If you read their findings, Da Silva and Tehrani only claim an indo-european origin for one fairy tale ATU 330, with a 0.54 probability. A few dowzens showing positive correlation for proto-germanic, proto-balto-slavic, etc. It is nonetheless erroneous to use such findings to claim all fairy tales are older than x or y.
Also, I'd argue that their methodology is severely flawed, it doesn't account very well for diffusion : It looks at where a fairy tale has been found, regardless of whether it seems that it was only recently been imported.
See the study : rsos.royalsocietypublishing.org - Comparative phylogenetic analyses uncover the ancient roots of Indo-European folktales
A really strong case for ATU 330 having only recently traveled can be made. First, in western europe it's heavily christianized, and often in the same way, with the same characters (Jesus, Saint Peter, the episode of the gate of Heaven, etc.) which leans in favor of a diffusion after it was christianize. Second, each tale resembles the other. When Dumezil identified Indo-European motives, he noticed huge differences and only a careful intepretation could link the three steps of Vishnu and the fight of Vidar against the Wolf Fenrir. In one case, the myth was taken over by Vishnu a really popular god, in the other, next to nothing is known of Vidar ; both are signs of decays. We would expect, if ATU 330 is so old, to find in in varying contexts : here, as an episode of a larger national epic, here as completely deformed, here with roles having completely changed over the course of the century. But the extreme similarity of the tale accross Europe, the fact that a lot of characters share the same names (e.g. Bonhomme Misère), the fact that it was published in print and heavily advertised as such, and obviously crossed linguistic lines (de la Rivière himself claimed that he was adapting an italian tale in 1718)
all make a stronger case for a recent diffusion of the tale
that Da Silva and Tehrani's model takes as proof of ancienty. Furthermore the tale is present, according to the ATU in a lot of languages that are not indo-european, forgetting Hungarian and Finnish because of their indo-european surroundings, it's also marked as present in China, Corea, Palestine and Georgia, which their model simply doesn't take into account, although it should be clear proof of large-scale diffusion of the tale.
Also, their model is sometimes simplistic in the extreme, for example, they associate each "country" with one unique language and calculate the linguistic distance accordingly, and the geographic distance from one point roughly in the middle of the geographic extension of the language, and this despite two problems.
1) There's a lot of countries where a multitude of languages are spoken.
2) The Aarne-Thomson-Uther index does not classify tales by language, often because we're not always sure of the language in which they were collected, but by country.
This leads to absurdities as their graph for ATU 330 show the tale present in Hindi but absent in Nepalese.
I don't have the Aarne-Thomson-Uther index available, but if you look at eh Aarne-Thomson index, its predecessor, it claims the story is present in India, and references : Indic Oral Tales
(FFC 73.2 n°180 Helsinki 1960).
…Which references in turn…
Henry Parker, Village Folk-Tales of Ceylon, London, 1910-14 (3 vol.) vol. III, pp. 339-342.
Singalese is not too distant from Hindi, I'll admit, but still, it's Sri Lanka, and a little off center from their point of reference for India. (and this story has more elements from Grandfather Death or the Spirit in the Botte)
More interestingly in 1989, Heda Jason published an updated version of the Indic Oral Tales
, in which he added a reference to AT 330 being :
Sharma Nagendra, Folk Tales of Nepal
, n°1 "Why We Can't See Death", pp. 11-15, Sterling Publishers, Delhi, 1976.
In Nepal ! It was classified as *indic" and Nepalese tales do share a lot with indian ones, so it makes sense to classify them as such, but by not going through with their sources, Da Silva and Tehrani assume that the tale is nonexistent in Nepalese but existent in Hindi (and Singalese is not on their graph) !
But I went through with the research and found out that they were accidentaly right : the tale does seem to exist in India, and more precisely in Gujarat.
Cf. Beck et al Folktales of India 1987:283-5 n°94 “The Carpenter’s Tale”. Note that it was made in 1987 and was not mentioned by Jason in 1989. Also you'll notice 1976 is an extremely late date if you want to prove that the tale you've collected go back to prehistory.
But you'll notice that neither Beck et al, nor Nagendra Sharma nor Henry Parker makes mention of a blacksmith, although Da Silva and Tehrani claim that their research proves something about the metallurgy of the proto-indo-europeans.
And from a glance at Da Silva and Tehrani's graphs, methds and bibliography they have no idea that they are accidentaly right (see: the omission of Nepalese) and many more objections could be made to other points.
Although I'd like to see more rigorous studies like this one in the future, you know, where they actually look at their data before putting it in an excel table.