Vous êtes sur la page 1sur 24

See discussions, stats, and author profiles for this publication at: https://www.researchgate.


2014-2015.. Polyphemus, a Palaeolithic Tale. -
The Retrospective Methods Network
Newsletter, Winter 2014-2015, 9: 43-64.

Article · December 2014


0 138

1 author:

Julien d'Huy
Université de Paris 1 Panthéon-Sorbonne


Some of the authors of this publication are also working on these related projects:


All content following this page was uploaded by Julien d'Huy on 04 January 2016.

The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document
and are linked to publications on ResearchGate, letting you access and read them immediately.

Polyphemus: a Palaeolithic Tale?
Julien D’Huy

To cite this version:
Julien D’Huy. Polyphemus: a Palaeolithic Tale? . The Retrospective Methods Network
Newsletter, 2015, pp.43-64. <halshs-01170837>

HAL Id: halshs-01170837
Submitted on 6 Jul 2015

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est
archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents
entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non,
lished or not. The documents may come from émanant des établissements d’enseignement et de
teaching and research institutions in France or recherche français ou étrangers, des laboratoires
abroad, or from public or private research centers. publics ou privés.

Nordisk kultur 8. The Prose Contexts of Poetry of the Scandinavian Middle Ages 1. Eddica Minora. Pp. This structure research has an empirical and positivistic visualizes the inferred evolutionary approach to using the so-called Historical.).e. “Haralds saga (PhD).). Les objets qui posent à l’ethnologue un motif’s history and geographical spread.).) which may hold some potential for The objects that pose a problem of revitalizing this type of research. Andreas. 1944. rigoureuses. From Snorri Sturluson’s. In In Copenhagen: Møllers. In Harmondsworth: Penguin. Skaldic Leslie. Diana Whaley.). Dissertation for the degree of philosophiae doctor Sveinbjörn Egilsson (ed. Institute of the African World (IMAF). and to trace each two or more descendant lineages (i. “Rekstefja”. Eddic Poetry: Primarily in the Fornaldarsǫgur. XXVIII. Saga- Anteckningar till Edda och skaldediktning I– Book 26: 38–66. “Haraldsstikki”. 1923–1944. fornfræða félags VI. auprès de reconstructive ambitions of the Finnish disciplines qui travaillent sur les mêmes School have been strongly criticized. Sveinbjörn Egilsson & Finnur Jónsson. Copenhagen: Popp. the Kings’ Sagas I:2. Polyphemus: A Palaeolithic Tale? Julien d’Huy. This problèmes à une échelle incomparablement method was conceptualized long before the plus grande et avec des méthodes plus development of computer-assisted methods. Elizabeth Ashman. Magnus Magnusson & Hermann Pálsson (trans. Wilh. Rowe.) organisms could be represented with a The Finnish School of comparative folklore branching diagram or ‘tree’. 893–939. Ed. Matthew (ed. raisons de chercher des enseignements peut. Paris I Sorbonne Abstract: This paper presents an analysis of 56 variants of European and North American examples of the so-called Polyphemus tale (international tale type ATU 1137) using phylogenetic software according to 190 traits. Turnhout: Brepols. Flateyjarbók I.). 2009. The ethnologist study of myths and folktales. Jón Helgason. Sigurður Nordal. certainly for stimulation. 3–179. Hardráda”. 1. The present classification to the ethnologist are certainly article considers the potential use and value of less numerous than those brought to the applying modern phylogenetic tools for the attention of the naturalists. L’ethnologue n’en a que plus de the attested versions ultimately originated. & Wilhelm Ranisch (ed. Litteraturhistorie B: Norge og Island. Rolf (ed. Each node from which of this school tried to collect all variants of a branches of the tree stem represents a tale. 2012 [2013].Heusler. which species based upon similarities and was recently discussed by Frog in an earlier differences in their physical or genetic volume of this journal (2013b). Gade 2009b: 807–808. Helen F. (eds. to analyse the diffusion and frequency of speciation event in which a lineage splits into each of its individual traits. The scholars characteristics. 43 . Ed.). 1903. Despite an initial enthusiasm. Ruhfus. Stavnem. Dortmund: Druck und Verlag von King Harald’s Saga: Harald Hardradi of Norway: Fr. In Fornmanna sögur eptir gömlum Lexicon Poeticum: Antiquæ Linguæ Septentrionalis – handritum útgefnar að tilhlutun hins norræna Ordbog over det norsk-islandske skjaldesprog. Bonniers Förlag. 1953. Heimskringla. Notationes Norrœnæ: Literary Adaptation of Myth and Legend”. 1825–1837. (My translation. from The Biological Model disciplines that work on the same problems A great advance in biology occurred when on an incomparably larger scale and with researchers realized that the lineage of more rigorous methods. They problème de classification sont certes moins also tried to reconstruct the ideal primeval nombreux que ceux soumis à l’attention des form of the tale (Urmärchen) from which all naturalistes. 1931. “Norges og Islands digtning”. des stimulations certainement. (Lévi-Strauss 2002: 311. 2002. Townend. Sigurðr. Norway: University of Bergen. In Poetry from Lund: Gleerup. Discussion addresses a number of points of comparative methodology while considering the historical implications of a relationship between different versions of this tale type recorded in diverse cultures. Nordal. relationships among various biological Geographic Method and its variations. Lunds Universitets årsskrift new ser. et al. Ernst Albin. Stockholm: Albert Akraness: Flateyjarútgáfan. Pp. Ed. also has all the more reasons to look perhaps for lessons. the être. 2012. “Sǫrla þáttr: The Kock. 1966.

g.). or evolutionary relationships following a long tradition: see Hafstein 2005). The most important Initial published attempts using of these is the fact that both are formed by phylogenetic software to study mythology and discrete heritable units which evolve folktales may date back to 2001.g. The use of a folktales. point mutation and Innovation (e. This can only work with only a limited number of explains why so many individual variants of functions (only 20 types of amino acids exist) tales differ from the abstract tale-types with and results were limited to only very short which they are identified. geographically genome informatics to Propp’s “sequence of and temporally. relationships between examples of myths and fairy tale 3 = ADPHW. Jun’ichi Oda progressively with time. innovation. The more two related applied an alignment program used for species or two myths diverge. Wilhelm von Sydow (who himself was structural. As early as 1909. Discrete heritable units (e.g. Propp’s relationship probably is. societal trends variants of the trait may survive and reproduce and conformist traditions) more than individuals with other variants) Allopatric or sympatric speciation Geographical or social separation Hybridization Mixture of two or more myths or tales Horizontal transmission defined to be the Extralineal borrowing or imposition movement of genetic material between bacteria or within the genome other than by descent in which information travels through the generations as the cell divides (e.g. intergenerational transmission. Table 1. the more distant their genetic functions” concerning 45 fairy tales. Under such elements should be studied “comparatively. as Gennep 1909: 84). According to Carl may be a consequence of functional. each of which was in genetic relationships and relatedness can turn coded as a given amino acid (e. program could then arrange the sequences of Applying the biological model to myths functions to identify regions of similarity that and tales is not new.g. fairy tale 2 = ACFDEF. circumstances.g. 1948 [1965]: 238–239): they approach presents some difficulty owing to tend to adapt to their environment and they limits of the genome model: the researcher evolve by means of natural selection. genes and individual phenotypes) motifs. software developed for assessing model to 16 functions. teaching. variation. Equivalence of elements and features in the comparison of genetic systems and of myths / folktales. the four nucleotides. writing (more recent) Mutation (e. learning and imitation reproduction Slow rate of evolution Fast or slow rate of evolution Parent–offspring. fairy potentially be applied to assess corresponding tale 1 = ACDEF. This Sydow 1927. mythemes. viruses. slippage. lines of code for each sequence (from 4 to 12 Arnold van Gennep stated that folkloric functions / amino acids). summarized in Table 1. occasionally clonal Parent–offspring. mobile DNA) mistakes) Natural selection of traits (individuals with certain Social selection of traits (e. tale-types) Mechanisms of replication by transcription and Teaching. Biological and mythological with the aid of the biological method” (van entities have many traits in common. Observing these sequence was reduced using the Greimas parallels. it is highly possible that convergent evolution could produce apparent 44 . codons. between the sequences in the same way it folktales are like biological beings (von worked for amino acids in a genome. transposons) Geographic cline Mythological transformations Fossils Ancient texts Extinction Disappearance branches).g. Genetic Systems Myths / Folktales Discrete heritable units (e. etc.

2013a–c. Polyphemus. and whether HGM. such as ‘Classic HGM’ (esp. It does not involve qualitative evolution (d’Huy 2012b. Greenhill and Quentin D. up-to-date statistical and phylogenetical especially as it became internationally known methods applied.g. Returning to (d’Huy 2012a–c. historically shared innovations. Tehrani 2013. This work has been then in the first half of the 20th century or the continued by other researchers. An world. Oda’s work holds a position in the 2013d. Atkinson for which it received heavy criticism (Frog (2013). At its 1. d’Huy & Dupanloup early could make tales appear to correspond 2015). assessment of the features of variants and thus 2014a.similarity between functions that are relations of variants of a tale or tales in a evolutionarily unrelated. another. the function of a donor occurs 2013e–f. the Cosmic Hunt. More generally. This makes it called tale-type or motif indeed form a unity possible to model the evolution of a tale or should better be regarded as divided into inside a tree statistically. It is impossible to reconstruct the tale as it most basic. the phylogenetic in their formal sequence of functions that approach offers new resources for considering otherwise have nothing to do with one how folktales evolve (d’Huy 2013a. This approach is phylogenetically distinct international types similar to the formal studies of the Classic (d’Huy 2013e. from a number of methodological problems Simon J. I used as large a sample of Confronting Methodological Problems versions as possible and multiplied the most The Historical-Geographic Method (HGM). Oda’s functions. Tehrani 2013). 2013e–f. which may be produced by be questioned whether the members of a so. to tackle many of the historical relationships behind their various remaining problems with this sort of approach forms (Frog 2013b: 21–22). When the tree-like the statistical reconstruction is essentially a mathematical outcome of the correlation of 45 . More research would database of mythology appear interpretable as be necessary to arrive at any certainty that is reflecting its historical spread through the impossible to obtain with this method. Tehrani 2013). each time (vs. I was the first. It can offer answers to a lot of questions. 2013). Rather be interpreted as reflecting the relative than shared mutations. the reviewed here: phylogenetic approach is very interesting. 2013e–f. Ross et al. it becomes possible to consider additional problem is that the results depend whether this correlates with reconstructions of on the specific order of sequences. for instance latter half of the 20th century and research by changing the dataset or the method in paradigms changed. It can common. but the effectiveness of of the Finnish School fell out of favour in the Oda's method was never tested. the Dragon. in the questions about the history of tales and the beginning of 2012. the degrees of formal contributions of vertical and horizontal relatedness are hierarchically organized in a processes in folktale evolution (d’Huy 2012a– tree according to variations that they hold in b. 2013). and this can of formal relatedness between items. 2013a–e). whereby e. 2013a–c. its critics (d’Huy 2013a. Krohn 1926). I studied many these questions now with the support of families of mythological narratives and modern phylogenetic tools has the potential to folktales using different datasets of mythemes produce new knowledge.1 Indeed. folktales and myths moved away from As far as I know. extent to which examples of a given folktale Phylogenetic tools statistically assess degrees exhibit a tree-like set of relations. Ross et al. Phylogenetic methods have been used 2013b). Phylogenetic tools have the potential to study many folktales and myths. including to resolve a lot of the problems addressed by Pygmalion. 2013c. it can be used to explore the was first composed and told to others. Little Red Riding Hood and the Several of these issues will be briefly Kind and the Unkind Girl. suffered Jahmshid Tehrani (2013) and Robert Ross. 2013a–c. folkloristic research on order to control the results. As the approaches history of research. Tehrani 2013). but uses a computer rather than graph we can reconstruct the proto-tale and its paper. in which human migrations that might be responsible case a variation in a conventional plot for that spread (d’Huy 2012a–c. see the definition below).

Phylogenetic tools can process of divergence to occurr more quickly easily accommodate incomplete phylogeny.g. but that is another level of analysis and interpretation. The evidence possible that variants could be grouped of individual tales has not been evenly together owing to a concentration of formal collected among all cultures and the narrative similarity in the co-occurrence of motif has the potential to be transmitted across elements in one episode even though the different areas. However. with a broader system of material (e. carried via contact networks overall narrative form and structure was close and population mobility. which is roughly The Classic HGM’s focus on the presence or like interpreting variation across dialects of a absence of story details neglected the logical language as reflecting a sequence of relationships evident between different developments based on the language’s versions of a same myth. history of cultures from which examples were collected. Moreover. spread” (Frog 2013a: 117).g. the could make tracing locations of origin and more distant their genetic relationship. in one region and more slowly in another. This process of to that of another set of variants. hierarchically structured according to larger Any attempt to find the place of origin of tales units of narrative.g. with religious their geographical distribution. it is (hypothetically) seems to be doomed to failure. some tools (such as midpoint use of phylogenetic tools also can take into rooting) may enable the essential features of a account the fact that the tradition in one tale from which all of the variants derive to be cultural area can undergo an abrupt and established. tale to the geographical area of its origin. phylogenetic tools allow for the cultural mythology). A researcher change). These tools assess formal may take the information produced in a relationships: the interpretation of the history phylogenetic analysis and compare that with behind that formal relatedness is a subsequent the geography of formal distribution and the analysis by the researcher. At least two progressive spread to new locations. variations of a group. a However. The reconstructive approach identified this method makes this statistical assessment variation with dispersal and reconstructing quantitatively on the basis of the number of the historical form of a tale was thus linked individual elements without being to identifying its location of origin. This processes of geographical spread problematic formal distance seems normally to correlate and most often impossible without support with geographical and/or temporal distance of from other types of evidence or association the examples (e. The tale whole story. the geographical emphasis 2. 2013). The Classic HGM could not show how two developed from “confusing a continuum of or more seemingly different themes could typological similarities [in the distribution of stand in a structural transformational variants] with a historical progression of relationship to each other (Lévi-Strauss developments accompanying geographic 1968: 185). moderating compatible with both evolutionist and and negotiating variation. Where formal relatedness of one may also simply drop out of use in some areas example does not align with other shared without leaving evidence of the local form. Insofar as 3. phylogenetic tools treat radical transformation that rapidly becomes formal relationships between texts and not socially dominant (e.similarity of individual elements. 46 . the elements chosen for each motif repeated displacement of earlier local and need to be shared equitably throughout the cultural forms as a historical process. This phenomenon structural treatments: the more two myths of cultural adjacence (Frog 2011: 92–93) diverge from each other or transform. Such additional principles (variation and selection) continua may be better understood as related in the process of folklore transmission are to contact networks in interaction. For this transmission has the potential for even the reason. The Moreover. Ross et al. the software makes this and there may not be any evidence to link a observable as a conflict in the data.

The representativeness of sources. The concern However. (example in Frog 2011: 82–83).4. principle that analysis should be based on an The problem of the representativeness of as extensive and exhaustive a corpus as sources is a question of whether isolated 47 . contemporary source-critical standards (cf. It does 6. because the validity of sources. The issue of decontextualization can impact on the encoding of specific traits for then be in part mitigated by the coverage of phylogenetic analysis if ‘the worst translation’ the maximal amount of text for each example alters surface details of images and motifs (in the present case. text-critical strategy for tracing the history of omission or alteration of elements in e. The effect of The decontextualization of sources is horizontal transmission (i. than primary sources in a data-set to be The translation could nevertheless have an analysed. versions used in analysis is therefore necessary. for example. not isolating through which the mythic message is the motif of the escape from Polyphemus’ communicated. but then those materials were attention to oral tradition. none are in an whether these may have been introduced by a intermediate position). (2010). which analyze taxa as brothers or corpus to assess the probability that certain cousins rather than assessing them as a traits in variants of low source-critical quality lineage per se (each example is at the top of accurately reflect local or cultural tradition. Source-critical problems.e. Phylogenetic methods collector/author/redactor. this attitude valuations of individual variants and cannot could have consequences for handling be purely quantitative. materials owing to the number of languages this type of reconstruction presents a accessible to any one researcher. These criticisms were in part associated with Some such comparative analyses presumed a changes in source-critical standards more historical relationship and thus parts recorded generally (Frog 2013b) but a significant factor in different tales might even be first combined in broad comparative research was and as a reconstruction of the historical local or remains reliance on edited and translated cultural tradition for comparison. A selection among the cave but also all of the surrounding tale). (2009) frequently cut from their context in more and by Curie et al. the introduction. As noted. like discarding masses of variants. infer a lineage based on the proximate relation of many elements at the same level. This was particularly problematic in motif analysis but also in tale- 5. However. or the stemmatic tree of relations. as the outcome of analysis will be dependent on was done for example by Jan de Vries the quality and representativeness of the data. it may be noted that phylogenetic is unwarranted when using phylogenetic analysis could be used as a tool in a larger methods. longer story or integrating it into that plot. Lévi-Strauss hermeneutical problem and such synthetic (1958: 232) notes that a mythic message is reconstructions should not be included rather preserved even through the worst translation. for example. which it sought assessed and sorted according to to distinguish from literacy influence. type analysis where. The Classic HGM advocated the 7. However. a text-type transmission according to which translation of low source-critical quality may these variants would create an inaccurate affect results in a phylogenetic analysis. It was cultural traditions and only partly reflect problematic in earlier research because inherited culture) has been addressed in an sections of text relevant for comparison were optimistic fashion by Greenhill et al.g. Early research gave preponderant possible. The decontextualization of sources and not need to presume a gap between the true presumptions of relatedness folktale and literary adaptations. complex narratives. This impression if treated as conventional of the approach needs to maintain qualitative inherited oral tradition. certain Criticisms against the HGM in the latter part traits of a tale were clearly outcomes of of the 20th century included issues raised by adapting the narrative to the context of a the sources used and source-critical standards. This emphasis on oral sources was part of the point 4 above). if literary normally now thought of in terms of isolation adaptations draw on elements from other from a performance context.

which could be parallels by looking for them: what qualifies born/borrowed at different times. the proto-myth. This is a broad irrelevant. Researcher interpretation in type. and this is complemented by over whether an example is historically rooted the problem of researcher subjectivity in in one culture as opposed to borrowed determining which elements are relevant for through contacts with another may be observation and which are not.examples can be considered representative of Krohn’s (1926: 28–29) conception that each a local or cultural tradition. 1). If this sort of identification is 1904). hermeneutics: to what extent does looking for I have previously applied phylogenetic parallels produce parallels and their methods to the historical reconstruction of the justification? In the background of this Polyphemus Tale elsewhere (d’Huy 2012a. Tale The reconstruction of the Polyphemus tale is a 8. The problem of the such loose groupings may not in all cases be physical. 2013f). A solution alleviated when focus is calibrated to a could be to determine the maximal number of broader scope: for example.13). the Americas and individual examples are considered in relation to those broad patterns Phylogenetic Analysis of the Polyphemus (cf. statistical tools. Frog 2013b: 27. It unnecessary to resolve whether a Sámi should also be noted that varying the number example reflects a borrowing from Russian or and categorisation of elements subject to Norwegian tradition if comparison is between analysis often does not change the overall European/Eurasian traditions and traditions in result (d’Huy 2013c. The earlier reconstructions of identification. such as the Sámi examples subject. This whether the variants present different locally problem is similar to the issue of producing established forms. which rejects particularly relevant to phylogenetic analysis the possibility of ‘multigenisis’ of narrative on the basis of individual formal traits. question appears to be the criticisms of Kaarle 48 . This study suffers from a total lack of considered justified. it is then followed by the explanation for the criteria used to limit the problem that many similar cases remain number of versions included in the corpus unidentified and the additional problem that (Calame 1995: 143). 84. and of the significance which The researcher’s identification of an example lies at the root of the story. Latvian and statistically evaluated for each motif thanks to Greek discussed by Frog (2011: 81. The researcher’s Greek/Homeric variants in Fig. This is motif has a single unique origin. Some elements (cf. geographical origin of this story also valid. This identification of motifs / elements of the text problem requires a close analysis to establish may nevertheless remain a problem. This becomes a problem of seems unsolvable (Glenn 1978). for example by estimating This is particularly significant for the types of how many founder events are necessary to interpretations discussed when different explain the diversity of a studied corpus. or if a local as presence/absence or ‘the same’/‘different’ teller know both the traditional and an remains dependent on researcher anomalous tale at the same time. The most complete attempt to reconstruction of their relatedness to the reconstruct the proto-version of Polyphemus abstract tale-type rather than purely on the was O. and the reviewed in Frog (2011: 81) that are identified reader may consult Justin Glenn (1978) for an with ATU 1148b on the basis of the historical introduction. A variants of a tale from a single cultural group solution may also be to search for a do not systematically group together as more sufficiently complex set of traits that could closely related to one another than to those of not be the product of many independent other groups (cf. the distribution of Sámi and inventions around the world. below). Hackman’s analysis based on a basis of formal features of the individual Historical-Geographical approach (Hackman examples. 87). textbook case. it becomes elements for each text subjected to analysis. Concern interpretation. too wide to be reviewed here. 31n. can be safely or group of examples could be inaccurate or dismissed as erroneous. This is simple examples of this are the examples of a very controversial issue that could be ATU 1148b attested in Sámi.

Karel 1907: 38–39) Lithuanian people (Richter 1889: 87–89) Germanic 2 English people (Baring-Gould 1890) West Highlands people (Campbell 1860: 105–114). I used a corpus of examples of this will test my earlier results. Simms 1903: 295–297) Southern Athabaskan 5 Jicarilla Apache people (Goddard 1911: 212–214. Li romans de Dolopathos Gascon people (Bladé 1886. K602: “Noman”. Wissler & Duvall 1908: 50–52) Iroquoian 3 Crew people (Lowie 1918: 216–217. Opler 1938: 256–260) Kiowa Apache people (McAllister 1949: 52–53) Lipan Apache people (Opler 1940: 122–125) Chiricahua Apache people (Opler 1942: 15–18) Tanoan 2 Kiowa people (Parsons 1929: 21–24. Dumézil 1965: 55–59) Uralic 3 Hungarian people (Stier 1857: 146–150) Sami people (Poestion 1886: 122–126. 218–220. tradition to explore the potential of Natural Stith Thompson (1961) counted five Language Processing software for identifying traditional elements or motifs in Polyphemuss motifs (d’Huy 2014c).1: Escape by dressing in versions according to 98 traits in d’Huy animal (bird. Uther (2004) 49 . human) skin. Indo-Iranian 1 Ossetian people (Dirr 1922: 262) Caucasian 2 Abaza people (Colarusso 2002: 200–202. I journal. The Odyssey (book IX) Modern Greek people (Athens: Drosinis 1884: 170–176. 44 the eye). Kartvelian 1 Mingrelia people (Frazer 1921: 449–450) Turkic 2 Oghuz Turks people (Book of Dede Korkut) Kyrgyz from Pamir (Dor 1983: 34–36) Afro-Asiatic 6 Berbers (Germain 1935. attempts to reconstruct the evolution of K1011: Eye-remedy. I here increase the number of versions K603: Escape under ram’s belly. 25–26) Greek 4 Homer. In this paper. Polyphemus. Examples and sources used in the phylogenetic analysis. Vinson 1883: 42–45. K521. Cappadocia: Dawkins 1916: 551. 303–304) Basque people (Cerquand 1992. 152–154). the trickster blinds the the initial sample sizes (24 versions analysed dupe (Often with a glowing mass thrust into according to 72 traits in d’Huy 2012a. My first preliminary tale-type: G100: Giant ogre. Webster 1879: 4–6) 2013a) and. Frobenius 1996. Krauss 1883: 170–173) Russian people (Ralston 1873: 178–181. Dardy 1884) Romanian people (Grimm 1857: 15–16) Valais people (Abry 2002: 58) Balto-Slavic 3 Serb people (Karadschitsch 1854: 222–225.Table 2. Under the pretence of Polyphemus faced major problems owing to curing his eyesight. Language / № of Langauge Family Variants Sources Algonquian 4 Ojibwa people (Desveaux 1988: 83) Atsina people (Kroeber 1907: 65–67) Niitsitapi people (Spence 1914: 208–212. in the preceding issue of this (56) and traits (190) studied. 2013a). 38–41) Palestinian-Israelian people (Patai 1998: 31–32) Syrian people (Prym et Socin 1881: 115) Language isolates 5 Kootenays people (Boas 1918: 213–219. Chios: Ludwig 1863: 287–289) Albanian 1 Albanian people (Comparetti 1875: 308–310) Italic 10 Abbruzzian people (Nino 1883: 305–307) Sicilian people (Crane 1885: 89) Jean de Haute-Seille.

An ordinary Markov Chain Monte significant editing for the earlier publication Carlo analysis for 20. Deception through false doctoring. but Madisson 2011). Figure 1.1: Magic objects been analysed individually. The fact that a stationary reasonable. each version.75 (Madisson & shepherd. The coding also incorporated a tale in which a person gets into the homestead symbol (?) for uncertainty in the data. Figure 1. The present substitution (the GTR) with gamma- study is founded on the premise that the texts distributed rate variation across sites.2. published sources in several languages left column). Tree under the maximum parsimony and consensus criterions (right) and bayesian tree (right).2 summarized into one – consensual – tree The versions are drawn from diverse (strict consensus. These motifs can be found in sentences have then been added to an index to disparate ways in other tales. in contain. These escapes. representative of the traditions of the cultures with relative burn-in discarding the first 25% in question to make phylogenetic analysis of sampled trees. a simple model to calculate the hero escapes by holding on to to the fleece the 100 more parcimonious trees was used. I will only consider the motif of to their presence in (1) or absence (0) from the escape from Polyphemus’ cave (K521.000 generations. K1010: representativeness of the corpus. before being with a living animal.5.000 generations or could reflect summaries and paraphrases with 4 chains was run. or fur of an animal who is going out.1 (English. and each of compare the mythological versions they them has its own evolutionary story. adds five additional motifs: F512: Person of the results remains conditional on the unusual as to his eyes. Ronquist & the sources used were not available in forms Huelsenbeck 2003). Give alarm when fugitive the shortest possible sentences. This also means that the reliability distribution of values had been reached was controlled with Tracer 1. F531: Giant. the posterior distribution that are up to modern source-critical standards of phylogenetic tree for all the versions was and may have potentially been subject to inferred. in order to produce a binary K603) and I define the Polyphemus type as a matrix. Some of (Huelsenbeck & Ronquist 2001. the host wants to kill the hero.0 (Drummond & 50 . French. breaking it into betray fugitive. K521: Each version of the Polyphemus Tale has Escape by disguise. Then each tree was rearranged by subtree concealing himself under an animal’s skin or pruning and regrafting. With MrBayes 3. German. using a model of DNA (although see discussion above).000. The sentences were coded according this study. of a master of animals or of a monstrous With Mesquite 2. D1612. Italian). The forming the corpus are sufficiently trees were sampled every 5. So. treelength: 608.

Maximum tree with a midpoint rooting. Both runs produced Amerindian versions found in the Valais 8.002 trees. right column). Valais versions (Figure 2). This means that a shared form of the tale as a lord of animals. Homer drew on a primeval mythological To root the trees. At the end of the run.002 were sampled. On the both legends and mythology (cf. I used a midpoint tradition older than the Indo-European solution with the MrBayes tree. the between the European and Amerindian average standard deviation of split corpora. As pointed out by Frog (p. with a lord of wild animals similar to frequencies was 0. as polytomy is introduced if a particular split in the Valais3 and North American variants. 2010: 168). It is likely that the European version The tree obtained is a consensus tree from all exhibits the most archaic features. 2013a). the Valais is formally intermediate. corpus.c. of which 6. samples (excluding the burn-in). The phylogenetic narrative traditions and the images of different link between both versions possess a very categories of imaginal being adapt and are strong confidence degree (0.). which places tradition that included a belief in a lord of the root directly between the Ojibwa and animals.97) and was shaped historically in relation to dominant systematically found in the previous livelihoods of the cultural environment in reconstructions (d’Huy 2012a.005. af Klintberg one hand. occurs in less than 50% of all trees and so the would be in agreement with Burkert’s program was unable to resolve this lineage statement (1979: 33) that the Cyclops in (Figure 1. Rambaut 2007). The supernatural shepherd of 51 . Figure 2. created by a Considering the monster in the earliest 50% majority rule.

European traditions is equivalent to the lord (see also Frog 2011: 91–93. on which the hero’s escape then it is improbable that the necessary is dependent. Figure 4.4 keeper of animals. narrative are historically related and the The plot of the Polyphemus tale is narrative was not carried to the Americas by structurally dependent on the monster being a late medieval colonization by Europeans. If the 2012. which discusses a motif associated with North American and European versions of the the lord of animals also adapted to livestock). also Tolley which the attested versions derive. and which would account for its contact and exchange relevant to the spread of long-term stability as an element of the plot the European version with sheep antedated 52 . Figure 3. 2014). Principal coordinates analysis (Cosine). Principal coordinates analysis (Jaccard). It is of animals in his control of resources while therefore probable that this feature of the tale the resources concerned are connected to was established already in the form from different kinds of livelihoods (cf.

On the other the adversary was most likely a lord of hand. and confirm the intermediary Yet the local evolution of this tale shaped it situation of the Valais’ and Ojibwa’s versions like the (Palaeolithic) proto-form. analysis (transformation exponent: c = 2.62.3 unclear why it would be maintained only in and 4) and a non-metric MDS (Jaccard. It is therefore 10. 2D. Non-Metric Multidimensional Scaling (Jaccard). The lord of animals is attested among Similarity index: Jaccard.07. 2001) show a of animals in the Valais instance may not remarkably consistent pattern. The appearance of a lord Past 3. Figures 5 and 6) conducted with in only one variant. Figure 6. nearest before the domestication of livestock that was geographical versions tended to form sister maintained in isolation in Western Europe. Cosine: PC1: 35. explains its place in our analysis. which 53 . geographically reflect a historical continuity from such an era speaking (North America / Europe. fig.74. PC1: 29.0 (Hammer et al. Figure 5. Accordingly. PC2: 11. Non-Metric multidimensional Scaling (Cosine) the domestication of livestock. clades). the tradition area of Valais where it is attested Cosine. A principal coordinates construable form of the plot. PC2: several peoples in Europe.859. the Ojibwa’s branch is also situated in animals or equivalent figure in the earliest an intermediate place.

0. Admixture model.47% for the European cluster) and the Valais (0. number of examples from each culture in the as exhibiting formal distinction from these corpus may not be sufficient to reconstruct the larger groups. detect the true number of clusters (K.000 Burn-in period.4 conventional form of the tradition for any one (Pritchard et al.53% for the Amerindian cluster. it remains noteworthy more similar to each other than to the other that significant formal variations in the cluster (Parameters: 10. Using the software structure American branches of the tale. Magdalenian. Figure 7). The software also computes the probabilities of each version for each cluster. may be the earliest pictographic representation of the Polyphemus tale (Breuil 1930: 262). two main historical background behind the branching of clusters are identified among the variants in the Ojibwa (as well as the Crow) and Valais the way that the variants within a cluster are examples is unclear. Nevertheless. I also used Structure2. test for the phylogenetic analysis clearly shows 1 to 12 clusters) in the sample of versions distinct groupings of the European and North studied. this again suggests that these two versions are in the middle ground between European and Figure 7. 0. The limited clusters.72% for the European cluster). Although the Harvester (Earl & von Holdt 2012). Delta K's score associated with 1 to 12 Amerindian developments. recessive alleles model used for of the traditions rather than away from it at random. This makes it appear less likely that Figure 8a. 54 . 0. 2000.3. Falush et al.000 MCMC Reps after burn-in.28% for the Amerindian cluster. number of inclined toward the center of shared features iteration: 10. The ‘Petit Sorcier á l’Arc Musical’ [‘The Sorcer with the Musical Bow’] in the Cave of the Trois-Frères in Ariège. European and North American clusters are 50. The data align perfectly with the Amerindian/European distinction. 2003) to culture in a dependable manner. southwestern France. The probability is by far the lowest for the Ojibwa (0. Ancestry Model.

The images of Figure 8a that may be might be random. but can be compared to the within a dense. If this set of images elements indeed belongs together. This bow in his hand (on which see further enables the hero to escape the monster who Demouche et al. large looking at the animal within the context of a anus / vulva (Breuil 1930: 261. depiction of a historical event. This figure is striking controls the beasts from his dwelling. rather than would appear to parallel elements that stand at those that are different from it.e. the that they represent hunters in animal disguise escape of the hero. The narrative depicted Figure 8b. for the relationship to the upright bison-man Gourhan 1971: 97) and a very detailed. The bison-man could be current in some form in the Palaeolithic era. interpreted as some type of magical hunter. them as their protector. for the prominence of the anus / vulva on the human. Comparative evidence supports interpretation is that the figure of the human. it would also account (see discussions in Breuil 1930: 263. Magdalenian period. guardian or other Avantès. representation of a herd (Figure 8a). legs and holding or pointing a bow.the two major branches of this complex the one used by the North American Lakota narrative emerged independently of one hunters approaching their prey. A popular or relevance. Vialou 1987: herd: the bison-man would then fill the role of 116). they features he has in common with it. Interpreting narrative through image systems of a remote earlier period is inevitably problematic and speculative. human thigh and prominent rear orifice is The potential case is included as a scene equally obscure. superimposed and complex Amerindian versions of the Polyphemus tale. they can reasonably be presumed to reflect some sort of a narrative through its constituent elements. a flute or a musical bow (Bégouën An illustration of the Polyphemus tale can & Breuil 1958: 58). The bison-man would also (Demouche et al. Sorcier à l’Arc Musical’ has also be described as a man with a bison head playing an An Example from Palaeolithic Rock Art? instrument. The in which the hero often hides inside an animal scene in question depicts a bison-man with a itself by entering through its anus. depicted animal and the co-occurrence of this This being observes one of the animals which with the peculiar feature of a human thigh on – if correctly interpreted – has a human thigh the animal. a supernatural guardian of a herd watching for the hero who escapes by hiding within one of the animals. in the French Ariège département agent and representative (Clottes & Lewis- and the cave drawings appear to date to the Williams 1996: 94). the herd of animals and suggests his identity Provided that the set of image-elements have is somehow connected to the herd by the been more or less accurately interpreted. hunter or predator but rather aligned with Frères. in a way similar to be consistent with the proposed evolution of 55 . long before the first The peculiar image of the animal with a domestication of animals. but its choice Breuil’s drawing). Some believe the core of the Prometheus tale – i. local or reflect an imaginal relevant to the Polyphemus tale (Breuil 1930: 262. A motif in that it appears to be a rather detailed of the hero hiding in this way would account representation of a bison standing on its hind. Another possibility is that potentially be interpreted from the this figure is not separated from the herd as a Palaeolithic cave drawings found in the Trois. Leroi. the probability that the Polyphemus tale was bison is a ‘shaman’. as seen more clearly in Figure 8b. In addition. The ‘Petit another. 1996). as a subject for representation could also be Interpreting such images is necessarily connected to some type of social prominence speculative and problematic. 1996). This cave is located in Montesquieu. and its longue durée is a relevant indicator but the bison-man head identifies him with that it held social interest and relevance.

A real conflicting signal Mesquite tree and 0.12. The main delta-score is here 2010).63 for the Bayesian tree between versions (box-like structures) was (Jukes-Cantor model. evaluates the degree to which a trait is shared characters transformation: Jaccard. An RI of 1 This algorithm makes it possible to see indicates that the tree shows no borrowings.63) are 0. whose mean RI is 0. Figure 9). NeighborNet graph of the Polyphemus variants. Bryant et al. for the model of its history. 2005. the tale’s protagonist in Europe/Eurasia from broadly the same as the mean RIs for the a guardian of animals into a herdsman of biological data sets presented by Collard et al.57 and 0. similar to those found in both trees instance. the Retention Index (RI) has been The results also should be controlled with calculated for both our trees. Index is a traditional tool in cladistics and Bryant & Moulton 2004. noise.57 for the between the taxa. The Retention NeigbhorNet (implemented in Splitstree4. Indeed. Thus the vertical speculative.Figure 9. of the stories. doubt.3422. uncertainty as while an RI of 0 indicates the maximum webbing. were notably a herd animal). conflicting data. These indices indicated that most brings the major part of these versions of the mythemes were shared through together into coherent geographical or cultural common descent. However. in completely significant rather than. and proposes good representations amount of borrowings that is possible.61. then. However. The may be guardian of a particular species. high RI values (for clusters. test this. biological data sets of Collard et al. This interpretation is structured by speciation. explaining it as an interpretive bias rate of borrowing of mythemes to be low. NeighborNet correctly simulated). To in selecting. through common descent.60) usually show a (see below). The RI about both clusters and evolutionary distances calculated with Mesquite was 0. suggesting a good conservation low horizontal transmission (Nunn et al. folktale data. greater than 0. labelling or interpreting data. Both RIs obtained (0. 1000 characters found. The delta-method scores individual 56 . we would expect the example. note that the RI for the Polyphemus myth does not look The Tale’s Retention Index sufficiently high enough to consider it If Polyphemus is a Palaeolithic tale. domesticated livestock (noting that here he (2006). but it is not unreasonable and is transmission (from mother to daughter worth putting forward owing to what we populations) should be the dominant know of the tale and can infer about its evolutionary process in both biological and history.

area across millennia. If the Valais variant is left aside as an exceptional outlier in the data The Reconstruction of a Protoversion and the branching of the Syrian. These phylogenetic reconstruction methods The enemy is a completely solitary figure. this new version where the immovable or a large door. can been reconstruct the probable form of the first reconstructed: Palaeolithic state of the Polyphemus family. Greek 2 and Two phylogenetic comparative methods Abaze variants’ cluster treated as the root (Maximum Likelihood with model Mk1 and point of the European variants' stemma (fig. escape. The monster possesses a been retained.] Then he falls asleep and a vengeance occurs that is [The enemy is a completely solitary figure. the hero clings to a living animal. of sheep is estimated to fall between nine and eleven thousand years ago in Mesopotamia. which may Mesquite 2. In the text. the following text. mythemes with herd of domestic animals (sheep). 2005) and probably including K603. Note that this model is from stemmatic models developed by other linked to features that are also correlate with means. claims (2013). in blue) 0. [To enters in the monster’s house [which is a hut. connected with fire. the hero clings to a living animal. a stemma for Germanic languages should appear more or less the same even if we were 57 . and is are applied to each mytheme of the family. 3. affronted alone by the hero. The monster waits for who is affronted alone. with more than a know whom he will meet]. Thus. the Amerindian traditions. Their average is migrated toward the South (fig. it versions and another choice of traits to study may have been disseminated through the tale.] The According to the reconstructed origin of the monster possesses herd of wild animals. He enters the 50% probability using both methods – have monster’s house. A human Then mythemes reconstructed with a high [perceives a light in the distance and does not degree of confidence – i. Greece / Syria / traps the man and his own animals with an Abaze. the domestication date this story.75 have been used with maximum approximate the features of versions of the parsimony and consensus criterions tree to European Neolithic proto-tale.taxa from 0 to 1. where reconstructed makes the outcome appear relatively traits are not necessarily those which occur consistent with what would be developed most frequently.] Then he waits monster was in a shelter and the animals were for the man near the entrance to kill him. if not more so.’s during the Last Glacial Maximum (Pala et al. [To sheep may go back to about the domestication escape. Parcimony reconstructions) implemented in tree at the left). Whichmann very hypothetical reconstruction. It could be the Palaeolithic myth of successive migrations from the Mediterranean the first appearance of game on Earth.3113. this is a in the data (Holland et al. yet they can propose remains almost the same (Figure 10). This statistical reconstructions. [The hero does not know whom he will meet. a vengeance occurs that is connected with fire. (2011) calculates delta scores across the the palaeolithic populations may have world’s language families. 2012.3. et al. In Europe. [He European type (Figure 3.e.] a house or something similar].] In of the species.] A human hunter the man near the entrance to kill him. in blue). 2002). immovable or a large door. However. Peričić et al. version of the story in which the monster was a master of animals. some families of folktales. a relatively high delta score We can attempt to correlate the trees with a (close to 1) shows a strong conflicting signal model of historical spread. Phylogenetic methods cannot discover the This model has been tested by removing original form of a story in the sense of an the Amerindian data: the Bayesian tree Urform with certainty. [He traps more than 75% probability have been the man and his own animals with an underlined. If This abstract is very close to what has been the new Polyphemus’ tale type was linked to found previously (d’Huy 2013a) using fewer the early stages of animal domestication. a giant who has one eye in the forehead. Indeed. because the whole branch is stable (cf. are at least as tree-like as preserved at least partially a reconstructed languages. contrary to Ross and al.

one of the first steps of diffusion in Europe includes Basque. I have observed a similar Following the working hypothesis that cluster previously (d’Huy 2013) with the use complex narratives of the escape from of other mythems to study the Polyphemus' Polyphemus are unlikely to emerge tales. It could be easy to explain. For cultural and population histories. During the independently of one another. Celtic languages were could only have spread across Eurasia and spoken across much of Europe. it Basque versions may be a borrowing from the becomes necessary to account for the tale. these stories first millennium BC. neighbouring Celtiberian (spoken in ancient 58 . the Black joined present day Alaska and eastern Siberia Sea and the Northern Balkan Peninsula. Burkert’s hypothesis that an earlier A 10. For instance. Yorkshire and Trends of Stability and Contrasts the West Highlands.000 year model of population form of the tale incorporated a belief in a lord movements and cultural changes have of animals requires the Amerindian branch of probably had transformative effects on data in order to advance beyond speculation traditions across Europe and America for to have empirically based support. unaware of a connection to Indo-European). the Pyrenean area. In this case. example. conditional millennia. Figure 10. Oghuz Turks. The during the Pleistocene ice ages. type’s distribution in two very large areas that Yet this tree alone would not resolve which are geographically remote from one another features in the primary split should be and diversity within these different areas considered probable for an antecedent form which only partly seems to correlate with other than those shared across that split. including North America when a former land bridge Great Britain. one can propose that on the improbability of multigenisis. Bayesian tree calculated without the Amerindian versions.

significant evolutionary change occurs in a it would suggest that different versions of the species. Pagel et al. there should number of speciation events (Webster et al.] in fits and starts. The results have been far superior to those 59 . The mean Amerindian data and 0. Gould & Eldredge receding in the wake of history. the result may imply that the Gallic invasion of the Balkans in 279 BC. tale.79. which would be followed by reconstructable version of the tale with any extended periods of stability with little net particular geographical space. each speciation events (nodes) have occurred.77E-13). p = linear relationship (Pearson + Spearman5) 0. The link between the Pyrenean area results about a European folktale). If an analogy may be drawn. 2006). be low relationship between geographic 2003.15E-22.043) in the of the tree to each final version. If this has happened repeatedly.4.” One sign of the coefficient matrix (permutation: 10000) on punctuational evolution of myths is the individual version data with SAM v. p = 0. which would be borrowing without However. it is generally restricted to rare and story have been ‘seeded’ through Europe very fast events of branching speciation again and again. could explain the clade languages. 2013c–d).008.0 correlation between branch length and the (Rangel et al. superseding one another and (Eldredge & Gould 1972. 2013 for higher Goidelic. Berber and Russian area. diffusion of versions could be more More precisely. there new version taking the place of older should be more total genetic change (longer versions. with many waves of was 0. and the West Highlands is included in versions (see Ross et al. so this low result estimate of the punctuational effect on the suggests a very complex evolution for the clock-like behaviour of these trees. included in the books and papers.07. The correlation square between path lengths and nodes in the coefficient detects only linear dependencies MrBayes’ tree has been used to give an between two variables. evolutionary change. and not necessary from the root variance (r² = 0. or what Frog (2011: 91) To test the multiple migration hypothesis. p(uncorr): 2. The geographical locations of each predicts no relationship between node and version were estimated using information path lengths. Yorkshire belongs to the Brittonic joining closely Israeli. establishment of a branch of the tradition How could the Polyphemus’ tales – and could not be shown to be ‘true’. newly consequently seem to make it difficult to mythological sister versions would tend to correlate the earliest. I has described as “the evolution of tradition realised a Mantel test using a Jaccard\s [. one of the phylogenetic (only the existence of a parental three tribes who settled Galatia (an area in the version needs to be taken into account) than highlands of central Anatolia) ca. I adopted The mean length has been calculated for the centroid of geographical coordinates for each branch of the Bayesian tree (Figure 2) each language area when no precise from the final version to the first polytomy – geographical information was available (using more than two based branches. The theory of population movements and cultural changes punctuated equilibrium states that when of traditions. Another hypothesis could be a came from southern France and could very good conservation of the structure of the potentially be the vector of transmission. p(uncorr): 6. but as a other tales – evolve and survive from the possible but indemonstrable explanation that Palaeolithic period? Biology may propose a would be the outcome of the effect of model (d’Huy 2013a.85 (Pearson: 0. an account of the Celtic major modifications. Where many distance and similarity between versions. diffusions (rather than a single one). If correct. long-distance influence of certain versions. geographical.3) among European data. A gradual model of evolution diffusion. which is also a the websites Glottolog and Wals). palaeolithic diverge rapidly. In these and Oghuz Turks could be explained by the conditions. I found that sign of punctuational evolution (Wagner & geographical distance explains 7% of the Erwin 1995).91..8% (r² = 0. and the Spearman’s rs: 0. 2010). This would 1977). breaking the continuity of linear path lengths).. The result European versions. the Tectosages. 270 BC.times in the Iberian Penninsula) or Gaulish such the Homeric one.

69. which increase the rate of analysis based on the textual surface of a heterogeneously written corpus. 2006) or in languages (10– Acknowledgements: The author gratefully 33% being the overall vocabulary differences acknowledges comments from Frog that played a significant role in developing this paper. artifact’) may lead us to believe in a false with a common ancestor. see Atkinson et al. while an R² importance of experimental replications and closer to 0 indicates a regression line does not using a larger database. have been corroborated here. Another approach.75). the Dragon: d’Huy 2013e. Punctuation may formal similarity in the co-occurrence of motif elements in one episode even if the overall narrative thus reflect a human capacity to enhance both form and structure was close to that of another set the group identity and the identification of of variants. Aix-Marseille (Venditti et al. The considerations of the past behind the remaining variation in path could be documented evidence.0 indicates that a in previous studies. previous works. Pagel et al. d’Huy 2013c: 100). 2008). a be to use the tools belonging to the field of Natural story of the origin of fire was told by an Language Processing. the closer Amerindian to offset another story by an the contents of two narratives (as reflected through their surface texts). An R² near 1. sentences. Here. to consider how to prevent ambiguity in the identification of unique terms and terms with many 60 . Centre punctuational effect in biological species Malher. not used here. However. This coding would concern and people belonging to a certain tribe the whole text and avoid the pre-selection of traits explained that another tribe with whom it (which is perhaps not so significant: see d’Huy shares many myths did not know how to tell 2013f). 2013. It is also necessary change (for an example. which shows the regression line fits the data well. 2013f. into how a tale evolves. d’Huy & Dupanloup 2015.Paris1/EPHE/Aix-Marseille Univ-AMU). UMR 8171). the shorter the distance between Indian of another tribe (Goddard 1904: 197). 2013a.obtained from biological data (r = from 0. It should contribute Notes 75% to the evolution of the Polyphemus tale. 75004 Paris. groups of founder effect. small social communities tend sentences. 1. 2013c. the trees obtained are the coefficient of determination (R²) has been better and more coherent than those obtained calculated. the Kind and the Unkind Girl: Ross et al. d’Huy Ethnology provides a model that could 2014b. Trees also did not show the curvilinear trend Julien d’Huy (dhuy.julien[at]yahoo. 2006). see d’Huy 2013f).fr) Institute of the that characterizes the node-density artefact African World (IMAF.22 Conclusion to 0. the Cosmic a result close to what was obtained for Hunt: d’Huy 2012b.18. Polyphemus: d’Huy another tale-type: the Cosmic hunt (84%: 2012a. also found in R² with a logarithmic regression (R² = 0. France. the R² with a reconstruction and the punctuational linear regression (R² = 0. could individuals with this group. and to what extent the reconstructions (the so-called ‘node-density versions studied belong to a same tale-type. 2006). rue Malher. Concerning the punctuated equilibrium effect. (22%. parts of text or structural formations) and to lose part of their mythological complex and whether certain words. 2013a. Paris I Sorbonne. Traits were selected for the whole tale in order to Folktale variations are largely defined by avoid the possibility that variants could be grouped people drawing a line between ‘us’ and together in analysis owing to a concentration of ‘them’ (Ross et al. explain these mythological punctuations. 2014a. Little Red Riding Hood: Tehrani 2013. into the tale’s A well-known artifact of phylogenetic possible prototype. 2013). To conclude. among languages within a language family. 9. phylogenetic and statistical tools showing a greater change of the tree length used to study folktale allow us to return to attributable to punctuationnal effects. mean R² = 0. They can offer insights explained by independent gradual effects. For instance. paragraphs or the experience something similar to founder whole text should be rewritten to facilitate the events and drift. With these tools. Pygmalion: d’Huy 2012c. such an approach would require them (Désveaux 2001: 85). 2. such as asking which elements should be effect also could be due to a mythological compared (individual sentences.83) is higher than the evolution of the folktale. the narratives would be. The proto-myth fit the data very well. family of this folktale. The punctuational taking many precautionary measures (d’Huy 2014c). (CNRS/IRD/EHESS/ effect for this folktale is stronger than the Univ. To avoid this. Pagel et al. The punctuational University.

Thomas Frederick. long and short 65.. Nart Sagas from the Caucasus. Pp. “Omer m’a tuer… ou moi-même en Traits”. Henri. The potential for these Volp: TroisFrères – Tuc d’Audoubert. “Un dessin de la grotte des Trois avalanche. Ancient Greece. S. Bordeaux: Auberon. “Untangling in the form of the protagonist from an Urform of Our Past: Languages.J. In The Evolution of Cultural Diversity: 4. 1860. Simon Greenhill & Ruth Mace. The Craft of Poetic Speech in America). Arts et tools. Quentin D.edu/hopper/text?doc=Pers Loescher.C. D. Légendes et récits variables are independent. versions. Ed. 1890. 2004. In La Mythologie et l’Odyssée: Society B 365: 3903–3912. “The Giant of New Mills. 2010. Popular Tales of the Spearman correlation is sensitive when two West Highlands Orally Collected I. the Clottes. Note that. Ithaca: Cornell 5. Geoffrey Lewis.. be compared if the proximity of elements Bégouën. Jean. rev.-F. Novelline popolari Chief Gregory R. Pearson's and populaires du Pays Basque. Bureau of American of the story: the dwarf (structural inversion of the Ethnology. J.. Americas in the late medieval colonization by London: UCL Press. J. The Odyssey. & H. les grottes ornées. 1930. but that is a Bladé. R. Paris: P. Folk-Lore 1: 130–131. in Comparetti. 2002. 67–84. In Comptes- traditions. & trans. & J. 1891. widespread dissemination of the motif in North Calame. tales collected on site or tales collected Atkinson. The Pearson correlation coefficient is sensitive only University Press. Samuel Butler Populations”. Walter. 2005. C. F. A. Breuil. giant) tries to punish the hero by creating an Breuil. Hurst & F. had been kept indoors too long. David Lewis-Williams.. they opened the stables for sheep that inscriptions et belles-lettres 74(3): 261–264. The Book of Dede Korkut. épiques.0218:book=1:card=1. such as special ceremonies or evening around the Baring-Gould. and b) because of the University of California Press. Mace et al. et al. Valais story as representing a historical continuity Bryant. certain giants protected the singing birds rendus des séances de l’année 1930: Académie des and sheep. 1958. Perseus Digital Library. Trees. sud- international de Grenoble 20–22 mai 1999. Orion. John.57– transcriptions and translations. “Branching. The hypothesis that the tale was carried to the Phylogenetic Approaches. Crane. the Campbell. Ed. “Languages Evolve from Westernized people under different conditions in Punctional Bursts”. 1856. Jean de Hauteseille. Moulton. Abazas. 61 . Les Cavernes du within a text is a factor. Trans. Les coefficient is 1 if the variables are perfectly chamanes de la préhistoire: Transe et magie dans correlated. Abkhaz and Ubykhs. Li romans de Dolopathos. M.. Collard. related. Evolution and Human Behavior 27: (London: A. 1900). 2002. 1974.. Letoublon. F.01. Jannet. Fifield. and Gregory Nagy. Europeans is unlikely because of a) the coherent Burkert. An initial potential indication that the Valais variant Boas.J. Thomas E. Spearman's correlation coefficient is 0. Structure and History in Greek clustering of European variants on the one hand and Myth and Ritual. Jean-Francois. Trans. Science 319: 588. Yet this motif Phylogenetic Networks”. Agglomerative Method for the Construction of avalanches (Rohrich 1976: 142–195). Edinburgh: variables being compared are monotonically Edmonston & Douglas. Timothy Power 169–184. John-Francis. 1979. tale). 1992. Shennan & J. needs to be explored further. and the Evolution of Harmondsworth: Penguin.perseus. Anthologie populaire de l’Albret. Pp. Genève: Droz. 2008.. 1885. ouest de l’Agenais ou Gascogne landaise II: Contes populaires. “Is Horizontal Transmission Really a Problem for Phylogenetic Comparative Methods? – Literature A Simulation Study Using Continuous Cultural Abry. Ed. hommage à Gabriel Germain: Actes du colloque Dardy. L. set free badly Bryant. 1875. Paris. C. 2014c. Contes de Gascogne I: Contes matter for another paper. the tale. partially explored in d’Huy 2013e and d’Huy Métiers: Graphiques.tufts. Gray. possible significations as well as how original Ed. Domenico. Berkeley / Los Angeles / London: American on the other. Colarusso. 3. Crane. Leclerc. Filimon & R. Currie. according to old Tyrolian frères (Montesquieu-Avantès) Ariège”. Works Cited from the Circassians. to a linear relationship between two variables. “Neighbor-Net: An treated cattle and punished cruel people with. Sources Princeton / Oxford: Princeton University Press. D. Available at: Italian. 1995. Tehrani. 1996. Rome / Turin / Florence: Ermanno http://www. Paris: Seuil. Cultural Smilarities and Differences among Human Homer. Paris: Maisonneuve & Ch. Blending. Agen: Michel & Medan. campfire (which can influence the content of the Sessay”. Molecular Biology and is not strong support for the lord of animals in the Evolution 21 (2): 255–265. 1886. 1918. S.D. Charles Brunet & Anatole de Montaiglon. Splits and Networks”. & V. Washington: might maintain archaic features appears at the end Smithsonian Institution. Italian Popular Tales. H. eus:text:1999. even if their relationship is not linear. Ed. Philosophical Transactions of the Royal personne”. 2006. London: Macmillan. Kutenai Tales. If the Cerquand.

Texts. Quadratura Americana: Essai Essais d’ethnographie et de linguistique (deuxième d’anthropologie lévi-straussienne. “Un ours dans les étoiles.. with Grammar. 2nd rev. Paris: Publications Orientalistes de 155. http://palaeo- Frobenius. 101– Préhistoire du sud-ouest 20(1): 91–106. 1921. 2012. & Bridgett M. & Niles Eldredge. Frog & Pauliina Mythologie Française 248: 47–59.T.1007/s12686-011-9548-7. “STRUCTURE HARVESTER: A Website and Gray. Rémy (ed. & A. 1996. Grimm. Richard McGillivray. La Maison des Sciences de l’Homme. “The Polyphemus Myth: Its Origin Dor. 1965. Chên-Bourg: série). Latvala with Helen F. Religions. Stephen Jay. 2014. In d’Huy.. M. Thomas E. Analyzing Phylogenetic Distance Data”. Boltz. Harper & P.D. University of Press. mœurs et légendes: Désveaux. II.A. A. E. Pp. A.P. Leslie. Karina Cappadocia and Pharasa. Doi: 276(1665): 2299–2306. Hafstein. 1904. Sahara 23: 49– Method(s)”.P. 131. & J. Molecular Frog. et al. “Le motif de Pygmalion: Origine Frog. V. Trans. “Ulysse. Helsinki: temporalite chez les Ojibwa septentrionaux. 1978. Le livre des heros: Legendes Evolution Reconsidered”. L. Leipzig: Friedrich. Iena: Revue de Littérature Comparée 15: 573–623. In Limited Sources. Archaeologia Baltica 15: 78–98. Currie & Russell D. Georges. Goddard. Dirr. Hupa Texts. James George. Die Polyphemsage in der Gradualism”. J. BMC Evolutionary Biology 7: 214. phylogénétique sur un mythe préhistorique”. Adolphe. 1857. 2013a. “BEAST: Berkeley University Press. D. Ed. J. Georg Editeur. Eldredge. Rambaut. Ed. Retrospective Methods. Helsinki: Academia Scientiarum Fennica. Berlin: Konigl. Schopf. “δ Plots: A Tool for Edisud. 1935. B. Alan Dundes. “Does Horizontal Transmission Program for Visualizing STRUCTURE Output and Invalidate Cultural Phylogenies?”. Frog & Sakari Katajamaki. “Le Conte-Type de Polyphème”. New York: Trustees. Helsinki: Folklore Studies. 1904. Greenhill.. Paris: Academia Scientiarum Fennica. Pp. “Punctuated Equilibria: An Alternative to Phyletic Hackman. 1988. Trans. 1977. Cambridge: University 7. 10. 2012c. Demouche. Folklore Theory: An Essay in the History of Ideas” Falush. Monstrueux. “Germanic Traditions of the Theft of the “Nuovelle approche de la gravure du ‘petit sorcier’ Thunder-Instrument (ATU 1148b): An Approach to à l’arc musical de la grotte des Trois Frères Þrymskviða and Þórr’s Adventure with Geirrøðr in (Ariège)”. 1996. Contes Kabyles II: Le electronica. In Models in Paleobiology. Glenn. RMN Newsletter Translations and Glossary. Genetics 164: Hammer. Bayesian Inference of Phylogeny”. 2012b. M. France. le cyclope et les Berbères”. 407–435. 2001. Kaukasische Marchen. 18–34. Helsinki. Pritchard. 1983. Apollodorus: The Library Package for Education and Data Analysis”. Bayesian Evolutionary Analysis by Sampling Gould. In New Focus on Méditerranéennes 5: 35–37. Goddard. Lukin.htm. Berkeley: Drummond. A. Ed.org/2001_1/past/issue1_01. FF Communications 307. Akademie der Wissenschaften. Ed. Challenges of Oral and Written Texts. 2003. San Francisco: Freeman Cooper. & Stephen Jay Gould. Pliny Earle. 2009. Pp.M. Justin. d’Huy. Ländlische Briefe von Anthropological Papers of the American Museum Georgios Drosinis: Land und Leute in Nord-Euböa. Wilhelm. Correlated Allele Frequencies”. London / New Multilocus Genotype Data: Linked Loci and York: Routledge. Préhistoire Anthropologie Circum-Baltic Perspective”. recherche Academiae Scientiarum Fennicae Humaniora 368. 120–162. Germain G.K. Die Sage von Polyphemus. 2012a.. Conservation the Royal Society B: Biological Sciences Genetics Resources 4(2): 359–361. Aix-en-Provence: Holland. “Biological Metaphors in 82–115. Fetta. J. Contes Kirghiz de la steppe et and Interpretation”. Georgios.. 2013b. 62 .). 1909. 2001. Verlegt bei E. Trees”.R. “Punctuated Equilibria: The Tempo and Mode of Dumézil. 2001. 2007. 1922. Drosinis. “PAST: Paleontological Statistics Software Frazer.J. F. Jicarilla Apache Texts. Ed. 1972.E. “The Parallax Approach: Situating Bioinformatics 17: 754–755. D. Approaching Methodology. Eldar Heide & Karen Désveaux.T. Stephens. Ronquist.R. Pp. Volksuberlieferung. E. 2002. “MRBAYES: 1148b)”. Dent A. 151. 1911. “Circum-Baltic Mythology? – The Strange Biology and Evolution 19: 2051–2059. 1884. Greece and Rome 25(2): 141– de la montagne. T.J. London: William Heinemann / New York: G. Boundless 58. Paris: Mercure de France. Helsinki: [Hackman]. Paris: Gallimard. Proceedings of Implementing the Evanno method”. & F. Niles. Modern Greek in Possibilities: Textual Scholarship and the Asia Minor: A Study of the Dialects of Siĺli. Earl. 1567–1587. Ryan. Sous le signe de l’ours: Mythes et Bek-Petersen. Paleobiology 3(2): 115– sur les Nartes. Simon J. Frog. Ø. Pp. 2005. Palaeontologia Electronica: 4(1). Available at: Putnam’s Sons. Frog. Leo. van Gennep. von Holdt. Annales d’Huy. In Folklore: Critical Concepts in Literary and “Inference of Population Structure Using Cultural Studies. edn. Deflandre. Case of the Theft of the Thunder-Instrument (ATU Huselsenbeck. “Revisiting the Historical-Geographic afrasienne et diffusion en Afrique”.. of Natural History 8. 1916. Slimak & D. J. O.Dawkins. Traditions in Long-Term Perspective”. P. Diederichs. 2011.

P. 2014b. dupanloup-d-afrique-en-amerique%C2%A0-la. Mythologie Française 254: 14– Bioinformatics”. J. Stephens & P. Matthews. Kraus Reprinting. Peričić. Leipsic: W. Antonio. Robert H. 2006. Myths and Tales http://nouvellemythologiecomparee. 2015. J. Multilocus Genotype Data”. Philosophical Reconstructions. Borgerhoff http://pm. of the Jicarilla Apache Indians. Oslo: Poestion. L’origine des manières de 959.C. 1137): A Lowie. The Sky Is My Tipi. table. G. Available at: Mythology”. Donnelly. Morris Edward. Myths and Tales Statistical Approach”.75. 8: 13–29. J. 1– Préhistoire du sud-ouest 21(2): 195–215. Phylogenetic Reconstruction of Palaeolithic Version 2. Berlin: R. Kiowa Tales. Pritchard. Occidental.K.L. and D. “Les contes peuvent-ils servir à Folktale: Using a Multiple Alignment Program of remonter le temps”. 153–174. Nouvelle Mythologie Comparée / New Kraus Reprinting. 1940 [1969]. 2012. “Inference of Population Structure Using Paris: Plon. Erinnerungen und Mittheilungen Mythology and Its Archaeological Consequences”.org. “Description of the Structure of the d’Huy. 2013f. 2002. 1863. 1968. J. Claude et al. Senri Ethnological Studies 55: 16. Claude. 312. 2013c. d’Huy. Die folkloristische Populations”. Lapplandische Marchen. Pagel. 1938 [1969]. Mazenod. J. Pp. Maddison. London: 314(6): 119–121. Russian and Bulgarian Divergence at the Molecular Level”. Sagen und Märchen der Parsons E. “Les mythes évolueraient par McAllister. André. Rathsel und Sprichworter. Raphael. Syrische Sagen Lévi-Strauss. Strickland. Cross-Cultural Comparison”. Ed. American Journal of Human Academia Scientiarum Fennica. Préhistoires Méditerranéenne 4. Kraus Reprinting. 1971.C. “Polyphemus (Aa. Available at: Opler. Leroi-Gourhan. Kaarle. New York 1969: archive/2015/02/19/julien-d-huy-et-isabelle. d’Huy. Trans. 1918. “Gros Ventre Myths and Tales”. 308.. Helsinki: Eastern Refugia”. Biological Sciences 365(1559): 3807-3819 Mythologie française 256: 17–23. Dupanloup. Jaronomir Erben. Paris: L. J. M. Lévi-Strauss. Contribution of Speciation to Evolutionary Karel. 1883. Episodes of Paternal Gene Flow among Slavic Krohn. & Albert Socin. Th. “A Phylogenetic Approach of Ludwig. Mark et al. et al. 2013b. M. Volkssagen. J. Molecular Biology and Evolution Arbeitsmethode: Begründet von Julius Krohn und 22(10): 1964–1975. “Kiowa-Apache Tales”. Gaertner. Rock Art Research 30(1): 115–118. “Il y a plus de 2000 ans. New York 1969: vol. “A Cosmic Hunt in the Berber Sky: A a Modular System for Evolutionary Analysis. 1883. “Myths and Traditions of the Phylogenetic Reconstruction of a Prehistoric Tale”.d’Huy.com/ of the Chiricahua Apache Indians. C. Detroit: Wayne State University Press. 2000. Aschehoug. 2014a. J. Oda. J. évolution et origines Transactions of the Royal Society of London B: paléolithiques du combat contre le dragon”. paléolithique: Mythologie et archéologie”.L. The RMN Newsletter May. Eugen. 1907. Barbera. Pala.hautetfort. Usi e costumi abruzzesi III.org/814. and Israel. d’Huy. Gilbert. Boatright. http://mesquiteproject. Ross. Paris: Plon. d’Huy J. 1929 [1969]. A. “High-Resolution Phylogenetic Anthropological Papers of the American Museum Analysis of Southeastern Europe Traces Major of Natural History 1(3): 55–139. FF Communications 300. 1886.P. Crow Indians”. le mythe de de Nino.. W. “Mitochondrial DNA Signals of af Klintberg. “Classification and Maerchen aus dem Volksmunde. Anthropologie structurale. Vienna: C. 141.. Denton: University of North Texas Press.. 1942 [1969]. Claude. & I. Morris Edward. 1998. Arab Folktales from Palestine Volkergruppen I. Kroeber. J.revues. 2013e. “D’Afrique en Opler. Friedrich S. Mulder 2010. 63 . d’Huy. W. 1958. Préhistoire de l’Art Gerold. 2001.html. Science Folk-lore Stories. “Mythologie et statistique. Gottingen: phylogénétique du vivant”. 1949. 2013d. 2011. 1926. Anthropological Papers of Nouvelle Mythologie Comparée (New the American Museum of Natural History 25(1): 1– Comparative Mythology) 1: 3–18. 1881. M. J. Mody C. Standring. “Simulating Trait Evolution for d’Huy. Florence: Tipographia di G. J. Genetics 90: 915–924. und Marchen der ubrigen indogermanischen Patai. 2014c. “Large Punctuational 5562363.W. Mesquite: d’Huy. 2005. L. Maddison. Arnold. “Motifs and Folktales: A New Opler. Les Cahiers de l’AARS 16: 93–106. New York: 480)”. Mythologie Française 252: 8–12. New York: Sudslaven in ihrem Verhaltnis zu den Sagen Kraus Reprinting. weitergeführt von nordischen Forschern. Friedrich. Morris Edward. Bengt. C. The Types of the Swedish Late Glacial Recolonization of Europe from Near Folk Legend. Pygmalion existait en Afrique du nord”. In ponctuations”. aus Griechenland. 1907. L’Homme 162: 309– Vandenhoeck & Ruprecht. Prym. J. et al. Available at: Nunn. Krauss. “Le motif du dragon serait Publications of the Texas Folklore Society 22.R. d’Huy. Myths and Amérique: La bonne et la méchante fille (ATU Legends of the Lipan Apache Indians. Comparative Mythology 2. Genetics 155: 945– Lévi-Strauss. 2010. 2013a.

82–119. Mythology of the 137. Ungarische and Speciation”. Karadschitsch. A. Anthropological Papers of the Tehrani Jamshid J. Chris et al. In Mythic Discourses: Finno-Ugrian Studies in Oral De situ linguarum fennicarum aetatis ferreae. Anna-Leena Siikala & Eila Folk-Tales. L’Art des Cavernes: Les Ross. Zeitschrift fur Volkskunde 1: 87– Thompson: Tales of the Stupid Ogre. The aim of this article is to provide the first of Finnic. 219–242. Freiburg: Herder. 1961. edn. Pp. Studia Fennica Folkloristica 20. 2nd rev. Field Columbian Museum. The Types of the Folktale: A Classification and Bibliography: Antti Aarne’s Internet Sites Verzeichnis der Märchentypen Translated and Glottolog. Le folk-lore du pays basque. Ed.11: e78871. Sage und Märchen: Erzählforschung Tiedeakatemia. Crowell Company. and Change 1: 205–240. Tolley. Language Dynamics von Sydow. 1883. Lewis. Stith. New York: Thomas Y. 2003.). London: Griffith & Studies and Philology: Some Points of View”. “Traditions of the Wagner.M. Dundes. Pp. “Folksagor. New York: Columbia University Indian. S. Helsinki: Academia Scientiarum Fennica. “Detecting the Nodedensity Ronquist. WALS = World Atlas of Language Structures. 2011. 1889. L. Vialou.. D. Denis. William Ralston Shedden.. Publication 85. Folktales: A Classification and Bibliography.. in Linguistic Phylogenies”. Englewood Wichmann. & D. York: American Museum of Natural History. T. http://wals. For three maps of the Uralic-speaking peoples in reasons of length. FF Communications 3. Elder and Co. Thompson. 1914. Pp. Press. Russian Tradition. Gaals Georg. Vinson. Systematic 3: Bayesian Phylogenetic Mixed Models”. Carl Wilhelm. “Folktale Chiefly in the Labourland. Enlarged.F. Erwin. Tochter. heute. 1987. “The Phylogeny of Little Red American Museum of Natural History 2. Cultural Geography of a Folktale in Europe”. Simms. This opening article of the series presents a heuristic cartographic model of estimated locations of groups speaking Finnic languages and their neighbours in ca. Rangel. In Farran. 1873. 1908. London: Smith.L. AD 1 and a map materials and present different issues has indicating the linguistic Urheimats of required presenting the investigation as a reconstructed intermediate proto-languages series. Blackfoot Indians.H. Andrea J. Robert Payne & Mark Pagel. 1976. The Myths of the North American R. “Molecular Phylogenies Link Rates of Evolution von Stier. Folkminnen och Folktankar 14: 105– Wissler. “On the Trail of Þórr’s Goats”. New Riding Hood”. Berlin. biology 55(4): 637–643. 1927. Proto-Sámi. 2013. Sciences 280: 1756. and Formula Tables. 87–122 won Stephanowissel.. Application for Spatial Analysis in Macroecology”.1. Chicago: Field Models”. 2012. Richter. et al. Hans-Jörg. forskningen”. 2004.C. 1879. Volksmarchen der Serben. PLoS ONE 8.6. 2010. with W. Julien. 1854. In New Approaches to Studying Columbian Museum. Basque Legends: Collected. Ed. Pp. etc. Pesth: Heckenast. Erwin & Spence. Peter J. F. University of Helsinki Abstract: This article is the first part of a series that first employs a descendant historical reconstruction methodology to reverse-engineer areas where Finnic languages were spoken especially during the Iron Age (500 BC – AD 1150/1300). Crow”. Paris: Le Rocher. Science 301: 478. Les Proceedings of the Royal Society B: Biological littératures populaires de toutes les nations 15. “MRBAYES Artifact in Phylogeny Reconstruction”. Paris: Maisonneuve. Frog. Helsinki: Suomalainen Rohrich. von Sydow. Stephen Chapman. The Study of Folklore. 2003. Ralston. http://glottolog. “Population Structure and Sanctuaires de la Préhistoire. 1948 [1965]. et al. Anecdotes and 93. Jokes.info/.P. 1857. Volksmarchen. “Litauische Marchen: Der Based on the System of Antti Aarne and Stith einaugige Riese”. Webster. Duvall. Speciation in the Fossil Record. Carl Wilhelm. Uther. Anstey.org/. R. et al.. Huelsenbeck. Clark. 2006. Fr von. Ed.. 281–324. Pars I Frog and Janne Saarikivi. the aim of providing three Northwest Europe in three approximated different maps which involve different periods: ca. 2013. & Douglas H. “Phylogenetic Patterns as Tests of Speciation Anthropological Series 2. “SAM: A Comprehensive Helsinki: Finnish Literature Society. AD 1000. 1995. Proto-Mordvin. AD 1000. Bioinformatics 19: 1572–1574. & J. Webster Wentworth. The present article is only the first part within the Uralic language family (Proto- 64 View publication stats . “Correlates of Reticulation Cliffs: Prentice Hall. Venditti. 1903. Stepanova. Clive. The Types of International Ecography 33: 46–50.