There are a few things I find myself asking when I meet people from previously unheard of parts of China; How do you say 你好，谢谢 and 再见? Then I ask for numbers one to ten. I’m sure I’ve annoyed my share of people with these seemingly useless questions, but for me they’ve always given me a good idea of the sound of a dialect. Furher, numbers always seemed important to me as some of the most common and useful words to learn.

The following is a list of numbers from one to ten in the dialects running from Shànghǎi to Nánjīng with the last four columns covering Hángzhōu, Qúzhōu/Jiāngshān, Línhǎi and Chánglè dialects all from Zhèjiāng province. Holes in the table do not represent an absence of that number but rather an absence of data in my possession. Note I’m missing a bit of Suzhou dialect. If anyone is willing to send a clear recording of those numbers by a native speaker, I would be grateful and your name would appear in pixely lights on the site.

A couple notes: Due to its greater frequency over 二, 两 has been given for most dialects. While 二 is still used frequently, e.g. the “twenty two” in “two hundred and twenty two”^‡, it will be replaced in most. So, for example, an apartment numbered 2210 will be read as “两两一零”.

In some cases I have only partial data and no speakers handy of whom I can ask the favour. The number 7 in Hangzhou, for example, I know to be a high tone and terminating with a glottal stop. However the specific text from which I’ve taken that information did not provide information on the vowel and guessing didn’t seem appropriate.

The Wǔjìn set was taken from one of many villages in the Wǔjìn area. It’s a slightly more rural sound than you may get closer to downtown Wujin or Changzhou. The most notable difference between the list for Wujin and urban Changzhou dialect is 三 becomes seiʔ⁵.

This took quite some time to compile using a number of sources. Most are from recordings I’ve made and transcribed myself. Some are my own transcriptions from outside recordings. Most of the Zhejiang data were compiled from a number of academic papers on Wu, mostly available through the Wu Association page. If you’re curious about specifics let me know and I’ll get the details.

Some of the tone numbers for recorded samples have been taken from various of sources including work by the ubiquitous YR Chao and Glossika’s excellent collection of tone tables. Trusting their research over my own ears I’ve done my best to match the sounds I heard to the tone values given. I reserve the right to have completely missed the mark on the tones and transcription.

This is especially true with the Gaochun tones. Due to the age of my speaker there is a notable discrepancy with what is available elsewhere, mostly because what’s elsewhere is at least a generation older. If you look hard enough you may find the table to the right, scanned from some text of dialect tone tables. I imagine it’s where the Glossika tones came from and unfortunately there’s only the one set. There are some notable and consistent differences between my speaker and the tables, so perhaps we can call the Glossika numbers 高淳老派 and these 高淳新派.

Of less academic value but greater practical value, I offer the following table. A large part of my motivation for compiling this data was to get a better sense of the ἰδέα^҂. What follows is my best guess at a collection of what should be universally understood numbers. Superscript corresponds to Mandarin tones, not tone values as in the previous table.

For what it’s worth I’ve had some success with this in Changzhou and Wujin as well as with one Shanghai transplant and one from Danyang.

- – -
† I’ve given two different pronunciations for 五 in Gaochun dialect. The first (ŋ̩⁵) is used by the older generations while the second (ʋɯ⁵) is used by younger speakers. The influence of the neighbouring Mandarin is all too apparent.
‡ What’s more confusing is the number 222 (read [liã³⁵ pɐʔ⁵ ɲiɪ¹¹³ n²] in Linhai dialect) which like most Wu dialects makes good use of 廿 for 20. The northern equivalent (rendered in Pinyin style spelling) would be roughly “liang be nei ni”.
҂ As in Plato.