I am wondering if there exists a way to get a numerical estimate of the probability of whether or not a given society will evolve a story with certain characteristics within a certain amount of time. For example: what is the probability that the Mayans, given 1000 more years, would have evolved some kind of story like Atlantis (i.e. involving a sinking city)? I'd also be interested in a cumulative probability that predicts how often you'd expect a type of story to evolve in any culture in the world ever (ex. how many Atlantis stories would you expect to be independently evolved worldwide within 1000 years?). I don't want to calculate the probabilities using just historical comparison, because I don't think enough mythological evolution has taken place in history to achieve statistical regularity for every type of story (i.e. it's too small of a sample size to use for reliable predictions).

Does the kind of measure I am looking for exist?

EDIT: my motivation for this question, (in case it somehow helps) is to find out the likelihood that a myth would have evolved with the characteristics "contains unusual event", "has important practical implications", and "claims to have been witnessed by a approximately every member of the society (at the time) which believed in the myth" (an approximation of that probability of import to me because I am looking into a claim that said that the probability for this is so low that it shouldn't have happened even once in any society in history)

  • This is more about the evolution of storytelling. This isn't exactly what you're looking for, but might lead you in a useful direction. I don't know. ncbi.nlm.nih.gov/pmc/articles/PMC5763351
    – DWKraus
    Commented Aug 28, 2020 at 22:19

1 Answer 1


In myths, humans follow a limited number of patterns, regardless of their cultural background. This field of study is called Comparative Mythology. This has uncovered a number of parallels, or archetypes, between the myths of different cultures, including some very widespread recurring themes and plot elements, like:

  • Creation of mankind from clay
  • Acquisition of fire for the benefit of humanity
  • Flood myths
  • Dying/Mortal gods & resurrection
  • Creative sacrifice
  • Axis mundi
  • Titanomachy
  • Giants
  • Dragons and serpents

Comparative mythologists come from various fields, including folklore, anthropology, history, linguistics, and religious studies, and they have used a variety of methods to compare myths:

  • Linguistic
  • Structural
  • Psychological
  • Phylogenetical

The probability of a culture developing a myth on these archetypes is high, since these archetypes have been found all over the world.

There are no mathematical models to predict probability of this AFAIK.

EDIT: Answer to added motivations & comment question:

Phylogenetics will give you models for the origin and evolution of languages, that is, the vocabulary and grammar. The grammar of a language will have some influence on the myths generated by that language, but not to the extent you ask for in your comment. A simple example would be:

If a language's grammar mandates a subject for a verb in a sentence (e.g. "it rains"), then what we see in mythologies stemming from such a language is a higher percentage of creator & intervening deities than in languages that do not have this rule.

As to your motivation criteria:

  • contains unusual events: myths by definition are about unusual events, otherwise they are just stories.

  • has important practical implications: on what do you base this? Myths, other than a source of religious control by a clergy, have very little practical implications

  • claims to have been witnessed by a approximately every member of the society (at the time) which believed in the myth: we know from anthropology and literary criticism that eye-witness and hearsay accounts are the most unreliable sources of historical information available to us. And regardless of that fact, there are - to my knowledge - no myths that entire societies claimed to have witnessed.

  • Maybe this is a stupid question, but wouldn't you be able to extrapolate probabilities based off of a phylogenetic model (because of the fact that evolutionary models make predictions based off of statistics)? Oh also I probably should have done this originally, but I added my motivations to the original question.
    – Certusic
    Commented Sep 1, 2020 at 18:37
  • Added response to your motivations and question to my answer
    – Codosaur
    Commented Sep 2, 2020 at 12:50
  • Thank you, I was unaware that phylogenetic models work through language features, I only heard of it and saw words like 'bayesian' along with it, so I made a mistaken assumption. Also, good point that the 'unusual event' criterion is redundant. As for the other two, I was thinking about putting in parenthesis the fact that the second one limits you to founding stories of religions, despite that not being an important factor to me, and the third one basically limits you to the Ten commandments. (If you are wondering why I care, look up the kuzari principle, which is what I was researching)
    – Certusic
    Commented Sep 4, 2020 at 17:14
  • There were no direct witnesses to the supposed origin of the 10 commandments besides Mozes,,so that myth also does not qualify to your third motivation.
    – Codosaur
    Commented Sep 5, 2020 at 8:58

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.