On "Chinese languages"

If you've been involved with the intersection of "Chinese" and "language study" (not even "linguistics") for more than five minutes, you must have seen the debate of "Chinese dialects" vs. "Chinese languages" (or "Sinitic languages" as snobs like me would call it [insert Twilight Volturi family meme here]). I don't think I'm particularly educated in Chinese linguistics (my whole academic linguistics upbringing is in IE languages) to confidently say I know 100% of what's going on, but I think it's fair to say I know more than most people who hold opinions as strong as mine.

By the way, my opinion is this: Sinitic languages are languages, not dialects. But that's not what I'm here to convince you of, because I don't think there's anything to be proven; I'm just here to rant about the "debate" itself and certain wild statements I constantly see.

The question of "language" or "dialect" is scientifically meaningless: the status of a "language" versus a "dialect" does not tip the scale of people's attention—cultural, political, or academic—at all. AAVE, universally considered a dialect, does not stop countless scholars from publishing research in areas of syntax, phonology, semantics, sociolinguistics, and more. Speaking to a large extent the same language¹ does not stop North Koreans and South Koreans from pointing their guns at each other. So I won't feel any insult for my education to shrug and say, "sure it's a dialect, whatever".

Instead, this question is always a question of politics and politics alone. To quote Max Weinreich (like every article on this matter is obliged to):

A language is a dialect with an army and a navy.

There are ample sources online interpreting this quote, but I'd like to tweak it slightly: a language is the army and the navy for a dialect. They are the discrete borders drawn on a map of continuum to separate "them" from "us". When new borders are established, new languages appear (see: Serbian and Croatian); within the same borders, languages are reduced to dialects. This is just one of the plots in the grand scheme of establishing "national identity", so I don't consider the debate to be any more factual and concrete than "is [insert your controversial region here] a part of [insert the country here]", "do [insert your controversial demographic here] belong in [insert the country here]", or "does [insert public figure here] represent the [insert the country here] people"². I have neither the guts nor the energy to start a fumed debate on politics; you are entitled to your opinions, and I am entitled to mine.

What I'm not okay with seeing, is these questions masquerading as "scientific" and secretly (often knowingly) weaponized for a political agenda. Let me at least point out some ways in which the "dialect" theory doesn't stand up to scrutiny (again—I have no intention to convince you of the converse).

By far the most common argument I see is:

The Western framework doesn't apply to us because our writing system works differently!! All Chinese speakers are mutually intelligible through writing (笔谈)!!

1. Writing systems are second-class citizens in linguistics. Many languages roam around just fine with no writing systems, others with multiple writing systems. They are also arbitrary: Vietnamese, after dropping Chinese characters and moving to the Latin script, did not suddenly become a different language.

2. 笔谈 is dead ever since 文言 (literary Chinese) was. There was a time when people did say the same string of words just with different pronunciations, but the 笔谈 we know nowadays are just non-Mandarin speakers being conditioned to mentally translate a transcribed language into their own language.

3. It's also false that the languages all use the same writing system. Sure, they are all Hanzi, but that's tantamount to English, German, and French all using the Latin alphabet. In their correct orthographies (which few Wu speakers still master, but most Cantonese speakers do), the character sets are drastically different, and understanding transcribed Cantonese for a Mandarin speaker is as difficult as a French speaker reading Spanish. As a simple example: the third person singular pronoun is 他 (tā) in Mandarin, 佢 (keoi5) in Cantonese, and 伊 (i1) in Wu. The latter two both derive from 渠, while the former has a completely different etymology, but they are phonetically and orthographically different enough to be considered different words in different languages.

   佢真係冇文化，連呢個都讀唔明³

There are too many identical words in these languages for them to be different!!

This is the same as saying "Spanish and Italian are the same language because they have many identical words". Congratulations on discovering the idea that "languages descending from the same ancestor language share cognates". It's also undeniable that there are many words that aren't shared across these languages.

There are many arguments in a similar vein, such as "all these varieties share the same cultural and literary heritage". Instead of addressing each of them individually, I have the following general response: if your criterion decides that Spanish and Portuguese are the same language, then it's not a good criterion unless you actually do believe so.

This is a separatist agenda!!

Disclaimer: That's absolutely not what I mean. Plenty of countries hold together without speaking the same language; plenty of countries fall apart while speaking the same language. Again, languages are used to justify the borders, but the borders aren't defined or created by languages.

Chinese is a macro-language, just like Arabic and German; having mutually unintelligible varieties doesn't preclude the language status of the whole.

To be fair, I can buy this argument, but I don't particularly entertain the idea of "macro-languages" either—it is itself a compromise between political and linguistic considerations. By saying that Chinese is a "macro-language", one at least acknowledges its different status from commonly-accepted languages like English or Japanese. Similarly, I can accept the idea of Abstand vs. Ausbau languages, but it should be understood that the latter has no say of linguistics.

My biggest qualm with the "dialect" theory as applied to Chinese is that it bends the language classification framework in a way that makes it less useful for no apparent gains. If 粤语, 吴语, 闽南语, and 普通话 are all dialects of the same language, then what's 广府粤语 and 莞宝粤语? What's 太湖片吴语 and 台州片吴语? If we call mutually unintelligible languages with distinct lexical, phonological, and syntactic features the same "language", then what term is left for actually mutually-intelligible varieties?

Looking in the other direction of classification: we know that French, Spanish, and Italian are all Romance languages; if we assume that "French" is of the same status as "Chinese", then what's the equivalent of "Romance languages" for Chinese? It appears that immediately above "Chinese" is the whole Sino-Tibetan language family (which is the equivalent of "Indo-European languages")—we made a huge jump from a major language family to a concrete language and then to its "dialects", with no intermediate levels permitted. Scientifically, I don't find that an aesthetically pleasing taxonomy.

Sino-Tibetan         <-- This is a language family
└── Chinese          <-- This is a "language"
    └── Yue          <-- This is a "dialect"
        ├── 广府粤语  <--
        └── 莞宝粤语  <-- Then what are these?

Indo-European        <-- This is a language family
└── Romance          <-- This is a branch (many intermediate branches omitted)
    └── French       <-- This is a language
        ├── Parisian French  <--
        └── Quebec French    <-- These are dialects

To conclude: if you want to make a sociopolitical statement about what is a language and what is not, then you are completely entitled to your opinion. However, once you try to bring up structural and scientific arguments, think about whether they make sense in the context of other languages and are consistent with your own beliefs. If they don't, then probably stop smuggling politics under the banner of science.