly drops out of the group of animal tales. Where does it belong? It is obviously not a tale of everyday life either, for where in everyday life does one find a harvest divided in such a way? Yet this is also not a tale with a fantastic content. It does not fit at all within the described classification.
Nevertheless, we shall affirm that the above classification is basically correct. Investigators here have proceeded according to instinct, and their words do not correspond to what they have actually sensed. Scarcely anyone will be mistaken in placing the tale about the firebird and the grey wolf among the animal tales. It is also quite clear to us that even Afanás'ev was wrong concerning the tale about the goldfish. But we see this not because animals do or do not figure in tales, but because fairy tales possess a quite particular structure which is immediately felt and which determines their category, even though we may not be aware of it. Every investigator who purports to be classifying according to the above scheme is, in fact, classifying differently. However, in contradicting himself, he actually proceeds correctly. But if this is so, if in the basis of classification there is subconsciously contained the structure of the tale, still not studied or even delineated, then it is necessary to place the entire classification of tales on a new track. It must be transferred into formal, structural features. And in order to do this, these features must be investigated.
However, we are getting ahead of ourselves. The situation described remains unclarified to the present day. Further attempts have not brought about any essential improvements. In his famous work The Psychology of Peoples, Wundt proposes the following division: (1) mythological tale-fables (Mythologische Fabelmärchen); (2) pure fairy tales (Refine Zaubermärchen); (3) biological tales and fables (Biologische Märchen and Fabeln); (4) pure animal fables (Reine Tierfabeln); (5) "genealogical" tales (Abstammungsmärchen); (6) joke tales and fables (Scherzmärchen and Scherzfabeln); (7) moral fables (Moralische Fabeln)
This classification is much richer than the one previously quoted, but it, too, provokes objections. The "fable" (a term which one encounters five times in seven classes), is a formal category. The study of the fable is just beginning. It is unclear what Wundt meant by it. Furthermore the term "joke tale" is in general unacceptable, since the same tale might be treated both heroically and comically. Still further, the question is raised as to the difference between a "pure animal fable" and a "moral fable." In what way are the "pure fables" not "moral" and vice versa?
The classifications discussed deal with the distribution of tales into categories. Besides the division into categories, there is a division according to theme.
If a division into categories is unsuccessful, the division according to theme leads to total chaos. We shall not even speak about the fact that such a complex, indefinite concept as "theme" is either left completely undefined or is defined by every author in his own way. Jumping ahead, we shall say that the division of fairy tales according to themes is, in general, impossible. Like the division into categories, it too must be placed on a new track. Tales possess one special characteristic: components of one tale can, without any alteration whatsoever, be transferred to another. Later on this law of transference will be elucidated in greater detail; meanwhile we can limit ourselves to pointing out that Bába Jagá, for example, might appear in the most diverse tales, in the most varied themes. This trait is a specific characteristic of the tale. At the same time, in spite of this characteristic, a theme is usually defined in the following fashion: a part of a tale is selected (often haphazardly, simply because it is striking), the preposition "about" is added to it, and the definition is established. In this way a tale which includes a fight with a dragon is a tale "about fights with dragons"; a tale in which Koscéj appears is a tale "about Koscéj," and so on, there being no single principle for the selection of decisive elements. If we now recall the law of transference, it is logically inevitable that the result will be confusion, or, more accurately, an overlapping classification. Such a classification always distorts the essence of the material under examination. To this is added an inconsistency in the basic principle of division, i.e., one more elementary rule of logic is violated. This situation has continued to the present day.
We shall illustrate this situation by giving two examples. In 1924 there appeared a book on the tale by Professor Vólkov of Odessa. Vólkov states, from the very first pages of his work, that the fantastic tale comprises fifteen themes. These are as follows: (1) about those unjustly persecuted; (2) about the hero-fool; (3) about three brothers; (4) about dragon fighters; (5) about procuring brides; (6) about a wise maiden; (7) about those who have been placed under a spell or bewitched; (8) about the possessor of a talisman; (9) about the possessor of magic objects; (10) about an unfaithful wife; etc.
How these fifteen themes were arrived at is not indicated. If one looks into the principle of this division, one obtains the following: the first class is determined by the complication (what the complication actually is we shall see later); the second class is determined by the character of the hero; the third, by the number of heroes; the fourth, by one moment in the course of the action, and so forth. Thus, a consistent principle of division is totally lacking. The result is actually chaos. Do not tales exist in which three brothers (third category) procure brides for themselves (fifth category)? Does not the possessor of a talisman, with the aid of this talisman, punish his unfaithful wife? Thus, the given classification is not a scientific classification in the precise sense of the word. It is nothing more than a conventional index, the value of which is extremely dubious. Can such a classification be even remotely compared with a classification of plants or animals which is carried out not at first glance, but after an exact and prolonged preliminary study of the material?
Having broached the questiop of the classification of themes, we cannot pass over Aarne's index of tales without comment. Aarne is one of the founders of the so-called Finnish school. The works of this school form the peak of studies of the tale in our time. This is not the place to give due evaluation to this movement. I shall only point out the fact that a rather significant number of articles and notes on the variants of individual themes exist in scholarly literature. Such variants are sometimes obtained from the least expected sources. A great number of them have been gradually accumulating, but they have not been worked over systematically. It is chiefly to this that the attention of the new trend is directed. Representatives of this school seek out and compare variants of separate themes according to their world-wide distribution. The material is geo-ethnographically arranged according to a known, previously developed system, and then conclusions are drawn as to the basic structure, dissemination, and origins of the themes. This method, however, also evokes a series of objections. As we shall see later on, themes (especially the themes of fairy tales) are very closely related to each other. In order to determine where one theme and its variants end and another begins, one must first have made a comparative study of the themes of the tales, and have accurately established the principle of the selection of themes and variants. However, nothing of the kind exists. The transference of elements is not taken into account here either. The works of this school proceed from the subconscious premise that each theme is something organically whole, that it can be singled out from a number of other themes and studied independently.
At the same time, the fully objective separation of one theme from another and the selection of variants is by no means a simple task. Themes of the tale are so closely linked to one another, and are so mutually interwoven, that this problem requires special preliminary study before they can be extracted. Without such study the investigator is left to his own taste, since objective extraction is not yet possible.
Let us take one example. Among the variants of the tale "Frau Holle," Bolte and Polívka quote tale No. 102 from Afanás'ev (the well-known tale, "Bába Jagá). They also include a number of other Russian tales—even those in which the witch is replaced by mice or a dragon. But they do not include the tale "Morózko." Why not? For here we have the same expulsion of the stepdaughter and her return with gifts, the same sending of the real daughter and her punishment. Moreover, both "Morózko" and "Frau Holle" represent the personification of winter, even though in the German tale we have the personification in a female form, and in the Russian one, in a male form. But apparently "Morózko," because of the artistic vividness of the tale, became subjectively fixed as a special type of tale, a special independent theme which can have its own variants. In this way we see that there are no completely objective criteria for the separation of one theme from another. Where one researcher sees a new theme, another will see a variant, and vice versa. I have given a very simple example, but difficulties increase with the extension and augmentation of the material.
Be that as it may, the methods of this school, first of all, needed a list of themes. This was the task undertaken by Aarne.
His list entered into international usage and rendered the study of the tale an enormous service. Thanks to Aarne's index, a coding of the tale has been made possible. Aarne calls themes types, and each type is numbered. A brief, conventional designation of tales (in this instance: by reference to a number in the index), is very convenient. In particular, the Folktale Commission could not have described its material wtihout this list, since the synopsis of 530 tales would have required much space, and in order to become acquainted with this material it would have been necessary to read through all of the synopses. Now, one need only look at the numbers and everything is clear at first glance.
But along with these commendable features, the index also reveals a number of real insufficiencies. As a classification it is not free of the same mistakes that Vólkov makes. The basic categories are as follows: (1) animal tales, (2) tales proper, (3) anecdotes. We easily recognize the previous devices changed to a new form. (It is a bit strange that animal tales are apparently not recognized as tales proper.) Furthermore, one feels like asking, "Do we have such precise knowledge of the concept of the anecdote to permit our employing it with complete confidence?" (Cf., the term "fables" used by Wundt.)
We shall not enter into the details of this classification, but shall consider only the fairy tales, which Aarne places in a subclass. I should note here that the introduction of subclasses is one of the services rendered by Aarne, since until his time there had been no thorough working out of a division into genus, species, and varieties. The fairy tales comprise, according to Aarne, the following categories: (1) a supernatural adversary; (2) a supernatural husband (wife); (3) a supernatural task; (4) a supernatural helper; (5) a magic object; (6) supernatural power or knowledge; (7) other supernatural motifs. Almost the same objections pertaining to Vó1kov's classification can be repeated here. What, for instance, of those tales in which a supernatural task is resolved by a supernatural helper (which occurs very often), or those in which a supernatural spouse is also a supernatural helper?
True, Aarne does not really attempt to establish a scientific classification. His index is important as a practical reference and, as such, it has a tremendous significance. But Aarne's index is dangerous for another reason. It suggests notions which are essentially incorrect. Clear-cut division into types does not actually exist; very often it is a fiction. If types do exist, they exist not on the level indicated by Aarne, but on the level of the structural features of similar tales, about which we shall speak later. The proximity of plots, one to another, and the impossibility of a completely objective delimitation leads to the fact that, when assigning a text to one or another type, one often does not know what number to choose. The correspondence between a type and' a designated text is often quite approximate. Of the 125 tales listed in the collection of A. I. Nikíforov, 25 tales (i.e., twenty percent) are assigned to types approximately and conditionally, which Nikiforov indicates by brackets. If different investigators begin to attribute the same tale to various types, what will be the result? On the other hand, since types are defined according to the presence of one or another striking incident in them, and not on the basis of the construction of the tales, and since one tale is capable of containing several such incidents, then one tale can sometimes be related to several types at once (up to five numbers for one tale). This does not at all indicate that a given text consists of five tales. Such a method of delineation is, in reality, a definition according to components. For a certain group of tales, Aarne even departs from his principles and quite unexpectedly, and somewhat inconsistently, switches from a division according to themes to a division by motifs. This is the manner in which he designates one of his subclasses, a group which he entitles "About the stupid devil." But this inconsistency again represents an instinctively chosen correct approach. Later I shall try to show that study on the basis of small component parts is the correct method of investigation.
Thus we see that the problem of classification of the tale finds itself in somewhat sorry state. Yet classification is one of the first and most important steps of study. We need merely recall what a great significance Linnaeus' first scientific classification for botany. Our studies are still in their "pre-Linnaen" stage.
Let us move on to another most important area of tale investigation: to its factual description. Here we can observe the following picture: very often the investigators, in touching upon questions of description, do not bother with classification (Veselóvskij). On the other hand, classifiers do not always describe a tale in detail, but study only certain aspects of it (Wundt). if an investigator is interested in both approaches, then classification does not follow description, but description is carried on within the framework of a preconceived classification.
Veselóvskij said very little about the description of the tale, but what he did say has enormous significance. Veselóvskij means by "theme" a complex of motifs. A motif can be ascribed to different themes. ("A theme is a series of motifs. A motif de. velops into a theme." "Themes vary: certain motifs make their way into themes, or else themes combine with one another." "By theme I mean a subject in which various situations, that is, motifs, move in and out.") For Veselóvskij, motif is something primary, theme secondary. A theme is, for him, a creative, unifying act. From this we realize that study must be concerned not so much with themes as with motifs.
Had scholarship concerning the tale acquainted itself better with Veselóvskij's precept—"separate the question of motifs from the question of themes" (Veselóvskij's italics)—then many vague matters would already have been done away with.
Yet Veselóvskij's teaching on motifs and themes represents only a general principle. His concrete interpretation of the term "motif" cannot be applied anymore. According to Veselóvskij, a motif is an indivisible narrative unit. ("By the term 'motif' I mean the simplest narrative unit." "The feature of a motif is its figurative, monomial schematism; such are those elements incapable of further decomposition which belong to lower mythology and to the tale.") However, the motifs which he cites as examples do decompose. If a motif is something logically whole, then each sentence of a tale gives a motif. (A father has three sons: a motif; a stepdaughter leaves home: a motif; Iván fights with a dragon: a motif; and so on.) This would not be so bad if motifs were really indivisible; an index of motifs would then be made possible. But let us take the motif "a dragon kidnaps the tsar's daughter" (this example is not Veselóvskij's). This motif decomposes into four elements, each of which, in its own right, can vary. The dragon may be replaced by Koscéj, a whirlwind, a devil, a falcon, or a sorcerer. Abduction can be replaced by vampirism or various other acts by which disappearance is effected in tales. The daughter may be replaced by a sister, a bride, a wife, or a mother. The tsar can be replaced by a tsar's son, a peasant, or a priest. In this way, contrary to Veselóvskij, we must affirm that a motif is not monomial or indivisible. The final divisible unit, as such, does not represent a logical whole. While agreeing with Veselóvskij that a part is more primary for description than the whole (and according to Veselóvskij, a motif is, even by its origin, more primary than the theme), we shall eventually have to solve the problem of the extraction of certain primary elements in a different way than does Veselóvskij.
Other investigators have proved as unsuccessful as Veselóvskij. An example of a methodologically valuable approach can be found in the methods of Bédier. The value of Bédier's methods lies in the fact that he was the first to recognize that some relationship exists in the tale between its constants and variables. He attempts to express this schematically. The constant, essential units he calls elements, giving them the sign Ω. He labels the variables with Latin letters. The scheme of one tale, in this manner, gives Ω+a+b+c; another, Ω+a+b+c+n; a third, Ω+m+1+n; and so forth. But his essentially correct idea falls apart in its inability to specify the exact meaning of omega. What Bédier's elements are in reality and how to separate them remains unclarified.
The problems of the description of the tale have been relatively neglected in favor of the concept of the tale as something finished, or given. Only at the present time is the idea of the need for an exact description growing ever wider, although the forms of the tale have already long been discussed. And actually, at a time when minerals, plants, and animals are described and classified precisely according to their structure, at a time when a whole range of literary genres (the fable, the ode, drama, etc.) have been described, the tale continues to be studied without such a description. Sklóvskij has shown to what absurdities the so-called genetic studies of the tale have sometimes gone to when they fail to consider its forms. As an example he cites the well-known tale about the measurement of land by means of a hide. The hero of the tale obtains permission to take as much land as he is able to encompass with an ox hide. He cuts up the hide into strips and encompasses more land than the deceived party expected. V. F. Mílller and others tried to detect here the traces of a judicial act. Sklóvskij writes: "It appears that the deceived party (and in all its variants the tale is concerned with deception) did not protest against the seizure of the land because land was generally measured in this manner. The result is an absurdity. If, at the moment of the supposed performance of the tale's action, the custom of measuring land 'by as much as one can encircle with a belt' existed and was known both to the seller and to the purchaser, then not only is there no deception, but also no theme, since the seller knew what to expect." Thus, the relegation of the story to historical reality, without taking into account the particulars of the story as such, leads to false conclusions, in spite of the investigators' enormous erudition.
The methods of Veselóvskij and Bédier belong to a more or less distant past. Although these scholars worked, in the main, as historians of folklore, their methods of formal study represented new achievements which are essentially correct but which have not been worked out or applied by anyone. At the present time the necessity of studying the forms of the tale evokes no objections whatsoever.
Yet present-day scholarship sometimes goes too far in this regard. In the above-mentioned book of Vólkov, one finds the following mode of description: tales first of all decompose into motifs. Qualities of the heroes ("two wise sons-in-law and the third a fool"), their number ("three brothers"), the deeds of heroes ("the injunction of a father for someone to keep watch over his grave after his death, an injunction which is carried out by the fool alone"), objects (a hut on chicken legs, talismans), and so forth, are all considered to be motifs. Each such motif is given a conventional sign—a letter and a number, or a letter and two numbers. More or less similar motifs are marked by one letter with different numbers. At this point just how many motifs does one obtain by being really consistent and marking the entire content of a tale in this way? Vólkov gives about 250 designations (there is no exact listing). It is obvious that there is much omitted and that Vólkov did do some selecting, but how he did it is unknown. Having isolated motifs in this manner, Vólkov proceeds to transcribe tales, mechanically translating motifs into signs and comparing schemes. Similar tales, it is clear, give similar schemes. Transcriptions fill the whole book. The only "conclusion" that can be drawn from this transcription h is that similar tales resemble each other—a conclusion which is completely noncommittal and leads nowhere.
We see the nature of the problems investigated by scholars. The less experienced reader may ask: "Doesn't science occupy itself with abstractions which in essence are not at all necessary? Isn't it all the same whether the motif is or is not decomposable? Does it matter how we isolate basic elements, how we classify a tale, and whether we study it according to motifs or themes?" Involuntarily one feels like raising more concrete, tangible questions, questions closer to the average person who simply likes tales. But such a requirement is based on delusion. Let us draw an analogy. Is it possible to speak about the life of a language without knowing anything about the parts of speech, i.e., about certain groups of words arranged according to the laws of their changes? A living language is a concrete fact—grammar is its abstract substratum. These substrata lie at the basis of a great many phenomena of life, and it is precisely to this that science turns its attention. Not a single concrete fact can be explained without the study of these abstract bases.
Scholarship has not limited itself to the problems dealt with here. We have spoken only of those questions related to morphology. In particular, we have not touched upon the enormous field of historical research. This historical research may outwardly be more interesting than morphological investigations, and here a great deal has been done. But the general question of the origin of the tale is, on the whole, unresolved, even though here too there are undoubtedly laws of origin and development which still await elaboration. Instead, all the more has been done on specific questions. The mere enumeration of names and works makes no sense. We shall insist that as long as no correct morphological study exists, there can be no correct historical study. If we are incapable of breaking the tale into its components, we will not be able to make a correct comparison. And if we do not know how to compare, then how can we throw light upon, for instance, Indo-Egyptian relationships,. or upon the relationships of the Greek fable to the Indian, etc.? If we cannot compare one tale with another, then how can we compare the tale to religion or to myths? Finally, just as all rivers flow into the sea, all questions relating to the study of tales lead to the solution of the highly important and as yet unresolved problem of the similarity of tales throughout the world. How is one to explain the similarity of the tale about the frog queen in Russia, Germany, France, India, in America among the Indians, and in New Zealand, when the contact of peoples cannot be proven historically? This resemblance cannot be explained if we have wrong conceptions of its character. The historian, inexperienced in morphological problems, will not see a resemblance where one actually exists; he will omit coincidences which are important to him, but which he does not notice. And conversely, where a similarity is perceived, a specialist in morphology will be able to demonstrate that compared phenomena are completely heteronomous.
We see, then, that very much depends upon the study of forms. We shall not refuse to take upon ourselves the crude, analytical, somewhat laborious task which is further complicated: by the fact that it is undertaken from the viewpoint of abstract, formal problems. Such crude, "uninteresting" work of this kind is a way to generalize "interesting" constructions.
“Propp's work is seminal...[and], now that it is available in a new edition, should be even more valuable to folklorists who are directing their attention to the form of the folktale, especially to those structural characteristics which are common to many entries coming from even different cultures.”
“It was primarily Claude Lévi-Strauss who made Propp's book popular outside the small circle of Western Slavicists: he immediately recognized the importance of Propp's methodology not only for the study of the fairy-tale, but generally for the study of narrative folklore. [Lévi-Strauss] expressed his admiration for all those”'who for a long time have been Propp's successors without knowing it."”
Times Literary Supplement
“Morphology will in all probability be regarded by future generations as one of the major theoretical breakthroughs in the field of folklore in the twentieth century.”