Modeling Middle Period Life Histories
Notes for roundtable discussion
Hilde De Weerdt
University of Oxford
1) the first set of questions we would like to you address:
o Are there important features in late Song and Yuan life history writing that differ
from preceding or succeeding periods?
o Does CBDB capture those features adequately? If not, what changes need to be
made for it to do so? Can you suggest new fields, new types, or new tables?
I went for a small set of biographical texts of particular genres (muzhiming and
xingzhuang written by and for Zhu Xi and Chen Fuliang) and scanned those closely for
information that would be useful a) for prosopographical purposes; b) for social network
analysis (which is where my primary interests lie). For those of you who attended the
Warrick conference or read the report I delivered there, my approach here is the exact
opposite of what I did there: there I analyzed the database data to gauge its immediate use
for social network analysis, here I start with a small set of primary sources to comment
on the database structure.
Huang Gan’s obituary (xingzhuang 行狀) for Zhu Xi:
In what seems to me to be a very unusual move, he outlined at the very end of his
lengthy obituary (37 pages in a modern Zhonghua shuju edition) how his obituary
differed from that of his contemporaries. In the process of showing us how his work
diverged from current conventions of biographical writing, he has some interesting things
to tell us about what may have been standard features of at least the xingzhuang type, a
type of biographical writing that also went into dynastic biographies—in fact the Song
shi biography for Zhu Xi very closely matches Huang Gan’s xingzhuang.
What conventions did Huang Gan take issue with?
Lack of dates
Conversations with the emperor: brevity
Quotations from memorials: brevity
Huang Gan here issues some interesting caveats about core primary biographical texts:
some of which (reserve) we may be familiar with, others (the lack of time references for
postings etc.) we may have seen but not fully appreciated. However, even though these
caveats are interesting, Huang Gan, while apologizing for the great lengths at which he
focuses on the scholarly side of things as well as on Zhu Xi’s persistent refusal to take
office and the stern tone of his exchanges with the emperor, in fact also abides by the
requirements of the genre—his apologies are a direct consequence of the strong hold the
conventions of the genre had on the imagination of biographers. Like the other more
extensive biographical accounts (muzhi ming and xingzhuang) I surveyed, his account
followed a particular pattern: moving from a discussion of Zhu Xi’s family background
(ancestors) and early signs of precociousness, official career and teaching, character,
philosophy and scholarship, death, writings, descendants, signature (name of author and
date of writing). [apologies follow] In fact when comparing this account to the earlier
account Chen Fuliang wrote of Xue Jixuan, some of the things that Huang Gan
apologized for were in and of themselves not unusual: Chen also quoted at length from
reports and memorials that Xue submitted—(focusing on local government, military
affairs and taxation).
When comparing the data that these authors put into their biographies to the current
tables (of codes, types and data), the following things came up: dividing them up by
1) studying and teaching:
Currently such information seems to be encoded through
Status (社會區分): teacher and teacher military (94 and 95)
In the table one can enter time for this (not evident from current layout)
More importantly, one would like to enter where these people taught (the address codes
should be applied here), as well as in what type of institution (school code table: could be
further developed) and which specific institution
The use of this should be clear as it would allow us to see who studied together in the
same institution (rather than simply same teacher).
--related to this: where would one enter information on establishment of
Similarly, different types of student status are included in the status code table (122-135
and 47, 90, 83, 100, 115), but at this point no tables include where (place), and which
school (type and name) they attended.
I wanted to add here as well, that the tables include status types, but status type tables (in
contrast to status code tables) did not come with the date we have.—the redone offices
type and codes are a good model for how this too could be redone.
2) another feature that figured prominently in xingzhuang/muzhiming of those I surveyed
(elite scholar-officials) is submitting memorials.
Should it be included?
1) if we want to reflect data relevant to biographers: yes
2) if we want to collect data on a society or at least on social strata that are so heavily
bureaucratic in nature, yes
What can we learn?
When and how frequently local officials made their voices heard up higher, or at
least more publicly
Times when, positions where, locations where this was more common?!
The emphasis on this aspect may reflect differences over time or differences of interest
among biographers and their subjects—that too could be of interest.
[The larger question of differences between Song/Yuan biographical writing with that
before or after, is a question I won’t be able to address, but a comparison with some of
the later reports (Nick Tackett’s on Tang-Five Dynasties period, for example) may help
us address this question over the course of the conference.
Where would this information go?
3) Writings/creative works:
--revise and edit list of genres (still featured on revised one in new interface)
--distinguish between different types of “scholarly production”
collated/commented on (not necessarily published)
[all of these came up in Xue Jixuan’s case]
4) assets table: include
book collection: size, location, time in existence, (accessed by), cataloged or not, cause of
5) Entry: examinations: specialized in (Classic, law, astronomy, maths, finance)
[Tang special exams]
1) I found a lot of association codes relevant to editing the work of others (compiled
poetry, scholarly works 341-342, 456-457; also prefaces and postfaces written or
solicited), but not:
*printed/circulated the works of
[below: Where? When?]
* Contributed to the printing of the works of
2) accessed the collection of
(borrowing not only means of access—reading libraries)
3) general lack of association by associations:
Wen 文 (Fushe etc.)
4) add place references for associations—time is included in original tables (not in
current search interface)
-Time? yes (year)
How is range defined here? (does it include time range: month to month, year to
-Place? (include place fields—in hierarchical order to allow for different kinds of place
If I could enter the data in bios on those types of associations (published/printed or
accessed the collection of), how would I find the association code?
At this point: here is how it looks:
[slide: types and then subtypes]
I had a similar problem last year when working on the mapping of correspondence
networks: correspondence associations could fall under writings/correspondences but also
My point here is that we need to think more carefully, as the number of associations
keeps growing, how we want to organize them.
This raises the difficult question of ontologies, which is I believe on the table, but worth
further discussion at this conference. I am generally in favor of a more developmental
model that allows for the abstraction of associations found as we delve through
biographical sources—we may have reached a point where it is worth rethinking the
original structure (which clearly also developed as Hartwell added on new association
types that reflected his interest in particular topics, like finance).
I didn’t attempt to draw up an alternative model, but
I do have a few other things, in response to the second set of questions:
o What would be the best way to harvest data from Song and Yuan biographical
o What are the most important/most extensive collections of materials? Are these
available in searchable digital formats?
those mentioned last time: modern indexes
(see biographical sources in Song Tools)
In addition to SRZJZLSY under development, also
Li Zhiliang 李之亮 which includes postings data at the court, circuit and prefectural
levels for the duration of the Song Dynasty
Songdai jingchao guan tong kao 宋代京朝官通考 (Chengdu : Ba Shu shu she, 2003), 5
vols.; Songdai lu fen zhangguan tong kao 宋代路分长官通考 (Chengdu : Ba Shu shu she,
2003), 3 vols.; Songdai junshou tong kao 宋代郡守通考(Chengdu : Ba Shu shu she,
2001), 10 vols.
For membership on the Council of State for the entire Sung dynasty, including documents
of appointment and some biographical and anecdotal information, see the punctuated,
宋宰輔編年錄校補 (Chronology of Sung chief and assisting councilors, revised).
Wang Jui-lai 王瑞來
This is a biographical index of over 6100 Song monks, indexed by Buddhist names.
Li Guoling 李国玲
NOTE: uses model of SRZJZLSY!!
Requiring more of an effort in terms of analysis:
宋人年譜叢刊 (dates for everything!!; esp. relevant if Huang Gan is
right that Song people did not like to put dates in funerary
This collection of chronological biographies (nianpu) includes 163 biographies of Song
individuals published between Song times and the present. All are punctuated.
List of these and muzhiming sources in updated version of Song tools. Brings us down to
primary sources, rather than indexes:
--structured texts: muzhiming and xingzhuang
[Quan Song wen!!!]
--small-scale projects, manual tagging—result in better tools for automated harvesting
e.g. Zhu Xi’s letters: dating; experimenting with tagging and developing terminological
indexes that are associated with particular themes
[rich qualitative data]
(pilot projects with other punctuated and collated wenji: Zhu Xi via GXBD) 《朱熹集》
宋·朱熹 131 卷.〔明嘉靖十一年福州府学本〕
To fit specific research interest—we will all have to do some of our developing—and it
may be then be more helpful to know what tools people have developed to come to their
e.g. depository for dictionaries, tools used to mine and analyze Unicode texts.