Review of ‘Comparative phylogenetic analyses uncover the ancient roots of Indo-European folktales’

by Emma Parfitt

A review of ‘Comparative phylogenetic analyses uncover the ancient roots of Indo-European folktales’ by Sara Graça da Silva, Jamshid J. Tehrani


Image courtesy of:

Let us begin with a story…

A blacksmith is granted three wishes by Jesus and St Peter for his hospitality. The smith had made a previous bargain with the devil, obviously not a great idea! He decided to use his three wishes to get out of the bargain. He wishes that whoever sits in his chair be stuck until he permits them to rise. He wishes that whoever climbs into his tree remains there until he permits them to climb down. He wishes that any persons or things must climb into a sack when he wills it. When the devil comes to fetch the smith those three wishes come in very handy. The smith outwits the devil.

That was a summary of the tale The Smith and the Devil (Hansan 2002). The paper under review proposed that this folktale originated in the Bronze age. The methods used were in relation to folktale archives, archaeological, linguistic and genetic data. Here is an explanation of their terms and methods.

An explanation of terms

The paper under review used comparative phylogenetic methods on Indo-European folktales. So let’s look at these terms.

Indo-European folktales

Indo-European refers to a family of several hundred related languages and dialects including the major languages of Europe and parts of Western, Central and South Asia. In this instance the study investigated a subsection of folktales: 275 cases of Tales of Magic. This included 50 languages. 275 tales of magic were taken from a larger catalogue of over 2000 folktale types. Basically the heart of this study hinges on whether you agree with the classification and categorisation of these tales from the Aarne Thompson Uther (ATU) Index.

A number of academics have critiqued this Index and in some cases attempted to build and improve on it. The original index was published in 1910. In the 1950s Propp (2015) critiqued the way in which the folktales were categorised. In 1928 Thompson refined the categorisation into 2500 basic plots. Imagine processing so many folktales and plots without a computer. Dundes noted that there were typos, redundancies and censorship in the records (1997: 198). However, the index was, and remains, central to international studies of folklore (Dundes 1997: 200).

The index was finally updated by Uther in 2004. Uther presented several criticisms of the original index. He pointed out that Thompson’s focus on oral tradition neglected older versions of stories, even when written records existed. Uther thought that the distribution of stories was uneven. For instance, Eastern and Southern European, and many other folktale types were underrepresented.

Things to consider: the biases of the index, including gender and ethnic biases as it was invented and revised by White males.

To briefly summarise:

Aarne – Finish, White male, folklorist and university lecturer

Thompson – American, White male, Harvard graduate

Uther – German, White male, varied university career (University of Munich, University of Göttingen, University-Essen).

I am also assuming, individuals with enough funds to go to university and be educated in a field like folklore (please share in the discussion any knowledge which contradicts this).

Comparative phylogenetic methods

The method could be compared to a cake recipe as follows:

  1. Take 275 tales. These can be found in any specialist ‘Tales of Magic’ delicatessen or curiosity shop.
  2. Separate the yolks from the whites. In other words the presence or absence of these tales in 50 Indo-European speaking populations.
  3. Mix with a smattering of jargon and statistics. I found the paper to be written in a statistical analysis style, therefore not very accessible.
  4. While the cake is in the oven draw some language trees. These indicate how populations have historically come into contact over time (population dispersal and the diversification of linguistic lineages).
  5. Remove from the oven and cool. Ice the cake by skimming over debates of written versus oral narrative, and the distinctions or overlaps between them.

The written-oral debate acknowledges that there is a difficulty in tracing oral stories because tales are so easily transportable by travellers. There has been interaction of oral and written texts overtime, with one informing the other (Bottingheimer 2009, 2014). This basically means that what is an oral folktale, and each tales origins is unclear.

For example, different renditions of Arabian Nights indicated that translators created new stories inspired from oral and literary traditions (Warner 2012: 18). While this demonstrates some of the issues with the paper’s source materials it also illustrates how fearless Graça da Silva and Tehrani (2016) were trying to tackle the complex subject of oral folklore utilising new methodologies.

  1. To finish the cake, sprinkle the results with statistics to see whether the 275 tales were more likely to spread generation to generation or across geographical areas due to migration, trade, and encounters between groups of people.
  2. Eat the cake.

Findings of the paper

Graça da Silva and Tehrani (2016: 9) proposed that The Smith and the Devil could have arisen in the Bronze Age, further supported by the earliest archaeological evidence we have for metal work. It does them credit that they used a mix of new methods alongside conventional literary-historical, archaeological, ethnography, genetic and linguistic approaches. Mixed and diverse methodology can add unexpected things to our understanding of complex processes. Storytelling research, for example, indicates the challenge of using story to explore social phenomena (Parfitt 2014).

In considering whether or not to accept the results of this paper, here is a quote from The Matrix (Wachowski 1999):

‘it is not the spoon that bends, it is only yourself’

Are the patterns that Graça da Silva and Tehrani (2016) observed real indicators of the oral spread of folktales or was the data set flawed, resulting in a misleading pattern? The paper concluded:

The results of the D analyses suggested that a substantial number of tales (100 of 275) exhibit significant correlations with linguistic relationships that are consistent with vertical processes of cultural inheritance (Graça da Silva and Tehrani 2016: 6-7).

In plain English this means that 36% of the folktales indicated that tales were more likely to be passed from generation to generation than spread geographically. Is this a substantial number? However, in another part of the paper they clarify that only 28% of the folktales were robust enough to suggest this (76 of 275). Thus, I am not surprised that the final results contradict previous findings that geographical spread was higher than generation-generation ‘vertical cultural inheritance’ (Ross et al. 2013). Storytelling traditions might be strengthened through engagement with other cultures and weakened by cultural isolation (Wilson 2014: 129). There are so many tales that demonstrate remarkable similarity in different parts of the world. Cinderella is a well-known example (Cox 2012). We need to ask whether these tales were passed from generation to generation or by the movements of people and stories? It is likely, of course, that both forms of transmission occur. What is interesting is that Graça da Silva and Tehrani (2016) suggest that ‘cultural inheritance’ is more likely than geographical spread.

The validity of their findings will be proven or disproven over time. There are doubts concerning the linguistic categories being correct and unbiased. Especially as with the use of large data, patterns that are perceived might not be accurate. Non mathematicians might expect a larger sample size to make results stronger, this is not the case, after a point it creates the possibility of more error. It does not help that the paper cited the Brothers Grimm, as an exmaple, because of the way the Grimms’ tales of “Germanic origin” were continuously edited to be more German (Ellis 1985, Tatar 1987: 143, Claxton-Oldfield 1988: 52, Zipes 2006: 195). It depends whether we are talking about edited written versions of stories or orality, which also have a complex history with text (briefly mentioned earlier).

It is likely that folktales and their distribution may be related to population histories and geographical proximity (as different groups met and exchanged stories). However, I’m not sure that similarities in word usage described in this paper confirms this without margins of error. An opinion also shared by John Lindow, a folklorist at the University of California, Berkeley (Samoray 2016). Word usage and the way stories are told shift over time in response to a variety of changes: social, technological, political, educational, and so on. I am skeptical of assuming that we know anything about how stories were exchanged orally as what we can infer from that long ago is largely dependent on written records. The work provides an interesting stepping point for further investigation. A similar analysis has yet to be carried out on all the index to test the robustness of the findings.


If the findings can be replicated then the ancient roots of these 275 tales might be traced to  the Bronze Age. Storytelling, folklore, and fairy tale literature suggest that stories have been spread through travel, trade, conquest and literary texts. Dupont (1999) said orality cannot be capture by writing, because ‘all that writing can convey is the absence of a body, a loss of orality.’ Folk and fairy tale texts provide evidence of ‘relationship to writing’ not orality. Thus, basing oral “genetic” research on text may be flawed. The jargon may make it sound professional and scientific, but a little digging reveals that this is not the case. I remain sceptical of the findings. I don’t think the results support their claims, and the claims have been picked up by the press rather quickly without thorough consideration. However, I applaud Graça da Silva and Tehrani’s (2016) use of creative methods. I found it an interesting analysis, and eagerly await being proved wrong!


Bottigheimer, R.B. (2014). Magic tales and fairy tale magic: from ancient Egypt to the Italian renaissance. New York, NY: Palgrave Macmillan.

Cox, M. R. (2012). Cinderella tales from around the world. Nashville: SurLaLune Press.

Samoray, C. (2016). No fairy tale: Origins of some famous stories go back thousands of years. Statistical analysis of language evolution helps estimate storytelling dates. Science News. 19 January 2016. Available from: [Accessed: 1 March 2016].

Dundes, A. (1997) “The Motif-Index and the Tale Type Index: A Critique”. Journal of Folklore Research Vol. 34, No. 3, pp. 195–202.

Dupont, F. (1999). The invention of literature. From Greek intoxication to the Latin book. Translated by Janet Loyd from the French. Baltimore, Maryland: Johns Hopkins University Press.

Ellis, J. (1983). One fairy story too many: the brothers Grimm and their tales. Chicago: University of Chicago Press: 79. In: Tatar, M. The hard facts of the Grimm’s fairy tales. 1987, reprinted in 2003. Chapter 1. Sex and Violence: The Hard Core of Fairy Tales. Princeton, New Jersey: Princeton University Press. p. 3-38.

Graça da Silva, S. Tehrani, J. (2016). Comparative phylogenetic analyses uncover the ancient roots of Indo-European folktales. Royal Society.  20 January 2016. doi: 10.1098/rsos.150645 Available from: [Accessed: 26 January 2016]. Claxton-Oldfield, S. Butler, B. (1998). Portrayal of stepparents in movie plot summaries. Psychology Reports. 82. p.879-882.

Hansen, W. (2002) Smith and the Devil. In: Ariadne’s thread: A guide to international tales found in classical literature. Pp.405- 408. London: Cornell University Press.

Parfitt, E. (2014). Storytelling as a trigger for shared conversations. Exchanges: Warwick Research Journal. 1(2).

Propp, V (2015). Orginally printed in 1958. Morphology of the folktale. Eastford, CT: Martino Fine Books

Ross, R.M. Greenhill, S.J. Atkinson, Q.D. (2013) Population structure and cultural geography of a folktale in Europe. Proceedings of the Royal Society. B280: 20123065. p.1-9. doi:10.1098/rspb.2012.3065.

Tatar, M. (1987). The hard facts of the Grimms’ fairy tales. Princeton, New Jersey: Princeton University Press.

Uther, Hans-Jörg. Classifying folktales: The Third Revision of the Aarne–Thompson Tale Type Index (FFC 184).

Wachowski, A. Lana Wachowski, L. (Dirs.). (1999). The Matrix. US: Warner Bros.

Warner, M. (2012). Stranger magic: Charmed states & the Arabian nights. London: Vintage.

Wilson, M. (2014). Another fine mess. The condition of storytelling in the digital age. Narrative Culture. 1(2). p.125-144.

Zipes, J. (2006). Fairy tales and the art of subversion. 2nd ed. Oxon: Routledge.


Emma Parfitt is a PhD Researcher in Sociology at the University of Warwick.


Categories: Uncategorized

Tags: , , , ,

Leave a Reply

Your email address will not be published. Required fields are marked *