@ToMach,
What you'd provided me with the webpage link(
How likely are chance resemblances between languages?
-- How likely are chance resemblances between languages?) that explained the formula of calculating the probability (in terms of matched chance between two far flung languages)is still very lack of convincing me to be awared of the coincidence is not surprising part of comparative linguistics or happend a lot than we expected because the author only considered fixed examples.-you know what -for example the first part CVC it only makes sense when both comparative languages has the same number and kind of consonant. The page has no flaw in terms of calculating probablity method methmatically but thats only when there are *dozens number* of resemblance words plus the the whole sample number of lexicons just lomited by hundreds or thosands of numbers.
Anyways ,Let's just make the chance that we can find 1,250 matches in 100,000 words.
Let me explain about the meaning of 1,250 and 100,000 respectively.
1,250 is the matches number from the list what I mentioned (Yes the 5,000 list by two authors) by the way,I made the number become lessened because you rejected 4 out of 6 list therefore I pick the number 1,250 instead of 5,000 (although I do not very much agree with your refutation because you only showed historicalyl likely hapend evidence -you did not make your refutation in linguistic standpoint,I can do that too anyways let me discuss about that part later)
To make fair calculation which allows more yield to your advantage Tomach,
100,000 is the number of my Japanese-Korean dictionary lexemes.(how many number of the whole Japanese words that to be known so far? )
By the way ,we should make an exemption of the Japanese words that consist of "Chinese characters" and completely foreign borrowed ones.
But I'd just skip this part here -remember that I am yielding to sweep as many variables as possible ;-)
Here is the formula for making <the number of chances that we can find 1,250 matches in 100,000 words.>-
!: factorial
100,000 !/(1,250!*(100,000-1,250)!)=A=
(Probablity of a single match in two phonemes word between Korean and Japanese,by the way in this calculation I can only use modern known consonant and vowels due to my short linguistic knowledge)---B
(1-B)-----C
AB(100,000-1,250)C(100,000-1,250)=ABC(100,000-1,250)^ =<the number of chances that we can find 1,250 matches in 100,000 words.>
To get B -we need to arrange the whole consonant and vowel in both of Korean and Japanese but the way please forgive my limitation that only available to show modern ones.
We should take consideration the number of overlapped part as well.
modern Korean consonant (ㄱ[g] ㄲ[gg] ㄴ[n] ㄷ[d] ㄸ[dd] ㄹ[l] ㅁ[m] ㅂ
ㅃ[bb] ㅅ ㅆ[ss] ㅇ[ŋ] ㅈ[j] ㅉ[jj]ㅋ[k]ㅍ )19 number
modern Korean vowel(ㅏ[a]ㅓ[ﾂ??ㅕ[yu]ㅖ[ye]ㅗ[o]ㅘ[wa]ㅙ[we]ㅚ[oi]ㅛ[yo]ㅜ[wu]ㅝ[wo]ㅞ[wﾆ津テ]ㅟ[wi]ㅠ[yu]ㅡ[eu]ㅢ[eui]ㅣ)-21 number
Since I am not very familiar with linguistics,I just wrote those pronounciation mark blatantly loughly.
I am not sure about modern Japanese but I would just say
consonant number is about 14 and the vowel number is about 7.
(Again I can be wrong- i just can read and write 窶堙絶?堙ｧ窶堋ｪ窶堙 and 窶堋ｩ窶堋ｽ窶堋ｩ窶堙 but not so sure about classify them grammartically,I am just showing mathmatical method only here)
So here we just need to calculate the probability of 2 phoneme words correspondence between two languages.
Its simple (1/19+14-13)*1/21+7-7)*(1/19)
then we need to square this number.Thats B.
Can you get these calculated result number ?
Use your window calculator for engineering use-if that does not tell you anything -I recommend you to use matrix calculating programe.
The 5,000 number of resemblance is again too many to say put into coincidence list even if we just want to put prove it probablity methmatical method.
<the number of chances that we can find 1,250 matches in 100,000 words.>
Then you need to divide it by the number of every possible chances that you can make 2 phonemes words with using both languages' consonant and vowels as I mentioned above-thats the probability of <the number of chances that we can find 1,250 matches in 100,000 words.> towards every possible chance of making two.
By the way,does that really happend in reality?
How many number of the whole Japanese words that to be known so far?-we need to get this number first to design the method to make more reliable result.