In an earlier question here I used an example of animal names versus words (often verbs) that have rather similar spelling and can be linked by observations of behaviour or the functional use to humans. The reactions to the example differed from nonsense to coincidence, while I see a clear pattern with a logical explanation.

So again some examples: A hound hunts A dog digs A sheep is shaven (for its wool) compare German Schaff) A cat catches (rodents) A cow chews (grass a lot with 4 stomachs...) A fly... flies A rabbit is rapid A squirrel secures (this relates to the hiding of food in the ground for the winter.) A sparrow (passerine bird) which appears and passes on forraging for food; French ‘il se apparu’ meaning ‘he appeared’ A wasp (French il va se poignez (Va-se-p...) meaning ‘he is going to be poignant (sharp/fierce)) An otter in the water A beetle: voir cette bête aller An owl flies at night: vole (la nuit) A worm morrows (anagrammic antonym) A bug (bukken means ‘to bend’ in Dutch. The name seems to indicate bending to the earth to see what kind of bug it is.

When does nonsense or coincidence become a pattern in linguistics?

  • How about snake:sneak? You can also try it with plants. Some work, some may share deep roots (cow:chew), but most ignore etymological research. – amI Oct 13 '18 at 2:08
  • 2
    I'm voting to reopen this question: "when does coincidence become a pattern?" is a perfectly relevant linguistic question, and one that e.g. Proto-World enthusiasts would do well to ask. – Draconis Oct 13 '18 at 15:32
  • 1
    Me too......... – fdb Oct 13 '18 at 18:07
  • 1
    Trask's Historical Linguistics has a whole section on this problem, with multiple examples of different types. He also deals with the extreme views like Proto-World, also with multiple examples. – jlawler Oct 13 '18 at 21:15
  • 2
    My comment was supposed to be embodied in the reason for closure but sometimes it doesn't work out that way on the final tally. Unclear what you are asking. Can we delete everything but the last sentence, and would that then be your question? Is this distinct from the question "what is a significant generalization"? If this only about animal names, is this only about historical linguistics? How is this different from the two completely distinct questions "When does nonsense become a pattern" and "When does coincidence become a pattern"? – user6726 Oct 14 '18 at 21:17

As a general rule, coincidences and nonsense become a pattern when you can show a consistent rule that can explain a significant amount of data.

Historical linguistics offers many such patterns. One of the first noticed (which led to historical linguistics becoming a thing in the first place) was that /p/ in Latin and Greek corresponds to /f/ in Germanic: pisc- ~ fish, pod- ~ foot, pater ~ father, pyr ~ fire, and many more. This is now known as (one component of) Grimm's Law.

On the other hand, there's no regularity to the correspondences you've proposed. For otter~water, you're relying on rhyming in certain dialects of English—in RP, for example, they have different vowels. Sheep and shaven, on the other hand, have the same first phoneme and a labial somewhere after that, owl and vole have the same final sound in two different languages and vaguely-related meanings, and wasp corresponds rather tenuously to an entire phrase in a different language.

So the general standard is: if you can show a consistent pattern that can make predictions, rather than having to come up with separate explanations for each instance, then it'll be considered a pattern rather than a series of coincidences.

  • I don’t think it is one rule that can explain it, unless the rules are categorised as: To not be coincidence “The similarity in spelling must lead to an explanation that links the animal name to its features, use to humans, sound it makes besides geographical features like ‘mountain’ lion. There should be other options kept open. And in different languages the same rules should be applied if it is not a compounded foreign phrase as wasp which has the rule work in French, not English. – Ajagar Oct 13 '18 at 7:20
  • Now you can make predictions. I am sure this works for most if not all domestic animals in the Germannic countries. – Ajagar Oct 13 '18 at 7:23
  • Maybe simplify. Every name is a description existing of compounded roots or words. – Ajagar Oct 13 '18 at 7:35
  • 6
    Just for the record: "otter" and "water" really are cognates. – fdb Oct 13 '18 at 17:59
  • 4
    ...As are "fly" n. and "fly" v. – fdb Oct 13 '18 at 18:06

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