Thoughts on “Morphology of the Folktale”, by Vladimir Propp

I.

Морфология сказки is the Russian name of a book written by Vladimir Propp and published in 1928. It was first translated into English in 1958 with the title Morphology of the Folktale, despite the fact that the Russian word сказки means “fairy tale”, and the book is indeed about fairy tales, specifically, not folk tales in general. The second-edition translation I read, which is from 1968, points this out in the introduction but opts to keep the familiar title of the 1958 translation. Really, then, this book’s English name should be Morphology of the Fairy Tale. But I don’t think anybody uses that title, so for lack of a satisfactory English title, I’m going to refer to the book by its Russian name, Морфология сказки.

Anyway… this is a very interesting book, and it’s introduced me to a whole field of study which I never even knew existed, that of the formal analysis of folklore. I was a little surprised to learn that this kind of analysis was even possible, since folk tales are phenomena of human creativity. I thought it would be like trying to analyse something like literary fiction; such analysis is of course possible (if I understand correctly, it’s what literary criticism is all about), but it doesn’t seem to lend itself well to a formal approach. On the other hand, Propp’s analysis of fairy tales in Морфология сказки is unmistakably formalist. It displays one of the hallmarks of a formalist approach: the development of a special system of symbols which are essential for presenting the analysis. As an illustration, I’ve written out the structure of one of the fairy tales analysed by Propp below, using Propp’s formal language. The fairy tale here is from Afanasiev’s Russian Fairy Tales (1855-1863); its Russian title is Гуси-лебеди, and its English title in the edition I have is “The Magic Swan Geese”.

γ1 β1 δ1 A1 C ↑ [D ¬E1 ¬F]3 d7 E7 F9 = G4 K1 ↓ [Pr1 D1 E1 F9 = Rs4]3

Note that I have made some minor adjustments to the symbols used here, for added clarity. (It’s not the prettiest formal language I’ve ever seen, even with these adjustments.) Let me try and give you an idea of what these symbols mean.

Each of the units separated by spaces (γ1, β1, etc.) represents a particular function in the fairy tale. The word function here has a technical meaning: it refers to an action taken by a character which serves to advance the narrative, or, as Propp puts it:

Function is understood as an act of a character, defined from the point of view of its significance for the course of the action.

These functions are the fundamental building blocks of fairy tales under Propp’s analysis. Here is a partial list of them, including only the ones that are present in Гуси-лебеди.

  • γ is interdiction: the hero is told to do or not to do something. In this particular case (that of Гуси-лебеди), the hero, who is a girl, is told by her mother to take care of her little brother and not to leave the courtyard. The superscript 1 indicates that the command is negative, rather than positive (superscript numbers indicate particular ways in which an action can take place in Propp’s formal language).
  • β is absentation: one of the members of the family leaves home. In this case it is the parents who leave (to go to work), and this is indicated by the superscript 1.
  • δ is violation: the hero violates an interdiction. In this case the interdiction is the one explicitly made earlier (γ): the girl goes out to play in the street, leaving her little brother by himself. But the interdiction may also be implicit (in which case the name for this action should probably just be mistake).
  • A is villainy: the villain carries out a villainous act, which can take many different forms. The form taken in Гуси-лебеди is that of abduction of a person (hence the superscript 1): the little brother, without his sister to watch over him, is seized from the air by гуси-лебеди (“geese-swans”). (I don’t know what exactly these “geese-swans” are supposed to be; there is an animal whose English name is swan goose but it only seems to live in a small area of southeastern Siberia, so I’m not sure if it would have featured in Russian fairy tales, and its name in modern Russian appears to be сухонос, anyway, not гусь-лебедь.)
  • C is beginning counteraction: the hero agrees to or decides to take action against the villain. In this case, the girl realises her little brother is missing and sets off in search of him.
  • ↑ is depature: the hero leaves home.
  • The next sub-sequence of three units is enclosed within square brackets, with a subscript 3 after the closing square bracket. This indicates that the whole sub-sequence is repeated three times. That the number is three and not, say, two or four is due to the well-known rule of three.
  • D is the first function of the donor: the hero is tested (explicitly or implicitly) by a character called the donor, with a source of help (often magical) available if they pass the test. As this is part of the triple-repeated segment, there are three instances of such a test in Гуси-лебеди. First, the girl comes across a stove and asks it where the geese went. But the stove asks her to eat one of its rye-cakes first, and she refuses. Second, the girl comes across an apple tree and asks the same, but the apple tree asks her to eat one of its apples first, and she refuses. Third, the daughter comes across a river of milk with banks made of pudding and asks the same, but the river asks her to eat some of its pudding, and she refuses again.
  • E is the hero’s reaction to the donor’s test; in this case, the girl refuses to eat any of the offers made by the three donors. Because the reaction is negative a NOT sign (¬) is written before the E (Propp writes “E neg.” instead, but I changed the notation slightly).
  • F is provision or receipt of a magical agent, or, in this case, a missed opportunity to do so (hence the ¬) because the girl failed the tests.
  • After the three failed tests, another DEF sequence occurs involving a hedgehog donor; this time the girl passes the test, and is pointed in the direction of the geese. Now this is an interesting part of the tale, because it could perhaps be said that Propp is taking some liberties to fit this part of the story into his theory. All that happens is that the girl comes across a hedgehog and feels like nudging him (D), but decides against it because of the hedgehog’s spines (E), then asks the hedgehog where the geese went and receives an answer (F). But maybe this is an example of where Propp’s theory actually has some explanatory power. For there is the question of why the fairy tale has to include this line about the girl wanting to nudge the hedgehog. It doesn’t have any effect on the story; the girl could just ask the hedgehog straight away. Propp’s answer to this question would be that it is a way of making the encounter with the hedgehog conform to the DEF structure.
  • G is guidance: the hero is transferred, delivered or led to the object of their search. In this case, the hedgehog provides the guidance; the guidance and the help provided by the donor are identical, hence the equals sign connecting the F and G functions.
  • K is resolution (Propp does not give it a name, for some reason, but resolution seems appropriate): the initial problem caused by the villainy in A is solved. In this case, the girl comes across a hut standing on chicken legs, in which the hag Baba Yaga is sleeping, while the girl’s little brother is sitting on a bench and playing with golden apples. She takes him away.
  • ↓ is return: the hero starts to make their journey back home.
  • Pr is pursuit: the hero is pursued. In this case, the swan-geese reappear and start chasing after the girl and her brother.
  • During the pursuit, three more tests (DEF sub-sequences) occur, with the same donors (the river, the tree and the stove, in that order) giving the test; this time, the girl passes each test by consuming the substances that the donors tell her to consume, and in return they hide her and her little brother from the swan-geese.
  • Rs is rescue: the hero is rescued from pursuit. In this case rescues occur as the help provided by the donors in the final sequence of three tests, and so we write this unit with an equals sign connecting it to F.

If you want to read the actual fairy tale, the original text can be found here (I checked using a copy of Afanasiev’s book that the texts matched). There is also a lovely Soviet short animated film based on the fairy tale that was made in 1949, although the detail of the story differs in some aspects from Afanasiev’s original recorded version.

II.

Now, this focus on the functions of the actions taken by the characters in fairy tales is one of the key characteristics of Propp’s method of analysis. It contrasts with older methods of analysis such as that used in the Aarne-Thompson classification system. In the Aarne-Thompson system tales are classified in a hierarchical manner based on distinctive features called motifs, which are often to do with what kind of characters, objects or qualities appear in the story, rather than what kind of things happen. That is, motifs are often noun-like rather than verb-like. For example, a distinction is made in the Aarne-Thompson index between tales involving animals only and tales involving humans. But fairy tales tend to involve a great amount of anthropomorphism (consider, for example, the fact that the stove, tree and river in Гуси-лебеди are apparently beings with the power of speech, and nobody in the story at any point remarks upon the oddness of this), so that it is often possible to substitute an anthromorphised animal character in place of a human character in a fairy tale, or vice versa, without making the tale any less effective. And it does seem possible that such substitutions can take place as a tale is transmitted across space or time: Propp points out that there is a tale about the sharing of the harvest in which somebody is deceived, and in Russia the deceived one is a bear (an animal), while in the more westerly European countries the deceived one is the devil (a human—well, a being who is supposed to have the rational capabilities of a human). The things that Propp focuses on—actions that advance the narrative—seem like they would be much less susceptible to this kind of substitution, as changing them would often require the course of the narrative to change in a fundamental way.

Another key characteristic of Propp’s method of analysis is the attention he pays to the sequences of functions that appear. In relation to this, he makes the following rather startling assertion. In his words:

The sequence of functions is always identical.

That is, he proposes that the functions which may occur in a tale exist in a specific order. They can therefore be arranged in a kind of universal sequence, and every fairy tale, even though not every function may occur in it, will have the functions which do occur in it in the usual order, so that the sequence of functions within a given tale is a sub-sequence of the universal sequence. There will never be two functions which occur in one order in one tale and the other order in another tale.

As you read further into the book, you realise that this statement isn’t exactly true to the letter. First of all, Propp admits that in a few of the tales he studied there are functions not included among the 31 well-attested functions which he lists in the book, or functions out of the usual order. But, he says, these often appear to be elements transferred from tales of other classes (such as humorous anecdotes, or legends), and in any case they are rare.

Some of the extra functions can be considered to be “auxiliary” functions. Such functions may occur between any two main functions involving different characters, and serve to notify one character of another’s actions. For example, in one tale, after the hero accomplishes his task, the princess who set the task for him holds a feast, and when the hero turns up at the feast he is granted recognition. The holding of the feast is the auxiliary function here, serving to link the functions N (solution) and Q (recognition). Because these auxiliary functions occur in particular environments, they don’t make Propp’s thesis unfalsifiable, and hence Propp doesn’t consider them problematic. They do, however, allow for a certain degree of freedom in Propp’s method of analysis.

Likewise, there are some well-defined circumstances under which certain functions may appear out of order. For example, the DEF functions (testing of the hero by a donor, followed by potential receipt of help) may appear before A (villainy), and the ↑ function (departure) may precede A. But these, again, are not too problematic for Propp’s thesis because of the well-defined circumstances in which they occur. It is still more useful to talk about the universal sequence, and to explain aberrations from this sequence as a result of certain transpositions, than to talk about a number of completely different sequences.

There are also various complicating factors. Sub-sequences may be repeated, one after another; cf. the fourfold repetition of the DEF sequence in Гуси-лебеди. Quite large sub-sequences can be repeated in this way; when the bulk of the narrative is part of a repeated sub-sequence Propp refers to the repeated units as moves. For example, he mentions a fairy tale in which a girl is abducted by a dragon, and then each of her three brothers sets out, one after another, to rescue her, with only the third, youngest brother’s attempt proving successful. Here, the whole segment from A (villainy) to G (rescue) (not including these endpoints) is repeated in three moves. Propp talks about the repetitions of the first kind (applying to relatively small sub-sequences) as if they are a different kind of phenomenon from the repetition of moves, but as far as I can tell, they are not formally distinct and differ only in scale. There is also a set of functions which appear more than once in the universal sequence, once before ↓ (return) and one after it. D, E and F are members of this set, which is why we see two discontiguous occurences of the DEF sequence in the structure of Гуси-лебеди.

Since I haven’t looked at the data Propp bases his conclusions on, and I haven’t read any of the criticisms of his work that have been made, I can’t really make a fair assessment of the validity of Propp’s conclusions for myself. However, even if the supposed universal sequence might not be quite as universal or invariant as Propp makes it out to be, it might still be a useful framework for analysing the structure of fairy tales. The question is not quite whether Propp is right or not, but rather whether the theoretical framework he has developed is useful or not (which is to some extent a question of degree). The way to assess this question would be to look at analyses of fairy tales that make use of this framework and compare their explanatory power with that of alternative theoretical frameworks.

III.

This idea that every fairy tale has a structure which can be described as a subsequence of a universal sequence of functions is the central idea of Морфология сказки. But Propp does talk about some other, more secondary aspects of the fairy tale in his book, too.

For example, he talks about the different characters in the fairy tale. As a corollary of the existence of a universal sequence of functions, there is also a universal set of characters, or more precisely spheres of action, which are made use of in fairy tales. Spheres of action don’t correspond exactly to characters, because, for example, there can be multiple characters which act in the same sphere of action at different points in the story, and a character may occupy more than one sphere of action. There are precisely seven spheres of action, according to Propp. I would talk more about them, but I don’t want to go into too much detail here. Propp also talks about how the different functions are distributed among these characters, and about how new characters are introduced. One interesting remark he makes here is that the initial situation in a fairy tale always involves the introduction of the members of a family, so no two of the characters that are present in the tale from the start are permitted to be unrelated. A striking consequence of this is in fairy tales where one of the starting characters acts as the villain, the villain is always a member of the family; for example, one tale concerns Prince Ivan and his sister, who is a witch and intends to eat her brother. And there are lots of other little pieces of information in the book about what kinds of things tend to happen and what kind of things tend not to happen in fairy tales which might be interesting in their own right.

One thing that is worth noting is that if you wanted to write your own fairy tales, for whatever reason (conworlding, etc.), in as authentic a style as possible, then the insights about fairy tales found in Морфология сказки could be very useful. The book’s list of the functions that occur in a fairy tale could effectively function as a writing guide; all you’d need to do would be to decide on a sequence of functions to use and fill in the details.

IV.

Propp’s thesis raises some interesting questions. For example:

  1. How far does the thesis apply to non-Russian fairy tales?
  2. How far does the thesis apply to other kinds of folk tales?
  3. Why does the thesis hold? That is, why is there a universal sequence of functions out of which fairy tales are structured? Why don’t fairy tales display a wider range of structures?

No complete attempt is made at answering these questions in the book, but Propp does make some remarks relating to them.

Relating to question 1, he mentions that his analysis can be applied to the fairy tales in the famous collection of the Brothers Grimm, which do present “the same scheme in general”, but it is more difficult; they display “a less pure and stable form of it”, probably because they have not been preserved in a wholly authentic form. Propp says that “uncorrupted tale construction is peculiar only … to a peasantry … little touched by civilization”. So it seems Propp expects his analysis to be applicable to the fairy tales of other European nations, at least, if they are considered in their authentic forms.

As for question 2, Propp says that many “legends, individual tales about animals, and isolated novellas” can be fruitfully analysed in the same way as fairy tales, and so perhaps a new name should be sought for the folk tales which are amenable to Propp’s analysis, because “fairy tales” is too specific. However, he doesn’t propose any new name. Note that it is also conceivable that other classes of tales may be structured according to different universal sequences, a possibility about which Propp doesn’t say anything.

Propp addresses question 3 quite explicitly, although he is careful to state that he is essentially engaging in wild speculation when he tries to answer it. He basically says that the existence of the universal sequence of functions seems to indicate that all fairy tales come from a “single source”, although he is vague as to what that single source might be. To be honest, I don’t really understand his answer. But, as he says, it is only a tentative supposition; further study is needed to answer this question.

So, Морфология сказки has left me feeling like I know a lot more about the structure of fairy tales, and at the same time realising that there is a whole lot more that remains to be known. This is, I guess, what it generally feels like to be introduced to a new field of study. One thing I would be particularly interested in knowing is whether Propp’s framework or something similar is helpful for explaining how folk tales change over time—that is, whether it is helpful in a diachronic perspective as well as a synchronic perspective.

But, moving away the particular topic of folk tales, there is a more general reason I’ve found reading this book very interesting: by successfully applying formal techniques to analyse a phenomenon that I would have thought was not very amenable to formal analysis, Propp has increased my expectations that formal techniques could also be helpful in analysing other kinds of phenomena. That might not sound like the most exciting thing, so let me try and put it in more illustrative terms. Imagine a pitch-dark room in which the entire set of information contained in the universe can be found. A small candle glows in the corner, and the small region of the room that it illuminates contains precisely the information that is knowable to humans. The pleasure of reading a book like Морфология сказки is that after reading it, you take another look at the candle and realise that it is glowing a little brighter than you thought.

(Edited 07/10/15 with some typo fixes and changes of wording.)

Advertisements

5 responses to “Thoughts on “Morphology of the Folktale”, by Vladimir Propp

  1. RedHeadedBookLover

    This is such an amazing post. I really love your blog so much and in turn cannot stop reading all of your posts!

  2. I have a hypothesis about question 3. Maybe part of the reason why fairy tales share a structure is to… convey to the listener that they are hearing a fairy tale. Like, in English, we open fairy tales with “once upon a time”, and then the listener can say “ah, I’m hearing a fairy tale” and interpret the story differently in light of that information. (I realize that’s vague — sadly I don’t have a less vague idea right now about the differences interpretation, depending on what genre the listener thinks it is.)

    Anyway, maybe some of the shared structure can be thought of as “syntactic”. If the words in a sentence were all jumbled up randomly, it would be impossible to glean meaning from them; the syntactic structure helps us understand the meaning by explaining how the words are related to one another. Maybe a similar syntax is needed for fairy tales, and the story would otherwise be incomprehensible?

    This is all wild speculation. Anyway I greatly appreciated this post! I learned a lot from it. I still intend to read Propp some day, but you have given me enough knowledge about the book to tide me over until I finally find the time.

    • It’s funny that you reblogged this today; I was thinking about it again because I have to return the book to the library in a few days. I thought it might be worth making a note of the passage where Propp talks about question 3:

      As has already been indicated, this conclusion [of the total uniformity in the construction of fairy tales] appeared quite unexpectedly. It was an unexpected one for the author of this work as well. This phenomenon is so unusual and strange that one somehow feels a desire to dwell upon it, prior to going on to more particular, formal conclusions. Naturally, it is not our business to interpret this phenomenon; our job is only to state the fact itself. Yet one still feels inclined to pose this question: if all fairy tales are so similar in form, does this not mean that they all originate from a single source? The morphologist does not have the right to answer this question. At this point he hands over his conclusions to a historian or should himself become a historian. Our answer, although in the form of a supposition, is that this appears to be so. However, the question of sources should not be posed merely in a narrowly geographic sense. “A single source” does not positively signify, as some assume, that all tales came, for example, from India, and that they spread from there throughout the entire world, assuming various forms in the process of their migration. The single source may also be a psychological one. Much has been done by Wundt in this sphere. But here also one must be very cautious. If the limitation of the tale were to be explained by the limited faculties of human imagination in general, we would have no tales other than those of our given category, but we possess thousands of other tales not resembling fairy tales. Finally, this single source may come from everyday life. But a morphological study of the tale will show that it contains very little pertaining to everyday life. Certain transitional stages from the pattern of daily living to tales do exist, and this pattern is indirectly reflected in them. One such transitional stage is found in beliefs which arose at a certain stage in the development of daily life, and it is very possible that there is a natural connection between everyday life and religion, on the one hand, and between religion and the tale on the other. A way of life and religion die out, while their contents turn into tales. As indicated previously, tales contain such obvious traces of religious notions that they can be tracked down without the help of a historical study. But since such a supposition is more easily clarified historically, we shall cite a small illustrative parallel between tales and beliefs. The tale evidences three basic forms of Iván’s bearers through the air. These are the flying steed, the bird, and the flying boat. But it happens that these forms represent bearers of the souls of the departed, with the horse predominating among agricultural and herding peoples, the eagle prevailing among hunters, and the boat predominant among inhabitants of the seacoast. Frobenius even cites the representation of such a ship for souls (Seelenschiff) from Northwest America. Thus one may suppose that one of the basic elements of tale composition, i.e., wandering, reflects notions about the wandering of souls in the other world. This notion, together with certain others, could undoubtedly have arisen independently of one another throughout the entire globe. Cultural crossings and the dying out of beliefs complete the rest. The flying steed gives way to the more amusing carpet. But we have gone too far astray. We shall leave this to be judged by the historian. The tale has still been studied very little on the plane of its parallel with religion and its further penetration into the cultural and economic aspects of daily living.

      This is the most general, fundamental deduction of our entire work. It is no longer possible to say, along with Speránskij, that no generalizations exist in the study of the tale. It is true that the present generalization is only an attempt. But if it is correct, it should in the future bring after it a series of other general conclusions. Perhaps then the secret in which the tale is so deeply wrapped will gradually begin to unfold.

      I don’t really understand exactly what Propp is saying here about religion and such, but I think I have a better idea now of what he means when he says fairy tales must have a single, shared source. You could interpret this as saying that there was a single “proto-tale” in the beginning, which changed and split into different tales over time, as people transmitted it via tellings. Kind of like how different languages can evolve out of a single proto-language. I don’t think historical linguists would normally say that, for example, the Indo-European languages all share a common structure due to their common origin, because the first things people think of when you say “structure” in relation to languages are things that we know are relatively unstable over time: word order patterns, etc. But it is true that all IE languages have a common structure in a certain sense; for example the regular phonetic correspondences between them can be thought of as an aspect of this common structure. Or, perhaps a better way to put it is that every IE language can be characterised completely as PIE + changes, so, if you abstract away the changes you can say that all IE languages are the same. That seems to be what Propp’s doing when he says every fairy tale has the same structure: he acknowledges that the functions do sometimes appear out of order, but he thinks every instance of this can be explained as a secondary modification of the underlying structure. This doesn’t preclude the possibility that fairy tales might be changed vastly by these secondary modifications, perhaps to the point where their common structure is unrecoverable, although in the case of the fairy tales in Afanasiev’s collection the common structure is fairly transparent (they are all from a single tradition, after all). Propp’s remark about Grimm’s fairy tales make more sense in light of this: they have probably undergone more extensive secondary modifications than Afanasiev’s.

      But the source doesn’t have to be such a literal one as that, with an actual single “proto-tale”. I think Propp’s remarks about the source being psychological, or from everyday life, are his attempt to point this out. The relevant analogy might be the concept of Universal Grammar. I’m not that knowledgeable about the theoretical side of linguistics, but my understanding is that Universal Grammar says that all languages share a common structure, and the differences between languages are the result of different parameter settings. In other words, a language is composed of universal principles which are modified by specific parameters; it’s a sum of a universal and a specific component, Principles + Parameters, in the same way an IE language is PIE + changes. It’s clear that Universal Grammar can’t be the result of monogenesis, because it applies to sign languages as well as spoken ones; instead it’s thought to be the result of the mechanical constraints on the brain’s language faculties. So, it may be that there is no single “proto-tale”, but the single source exists in the sense that there are constraints associated with the nature of fairy tales which cause them to share a certain degree of structure.

      Anyway, those were my thoughts upon typing up that quote. I’m not sure they are up to my usual standards of coherency, but I’ll leave them here, since this is only a comment.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s