Save

Towards a Broad Concept of Punctuation

論廣義的標識符號概念

In: Bamboo and Silk
Author:
Matthias L. Richter University of Colorado at Boulder (科羅拉多大學波爾得分校) Boulder (波爾得)

Search for other papers by Matthias L. Richter in
Current site
Google Scholar
PubMed
Close
Open Access

Abstract

This article proposes a typology of punctuation devices based on functional criteria. It argues that a broad concept of punctuation – including not only non-“alphanumeric” marks but also layout and spacing as well as script features – is needed to do justice to the diversity of material features of manuscripts and their changes over time.

1 Methodological Considerations

Punctuation, like most material features of written texts, is something modern readers tend to take for granted.1 We intuitively process the information embedded in these features and attach much less importance to them than we do to the words of the text, which seem to convey its meaning entirely. However, successful communication of a text is not based on a string of words alone. In spoken language, the information about how the sequence of words that constitute a text are intended to be understood is conveyed by prosodic features, such as intonation, volume and quality of voice, as well as by situational factors, such as position, appearance, and gestures of the speaker. These features are necessarily absent from a written text, as long as it is not read aloud.2 And in modern times of predominantly silent reading practices, the text remains stubbornly silent even while being read. The written text is not an agent. It cannot, as Plato famously complains, speak for itself.3 It is the potential victim of innocent misunderstanding no less than of willful distortion.

Writing cultures have from early on devised means to compensate for this deficit. Conventions were developed – some of them short-lived, others extremely long-lasting – for encoding information about the text’s intended illocutionary qualities in its visual representation.4 Codicological features such as book format and material or graphic features such as type of script, size of character font, layout features, and multiple forms of punctuation all contain more or less narrowly defined information about the communicative intent of a written document. Of these material features, the one most immediately connected with the text is punctuation. Consequently, no other feature of a written text has as great a potential as punctuation to reveal how the text was understood by the people who wrote and read it at the time when the manuscript (or, in later periods, printed text) was produced and during the time span when the written text was used. Explicit indications of the internal structure and external delineations of the text can provide traces of its use or of the purpose for which it was written. If we aspire to read an early Chinese text not entirely based on the interpretive habits of the twenty-first century reader nor solely through the lens of transmitted early Chinese literature, i.e. from the late Western Han librarians’ perspective on their textual heritage,5 if we hope to get a sense of how a text spoke to people in earlier times, we need to learn to decode the fossilized imprints of their ways of approaching a text, which producers and users of early Chinese manuscripts left us in the form of punctuation. Yet, punctuation plays at best a marginal role in the study of early Chinese manuscripts.

The relative neglect of punctuation as a subject of serious scholarship is lamented not just in Chinese studies and not just with regard to the literature of the past. The problem has been remarked on even in the context of contemporary education. Spelling is generally given much more attention than punctuation as a criterion for evaluating the quality of someone’s writing. Orthographic standards tend to be more rigid than the rules of punctuation. Synchronically, orthography

is essentially a closed system […] a spelling is almost always either right or wrong […] Punctuation, on the other hand, is a relatively flexible system. It may use a relatively small number of marks but these can be used as dynamically as can any combination of words […] there is a choice involved and authors can consciously make these choices.6

In the Chinese case, the neglect of punctuation in scholarship is largely due to the fact that punctuation played a very minor role in the printed literature of traditional China.7 In all Chinese writing there is arguably less need for marking word divisions than in alphabetical writing, and in Literary Chinese the high frequency of initial and final particles makes sentence divisions relatively obvious. Further, as Imre Galambos points out, “the often parallel structure of sentences, its organic symmetry and rhythm” provide an important aid to reading.8 So it may have seemed unnecessary and uneconomical to insert signs between the characters, in order to mark syntactic units, as is done in modern computer generated script, where each punctuation mark takes the same amount of space as one character.9

The scarcity of punctuation in printed texts and the fact that manuscripts, which are usually punctuated, have only recently gained importance in scholarship have led to mistaken generalizations about literary Chinese as a language without any punctuation:

Les textes anciens n’étaient pas ponctués. Tout étranger qui a étudié la langue ancienne se souvient de ses premiers découragements face à des textes d’une seule pièce, ou la limite des phrases n’est pas marquée. On apprend à repérer ou commencement et ou finissent les phrases. Un lettré chinois trouvera plaisir à noter lui-même les points forts ou les pauses du texte en traçant avec soin des petits cercles à l’encre rouge dans les marges de ses livres. Pour les textes classiques, ou le rôle de l’enseignement oral était important, l’acquisition du rythme se faisait sans doute aisément; pour les textes administratifs, jurisdiques, techniques, des formules stéréotypées permettaient de s’y retrouver. La ponctuation est, en Chine, d’usage récent.10

Even more consequential conclusions have been drawn from the apparent lack of punctuation in traditional Chinese literature. The “absence of sentence function marking” was interpreted as an indication of an underdeveloped concept of sentencehood and, ultimately, the absence of a concept of truth in pre-Buddhist Chinese philosophy.11 More realistically, Harbsmeier and Bodde view the practice of leaving the punctuation of printed texts to their readers as rooted in social factors, such as a “proprietary attitude toward literacy and book culture.”12 The pleasure in punctuating a text oneself, mentioned by Viviane Alleton, may well have been derived not just from the fact that the text became more transparent to the reader in the process but also from the affirmation it provided of their identity as part of the educated elite.

Evidently, punctuation is easy to misinterpret or to overlook entirely, if we project preconceived notions of the nature and function of punctuation onto new sources with whose punctuation practices and underlying conventions we are not yet familiar.

2 Definition of Punctuation

Whether or not any written document can be entirely without punctuation depends on how we define the term. At first glance, what punctuation is seems obvious and trivial: a small set of marks – i.e., comma, semicolon, colon, period, hyphen, dash, parentheses – serving to divide the text according to its logical structure, i.e., based on syntactic criteria. This captures much of the inventory and function of modern punctuation, particularly in Indo-European languages, but it is an incomplete account in several ways: it covers neither the full range of punctuation devices nor all functions of these – and, most importantly, it does not account for historical changes. The Latin palaeographer Julian Brown gives a more accurate definition:

Punctuation is the use of spacing, conventional signs and certain typographical devices as aids to the understanding and correct reading, both silently and aloud, of handwritten and printed texts.13

Others, e.g., Geoffrey Nunberg, use the term “punctuation” in a narrower sense “to refer to a category defined in partially graphic terms: a set of non-alphanumeric characters that are used to provide information about structural relations among elements of a text, including commas, semicolons, colons, periods, parentheses, quotation marks and so forth.” This narrow definition obliges Nunberg to add that “[f]rom the point of view of function […] punctuation must be considered together with a variety of other graphical features of the text, including font- and face-alternations, capitalization, indentation and spacing, all of which can be used to the same sorts of purposes.”14

A narrow definition serves its purpose well when it refers synchronically to a writing practice based on largely uniform punctuation standards, such as they tend to exist in today’s modern cultures that enforce the same orthographic and other text production standards through a common, unified education system and, more importantly, a professional guild of publishers of printed texts, who (except for slight variations) all subscribe to the same orthographic and punctuation standards. In such circumstances, the forms, functions, and proper use of punctuation marks can be described fairly clearly as a set of non-alphanumeric signs and can as such be clearly distinguished from, and thus be related to, the alphanumeric signs representing the words of the text. These two areas, in turn, can be neatly distinguished from the communicative functions of layout features and various paratextual devices of the document in question. If, however, we take a diachronic approach or refer to diverse punctuation practices even within the same historical period, we cannot narrowly define punctuation in contradistinction to alphanumeric signs (i.e., script, including that of non-alphabetic writing systems), nor separate it from text layout. Not only do these three areas interact in every written document, their functions and the elements performing these functions often transition from one area to another. Consequently, no absolute distinction between the areas of script (alphanumeric signs representing language), punctuation (narrowly defined as non-alphanumeric signs), and layout can be upheld.

The necessary widening of the area under investigation raises the question of whether the term “punctuation” is still justified. This can be answered in the affirmative not just for the pragmatic reason that an established term is always to be preferred to a newly invented one, unless the former would obscure a truly novel concept. The term “punctuation” has always stood for much more than the punctus, the marker of spaces between words in Latin texts. Hence, no customary literal understanding of the word should stand in the way of the broad definition that I suggest for this treatment of the topic: the term “punctuation” refers to all visual features which, in addition to the immediate written representation of words, aid the reader by disambiguating or otherwise specifying how to adequately vocalize (and ultimately interpret) these representations of words, either internally in the process of silent reading or externally in reading the text aloud. This function goes beyond the recognition of words but applies equally to the levels of the sentence and the text.

3 Areas of Punctuation

According to such a broad definition, punctuation comprises three main areas: (1) layout and spacing, (2) script features, such as capitalization, italics, bold script, size and alignment of script (e.g., super- and subscript), and (3) non-alphanumeric elements of writing, which may be loosely termed “marks.”

The boundaries between these categories are permeable: For example, early Latin manuscripts used marks (specifically the punctus) to indicate word divisions, while the same function was performed by spacing in later periods, in handwritten texts as well as in printed or typewritten ones.15 Modern writing technology has reintroduced some marks that had become obsolete for some time: Computer programs usually allow the writer to display a punctus (∙) for word divisions and a paraph (¶) to indicate the end of a paragraph. On the printout of the same text these divisions are only reflected in the form of spacing. That these signs are distinguished from the text proper by displaying them on the screen in a lighter color and that even PDF versions of the same text do not display them at all shows that they are understood to aid the writer in preparing the text for readers, but readers are not expected to rely on these devices that do not anymore belong to our intuitive concept of punctuation.

Moreover, the division between non-alphanumeric marks and elements of writing that represent language, viz. letters, or in the Chinese case characters, is not necessarily permanent either. The paraph itself as a mark for paragraph divisions is an example for a transition between different areas of punctuation: It is a letter that developed into a sign. The paragraphus – a Γ, γ, §, or a similarly shaped sign – marked the beginning of a new paragraph and signaled to the rubricator where he had to fashion the litterae notabiliores, the flourished capital letters at the beginning of the new paragraphs. This sign was increasingly replaced by a nota K (standing for kaput or kapitulum), which was later replaced by the letter C (for the variant spelling capitulum) to which a vertical stroke was added, resulting in the paraph ¶.16

The replacement of the umlaut indicator superscript “e” by two dots is another example of a letter mutating into a symbol. The symbol &, derived from a stylized form of the Latin word et, came to be used even in languages other than Latin, which proves that the symbol functions independently of its original phonetic value. A similar phenomenon occurs in 13th and 14th century Uighur manuscripts when they mark the end of texts with a ligature of the Chinese characters “了也.” The two separate characters stand for liao ye (“it is finished”) in Chinese, but the language of the manuscript text after which they appear is not Chinese.17

All three areas of punctuation – layout and spacing, script features, and marks – play a significant role in early Chinese manuscripts. In the context of Chinese writing, it would clearly be mistaken to describe marks as “non-alphanumeric.” Hence, I define “script” here narrowly as Chinese characters that stand for words (including numbers), while I will simply call everything else that is applied to the writing surface, in ink or otherwise, “marks.” Layout and spacing naturally concern marks just as much as script. I will not attempt to match the terms I propose with Chinese terminology. The most commonly and broadly used term in Chinese, biaodian fuhao 標點符號, usually refers almost exclusively to marks.18 Other features of text organization – such as the placement of titles, numbering of text sections or bamboo slips, or marking off of catalogs – are usually treated under the rubric of tiji 題記.19

Marks are clearly the most frequent and most widely distributed means of punctuation in all genres of writing throughout the Warring States and early imperial periods. Among these, the most frequent are a double or single short dash below the bottom right of a character. In a similar position there are thicker such short dashes or small black squares or small rectangular hooks. Larger hook-like marks of varying shape occur at the end of texts, usually not as close to the preceding character as the aforementioned marks. Round dots are usually placed either between characters in mid-column or on top of a column or in the margin. Thick horizontal bars can reach across the entire width of the column. Especially administrative documents feature a wider variety of differently shaped marks.

Layout and spacing, just like in other manuscript cultures, play a greater role in the diplomatic context than they do in literary texts.20 Specially large character spacing or large blank spaces after portions of text are the most frequently employed means in this area. Less often we find different orientation of text, placement of text on the verso of the manuscript or in margins, elevated text columns, or writing in several registers.

Special script features are the least frequent means of punctuation. Elongated vertical strokes occur mostly in Han administrative documents. Some manuscripts of earlier periods use stylized, ornamental types of script that are reminiscent of inscriptions on prestigious objects. Such properties of script usually coincide with special layout features.

To understand the function of all these and more punctuation devices that occur in early Chinese manuscripts and thus to recover their informational value for us, who study these manuscripts today, it is important to remind ourselves of the great diversity and fluidity in historical punctuation practices, so that we won’t seek explanations within a too narrowly defined range of possibilities when we explore the not yet fully understood punctuation practices in ancient Chinese manuscripts.

4 The Semantic Value of Punctuation

Modern readers generally expect punctuation to reflect the logical structure of a text. This perception is owing to a largely syntactical punctuation practice that has enjoyed a relatively stable existence in Western print culture for some considerable time. However, as Malcolm Parkes observes,

[t]hroughout the history of punctuation attitudes to the use of the symbols have been bound up with developments in traditional attitudes to discourse. These are reflected principally in discussions of punctuation based on different modes of analysis: grammatical analysis and rhetorical analysis […] Grammatical analysis has been concerned with the application of punctuation to identify the boundaries of sententiae (later, ‘sentences’) and the units of sensus or grammatical constituents within them. Rhetorical analysis has been concerned with the ways in which punctuation reflects the periodic structure of a discourse, […] With its emphasis on pauses for breath this mode of analysis has been preoccupied with bringing out correspondences between the written medium and the spoken word.21

For Western alphabetic writing, the general assumption is that of a historical shift of punctuation “from elocutionary principles to syntactic principles.”22 This development is linked with changes in reading practices – particularly the practice of silent reading is seen as a major factor that led to the increasing dominance of syntactic principles of punctuation:

Punctuation is a complicated and much debated matter, connected as it is with such different subjects as grammar, prosody, rhetoric, liturgical practice and music, and because it is immediately concerned with the sense of the text. […] Modern punctuation is of a syntactical nature, that is, it indicates the grammatical structure of the text. Mediaeval punctuation had partly the same function, but was to a great extent rhetorical, in other words it underscored the structure of the text (its rhetorical units) as it was read aloud. It not only marked the pauses the reader had to observe while pronouncing the written text and their length, but also the pitch.23

To assume a complete shift from a previously purely rhetorical punctuation practice to likewise purely syntactic principles would, however, not do justice to the complexity of writing practices of any historical period. It is advisable for a study of any culture and any period to realize that “while punctuation can at particular moments be specifically rhetorical or syntactical, much of the time both kinds of marking coincide.”24 For performative texts, such as liturgical ones (gospels, lessons, epistles, prayers, blessings, and psalms), punctuation did not just indicate pauses for breath but more specifically caesurae and cadences along with corresponding changes of pitch. This system was to some extent carried over into and perpetuated in the punctuation of literature.25

5 Who Punctuated?

The function of punctuation in any document is, of course, inseparably connected with the question of its originator. Punctuation can originate with the author or editor of a text, with the scribe who first put it into writing, with a copyist who copied an earlier written version of the text, with a corrector, or with a later reader, or all of the above. The person who punctuated a text potentially had a great influence on its interpretation and later transmission. Punctuation and its interpretive effect on the text do not necessarily coincide with the authorial intent. In the European tradition, punctuation was usually a matter of book production, not the business of authors.26 In an age when, firstly, texts are usually composed in writing and, secondly, this first written form comes to be understood as a preliminary representation of the text – a “manuscript” (or “typescript”) in the sense of a draft – before it is disseminated in print, editors have to a considerable extent appropriated the text by conforming it to their standards, before they release it into print. Authors usually accept this division of labor; they possibly adapt their mode of writing accordingly and may even take a certain laissez faire attitude towards orthography and punctuation and other formal features of the text, since they can rely on the editor for this part of the work.27 The advance of modern computer-based word processing appears to have led to a decrease in editorial input and an increase in the author’s control over the text in its final form, including punctuation.

This development makes it all the more important for us not to confuse the interpretive force of punctuation in historical sources with authorial intent. Recognizing punctuation as a phenomenon not so much of text production but rather of book production and the reception of texts, however, leaves us with some uncertainty as to whose approach to and use of the text the punctuation in a given manuscript reflects. While two of the three areas of punctuation – layout and spacing as well as script features – are clearly subject to manuscript production, the third and most discussed one, i.e., the non-alphanumeric marks, can go back to either the production or the later use of a manuscript or both. In most instances we can at best cautiously speculate which is the case. Spacing occasionally allows assumptions as to whether the marks were inserted during the process of writing the text or afterwards.28 Even in the latter case, the marks can still have been added by the same person as the one who wrote the words of the text.29 Hence, it is virtually impossible to decide with confidence that marks are traces of the use rather than of the production of a manuscript. If they are part of the production process, the marks can still have multiple origins: their nature and position may have been determined by a commissioner of the manuscript (who theoretically could be identical with the author of the text) or by the scribe or by someone who was specially given the task of inserting punctuation. Such a division of labor (aside from the preparation of the stationery, which was most likely taken care of by persons other than the scribes or copyists) is by no means a far-fetched assumption. In Western manuscript production not only the copyist, but also the rubricator, the illuminator, and the corrector all played their different parts in completing a manuscript.

Malcolm Parkes reminds us that punctuation may have accumulated in the life of a manuscript after it was initially written:

When considering copies as witnesses to the practice of a particular period in time, it is necessary to determine the status of the punctuation: for example, in a manuscript, whether it is that of the scribe in the same ink as the text, or has been added by a corrector or reader in ink of a different colour. Some copies of texts have several layers of punctuation which have been introduced at different times.30

We must also remember the possibility that some or all of the punctuation in a given manuscript was not determined by anyone who had any part in the production of this particular manuscript. Various features of the written representation of a text are often taken over from earlier copies that the manuscript in question is based on. Copyists are prone to take over features of style of script, orthography, spacing, or punctuation of the models they copy, even if these features go against a production standard they are otherwise committed to. This adherence to the copied model can either be inadvertent or the copyist interprets these features as necessary parts of the copied text that he feels committed to. Accumulated punctuation can obscure earlier understandings of a text or even become an impediment to reading.31

On the other hand, there are cases – as demonstrated by Fukuda Tetsuji in his study of the Guodian manuscript *Yu cong 語叢 3 – in which we can relate different forms of punctuation marks to different hands. Even though this shows that the punctuation marks were added in the process of manuscript production and not by later users and that the particular form of the mark reflects the choice or habit of the scribe and not the copied model, the position of the marks can still have been taken over from the copied model, which would mean that the punctuation in the manuscript under discussion does not reflect the scribes’ interpretation of the text but someone else’s.32

The extent to which punctuation reflects anyone’s interpretation of the text or contains an arbitrary element also depends on the degree and nature of the literacy of the scribe. Some scribes may have understood little of the texts they copied. They may have been indifferent to their contents or entirely unable to understand them.

6 Form vis-à-vis Function

For an adequate description of punctuation in early Chinese manuscripts it is of particular importance to maintain a clear distinction between the form and the assumed function of punctuation devices. A mixing of these two levels of reference would not only confuse the functional relations of elements of language and its written representation but also ignore the fact that punctuation devices frequently have multiple functions. This applies to historical changes in punctuation practices just as much as to relatively consistent punctuation conventions used in one and the same community at one point in time. In modern English, for example, a sentence is frequently defined as a unit of language that begins with a capitalized word and ends with a period. At the same time the function of capitalization is defined as marking the beginning of a sentence and that of the period as concluding it, thus resulting in an epistemological circle that invalidates either definition.33

The evidence we have so far from early China indicates that no general punctuation conventions existed. Nevertheless, we can recognize relatively consistent practices in manuscripts that were apparently produced in the context of the same scribal school, i.e., by a group of people trained to follow the same writing conventions. Since we can as a rule neither assume that the manuscripts we study were punctuated according to consistent rules, nor do we even know who determined the punctuation in a manuscript, we need to explore its function on a case-by-case basis, carefully observing all phenomena related to punctuation without any preconceived idea about their function. In no case can we claim that a means of punctuation in a certain manuscript must have the same function as in another manuscript, unless there are strong arguments for the two manuscripts in question having followed common production standards.

7 Typical Functions of Punctuation

In order not to discuss the function of punctuation in an impressionistic manner or exclude possible functions that do not intuitively occur to the student of early manuscripts, it seems advisable to follow – and, whenever needed, adapt – an ordering scheme and terminology that has proven useful in existing scholarship.

Most Chinese studies of punctuation in early manuscripts approach the topic from the form of marks. Studies of individual manuscript corpora list the marks occurring in the respective manuscripts and then explain their function. Systematizing studies use different ordering principles, depending on the length of the study. While Zhang Xiancheng describes the function of each mark across various pre-imperial and early imperial manuscript corpora, Guan Xihua’s more extensive study uses a corpus-based approach within a chronological framework.34 The scope of this chapter does not permit to exhaustively discuss the punctuation in the vast number of early Chinese manuscripts known today. In order to still get a sense of the wider range of possible functions than would result from a mere discussion of the most typical marks, we will use the functional aspect as an ordering principle and then use examples from different manuscript corpora to show which punctuation devices can fulfill these functions.

Li Ling proposes a fourfold functional classification of punctuation devices. He distinguishes between (a) pianhao 篇號 (‘text/chapter marks’), (b) zhanghao 章號 (‘passage/subchapter marks’), (c) judou 句讀 (‘sentence and phrase marks’ or rather ‘periods and pauses’), as well as (d) chongwen hao 重文號 (‘repetition marks’) and hewen hao 合文號 (‘ligature marks’).35 Li Ling obviously did not aim to theorize punctuation and propose a comprehensive classification. For such a purpose his categorization – although it certainly names the most frequent functions of punctuation in early Chinese manuscripts – would be too general and at the same time not comprehensive enough.

Cheng Pengwan’s classification of punctuation devices by function is more comprehensive and more fine-grained, distinguishing between eight functions and a catch-all category of “others.”36 He starts not with marks indicating structural relations among elements of a text (such as modern comma, colon, semicolon, period), as a modern reader might expect, but with the more frequent signs found in early Chinese manuscripts that indicate different types of abbreviation: (1) chongwen fu 重文符 (‘repetition signs’), (2) hewen fu 合文符 (‘ligature signs’), and (3) shengdai fu 省代符 (‘substitution signs’). The term used for his category (4), tingdun hao 停頓號 (‘pause marks’), suggests the function of rhetorical punctuation, i.e., that of indicating pauses for breath in reading the text, but he also defines them as means of segmenting and includes examples of delimiters marking the end of textual units.37 Category (5), cengci hao 層次號 (‘marks for levels [of the text]’), refers to marks delimiting and indicating the hierarchical relationship of parts of the text.38 (6) tiji biaozhi 題記標識 / tishi fu 提示符 (‘title/subheading indicators’) mark delimitations of the text on the paratextual level, in particular titles. (7) goujiao fu 鈎校符 (‘check marks’) occur chiefly in inventories and (8) jiege fu 界隔符 (‘separator signs) separate different categories of entries in administrative documents. Cheng Pengwan’s final category (9) qita fuhao 其他符號 (‘other signs/marks’) comprises mostly deletion and other correction marks but is also open to special cases such as the black and red marks of different shapes in the Shanghai Museum *Zhou Yi 周易 manuscript.

It is a particular strength of Cheng’s typology of marks that it is consistently based on criteria of functionality and that it is quite comprehensive. Despite its comprehensiveness, however it still leaves out some important functions, such as that of distinguishers and internal abbreviation marks – the latter probably because they are perceived to be part of writing proper and exist on the level of the written text and not the metalevel of how it is punctuated. As discussed above, it is crucial not to restrict ourselves to non-“alphanumeric” (or perhaps 非漢字性) elements of writing but also include script features as well as layout and spacing in our considerations.

8 Categories of Punctuation in Early Chinese Manuscripts

To capture the dynamic phenomenon of punctuation in early Chinese manuscripts, I propose a typology based on functional criteria, inspired by Nunberg’s distinction of three types of “text category indicators” as the central functionality of punctuation concerned with segmentation of the text. Since Nunberg’s discussion is limited to the levels of paragraph and sentence, rather than the entirety of written texts, features relevant to the lower levels of the word, such as abbreviations, and features concerning the higher level of the complete text (e.g., placement of titles, counts of characters or text sections) need to be added to his framework.39

1. Abbreviators

1.1 repetition marks

1.2 ligature marks

1.3 internal abbreviation marks

2. Indicators

2.1 delimiters (initiators and terminators)

2.2 separators

2.3 distinguishers

2.4 declamation aids

3. Other paratextual means of defining textual identity

3.1 chapter/section/character counts

3.2 slip/column numbers

3.3 titles

3.4 corrections

This scheme is systematic in that it progresses from the smallest unit, the word, to the level of the text and finally the entire manuscript. Beginning with abbreviators (1) also fits the reality that these play a prominent role in early Chinese manuscripts. The function of repetition marks (1.1) is simple, almost trivial, if we consider just one character: A₌ obviously stands for AA. It gets complicated when several such marks are used in a row and A₌B₌ can be AABB or ABAB. Here, reader competence becomes a relevant question, especially the longer the chains of characters with repetition marks become and the more difficult the parsing of sentences or clauses and resulting patterns of reading get. This means that reading either must have been cumbersome and slow, and the economy of using this kind of abbreviation was deemed worth putting up with those disadvantages; or, more likely, readers were familiar enough with the texts or at least with certain patterns in the texts that the ease of reading was not much compromised. This is significant since it could tell us how much the written text was the sole source of access to it for its users or how much it served auxiliary functions in the textual culture of the time. The type of abbreviator that Cheng Pengwan lists as a separate category, shengdai fu 省代符 (‘substitution signs’), while I consider it as a special case of repetition mark, does expect the reader to recognize patterns of the text: The Mawangdui text *Tianwen qixiang za zhan 天文氣象雜占 (Miscellaneous prognostications of astronomy and meteorology) has a catalogue of six sentences, all beginning with the phrase shi wei 是謂; this manuscript writes the full phrase only the first time 是胃秆彗兵起有年, replacing the character for wei {} in the remaining five members of the catalogue (帚彗有內兵年大孰 etc.).40 This clear pattern may not have required much intellectual competence on the part of readers, but in other cases what the repetition mark stands for cannot be made mechanically, based on its position alone, but requires semantic consideration: in the Tsinghua manuscript *Ming xun 命訓, the phrase 大命世罰少命 cannot be read intuitively as 大命世罰少命命身, but the reader must recognize that the mark stands for the verb fa , even though it does not follow the noun ming directly in the preceding sentence: 大命世罰少命罰身.41

Ligatures (1.2) are much more unproblematic in this regard, as they are a matter of convention, making their recognition similar to that of any single character. The convention in itself is interesting, however, since it often does not seem to have anything to do with economy of space or writing but potentially reveals what elements of speech were considered as particularly closely related. Some ligatures do not involve any abbreviation of the characters but merely place them closer together without any spacing or seemingly merge them into the space of one character: 一日₌, 之日₌ or 之月₌; in others, the bottom horizontal of the first character merges with the top horizontal of the second, thus making the two characters one: ₌ for 上下, or in 之歲₌ the first character gets integrated in the top component of the second. Other ligatures use the coincidence that one of the characters is fully contained in the other, more complex one: is fully present in , so ₌ can represent 大夫, just as ₌ can stand for 並立 (竝立) or ₌ for 孔子. Other ligatures integrate one of the characters into the other by omitting a component of the “host character,” such as when writing 君子 by placing the in the space of the omitted component of . In such cases, no ambiguity hinders the reading of the text. There are some ligatures, however, that must have slowed down readers unfamiliar with the text, if not invited misunderstanding: ₌ (with in its old form written as on top of []) could be read as 之人 or 之先 or 先人, or theoretically as 先先, interpreting the mark as a repetition mark. The context would normally disambiguate such cases easily enough. Cases such as ₌, however, which offered the possibility of 之心, 之志 or 志心 (excluding the unlikely reduplication 志志), needed more clarity on the part of the reader about the content of the text. The degree of required reader competence would increase with combinations of different types of abbreviations, which sometimes included abbreviations of a character that could be signaled by an internal abbreviation mark (1.3).

That this mark has remained conspicuously absent from the discussion of punctuation may be due to a perception that it is a feature of how a character is written and therefore not relevant to punctuation.42 Indeed, these marks are different from repetition and ligature marks in that they are not written after the character – i.e., placed in the space between characters – but are integrated into the space of the character proper. However, this method of abbreviation is fundamentally different from abbreviated ways of writing a character by simplifying graphic components of the character (writing them in a cursive form) or simply omitting them, but it uses a standardized mark (i.e., two parallel horizontal strokes) to indicate the omission or simplification of graphic components. This is another example of the need to expand the concept of punctuation beyond just “non-alphanumeric” marks to include also script features as well as layout and spacing.

Attention to this feature can throw more light on palaeographic details. For example, the fact that the word qiang < *N-kaŋ {/} is written with a combination of the graphs and with an internal abbreviation mark and occasionally an additional underneath that mark (/), makes it easier to see that the internal abbreviation mark signals the fact that the phonetic component (*kaŋ) is represented in the reduced graphic form of . The failure to recognize this abbreviation has led to mistaken interpretations of the graph for the word qiang < *N-kaŋ {/}, such as Xu Shen’s identification of (*ɢʷˤəŋ) or He Linyi’s 何琳儀 identification of (*kʷəŋ) as its phonetic.43

Occasionally, the three forms of abbreviation – by repetition marks, ligature marks and internal abbreviation marks – can coincide with a degree of complexity that again raises the question of how much familiarity with the text may have been required of a reader to even decipher its written form accurately. Slip 9 of the Shanghai Museum manuscript *Gu cheng jia fu 姑成家父 has the passage 公恩亡告₌₌, which stands for 公恩亡告。告強門大夫。強門大夫曰…, requiring the reader to decipher a sequence A₌B₌C₌D₌ as: … A. ABCdD. BCdD … (with d=, D=).

Indicators (2), in particular delimiters, are what we usually have in mind when we talk about punctuation. Nunberg distinguishes between (a) “delimiters, which mark one or both ends of a member of a given category type,” (b) “distinguishers, which set off a piece of text from its surroundings [sic] text in virtue of some distinctive properties of its inscription,” and (c) “separators, which are inserted between elements of the same category type.”44 In early Chinese manuscripts indicators do not seem to play a prominent role. Warring States manuscripts typically have a greater number of abbreviators than indicator marks, which is not to say that the function of indicators is absent.

Delimiter (2.1) functions can be performed on the level of layout. Instead of marks indicating beginnings or ends of texts, the initiator (2.1.1) function can be performed by beginning a new part of a text on top of a new slip and leaving additional space after the end of a unit of text. As a general tendency, it seems that the terminator (2.1.2) function plays a greater role already in the Warring States, occasionally performed by marks or by spacing or a combination of both. Some texts already distinguish between section terminators within a text and text terminators of a different shape at the end of the entire text. The terminator function can also be performed by paratextual elements such as placing a section count or title at the end of the text. Initiators become more relevant in documents of greater length, since they are more helpful to the reader in navigating the text. As initiators are typically placed near the top of the document, they are easier to find for a reader than terminators that can occur anywhere in the column, depending on where the text happens to end. Separators (2.2) are used to distinguish units of text that do not form a coherent text in the same way as paragraphs or sections or a longer text do. For example, a mere compilation of loosely connected or entirely unconnected proverbs, maxims, or didactic precepts can keep these individual items separate either on the level of layout through column breaks or by inserting dots between the textual units in the middle of the columns (see for example the Guodian manuscripts *Yu cong 語叢 or the Mawangdui text titled Cheng ).

Distinguishers (2.3) are a good example of how functions of punctuation can be performed by marks or layout or script features: On slip 90 of the Warring States Chu manuscripts from Baoshan 包山, a small dot is placed as a distinguisher mark behind the characters gong qiu 龔㤹 and gong you 龔酉, signaling that they are personal names. The Shanghai Museum Chu manuscript *Kongzi shi lun 孔子詩論 uses two long horizontal strokes on the bottom of the character ming to mark the nominal use of the word. The mark is similar in appearance and placement to internal abbreviation marks, so it is not clear whether we should understand it as a different way of writing the character or as an externally placed mark behind it. The important point to realize is that such categorical distinctions probably played no role in the perception of producers and users of the manuscripts in early China. A clear case of the distinguisher function being performed by variation of the character itself is the distinction of the word wen in its regular sense of ‘pattern’ from the name of the Zhou king by integrating an additional component in the character, reserving the unchanged form of the character for the Zhou king.

The distinguisher function can also be performed by script features without altering the structure of the character but writing it in a more elaborate, less cursive way than the style employed in the text. This is true of the Han manuscript Yin shu 引書 from tomb 247 of Zhangjiashan 張家山, where the character yin in the title on the verso of slip 1 is more elaborate than that of the same character in the text, and of the Han manuscript Shen wu fu 神烏賦 from tomb 6 of Yinwan 尹灣, where the character wu in the title (written on slip 132) is noticeably less cursive than the same character inside the text. In both manuscripts the titles are also distinguished from the text by their larger character size. One important aspect of recognizing that peculiarities of script may be intentionally introduced to perform a distinguisher function is that a failure to consider this possibility may lead to mistaken assumptions about different scribal hands and ideas about the mode of manuscript production that follow from these assumptions.

Elongated finishing vertical strokes of characters, as they become typical of the “clerical script” (li shu 隸書) in the early imperial period also ought to be considered as potential distinguishers. Most often this feature might just be a matter of fashion or any other kind of aesthetic preference, but it is at least possible that in cases such as , the character for ling, ‘command’, the exaggerated length of the vertical stroke was meant to underscore authority. In some Mawangdui manuscripts, the sentence final ye is written with the final stroke elongated vertically, instead of curved to the right, ending in a horizontal direction. At the end of a textual unit this may be a combination of distinguisher and terminator function. That in some sentences the last “full word” before the sentence final particle is distinguished in the same way may point to a possible function in aid of declamation of the text. More evidence is needed to establish such a function.

Declamation aids (2.4) can be shown to have been used, however, in texts from the Warring States into the early imperial period. Typically, a small black hook-shaped mark close to the right edge of the slip or column in silk manuscript to alert a reciter to a potential stumbling block in the text or to a point where the reciter was expected to underscore (phonetic or metrical) prosodic features. The exact function of such marks cannot always be established with certainty, but the most compelling cases are probably those where a mark is placed immediately before the concluding sentences of a catalogue or before a rhyme change, alerting the reciter of a text that a special feature lies ahead that needs attention.45 The majority of declamation aids seems to mark rhetorical units of speech (periods) in a way best described by Cheng Pengwan’s term tingdun hao 停頓號 (‘pause marks’).

If abbreviators – pertaining as they do to the level of the word rather than larger textual units – may be considered a marginal part of punctuation, whose core function is typically seen in the segmentation of the text to facilitate reading, the discussion of distinguishers departs from the core area of punctuation in the opposite direction – that of the entire document and its paratextual features, reaching from chapter, section, or character counts to slip numbers to titles. Such features until recently appeared to be rare in Warring States manuscripts and to get developed with increasing complexity and sophistication only in the early empire, as texts were combined either in the same physical object or in separate manuscripts that were understood to belong together to form what in modern parlance would be called a “book”. In recent years, however, more cases of multi-text manuscripts have emerged from the Warring States period, proving that the emergence of “books” in this sense is not just a linear development over time but relates also to the different status of different types of manuscripts intended for different uses.

The brief discussion of how the functions that punctuation can perform transition from the area of “non-alphanumeric” marks to layout to features of the script itself shows that the synchronic diversity and diachronic changes of writing practices make it imperative that we apply a broad concept of punctuation in our study of pre-modern documents. A narrow focus on marks alone would obscure too much of the actual developments in early Chinese textual culture.

References

  • Alleton, Viviane, L écriture chinoise (Paris: Presses Universitaires de France, 1970).

  • Altman, Rochelle, Absent Voices: The Story of Writing Systems in the West (New Castle, Delaware: Oak Knoll Press, 2004).

  • Baker, Nicholson, “The History of Punctuation,” in Baker, Nicholson, The Size of Thoughts (New York: Random House, 1996), 7088.

    • Search Google Scholar
    • Export Citation
  • Behr, Wolfgang, and Bernhard Führer, “Einführende Notizen zum Lesen in China mit besonderer Berücksichtigung der Frühzeit,” in Bernhard Führer ed., Aspekte des Lesens in China in Vergangenheit und Gegenwart (Bochum: Projekt Verlag, 2005), 142.

    • Search Google Scholar
    • Export Citation
  • Bischoff, Bernhard, transl. Dáibhí Ó Cróinín and David Ganz, Latin Palaeography: Antiquity and the Middle Ages (Cambridge: Cambridge University Press, 2001). (German original: Paläographie des römischen Altertums und des abendländischen Mittelalters [Berlin: Erich Schmidt, 1979]).

    • Search Google Scholar
    • Export Citation
  • Bodde, Derk, “Punctuation: Its Use in China and Elsewhere,” Rocznik Orientalistyczny 47.2 (1991), 1523.

  • Bosley, Richard, “The Emergence of Concepts of a Sentence in Ancient Greek and in Ancient Chinese Philosophy,” Journal of Chinese Philosophy 24.2 (1997), 209229.

    • Search Google Scholar
    • Export Citation
  • Brown, Julian auth., Bately, Michelle P. Brown, and Jane Roberts ed., A Palaeographer’s View: The Selected Writings of Julian Brown (London: Harvey Miller Publishers, 1993).

    • Search Google Scholar
    • Export Citation
  • Chen Mengjia 陳夢家, Han jian zhuishu 漢簡綴述 (Beijing: Zhonghua, 1980).

  • Chen Wei 陳偉, Baoshan Chu jian chutan 包山楚簡初探 (Wuhan: Wuhan daxue, 1996).

  • Cheng Pengwan 程鵬萬, Jiandu boshu geshi yanjiu 簡牘帛書格式研究 (Shanghai: Shanghai guji, 2017).

  • Clemoes, Peter, Liturgical Influence on Punctuation in Late Old English and Early Middle English Manuscripts, Occasional Papers: Number 1 printed for The Department of Anglo-Saxon, Cambridge, 1952. Reprinted in: Old English Newsletter Subsidia 4 (New York: CEMERS, SUNY-Binghamton, 1980).

    • Search Google Scholar
    • Export Citation
  • Derolez, Albert, The Palaeography of Gothic Manuscript Books: From the Twelfth to the Early Fifteenth Century (Cambridge: Cambridge University Press, 2003).

    • Search Google Scholar
    • Export Citation
  • Fukuda Tetsuji 福田哲之, “Ziti fenxi zai chutu wenxian yanjiu shang de yiyi: yi Guodian Chu jian ‘Yu cong san’ wei zhongxin字體分析在出土文獻研究上的意義:以郭店楚簡《語叢三》為中心, in Ye Guoliang 葉國良, Zheng Jixiong 鄭吉雄, and Xu Fuchang 徐富昌 ed., Chutu wenxian yanjiu fangfa lunwenji 出土文獻研究 方法 (Taibei: Taida, 2005), 189204.

    • Search Google Scholar
    • Export Citation
  • Galambos, Imre, “Punctuation Marks in Medieval Chinese Manuscripts,” in Sobisch, Jan-Ulrich, and Jörg B. Quenzer ed., Manuscript Cultures: Mapping the Field (Berlin: De Gruyter, 2014), 341357.

    • Search Google Scholar
    • Export Citation
  • Graham-White, Anthony, Punctuation and Its Dramatic Value in Shakespearean Drama (Newark: University of Delaware Press, 1995).

  • Guan Xihua 管錫華, Zhongguo gudai biaodian fuhao fazhan shi 中國古代標點符號發展史 (Chengdu: Ba Shu, 2002).

  • Guo Pan 郭攀 and Xia Fengmei 夏鳳梅, “Zhonguo gudai yuyan xue ‘ju’ gainian de yanjin中國古代語言學“句”概念的演進, Gu Hanyu yanjiu 古漢語研究 3 (2009), 3542.

    • Search Google Scholar
    • Export Citation
  • Hall, Nigel, and Anne Robinson eds., Learning About Punctuation (Portsmouth, NH: Heinemann, 1996).

  • Hansen, Chad, “Chinese Language, Chinese Philosophy, and ‘Truth’,” Journal of Asian Studies 44.3 (1985), 491519.

  • Harbsmeier, Christoph, Science and Civilisation in China, Vol. 7, Part I: Language and Logic (Cambridge: Cambridge University Press, 1998).

    • Search Google Scholar
    • Export Citation
  • He Linyi 何琳儀, Zhanguo guwen zidian: Zhanguo wenzi shengxi 戰國古文字典:戰國文字聲系, 2 vols. (Beijing: Zhonghua, 1998).

    • Search Google Scholar
    • Export Citation
  • Kennedy, Maev, “Pride, prejudice and poor punctuation,” The Guardian 23 October 2010, accessed January 3, 2011, http://www.guardian.co.uk/books/2010/oct/23/jane-austen-poor-punctuation-kathryn-sutherland.

    • Search Google Scholar
    • Export Citation
  • Kerouac, Jack auth., Howard Cunnell ed., On the Road: The Original Scroll (New York: Penguin, 2007).

  • Levinson, Joan Persily, Punctuation and the Orthographic Sentence: A Linguistic Analysis (PhD dissertation (Microfilm), The City University of New York, 1985).

  • Li Ling 李零, Jianbo gushu yu xueshu yuanliu 簡帛古書與學術源流 (Beijing: San lian, 2004).

  • Li Ling 李零, Jianbo gushu yu xueshu yuanliu (xiuding ben) 簡帛古書與學術源流 (修訂本) (Beijing: San lian, 2008).

  • Lin Suqing 林素清, Zhanguo wenzi yanjiu戰國文字研究 (Ph.D. diss., Taiwan daxue, 1984).

  • McDermott, Joseph P., A Social History of the Chinese Book: Books and Literati Culture in Late Imperial China (Hong Kong: Hong Kong University Press, 2006).

    • Search Google Scholar
    • Export Citation
  • Meng Yuelong 孟躍龍, “Qinghua jian Ming xun ‘Shao ming = shen’ de dufa: Jianlun gudai chaoben wenxian zhong chongwen fuhao de teshu yongfa清華簡《命訓》“少命=身”的讀法:兼論古代抄本文獻中重文符號的特殊用法, Jianbo 簡帛 13 (2017), 7177.

    • Search Google Scholar
    • Export Citation
  • Nehamas, Alexander and Paul Woodruff trans., Plato: Phaedrus (Indianapolis: Hackett, 1995).

  • Nunberg, Geoffrey, The Linguistics of Punctuation (Stanford: Center for the Study of Language and Information, 1990).

  • Parkes, Malcolm B., Pause and Effect: an introduction to the history of punctuation in the West (Berkeley: University of California Press, 1993).

    • Search Google Scholar
    • Export Citation
  • Richter, Matthias, “Textual Identity and the Role of Literacy in the Transmission of Early Chinese Literature,” in Li Feng and David Branner ed., Writing and Literacy in Early China: Papers from the Columbia Early China Seminar (Seattle: University of Washington Press, 2011), 206236.

    • Search Google Scholar
    • Export Citation
  • Richter, Matthias, “Punctuation, Premodern,” in Rint Sybesma ed., Encyclopedia of Chinese Language and Linguistics (Leiden: Brill, 2015).

    • Search Google Scholar
    • Export Citation
  • Shaughnessy, Edward L., “Varieties of Textual Variants: Evidence from the Tsinghua Bamboo-Strip *Ming xun Manuscript,” Early China 39 (2016), 111144.

    • Search Google Scholar
    • Export Citation
  • Shaughnessy, Edward L., “To Punish the Person: A Reading Note Regarding a Punctuation Mark in the Tsinghua Manuscript *Ming xun,” Early China 40 (2017), 303309.

    • Search Google Scholar
    • Export Citation
  • Truss, Lynn, Eats, Shoots & Leaves: The Zero Tolerance Approach to Punctuation (London: Profile Books, 2003).

  • Wei Yihui 魏宜輝, “Zailun Mawangdui boshu zhong de ‘shi=’ ju再論馬王堆帛書中的 “是=”句, Dongnan wenhua 東南文化 2008.4, 5657.

    • Search Google Scholar
    • Export Citation
  • Wingo, E[lvis] Otha, Latin Punctuation in the Classical Age (The Hague: Mouton, 1972).

  • Xiao Yunxiao 肖芸曉, “Shilun Qinghua jian shushou de zhi yu neng試論清華簡書手的職與能, Jianbo 簡帛 25 (2022), 6785.

    • Search Google Scholar
    • Export Citation
  • Yan Kejun 嚴可均 (1762–1843) comp., Quan shanggu Sandai Qin Han Sanguo Liuchao wen 全上古三代秦漢三國六朝文 (Beijing: Zhonghua. 1958 [1836]).

    • Search Google Scholar
    • Export Citation
  • Zhang Xiancheng 張顯成, Jianbo wenxian xue tonglun 簡帛文獻學通論 (Beijing: Zhonghua, 2004).

  • Zieme, Peter, “Liaoye – a Chinese Ligature in Uigur Manuscripts from the 13th and 14th Centuries,” in Michael Friedrich and Jörg Quenzer ed., Manuscript Cultures – Newsletter No. 2 (Hamburg: Research Group Manuscript Cultures in Asia and Africa, 2009), 1012.

    • Search Google Scholar
    • Export Citation
1

“The nine basic marks of punctuation – comma, dash, hyphen, period, parenthesis, semi- colon, colon, space, and capital letter – seem so apt to us now, so pipe-smokingly Indo- European, so naturally suited in their disjunctive charge and mass to their given sentential offices, that we may forgivably assume that commas have been around for at least as long as electrons, and that while dialects, cursive styles, and typefaces have come and gone, the semi-colon, that supremely self-possessed valet of phraseology, is immutable.” Nicholson Baker, “The History of Punctuation,” in Nicholson Baker, The Size of Thoughts (New York: Random House, 1996), 70.

2

Only in exceptional cases, such as theater plays, liturgical texts, or the scores for vocal music are these situational or prosodic elements verbalized (e.g., in stage directions) or represented by extralinguistic signs.

3

“[W]riting shares a strange feature with painting. The offsprings of painting stand there as if they are alive, but if anyone asks them anything, they remain most solemnly silent. The same is true of written words. You’d think they were speaking as if they had some understanding, but if you question anything that has been said, because you want to learn more, it continues to signify just that very same thing forever. When it has once been written down, every discourse rolls about everywhere, reaching indiscriminately those with understanding no less than those who have no business with it, and it doesn’t know to whom it should speak and to whom it should not. And when it is faulted and attacked unfairly, it always needs its father’s support; alone, it can neither defend itself nor come to its own support.” Phaidros 275d–e, quoted after Alexander Nehamas and Paul Woodruff trans., Plato: Phaedrus (Indianapolis: Hackett, 1995). 80–81.

4

Linguistic changes and changed cultural circumstances – new production techniques and materials – of course led to changes also in the material features of written texts. Yet, some features have an astonishingly long history, especially if they are based on constant anthropological factors: Rochelle Altman traces the modern US ‘legal size’ paper format of 8-1/2 by 14 inches back to 2nd millennium BCE Akkadian clay tablets. That the width of 8–9 inches remained so constant over many centuries is certainly due to the fact that it “is the maximum width for the scanning human eye to process and comprehend data,” but the durability of the 14 inches height, originally resulting from the fact that clay tablets of that size had a weight convenient to lift for reading, must result from a semantic value that has become firmly attached to this format. See Rochelle Altman, Absent Voices: The Story of Writing Systems in the West (New Castle, Delaware: Oak Knoll Press, 2004), 1–4.

5

The first known systematic effort to order the Chinese literary heritage was begun in the first century BCE. Experts in various areas of knowledge reconstructed pre-imperial texts by collating, sifting and rearranging the manuscripts that had been collected for the Imperial Library from all over the empire. The librarians Liu Xiang 劉向 (79–8 BCE) and his son Liu Xin 劉歆 (46 BCE–23 CE) left behind accounts of their work, which have in turn been reconstructed from fragments by the Qing scholar Yan Kejun 嚴可均 (1762–1843). See Yan Kejun 嚴可均 comp., Quan shanggu Sandai Qin Han Sanguo Liuchao wen 全上古三代秦漢三國六朝文 (Beijing: Zhonghua. 1958 [1836]), 38:336a–339b (for Liu Xiang’s Bie lu 別錄); 41:351b–353a (for Liu Xie’s Qi lüe 七略).

6

Nigel Hall and Anne Robinson eds., Learning About Punctuation (Portsmouth, NH: Heinemann, 1996), 8.

7

“[P]unctuation […] had it been perfected and popularized, would have substantially lessened the difficulties of handling the Chinese written language. As we shall see, a crude proto-punctuation can be found in certain Chinese texts going back as early as the third century BC, yet until recent times punctuation in China remained under-developed, sporadically used, and often looked down upon as a vulgar device unworthy of the true scholar.” See Derk Bodde, “Punctuation: Its Use in China and Elsewhere,” Rocznik Orientalistyczny 47.2 (1991), 15. I thank Wolfgang Behr for kindly providing me with a copy of Bodde’s article.

8

Imre Galambos, “Punctuation Marks in Medieval Chinese Manuscripts,” in Jan-Ulrich Sobisch and Jörg B. Quenzer ed., Manuscript Cultures: Mapping the Field (Berlin: De Gruyter, 2014), 341.

9

For economics of print culture, especially with regard to use of space, see Joseph P. McDermott, A Social History of the Chinese Book: Books and Literati Culture in Late Imperial China (Hong Kong: Hong Kong University Press, 2006).

10

Viviane Alleton, L écriture chinoise (Paris: Presses Universitaires de France, 1970), 61.

11

See Chad Hansen, “Chinese Language, Chinese Philosophy, and ‘Truth’,” Journal of Asian Studies 44.3 (1985), 491–519 (quoted phrase from p. 516). For further discussions of this issue, see also Richard Bosley, “The Emergence of Concepts of a Sentence in Ancient Greek and in Ancient Chinese Philosophy,” Journal of Chinese Philosophy 24.2 (1997), 209–229, as well as Guo Pan 郭攀 and Xia Fengmei 夏鳳梅, “Zhonguo gudai yuyan xue ‘ju’ gainian de yanjin” 中國古代語言學“句”概念的演進, Gu Hanyu yanjiu 古漢語研究 3 (2009), 35–42.

12

Bodde, “Punctuation: Its Use in China and Elsewhere,” 23. See also Christoph Harbsmeier, Science and Civilisation in China, Vol. 7, Part I: Language and Logic (Cambridge: Cambridge University Press, 1998), 178–181. A concise review of this debate is given by Wolfgang Behr and Bernhard Führer, “Einführende Notizen zum Lesen in China mit besonderer Berücksichtigung der Frühzeit,” in Bernhard Führer ed., Aspekte des Lesens in China in Vergangenheit und Gegenwart (Bochum: Projekt Verlag, 2005), 33–42.

13

Julian Brown auth., Bately, Michelle P. Brown, and Jane Roberts ed., A Palaeographer’s View: The Selected Writings of Julian Brown (London: Harvey Miller Publishers, 1993), 79.

14

Geoffrey Nunberg, The Linguistics of Punctuation (Stanford: Center for the Study of Language and Information, 1990), 17.

15

The practice of word division is very ancient – attested already in cuneiform writing. It is remarkable that in both Greek and Latin, albeit at different times, such a solidly established practice was given up for some time in favor of scriptio continua. Elvis Otha Wingo, Latin Punctuation in the Classical Age (The Hague: Mouton, 1972), 14–17.

16

See Malcolm B. Parkes, Pause and Effect: an introduction to the history of punctuation in the West (Berkeley: University of California Press, 1993), 43, 202–207, and 305.

17

See Peter Zieme, “Liaoye – a Chinese Ligature in Uigur Manuscripts from the 13th and 14th Centuries,” in Michael Friedrich and Jörg Quenzer ed., Manuscript Cultures – Newsletter No. 2 (Hamburg: Research Group Manuscript Cultures in Asia and Africa, 2009), 10–12.

18

Chen Wei prefers the more neutral term biaozhi fuhao 標識符識 (literally: ‘signs and symbols’), apparently because the term biaodian fuhao suggests the misleading connotation of the later practice of adding circular marks (dian ) to unpunctuated printed texts. Cf. Chen Wei 陳偉, Baoshan Chu jian chutan 包山楚簡初探 (Wuhan: Wuhan daxue, 1996), 22–28, as well as later publications of the same author. Chen Mengjia likewise avoids the anachronistic biaodian fuhao by using the even simpler biaohao 標號 (‘marks’). See Chen Mengjia 陳夢家, Han jian zhuishu 漢簡綴述 (Beijing: Zhonghua, 1980), 308–09.

19

Cf. Zhang Xiancheng 張顯成, Jianbo wenxian xue tonglun 簡帛文獻學通論 (Beijing: Zhonghua, 2004), 165–220.

20

The present article will use the term “literature” in a broad sense, not just for the narrative, but also for argumentative and didactic manuscript texts of a more pragmatic nature. My distinction between literature and administrative documents largely follows the customary distinction between shuji 書籍 and wenshu 文書 in Chinese scholarship on early manuscripts.

21

See Parkes, Pause and Effect, 3–4.

22

Hall and Robinson, Learning About Punctuation, 9. See also Bernhard Bischoff, transl. Ó Cróinín and David Ganz, Latin Palaeography: Antiquity and the Middle Ages (Cambridge: Cambridge University Press, 2001), 169–178. (German original: Paläographie des römischen Altertums und des abendländischen Mittelalters [Berlin: Erich Schmidt, 1979]).

23

Albert Derolez, The Palaeography of Gothic Manuscript Books: From the Twelfth to the Early Fifteenth Century (Cambridge: Cambridge University Press, 2003), 185.

24

Anthony Graham-White, Punctuation and Its Dramatic Value in Shakespearean Drama (Newark: University of Delaware Press, 1995), 46.

25

This development is explained in detail in Peter Clemoes, Liturgical Influence on Punctuation in Late Old English and Early Middle English Manuscripts, Occasional Papers: Number 1 printed for The Department of Anglo-Saxon, Cambridge, 1952. Reprinted in: Old English Newsletter Subsidia 4 (New York: CEMERS, SUNY-Binghamton, 1980).

26

“There is little evidence before the sixth century that guides to phrasing – punctuation – originated with the author. […] If authors supplied punctuation to a text it was as readers not writers. Because the work of scribes or amanuenses was ‘mechanical’, they confined themselves to reproducing as faithfully as possible what had been transmitted to them without any further interpretation; hence they did not supply punctuation to a text.” (Parkes, Pause and Effect, 9.) Even in relatively modern times authors had different attitudes towards punctuating their own texts. Byron, for example, refused to do his own final punctuation (cf. Hall and Robinson, Learning About Punctuation, 8) and, while “Charles Dickens when correcting proofs […] paid meticulous attention to punctuation, […] Jean Jacques Rousseau, William Wordsworth and Charlotte Brontë all asked their respective publishers to correct the punctuation in their manuscripts.” Parkes, Pause and Effect, 5.

27

In some cases, modern scholarship endeavors to recover the original style in which authors wrote before their texts underwent editorial grooming by their publishers. In the case of Jane Austen, whose prose is punctuated in a highly elaborate fashion, the British scholar Kathryn Sutherland holds that the highly irregular punctuation she finds in the author’s manuscripts “makes her more interesting, and a much more modern and innovative writer than had been thought. In particular, her use of dashes to heighten the emotional impact of what she is writing is striking: you have to wait for Virginia Woolf to see anything comparable.” See Kennedy, Maev, “Pride, prejudice and poor punctuation,” The Guardian 23 October 2010, accessed January 3, 2011, http://www.guardian.co.uk/books/2010/oct/23/jane-austen-poor-punctuation-kathryn-sutherland. For Jack Kerouac’s On the Road it took half a century until it was first printed in its original punctuation, advertised on the book cover as “rougher, wilder, and racier than the 1957 edition.” See Jack Kerouac auth., Howard Cunnell ed., On the Road: The Original Scroll (New York: Penguin, 2007).

28

If the spacing between characters in a manuscript is highly regular and the spacing is wider where marks occur, it is fairly safe to assume the marks were added in the process of writing rather than afterwards.

29

It is often observed that writers add punctuation after writing a certain portion of text. See Hall & Robinson, Learning About Punctuation, 35.

30

See Parkes, Pause and Effect, 5.

31

A prominent example of an ever-increasing amount of punctuation in the same texts are the plays of William Shakespeare. However, Graham-White shows that, although the punctuation in earlier versions of Shakespeare’s plays may have little to do with authorial intentions, a replacement of old punctuation by modern principles also obliterates traces of earlier understandings and performances of the text. See Graham-White, Punctuation and Its Dramatic Value in Shakespearean Drama. See also the anecdote about a director of Hamlet stripping the text of most of its punctuation before rehearsals, related in Lynn Truss, Eats, Shoots & Leaves: The Zero Tolerance Approach to Punctuation (London: Profile Books, 2003), 12.

32

Fukuda Tetsuji 福田哲之, “Ziti fenxi zai chutu wenxian yanjiu shang de yiyi: yi Guodian Chu jian ‘Yu cong san’ wei zhongxin” 字體分析在出土文獻研究上的意義:以郭店楚簡《語叢三》為中心, in Ye Guoliang 葉國良, Zheng Jixiong 鄭吉雄, and Xu Fuchang 徐富昌 ed., Chutu wenxian yanjiu fangfa lunwenji 出土文獻研究方法 (Taibei: Taida, 2005), 189–204. Xiao Yunxiao’s more recent, systematic study of the Warring States manuscripts of the Tsinghua University collection uses a much broader basis of analysis, which allows her to establish consistent punctuation practices in manuscripts that also share identical or highly similar features of handwriting, suggesting that groups of scribes made their own decisions about punctuation. See Xiao Yunxiao 肖芸曉, “Shilun Qinghua jian shushou de zhi yu neng” 試論清華簡書手的職與能, Jianbo 簡帛 25 (2022), 67–85.

33

Cf. the discussion in Joan Persily Levinson, Punctuation and the Orthographic Sentence: A Linguistic Analysis (PhD dissertation (Microfilm), The City University of New York, 1985), 6–16. Sentences can be defined differently, e.g., “syntactically (as a group of words ‘that contains a subject and a predicate’); or prosodically (as a group of words ‘that can be uttered by itself’ […]); or semantically (as a group of words ‘that expresses a proposition’ […]).” (Nunberg, The Linguistics of Punctuation, 21–22.) As for the latter, capitalization and periods can have different functions: e.g., the former also serves to mark off proper nouns and the latter to signal abbreviations. Geoffrey Nunberg (The Linguistics of Punctuation, 17–23) avoids this confusion between language and its written representation by distinguishing between “lexical grammar” to refer to the former vis-à-vis “text grammar” for written language.

34

See Zhang Xiancheng, Jianbo wenxian xue tonglun, 179–214, and Guan Xihua 管錫華, Zhongguo gudai biaodian fuhao fazhan shi 中國古代標點符號發展史 (Chengdu: Ba Shu, 2002). Neither of these authors is entirely consistent in using the same criterion after which they name punctuation marks. Although most terms for punctuation marks are based on form (e.g., da yuandian 大圓點 [‘large round dot’] or chang cu hengxian 長粗橫綫 [‘long thick horizontal line’]), some refer to the assumed function (e.g., jieshu hao 結束號 [‘terminator mark’]), some to both (‘queren hao 確認號 [‘ check-mark’]).

35

See Li Ling 李零, Jianbo gushu yu xueshu yuanliu (xiuding ben) 簡帛古書與學術源流(修訂本) (Beijing: San lian, 2008), 131–132 (and Li Ling 李零, Jianbo gushu yu xueshu yuanliu 簡帛古書與學術源流 (Beijing: San lian, 2004), 121–122).

36

See Cheng Pengwan 程鵬萬, Jiandu boshu geshi yanjiu 簡牘帛書格式研究 (Shanghai: Shanghai guji, 2017), 178–230.

37

He distinguishes a number of marks that can perform this function by their shape: 1. xiao duan heng 小短橫 (short dash), 2. gou zhi 鈎識 (hook mark), 3. xiao mo fangkuai 小墨方塊 (small ink/black square), 4. mo heng 墨橫 (ink/black horizontal stroke), 5. shixin yuandian 實心圓點 (filled circle), 6. kongxin yuandian 空心圓點 (hollow circle), 7. qita tingdun fuhao 其他停頓符號 (other pause signs). See Cheng Pengwan, Jiandu boshu geshi yanjiu, 188–197.

38

See Cheng Pengwan, Jiandu boshu geshi yanjiu, 178–230.

39

I have proposed the same categories in a brief lexicon entry, Matthias Richter, “Punctuation, Premodern,” in Rint Sybesma et al. ed., Encyclopedia of Chinese Language and Linguistics (Leiden: Brill, 2015).

40

Wei Yihui 魏宜輝, “Zailun Mawangdui boshu zhong de ‘shi=’ ju” 再論馬王堆帛書中的  “是=”句, Dongnan wenhua 東南文化 2008.4, 56–57.

41

For a discussion of this case, see Edward L. Shaughnessy, “To Punish the Person: A Reading Note Regarding a Punctuation Mark in the Tsinghua Manuscript *Ming xun,” Early China 40 (2017), 303–309; in which he corrects his earlier reading in Edward L. Shaughnessy, “Varieties of Textual Variants: Evidence from the Tsinghua Bamboo-Strip *Ming xun Manuscript,” Early China 39 (2016), 138. See also Meng Yuelong 孟躍龍, “Qinghua jian Ming xun ‘Shao ming = shen’ de dufa: Jianlun gudai chaoben wenxian zhong chongwen fuhao de teshu yongfa” 清華簡《命訓》“少命=身”的讀法:兼論古代抄本文獻中重文符號的特殊用法, Jianbo 簡帛 13 (2017), 71–77.

42

Internal abbreviation marks were mentioned and discussed by Lin Suqing as early as four decades ago. See Lin Suqing 林素清, Zhanguo wenzi yanjiu 戰國文字研究 (Ph.D. diss., Taiwan daxue, 1984).

43

See He Linyi 何琳儀, Zhanguo guwen zidian: Zhanguo wenzi shengxi 戰國古文字典:戰國文字聲系, 2 vols. (Beijing: Zhonghua, 1998), 647.

44

Nunberg, The Linguistics of Punctuation, 52–53.

45

For examples, see Matthias Richter, “Textual Identity and the Role of Literacy in the Transmission of Early Chinese Literature,” in Li Feng and David Branner ed., Writing and Literacy in Early China: Papers from the Columbia Early China Seminar (Seattle: University of Washington Press, 2011), 223–231.

Content Metrics

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 185 185 12
PDF Views & Downloads 258 258 22