I would like to conduct a quantitative (statistical) analysis of certain traits of Chinese characters, especially regarding the connection between character structure and phonetics / historical phonetics. Lists with frequency information would also be useful.
I cannot answer your question about a database, but remember that Mainland China uses "simplified" characters. Their structures may be less like the historical representations. In Taiwan, they use the traditional construction of the characters.
Given your research plans, I assume that you know this, but your analysis could be affected by which character structure set you examine, so I thought I would mention it.
:-)
Yes, I am a trained Sinologist with good knowledge in both modern and ancient Chinese. I am looking for digital resources that I can use for quantitative analysis. Most catalogues have both variants included.
Here are two online resources with information of Chinese historical phonology:
1. This one covers all the information you may need, including piyin, simplified and traditional orthography, IPA, explanation in Chinese, English translations, analyses of orthographical structures, historical pronuciations, and a few dialectal pronuciations, etc. However, note that the users need to be careful with the information and double check with the help of the other resources. Also, I have not found any way of downloading the database behind.
http://www.zdic.net/sousuo/
2. This one provides the historical categories and the historical reconstructions by different researchers ( Note that some researchers are Sinologists whose names are written in Chinese. Google translate would give their western names). This resource is very accurate.
http://www.eastling.org/tdfweb/midage.aspx
As for word frequency, the following resources are useful.
Cai, Q., & Brysbaert, M. (2010). SUBTLEX-CH: Chinese word and character frequencies based on film subtitles. PLoS ONE, 5(6), e10729. http://expsy.ugent.be/subtlex-ch/
Institute of Linguistics, A. S. (2012). Taiwan Mandarin Spoken Word List. http://mmc.sinica.edu.tw/resources_e_01.htm
Tseng, S.-C. (2013). Lexical coverage in Taiwan Mandarin conversation. http://mmc.sinica.edu.tw/resources_e_02.html
Thanks for these sources. Unfortunately, they are of limited use due to the fact that the data cannot be downloaded. The second source does not even display correctly in my browser and navigation seems to be broken the 上古音查询 resource seems to be working better).
Thank you nevertheless, this is already quite useful. Hopefully someone can provide sources that are also downloadable.
I don't know about database. But this Ted talk maybe helpful. http://youtu.be/troxvPRmZm8
Sadly, the TED talk is quite ridiculous from a scholarly perspective.
Rick Harbaugh, a professor at Indiana University in Economics, constructed an electronic Chinese character dictionary "Chinese Characters" "zhong1 wen2 zi4 pu3" that would allow you to search characters by components and build statistical models. Consider this database as an electronic version of Shuo1 Wen2 Jie3 Zi4 by Xu Shen, with modern pronunciation annotations. The organization of the database is clean and cross-reference accurate and complete. It's a great resource.
Here is the website:
http://www.zhongwen.com/
Thanks for the zhongwen.com link - I am of course aware of this site. However, as with many other ones, it seems problematic to download the corpus in order to conduct statistical analyses. I have also checked mdbg.net, dict.revised.moe.edu.tw, mandarintools.com, dgs.monash.edu.au, unicode.org, starling.rinet.ru, et al.
If you are interested in the reconstructed Old Chinese phonological values of the characters, the supplementary materials to Baxter and Sagart 2014 (Old Chinese: A new reconstruction, Oxford/New York: OUP) can be downloaded for free:
Link
http://ocbaxtersagart.lsait.lsa.umich.edu/
Thanks for the www.zdic.net link. Where exactly is the download for the Chinese character database? Thanks!
The Baxter-Sagart reconstruction DB mentioned above is a pretty cool resource! For structural information I think the current best resource is CJKV-IDS: https://github.com/cjkvi/cjkvi-ids
You may check Chinese Linguistic Data Consortium(CLDC). It's a big data for Chinese characters, phonetics and texts. If it doesn't work, you may enter into http://www.cncorpus.org/login.aspx.
I found the data in the ytenx repository to be super useful: https://github.com/BYVoid/ytenx/blob/master/ytenx/sync/dciangx/DrienghTriang.txt