Population numbers for languages

[Qwynegold]

I'm working with this one project and I'm trying to find out how many people speak certain languages. This is of course really hard because of various problems with counting speakers, but it would be nice to have at least some kind of figures. There are some languages for which I have found absolutely no statistics, so I'm wondering if anyone here might know something. Preferrably I want figures for both L1 and L2 speakers separated, but if there's no better data, I'll have to settle for combined numbers, or anything you might have.

• Hokkien (Quanzhang), all dialects, but excluding other varieties of Minnan. WP says 37 million, but it doesn't cite the source. I suspect they've pulled that number from Ethnologue, where it says that all varieties of Minnan together are spoken by 37 million people in mainland China.
• (Standard) Cantonese, excluding other varieties of Yue.
• Latin. The only thing I can find when googling is that apparently some pope or someone has said that he estimates that there are only 100 people who are as fluent in Latin as he is. But that seems like a very strict definition of fluency. I mean, just counting all the Latin teachers out there, there must be more than 100. Surely one must be able to say that at least they "know" Latin well.
• Ancient Greek. I haven't found the slightest guesstimate. Now, being dead languages, Latin and Ancient Greek probably don't have any L1 speakers, but you never know. There are supposed to be some people who are teaching their children Cornish as their first languages, and then there was that one case of a kid who was taught Klingon. But L2 numbers are of course more important for Latin and AG.

[bbbosborne]

I think I got what you want:
I found that the strictly cantonese portion of the yue dialect is mostly confined to guangzhou, hong kong, and macau (prestige dialect). looking at the demographic sections for those cities on wikipedia, half of guangzhou, 94.6% of hong kong, and 85.7% of macau speak cantonese. the guangzhou statistic is sourced to a chinese newspaper while the rest are census reports.
doing the math, the total number of cantonese speakers amounts to about 14,317,492 people. unfortunately lack of other info prevents me from getting separate L1 and L2 data.
note that this was very haphazard, so do take this with a grain of salt

[Vijay]

The Cantonese spoken in Hong Kong is not identical to the prestigious standard spoken in Guangzhou. However, the Yuehai dialects including Guangzhou dialect are apparently mutually intelligible. This says there are about 80 million speakers of all the Yuehai dialects combined, including Guangzhou dialect.

[Qwynegold]

I think I got what you want:
I found that the strictly cantonese portion of the yue dialect is mostly confined to guangzhou, hong kong, and macau (prestige dialect). looking at the demographic sections for those cities on wikipedia, half of guangzhou, 94.6% of hong kong, and 85.7% of macau speak cantonese. the guangzhou statistic is sourced to a chinese newspaper while the rest are census reports.
doing the math, the total number of cantonese speakers amounts to about 14,317,492 people. unfortunately lack of other info prevents me from getting separate L1 and L2 data.
note that this was very haphazard, so do take this with a grain of salt

Oh thanks! This at least gives me a sense of how many speakers there are.

[Qwynegold]

The Cantonese spoken in Hong Kong is not identical to the prestigious standard spoken in Guangzhou. However, the Yuehai dialects including Guangzhou dialect are apparently mutually intelligible. This says there are about 80 million speakers of all the Yuehai dialects combined, including Guangzhou dialect.

Thanks! I had overlooked that paragraph. But when I click on to the page about Yuehai, it says only 13 million. I'll have to investigate further when I get to a computer. I'm using my smartphone right now, and it's really cumbersome.

[Vijay]

OK, on second thought, 80 million is probably way off. See below for brainstorming that led me to this conclusion

More: show

But when I click on to the page about Yuehai, it says only 13 million.

I know, but it comes with a question mark, and that estimate is from 1998, whereas the other one is from 2009. The reference note reads:

"Guangfu" (apparently = all of Yuehai) in Olson, An Ethnohistorical Dictionary of China

Guangfu is (also?) another name for Guangzhou, so maybe the reference was really referring only to Guangzhou dialect. Idk.

Part of the problem with both of these estimates seems to be that we don't have any information as to where the sources themselves are getting their information.

It might be worth noting that Guangzhou's population in 1995 was 4,745,000, whereas in 2000, it was 7,330,000. Hong Kong's population in 1995 was 6.156 million; in 1998, it was 6.544 million, and in 2000, it was 6.665 million. So perhaps the total population in 1998 within China may have been about 13 million.

However, Guangzhou has been continuing to grow so much it is now almost at maximum capacity. In 2005, its population was 8,485,000, and in 2010, it was 9,620,000. In Hong Kong, the population in 2005 was 6.813 million; in 2009, it was 6.973 million, and in 2010, it was 7.024 million.

But even that is at most 16,644,000 people. Hmm, so perhaps the 80 million estimate isn't accurate, because it would require the vast majority of Cantonese speakers to be from outside the Cantonese-speaking parts of China, but how is that possible? Wikipedia's 2007 estimate for Yue Chinese as a whole at 60 million is 20 million less than this estimate for Cantonese!

[Qwynegold]

All right, thank you so much. Now I know at least that it probably has nowhere as many speakers as I first thought.

[Vijay]

I suspect they've pulled that number from Ethnologue, where it says that all varieties of Minnan together are spoken by 37 million people in mainland China.

Wait a minute, doesn't Ethnologue say that 25.7 million people speak Minnan in China (I presume this means mainland China)? Or is that outdated?

The Italian Wikipedia says Hokkien has 47.3 million speakers as of circa 2009, also citing Ethnologue, specifically this page. But I think that page now says 49.8 million. The Indonesian Wikipedia also says 49 million but followed by the words (date missing) in English. The Malay Wikipedia says "50 million (undated)."

Latin. The only thing I can find when googling is that apparently some pope or someone has said that he estimates that there are only 100 people who are as fluent in Latin as he is. But that seems like a very strict definition of fluency. I mean, just counting all the Latin teachers out there, there must be more than 100. Surely one must be able to say that at least they "know" Latin well.

I don't know to what extent this counts, but I know Latin well enough myself to use it regularly on another language forum. Using a dictionary, of course, and I usually have to double-check and make sure I didn't screw up the grammar; it usually works but not always.

More relevantly, though, I think you might find this Reddit discussion kind of interesting. Not exactly authoritative, but still something. Two quotes that caught my eye:

That all said, the community of people trying to actually speak and use a "living" Latin is growing, which is rather good. As to estimates for how many people in the world can speak truly fluent Latin, I'd say 100 is possibly a generous estimate, though there are surely many, many more who can speak Latin at a level below fluency.

While a lot of people have some Latin from their school days, most will not have learnt it as a living language and only a small minority will have learnt enough to be able to hold a spoken conversation let alone to fluency. That being said, there are a small number of institutions and programmes which do treat Latin as a living language or even as an everyday language and some of these are long enough or in-depth enough to produce fluent speakers. Given the number of these programmes operating, the opportunities to use Latin outside these programmes, the rate of decline in language skills over periods of disuse, and the length of the average lifespan, I would guess that there are at least a couple of thousand people fluent in Latin and possible as many as a few tens of thousands. I would also expect a majority to be in the Catholic church.

Ancient Greek. I haven't found the slightest guesstimate.

One tricky thing about estimating the number of people who know Ancient Greek in particular is this (towards the end of the (English ) Wikipedia article on Ancient Greek; this of course is a problem rather different from the problem of estimating how many people know Latin):

Ancient Greek is also used by organizations and individuals, mainly Greek, who wish to denote their respect, admiration or preference for the use of this language. This use is sometimes considered graphical, nationalistic or funny. In any case, the fact that modern Greeks can still wholly or partly understand texts written in non-archaic forms of ancient Greek shows the affinity of modern Greek language to its ancestral predecessor.

Now, being dead languages, Latin and Ancient Greek probably don't have any L1 speakers, but you never know. There are supposed to be some people who are teaching their children Cornish as their first languages, and then there was that one case of a kid who was taught Klingon. But L2 numbers are of course more important for Latin and AG.

There is indeed at least one claim I've found of children being raised speaking Latin. I'm not so sure about Ancient Greek; what I keep finding with regards to that is that Ancient Greek is intelligible to some degree to Modern Greek-speakers and that Pontic Greek, spoken by "as few as 5,000 people," is considered the one living variety that is closest to Ancient Greek. (Wikipedia has its estimate for the number of native speakers of Pontic Greek at "778,000 (2009-2015)").

[Qwynegold]

I suspect they've pulled that number from Ethnologue, where it says that all varieties of Minnan together are spoken by 37 million people in mainland China.

Wait a minute, doesn't Ethnologue say that 25.7 million people speak Minnan in China (I presume this means mainland China)? Or is that outdated?

Huh, that's odd. Now I don't see anything about 37 million. It does however say "27,300,000 (2015)" for me. Maybe I misread that 27 as 37?

Latin. The only thing I can find when googling is that apparently some pope or someone has said that he estimates that there are only 100 people who are as fluent in Latin as he is. But that seems like a very strict definition of fluency. I mean, just counting all the Latin teachers out there, there must be more than 100. Surely one must be able to say that at least they "know" Latin well.

I don't know to what extent this counts, but I know Latin well enough myself to use it regularly on another language forum. Using a dictionary, of course, and I usually have to double-check and make sure I didn't screw up the grammar; it usually works but not always.

Yeah, I think that's pretty good. So I think the number 100 sounds way too low.

While a lot of people have some Latin from their school days, most will not have learnt it as a living language and only a small minority will have learnt enough to be able to hold a spoken conversation let alone to fluency. That being said, there are a small number of institutions and programmes which do treat Latin as a living language or even as an everyday language and some of these are long enough or in-depth enough to produce fluent speakers. Given the number of these programmes operating, the opportunities to use Latin outside these programmes, the rate of decline in language skills over periods of disuse, and the length of the average lifespan, I would guess that there are at least a couple of thousand people fluent in Latin and possible as many as a few tens of thousands. I would also expect a majority to be in the Catholic church.

I saw that too. But this is an estimation of like 2,000-50,000 or something. That's too much variance. And it's not the least bit authorative. Oh well, I guess I'll have to do without that data.

One tricky thing about estimating the number of people who know Ancient Greek in particular is this (towards the end of the (English ) Wikipedia article on Ancient Greek; this of course is a problem rather different from the problem of estimating how many people know Latin):

Ancient Greek is also used by organizations and individuals, mainly Greek, who wish to denote their respect, admiration or preference for the use of this language. This use is sometimes considered graphical, nationalistic or funny. In any case, the fact that modern Greeks can still wholly or partly understand texts written in non-archaic forms of ancient Greek shows the affinity of modern Greek language to its ancestral predecessor.

Hmm, do you think modern Greeks understand most of Ancient Greek's vocabulary?

There is indeed at least one claim I've found of children being raised speaking Latin. I'm not so sure about Ancient Greek; what I keep finding with regards to that is that Ancient Greek is intelligible to some degree to Modern Greek-speakers and that Pontic Greek, spoken by "as few as 5,000 people," is considered the one living variety that is closest to Ancient Greek. (Wikipedia has its estimate for the number of native speakers of Pontic Greek at "778,000 (2009-2015)").

That's interesting. Thanks!

[Vijay]

Huh, that's odd. Now I don't see anything about 37 million. It does however say "27,300,000 (2015)" for me. Maybe I misread that 27 as 37?

Interesting, it sounds like you have access to a more updated edition of Ethnologue than I do. The estimate I cited was from 2001.

I don't know to what extent this counts, but I know Latin well enough myself to use it regularly on another language forum. Using a dictionary, of course, and I usually have to double-check and make sure I didn't screw up the grammar; it usually works but not always.

Yeah, I think that's pretty good. So I think the number 100 sounds way too low.

Be that as it may, is it possible to find an estimate for how many people speak/know any language to this level, especially around the world? That would seem like a difficult thing to calculate.

Hmm, do you think modern Greeks understand most of Ancient Greek's vocabulary?

I'm not sure about "most," but Greek also has a unique thing called Katharevousa, which is kind of an attempt at a revival of Ancient Greek and still used to some degree for practical purposes. One example that TY (Modern) Greek gives is that a white house in Modern Greek would normally be άσπρο σπίτι áspro spíti, where apparently, both words are borrowed from (Byzantine?) Latin. However, the White House in Washington, D.C., for example, would be Λευκός Οίκος Lefkós Oíkos where both words survive from Ancient Greek.

That's interesting. Thanks!

Varsågod.

[Qwynegold]

Huh, that's odd. Now I don't see anything about 37 million. It does however say "27,300,000 (2015)" for me. Maybe I misread that 27 as 37?

Interesting, it sounds like you have access to a more updated edition of Ethnologue than I do. The estimate I cited was from 2001.

Whuhuh? How can that even be? I don't have a subscription or anything to Ethnologue.

Be that as it may, is it possible to find an estimate for how many people speak/know any language to this level, especially around the world? That would seem like a difficult thing to calculate.

Yeah, but somehow they are able to offer some numbers. One should of course take these with a grain of salt.

Hmm, do you think modern Greeks understand most of Ancient Greek's vocabulary?

I'm not sure about "most," but Greek also has a unique thing called Katharevousa, which is kind of an attempt at a revival of Ancient Greek and still used to some degree for practical purposes. One example that TY (Modern) Greek gives is that a white house in Modern Greek would normally be άσπρο σπίτι áspro spíti, where apparently, both words are borrowed from (Byzantine?) Latin. However, the White House in Washington, D.C., for example, would be Λευκός Οίκος Lefkós Oíkos where both words survive from Ancient Greek.

Aha, that's interesting. നന്ദി!

[Io]

Greek speakers don't understand 'most' of Ancient Greek. There are certain things which are denoted with archaic words, por ejemplo, 'bakery' is αρτοποιείο from άρτον instead of ψωμί, wine labels use λευκός οίνος instead of άσπρο κρασί, similarly ζύθος for beer, shoe shops use υποδήματα instead of παπούτσια to refer to 'shoes', also in speeches some politicians say Ελλάς instead of Ελλάδα.

As for katharevoussa there's a newspaper called Estia which is still published in it and the Greek church also uses it, the Ecumenical Patriarch of Constantinople's website is in katharevoussa, also, there are varying degrees of ancientness but I don't really know much about it, katharevoussa is quite an interesting thing. I once bought a history magazine and it was full of quotations from historical figures who wrote their dictionary and stuff like that in katharevoussa, it's very confusing and it's what prompted me to get somewhat acquainted with Ancient Greek too.

Also, the number of Pontic Greek speakers seems highly inflated, but that's just a personal feeling, I don't have sources to quote on its actual usage.

[Vijay]

Ευχαριστώ, Io!

Huh, that's odd. Now I don't see anything about 37 million. It does however say "27,300,000 (2015)" for me. Maybe I misread that 27 as 37?

Interesting, it sounds like you have access to a more updated edition of Ethnologue than I do. The estimate I cited was from 2001.

Whuhuh? How can that even be? I don't have a subscription or anything to Ethnologue.

Oh, I dunno then. I don't, either.

നന്ദി!

Heh, you're welcome!

(Off-topic, sorry, but tbh, this word always felt a little contrived to me, maybe even like a neologism. Nowadays, the first thing I always associate നന്ദി with is signs that say the equivalent of "thank you for visiting _______!" We don't really have an equivalent for 'thank you' in Malayalam; we usually just say it in English AFAICT. I don't think we'd usually say 'thank you' in Malayalam for something like this, and I generally don't think we thank people in Malayalam as often as Westerners do in their languages).

[Qwynegold]

Ευχαριστώ, Io!

+1

നന്ദി!

Heh, you're welcome!

(Off-topic, sorry, but tbh, this word always felt a little contrived to me, maybe even like a neologism. Nowadays, the first thing I always associate നന്ദി with is signs that say the equivalent of "thank you for visiting _______!" We don't really have an equivalent for 'thank you' in Malayalam; we usually just say it in English AFAICT. I don't think we'd usually say 'thank you' in Malayalam for something like this, and I generally don't think we thank people in Malayalam as often as Westerners do in their languages).

Ah, I just used Google Translate.

[richard1631978]

Greek speakers don't understand 'most' of Ancient Greek. There are certain things which are denoted with archaic words, por ejemplo, 'bakery' is αρτοποιείο from άρτον instead of ψωμί, wine labels use λευκός οίνος instead of άσπρο κρασί, similarly ζύθος for beer, shoe shops use υποδήματα instead of παπούτσια to refer to 'shoes', also in speeches some politicians say Ελλάς instead of Ελλάδα.

As for katharevoussa there's a newspaper called Estia which is still published in it and the Greek church also uses it, the Ecumenical Patriarch of Constantinople's website is in katharevoussa, also, there are varying degrees of ancientness but I don't really know much about it, katharevoussa is quite an interesting thing. I once bought a history magazine and it was full of quotations from historical figures who wrote their dictionary and stuff like that in katharevoussa, it's very confusing and it's what prompted me to get somewhat acquainted with Ancient Greek too.

Also, the number of Pontic Greek speakers seems highly inflated, but that's just a personal feeling, I don't have sources to quote on its actual usage.

I've in the early 20th century the Greek Orthodox Church decided to translate the bible into Modern Greek to make it easier to understand, only for riots to break out because people thought it devalued the meaning of the text.

[Hakaku]

I ran into a similar issue a few years back when looking at population numbers for (Central) Okinawan. The Ethnologue gave a figure of 980,000 speakers in 2000, which doesn't make sense given that the total population was roughly 1 million around that time and it's well-known that the islanders primarily only speak Japanese today.

What I did was look for documents (linguistic or other) or other media that made age-based claims or gave personal anecdotes. For example, "Nowadays the language is only spoken by those over the age of X". Or, another example: a video showing a 60-year old woman relearning the language because she never had the chance growing up due to language policies that forbade her. Something like that can give you a rough guess of what age bracket the L1 speakers fall in, where the cutoff is, and in what age bracket you can guesstimate the L2 speakers.

The end result was this: https://www.jlect.com/downloads/Number- ... eakers.pdf

The document got picked up by the Endangered Language Project: http://www.endangeredlanguages.com/lang/5354

The numbers aren't perfect (they may still be too high), but it's much more representative of the reality than what Ethnologue reports.

[Qwynegold]

That's really interesting. Unfortunately I haven't really been able to find any clues at all. Anyway, I think the Chinese languages all have less than 50 m speakers which was one cut-off point I had decided for this project. Ancient Greek I haven't been able to find any kind of estimates for though.

I had one guest lecture by a guy who has been studying the Ryukyuan languages. I think it sounds incredible that there are even any speakers left, especially of the non-Okinawan languages. :/

[Vijay]

Anyway, I think the Chinese languages all have less than 50 m speakers

Huh? Mandarin has more than ten times that many. (Also, isn't this pretty inconsistent with the vague criteria you're using for "knowing" a language?).

I think it sounds incredible that there are even any speakers left, especially of the non-Okinawan languages. :/

Why?

[cabinet_cat]

I ran into a similar issue a few years back when looking at population numbers for (Central) Okinawan. The Ethnologue gave a figure of 980,000 speakers in 2000, which doesn't make sense given that the total population was roughly 1 million around that time and it's well-known that the islanders primarily only speak Japanese today.

What I did was look for documents (linguistic or other) or other media that made age-based claims or gave personal anecdotes. For example, "Nowadays the language is only spoken by those over the age of X". Or, another example: a video showing a 60-year old woman relearning the language because she never had the chance growing up due to language policies that forbade her. Something like that can give you a rough guess of what age bracket the L1 speakers fall in, where the cutoff is, and in what age bracket you can guesstimate the L2 speakers.

The end result was this: https://www.jlect.com/downloads/Number- ... eakers.pdf

The document got picked up by the Endangered Language Project: http://www.endangeredlanguages.com/lang/5354

The numbers aren't perfect (they may still be too high), but it's much more representative of the reality than what Ethnologue reports.

Thanks for your work, I hope your contribution will bring some more attention to the preservation of the language.

I think it sounds incredible that there are even any speakers left, especially of the non-Okinawan languages. :/

Eh, I don't know how strictly enforced the language restrictions were, but I'd like to assume the Ryukyu islands were only somewhat loosely administered by the government at the time. So I am definitely not surprised they are still spoken.