I just spent three days to count the population of the Kra-Dai language family (Almost completed, but there are still data from some small dialects of Southwestern Tai (SWT) that are to be collected for the input). The time and energy spent were not less than writing a short academic paper. I was a bit shocked after seeing the amount, even it hasn’t been fully counted. The population of the Tai branch is already more than 97.85 million, close to 100 million, accounting for about 96% of the total population of the whole Kra-Dai family, which is about 102 million.

​

The genesis was that I was writing a paper that needed to quote the Tai branch’s population, but I could see that the latest authoritative data was quite old, being more than 80 million in 1997, as in "the total number of Tai speakers is greater than 80 million" (Edmondson & Solnit 1997:1). I was struggling to find other data, but nothing I could trust, so I had to count it myself.

​

In this statistic, the horizontal rows first take the countries/regions, and then different languages as the unit; as for the vertical columns, they take the language branch, language group, year, academic reference, and reference websites as the unit. The so-called “language” basically refers to the division standard of ISO 639-3, that is, both mutually intelligibility and/or internal identification can form a single language group, such as Yang Zhuang [zyg] and Min Zhuang [zgm] which were separated from the "Dejing Vernacular" area of Southern Zhuang in the Chinese literature. This statistic also includes some small languages that have not yet been included in ISO 639-3 but have been studied in the publications with population data. For example, the so-called Nuoxi Yao (𦰡溪瑶话) in Hunan Province of China is actually a dialect of Kam-Sui language that is very close to the Kam language. Some languages are with codes in ISO 639-3, such as Tày Tac from Vietnam, but with no population data. They are temporarily included in the table for adding the data of population later. All data are believed to be native speaking population, excluding the population of second language acquisition. Therefore, the real data should be only more, not less, for the following reasons.

​

First, I do not count the population of Kra-Dai outside of China and the Indochina Peninsula. For example, there are a considerable populations of Tai in Western countries such as the United States and France, and they are Thai, Lao, White Thai, Black Thai, Phu Thai and some other Tai languages. Although many of them are still speaking their native languages, but we don't know the exact percentage of native Tai speakers in their total population, so they cannot be counted.

​

Second, from the languages I am familiar with, the current statistics tend to be lower than the real situation. Take Yang Zhuang which I am most familiar with for an example. Recent publications and authoritative website materials count its population from 0.77 million to 0.87 million in Jingxi, Debao and Napo Counties of Guangxi, China. But I clearly know that even the most scientific survey of Yang Zhuang to date, Jackson et al. (2012), lacked surveys on the following groups and populations belonging to Yang Zhuang: the main residents of Qiaoye 橋業, Babie 巴別 and Dongjing洞靖 Townships in Tianyang County of Guangxi are Yang Zhuang native speakers (although Tianyang is a Northern Zhuang dominant county); many Tianbao (Tien Pao) people in Funing County of Yunnan Province are of Yang Zhuang; There are many Yang Zhuang villages in Bantao Township of Tianlin County, Guangxi; there are also a large number of Yang Zhuang dialects in the west parts of Tiandeng and Daxin Counties of Guangxi, and their languages are obviously different from those (of Zuojiang Zhuang) of the main residents of this two counties. Not to mention Nung Giang in Northern Vietnam, which is linguistically identical to Yang Zhuang in Jingxi County of Guangxi, China, because in ISO 639-3 the Nung including Nung Giang is a "single" language. Since there is no exact statistics for these missing populations, I spent a long time to analyze and count the Yang Zhuang speakers of these towns and villages, and finally concluded that the population of Yang Zhuang is at least 1.1 million. This raises another problem, that is, these populations which are added to Yang Zhuang should be deducted from other Zhuang dialects’ populations which had been mistakenly counted including them.

​

Furthermore, I could only quote the data which have been included in the statistics, but in fact, data from many small languages have not been included because they have not yet been counted. For example, in the results of SIL’s survey of the Dejing Zhuang areas in Jingxi, Debao, and Napo Counties of Guangxi (Jackson et al. 2012), they conducted two ISO 639-3 codes for Yang Zhuang and Minz Zhuang, and have provided the detailed statistics including speakers’ population. But for the other small branches such as Nong'an, Zuozhou, Sheng, Rui, Jue, Ao, Dong, Yong, and etc., there is no last released population data even they also included them in the research. Therefore, these small branches are not counted in the list of the sixteen "distinct Zhuang languages" of ISO 639-3, causing the loss of their population data. In other words, although the official Zhuang ethnic group has a population of nearly 20 million, my combined Zhuang languages’ population is only more than 15 million. It may be close to the fact that many young generations in the urban areas have already given up their native languages, but it certainly includes these missing data of the small languages or varieties.

​

Moreover, the Tai population in Vietnam has the same problem of data loss. Because Tay and Nung are listed in ISO 639-3 as two single languages (in fact, they are surely not single languages, but for now I have to be based on the current standard of ISO 639-3, so they are only two languages in the table), their population was quite easily counted. However, the language of Thái, the second largest officially recognized minority (following Tay) in Vietnam, has no a single code in ISO 639-3, so it needs to be split into many distinct languages (namely Tai Dón [twh], Tai Dam [blt], Tai Daeng [tyr], Tai Pao [tpo], Tai Thanh [tmm], Phu Thai [pht], Tai Do [tyj], Pa Di [pdi] and etc.). Therefore, I could not adopt the Thái’s population directly from the official statistics of Vietnam. After I added up all these separated Thái groups, the total population of them is less than 1.74 million, which is 80,000 fewer than the whole population of the official Thái ethnic group (1,820,950 in 2019) in Vietnam. However, this can also be explained by the fact that some Thái people have given up their native languages.

​

Finally, several major Tai languages in Thailand and Laos have data that are too outdated or inconsistent and inaccurate. For example, according to the old data on Ethnologue and MultiTree, the mother tongue population of the four major regional dialects of Thai (within them Central Thai is the national language) is as follows: Central Thai 20.2 million (2000 Ethnologue), Northeastern Thai (the Lao varieties distributing in Thailand) 15 million (1983 SIL), Northern Thai 6 million (1983 SIL) and Southern Thai 4.5 million (2006 Mahidol University data). As a result, these four Thai dialects add up to a population of 45.7 million. However, the population of Thailand is now about 69.8 million (2020), that is to say, after deducting these four Thai major dialects, the remaining 24.1 million people are not native speakers of Thai dialects. This is far from the actual situation, for this result has a contradiction between the old language data and the nowadays Thailand total population. Therefore, I have repeatedly verified and summed these languages. Some details are as follows.

​

1. Central Thai (Siamese) has different amounts of data in different sources, ranging from 20 million to 36 million. Therefore, I had to first determine a comparison of credible population of Central Thai speaker and the whole population of Thailand in the same period. For example, in 2000, the total population of Thailand was 56,281,538, while the population of Central Thai was 20,200,000, that is, 0.358909879115244% of the whole population was native Central Thai. Therefore, because the recent demographics of Thailand’s total population is 69,799,978 (2020), assuming that the population growth rate is comparable to the language speakers, then the number of people whose native language is Central Thai should be at least 69,799,978 times 0.358909879115244%, that is, 25,051,902 people. In fact, along with the modernization, more and more young Thai citizens from minority ethnic groups switch to Central Thai language as their mother tongue, so in fact the population of the native Central Thai speakers should be much more than this number. Therefore, 25.05 million should be its minimum insurance data.

​

2. The population of Northern Thai speakers in most materials is generally based on the data of 6 million in 1983. For example, Ethnologue (18th ed., 2015) still adopts this old data. I don’t understand why it has not been updated after so many years. So, I adopted the Joshuaproject’s data, which is 7,361,000. Taking into account population growth and proportion, this data should be more credible.

​

3. The southern Thai language population in Ethnologue in 2006 was 4.5 million, but it is 14 years from now. In 2018, there were 9.454 million people in southern Thailand, and about 52% of the population in southern Thailand are the native Southern Thai speakers. Therefore, the data of the population of Southern Thai speakers should be about 5 million (in fact, some websites also list it as about 5 million).

 

4. The native language population data of the Lao language should be separated into Thailand and Laos, because the so-called Lao language includes the main ethnic language of Laos and Northeastern Thai (or Isan) of Thailand. According to the newer data, the total population of the Lao Language should be about 30 million ("Lao (Laotien)". Inalco. 20 January 2017), making Lao be the most populous language in the entire Kra-Dai language family. However, the population of the Lao native speakers given by Ethnologue is only 3 million in Laos, and it should be an outdated data. I have noticed that the total population of Laos is growing very fast. There were only 6.25 million people a decade ago, but now it has reached 7,275,560 (the latest data in 2020) (https://www.worldometers.info/world.../laos-population/ ). According to previous statistics, about 52% of the population in Laos is native Laotian speakers, so the population of Lao mother tongue speakers should be at least 3,783,291 (there are still no statistics of ethnic minority populations whose native languages have been replaced by Lao as their mother tongues). Then, the Lao (Isan) speaking population in Thailand should be 30 million minus 3.78 million in Laos, 23,000 in Cambodia is, and 17,532 in Vietnam, to equal 26,179,468. According to recent data published, the population of the entire Northeastern Thailand is about 32.88 million, so it is reasonable that there are more than 26 million Lao (Isan) speakers in this region. Therefore, I have inputted 3783,291 people for Lao population in Laos, and the 26,179,468 for Lao/Isan speaking population as the minimum insurance data for Thailand.

As a result, the total population of the four major Thai dialects in Thailand is 63,593,098, accounting for about 91% of Thailand’s total population 69,799,978 (2020). This is more realistic. In fact, this data is still low, because now the non-Thai languages and dialects in Thailand, except for the Malay languages in the southernmost and the Khmer languages in the southern edge of Northeastern Thailand, are mostly shrinking.

​

In short, the population of the Kra-Dai language family has exceeded 100 million, and it is obviously one of the large language families in the world; among them, the Tai branch (or the Tai-Yay Branch which is preferable for me) is the most developed language group. This statistic will be continued. On the one hand, there are still some small languages’ population data to be collected. On the other hand, a language’s population sometimes expands (for example, more and more ethnic groups in Thailand and Laos will collectively switch to Thai or Lao as the mother tongue), and sometimes declines (for example, the Kra-Dai peoples in China, Vietnam, Myanmar and other countries may discard their ethnic languages because more and more new generations are giving up their native languages and switching to their national languages), and there are increase or decrease of population everywhere, etc. Anyway, my data of this statistics will be changed all the time, and it should be updated regularly or irregularly for the future research.

​

Reference

Edmondson, Jerold A. & David B. Solnit. 1997. Introduction. In Edmondson, Jerold A. & David B. Solnit (ed.), Comparative Kadai: the Tai branch, 1-17. Summer Institute of Linguistics and the University of Texas at Arlington Publications in Linguistics, 124. Dallas: Summer Institute of Linguistics and the University of Texas at Arlington.