SERICA

Collections of specimen pages are a characteristically Chinese bibliographical genre, albeit one that doesn’t go back much more than a century.

The earliest example is thought to be a work by the famous bibliographer Yang Shoujing 楊守敬 (1839-1915), who in 1880 became an attaché in the Chinese embassy in Tokyo. While in Japan, he befriended local bibliophiles such as Mori Risshi 森立之 (1807-1885), and taking advantage of the low price that Chinese books were currently fetching on the market there, built up a collection of editions that were either rare or not extant in China, and eventually shipped them home. His notes on these books were published in 1901 with the title Riben fangshu zhi 日本訪書志.

Ten years later, the tracings of specimen pages that he made from the rare books he saw in Japan, together with others made in China (a total of 483) were published in twelve fascicles. Unsurprisingly – as he was a bibliographer and collector himself – there was a copy among Piet van der Loon’s books:

留真譜初編 不分卷 / (清)楊守敬編
清光緒辛亥[1911]宜都楊氏觀海堂刊本
線裝12冊 ; 31公分
Sinica 4545

This work is notable for three reasons. Firstly, it is the prototype of the genre; secondly, it is rare; and thirdly it is block-printed, and not produced photo-lithographically as all the later collections are, right up to the present.

The following illustration shows how remarkable this work is:

Although the page appears to have been produced carelessly, it is in fact a superb example of the block-cutter’s craft. The tilt is deliberate, as evidenced by the specimen on the right hand side of the leaf, which is squared up (not visible of course in this illustration); it has been made so in order to replicate exactly the copy from which it was traced. The smudges are also deliberate, and even the worm holes have been carefully reproduced.

A second collection (「二編」) in eight fascicles and containing a further 252 specimen pages was published posthumously in 1917. Kyoto has both parts, so does Stanford, but I have so far found none elsewhere.

The purpose of these collections, which continue to be produced, is primarily to show what the editions look like – Chinese printed pages are very variable in design and presentation, and are difficult if not impossible to visualise from normal catalogue entries. But specimen pages are also an invaluable tool for the cataloguer as an aid to identification.

This is best illustrated by the following example of a problem we experience whenever we set about cataloguing a Chinese book.

Here is a palace edition whose contents needn’t concern us, but it is described as “a synthesis of the doctrines of the Neo-Confucian school” by Fang Chao-ying (in Hummel, 474). It was compiled and printed at imperial behest in 1715 by a team of scholars working under the direction of Li Guangdi, and is absolutely typical of the genre. I catalogue it thus:

御纂性理精義 十二卷 / (清)康熙五十四年[1715]李光地等奉敕撰
內府刊本
線裝8冊 ; 28公分
Sinica 2834

The copy is rather good, being printed on fine white paper, and the preface bears two imperial seals. In the bibliography of palace editions compiled by the National Palace Museum and Liaoning Provincial Library in 1995 (清代內府刻書目錄解題, 301), the physical description of the edition reads as follows:

半葉大字八行，行十八字，小字雙行，行二十一字，中字八行，行二十字。四週雙邊，白口，單魚尾，無行格。版框高二二·五厘米，寬十六·二厘米。

I will translate and explain this for the sake of those who are coming to this sort of bibliographical description for the first time. What we have is a detailed description of the printing block, unusually long in this case, as the text comes in three sizes:

Each half-leaf has large text in 8 columns, with 18 characters in each column; there is small text in double columns, with 21 characters in each column; there is medium-size text in 8 columns, with 20 characters in each column. The block frame is a double line all the way round, the centre column is white, there is one fish-tail, and there are no column rulings. Each half-block is 22.5cm. high and 16.2cm. wide.

I think most of this is self explanatory apart from the fish-tail, perhaps. This is the black device in the centre column which is there as a marker for folding the leaves. Often there are two, and occasionally more. Sometimes there are none. This image will make things clear (apart from the medium-size text, which doesn’t appear on the first leaf):

Now here is a leaf from what a first sight appears to be the same edition:

But closer inspection will show that the impression comes from a completely different block. The edition is actually a copy of the palace edition made by a commercial publisher whose name is stamped on the printed label on the first fascicle. I describe it as follows:

御纂性理精義 十二卷 / (清)康熙五十四年[1715]李光地等奉敕撰
覆刊內府本
線裝8冊 ; 29公分
書簽有「書業德記發兌」印記
Sinica 2573

The problem is, the elaborate physical description above could equally be applied to this copied edition. Only the dimensions are slightly different, and even they are not critical to identification. A wooden block which is dry, perhaps having been stored for a long time (and some are stored for centuries) will have shrunk. When inked, it will begin to expand, so that the impressions taken at the end of the day are sometimes considerably larger in height than those taken at the beginning. The grain of a printing block always runs horizontally (as evidenced by cracks, which are never vertical), so the expansion always affects the height, not the width of the printed leaf, which stays the same. However, thin Chinese paper is very unstable, and easily distorts when moistened, so that the printed area of a leaf, first wet from the printing, then dry, is not necessarily identical with that of the block from which it was printed. And if the book is old and the leaves have been repaired and perhaps even laminated, the measurements will be of no use whatsoever for distinguishing editions.

All this leads me to question the value of these traditional Chinese book descriptions, especially when we can now produce scans of “specimen pages” 書影 so easily and make them universally available on the internet. We do not need to describe what can be seen at a glance, and the specimen page leaves no doubt as to which block an impression is taken from.

This is why modern online catalogues are increasingly attaching specimen pages to their records, and why we have developed this functionality for our own. Producing them is a little time-consuming, and I have not been able to make them for every record. But I try to do so at least when I’m unable to identify an edition, or when my confidence in an identification is less than total.

In the examples quoted above, it is quite clear which edition is which, because one is printed on fine white paper, the fascicles have indigo covers, and the preface has imperial seals. The other, by contrast, not only bears the seal of a dealer, but is of inferior workmanship and is printed on very yellow bamboo paper. But the identification of the palace edition is confirmed by comparing the impression with specimen pages from the many copies of this work in Peking University Library that appear in CALIS.