If I were to take 2 random people aged 18 or over from anywhere on the face of the planet and put them together, what’s the percentage chance that they both have at least one common language? Not necessarily fluent or high level, just good enough so they can each instruct the other to do some simple tasks (i. e put the red circular box on the green square box and place the the yellow statue of a dog on top) that would be almost impossible to relay accurately just by gestures.
Do you want a spoken language, a written language, or either? All educated Chinese can communicate in writing, but not necessarily in speech.
This is hard to estimate since it is very hard to estimate the total number of people who can roughly communicate in a language - those numbers don’t seem to exist. I’ll give you a rough approximation though, using the data from wikipedia on total number of speakers of various languages. With a world population of 7 billion, all I’m doing is looking at is the probability that two people, randomly selected out of 7 billion, would fall into the same category in that table. Pretty rough and lots missing in that estimate, but here goes:
- Mandarin Chinese: (1.151/7)^2 = 2.70%
- English (1/7)^2 = 2.04%
- Spanish (0.5/7)^2 = 0.51%
- Hindi (0.49/7)^2 = 0.49%
- Russian (0.277/7)^2 = 0.16%
- Arabic (0.255/7)^2 = 0.13%
- Portuguese (0.24/7)^2 = 0.12%
- Bengali (0.215/7)^2 = 0.09%
- French (0.2/7)^2 = 0.08%
- Malay/Indonesian (0.175/7)^2 = 0.06%
- German (0.166/7)^2 = 0.05%
- Japanese (0.132/7)^2 = 0.04%
- Farsi (Persian) (0.11/7)^2 = 0.02%
- Urdu (0.104/7)^2 = 0.02%
- Punjabi (0.103/7)^2 = 0.02%
Total: 6.53% for the top fifteen languages. The rest of the languages will provide a pretty insignficant portion of a final total - maybe another 3% or so, to pull a number out of thin air. Call it 10%.
I think these original numbers assume a higher level of fluency than you were originally thinking, so it might be closer to 20% or something. The point of this estimate is more that the number is probably between like 5% to 25% - not as small as you might think.
I think it may be even higher… just to take English as an example, if 1 Billion people speak it anywhere near passably, the number of people who know an odd word here and a weird one there – enough for very basic communication – could be as high as double that, perhaps (pulling numbers out of a place the sun don’t shine, of course…) – in which case, English alone will give you (2/7)*(2/7) = ~8%
Point being, that doubling the number of people who speak a language will not double the chance that 2 random people will speak it – it will **quadruple **it.
Now I suspect that the number of people who speak English really poorly is probably the highest analogous number for any language (i.e., there probably aren’t another Billion people who only know a basic 100 words of Mandarin, or 250 Million who know a minimum of Arabic; for most other languages it’s probably closer to black and white, you speak it or you don’t) – but changing the assumptions event slightly changes the results more significantly than linearly with the assumptions.
I highly doubt that list, since it doesn’t even have any mention of Cantonese, nor any other Chinese language other than Mandarin.
There aren’t that many Canotnese speakers as compared to Mandarin. Wikipedia gives 70 million, which would be less then the number given for Punjabi in spenczar’s list.
Another point to bear in mind is that some of the languages on that list (and off it) are mutually intelligible to some degree. Spanish and Portuguese, I believe, for instance. I would imagine some combination of Urdu/Punjabi/Hindi/Bengali too. I have managed to communicate with Italians in (my really appallingly bad) French. So that would bump the numbers up too.