Wenzhounese in Italy.

This Victor Mair post at the Log is fascinating for two completely different reasons. First is the “Devil’s language” aspect:

Wenzhounese is the most divergent variety of Wu and is considered a separate language by some. It is not mutually intelligible with other varities of Wu. It preserves words from Classical Chinese that are no longer used in other varieties of Chinese, and its grammar differs significantly. It also has the most eccent[r]ic phonology, and as a result is considered the “least comprehensible dialect” for an average Mandarin speaker. These feature[s] are a result of the geographic isolation of the Wenzhou area.

There are links to a number of Log posts about the dialect. The second reason is the situation of the Wenzhounese who have ended up living in Italy:

What prompted me to write this post were the answers I received when I asked my informants whether their Wenzhounese relatives in Italy, of whom there are many, learned Italian, and they said, no, they don’t have to. They said that the Wenzhounese in Italy are so numerous and dispersed throughout the country that there isn’t a need to learn Italian. The networks and support services available to them are so extensive that they can easily get by just knowing Wenzhounese.

That’s a common phenomenon in many situations, of course, but I wouldn’t have expected it for the speakers of an obscure Chinese topolect in Italy.


  1. Trond Engen says:

    Wenzhounese may be the “most divergent variety of Wu” and “”the least comprehensive dialect” for an average Mandarin speaker”, but reading further, that seems to be simply because it’s placed further down the coastal continuum towards Hokkien.

  2. David Marjanović says:

    Also, the Wu dialects apparently don’t have any exclusive shared innovations, so Wu is defined by the absence of the innovations of its neighbors.

    Rue Belleville in Paris is another Wenzhou colony.

  3. Wikipedia tells me that Wenzhounese is the most common Chinese topolect spoken in Europe generally, as well as being 90% in Italy.

    So many novels in the world, and what can I find from the Wenzhou diaspora? Google doesn’t know.

  4. Wikipedia tells me that Wenzhounese is the most common Chinese topolect spoken in Europe generally, as well as being 90% in Italy.

    Good lord, I had no idea.

  5. I had no idea Wenzhounese was so widespread in Europe either! And in answer to Tangent’s question, I suspect that the novel(s) that will immortalize in the world of literature the story of the Wenzhou diaspora in Europe and its ultimate fate will be written in the future…where, by whom, in what language(s), intended for which readership? Who knows…

    Trond’s guess that Wenzhounese is called divergent simply because of its geographical location (i.e. within the Wu dialect area it is almost maximally distant from the border with Mandarin) seems to be confirmed by the comparative vocabulary list here:


    Wenzhounese (fifth column from the right) seems quite unremarkable phonologically within Sinitic.

    I’ve also been reading an interesting dissertation on varieties of Sinitic, and in some respects Wenzhounese is a transition dialect, half-way between Central Wu or Mandarin varieties on the one hand and Min or Cantonese varieties on the other. So the “geographic isolation” is relative: Wenzhounese definitely does not look like a Sinitic counterpart to Sardinian in Romance or to Icelandic in Germanic.

  6. Etienne: On such a list no dialect looks particularly difficult. Wenzhou is difficult mainly because of some extreme sound changes it has undergone. For example, syllables with the rime -iang in Standard Mandarin has -iæ̃ in Northern Wu, -œŋ in Cantonese, -iũ in Southern Min, et cetera, et cetera, which are quite easy to grasp after some exposure. The corresponding Wenzhou rime is -i.

    To be fair, Southern Min has undergone even a more extreme set of sound changes, making it extremely dissimilar to all of Sinitic. However, the sound changes were ancient (earlier than 1000AD), and it has since then imported, like a non-Sinitic language, a set of character readings for literary words, from a Northern variety of Chinese not unlike Mandarin or Cantonese. It creates the typical Romance situation where it’s easy to understand “The Taiwanese people decides to elect a democratic government yadda yadda”, but you can’t get “Man eats fish”.

  7. J.W. Brewer says:

    Here’s a minor linguistic mystery. From poking around the google books corpus, it appears that prior to the dominance of pinyin in English-language publications, the usual English name for the topolect (as well as the related ethnonym) was “Wenchowese,” with the variant “Wenchownese” having but a single hit. More recently, “Wenzhounese” is dominant, with “Wenzhouese” the much rarer variant. What’s affecting the presence/absence of the extra N?

  8. David Marjanović says:

    Victor Mair’s preferences in euphony? No idea.

  9. January First-of-May says:

    What’s affecting the presence/absence of the extra N?

    Without knowing much (if anything) about Sinitic dialects (or English names for them), I’d say it probably has to do with the final sound being a consonant or vowel: the suffix would be -ese for consonant endings (such as “Wenchow”), and -nese for vowel endings (such as “Wenzhou”). Basically the same thing as the indefinite article (just in the opposite direction).

    This might be completely incorrect, however.

  10. Maybe interference from Shanghainese, Wenchownese’s big brother? (I was annoyed to check my copy of Ramsey’s The Languages of China and find that he devotes his entire Wu section to Shanghainese.)

  11. @ J.W. Brewer:

    You might also have picked up the unshu mikan, which came via Japanese.

  12. David Eddyshaw says:

    “Wenzhounese is the most common Chinese topolect spoken in Europe”

    This is presumably “Europe” in the gottverdammt Brexiteer sense; in the UK I’m pretty sure it’s Cantonese. On the other hand, maybe only a minority of European Chinese live in the UK, so both could be right.

    Wikipedia says 433,000 in UK versus 322,000 in Italy and about 300,000 in Spain and Germany together, so that may be it. It also has 700,000 in France, with Reunion; as far as I can make out, they’re Cantonese and Hakka by origin.

  13. David Eddyshaw says:

    Misreading on my part: looks like metropolitan France actually does have a substantially bigger Chinese-origin population than the UK, and that they are indeed at least half Wenzhounese.

  14. J.W. Brewer says:

    FWIW here’s an old LL post (with contributions from hat) on the “intrusive n” in such words, dating “Javanese” back to 1704. http://itre.cis.upenn.edu/~myl/languagelog/archives/002775.html

  15. David Marjanović says:


    Obsolescent; mostly replaced by the unstressed prefix Scheiß-.


  1. […] Hat reports on the Wenzhounese of […]

Speak Your Mind