Phrase-based alignment of classical Chinese and English

Paper presented at Greek and Latin in an age of Open Data:

Phrase-based alignment of classical Chinese and English
Donald Sturgeon and John S. Y. Lee

Abstract

Aligned parallel corpora are useful for a variety of purposes including machine translation and statistical studies, as well as making possible new and innovative digital tools for use in pedagogy and research. Alignments can be made at various levels of granularity, a common type being alignment of sentences. In the case of classical Chinese in particular, databases containing such alignments are also of direct utility to scholars and linguists due to the complex semantics of individual terms of the language, the limited size of the extant body of writing, and a lack of sufficiently comprehensive bilingual dictionaries. Aligned corpora make possible automated extraction of relevant linguistic data for arbitrary terms, while avoiding the prohibitively high cost involved in manual construction of an adequate bilingual dictionary.

While in many modern languages sentences are delimited in the written form by the presence of certain punctuation marks, classical Chinese was for many centuries written without any punctuation marks whatsoever, and later with punctuation that delimited only boundaries between phrases. Modern editions of classical Chinese texts include punctuation marks corresponding closely to (and greatly influenced by) modern English punctuation, but often disagree on the precise details of such punctuation, highlighting the degree of freedom present in adding such marks. Due to the grammar of classical Chinese, this freedom often extends to choices determining apparent sentence boundaries. Similarly and partly as a result of this, English translations of these texts often differ in the precise delimiting of sentences in the source text.

As a result of these linguistic and historical factors, sentence-based alignment of classical Chinese texts and their modern translations is problematic, as sentences of the source and target languages often fail to correspond exactly due to different choices made in punctuating the text, even where these do not correspond to significant differences in interpretation. By contrast however due to the much lower degree of freedom involved, different modern editions of early texts exhibit much less disagreement regarding the delimiting of phrases.

Motivated by these factors, this study investigates automated phrase-wise alignment of a corpus of classical Chinese texts and their English translations, comparing unsupervised machine-generated phrase-wise alignments versus sentence-wise alignments by means of human annotated results.

Download the full paper.

Posted in Chinese, Digital Humanities, Talks and conference papers | Comments Off

Chinese Text Project – Dictionary update

Update to the CTP:

The dictionary section of the site has been updated to make better use of English translations. Dictionary pages now cite English translations of example sentences together with the corresponding Chinese examples. Additionally, dictionary look-ups for passages of texts that have English translations now display these translations side by side with the Chinese text for easier comparison. If you prefer the old behaviour, please log in to your CTP account and change the “Dictionary display” setting to “No translations”.

Posted in Chinese, Digital Humanities | Comments Off

Knowledge and Language in Early Chinese Thought

An invited lecture given at the Taiwan Philosophical Association at Taiwan University.

Abstract

Early Chinese thinkers did not typically characterize knowledge in terms of sentential constructs nor consider these to be a fundamental constituent or theoretical foundation of knowledge. At the same time, relationships between language and knowledge were the subject of intense critical debate, in which thinkers recognized the possibility of such relations existing and the significance should they hold, but were in each case challenged by those skeptical of their generality. This paper will discuss early explorations of the relation between language and knowledge and attempt to explain why some Chinese thinkers came to be skeptical of the role that language might play in understanding and obtaining knowledge.

I shall begin by arguing that in classical Chinese there is an important linguistic disanalogy between knowledge, truth, and belief that would weigh strongly against attempting to account for knowledge in terms of sentential constructs. Instead, knowing was more typically thought of as consisting in objectively correct action and the correct use of words. Secondly, while those with a positive view of the role of language in explaining knowledge attempted to show that there are objective standards that govern the correct use of words, they found it difficult to fully account for the claimed objectivity and uniqueness of these standards. Thirdly, though early thinkers looked for strictly formal regularities in language, in doing so they made the discovery that language does not in fact follow such formal patterns. Finally, Daoists in particular suggested that words can be intentionally used in unconventional ways that force them to take on new and seemingly incompatible interpretations in different contexts, suggesting that words may not be used in accordance with any fixed objective standards at all.

Posted in Chinese, Philosophy, Talks and conference papers | Comments Off

Early Mohists and Logic

Presented at the History of Logic in China (HOLIC) 2014 conference.

Abstract

The extant text of the Mozi can be divided into several sections separated by quite clear divisions in both writing style and content. Of these, one section of six texts – often termed the “dialectical chapters (墨辯)” – stands out as focusing directly upon abstract issues in ethics, science, and logic. These chapters are generally thought to be of relatively late authorship, and their contents attributed to so-called “later Mohists”. By contrast, many of the core doctrines of Mohism itself are expounded in groups of essays that are likely of earlier authorship, often referred to as the “ten triads” or “core chapters”.

The later Mohist works include content that can be easily recognized as logical in character or as having clear logical significance, such as definitions of logical operators and discussions revolving around the correctness of groups of formally parallel linguistic expressions. While the earlier Mohist works by contrast do not contain such immediately obvious examples of logical theorizing, they are nonetheless of considerable utility in understanding Mohist logical thought as a whole: though diverging radically in content and style, the earlier and later works exhibit continuity regarding certain key ideas and interests. Crucially, in comparison to the notoriously terse, technical, and at times obscure writing style employed in many parts of the later Mohist works – which are in addition plagued by serious textual corruption and as a consequence often the subject of widespread scholarly disagreement regarding textual emendations and literal interpretation – the early Mohist works are written in a verbose and often repetitive style that is generally much less problematic to interpret. As a result, early Mohist writings may provide invaluable evidence about mainstream Mohist ideas and concepts that can usefully inform the interpretation of later Mohist logic, as well as suggest how the Mohists themselves may have come to develop and understand key ideas that we now identify as “logical”.

Posted in Philosophy, Talks and conference papers | Comments Off

Knowledge in Early Chinese Thought

The final version of my PhD thesis, titled Knowledge in Early Chinese Thought, is now available online.

Abstract

Early Chinese philosophical texts contain discussions of the nature, origins, and possibility of knowledge, in which both positive accounts and skeptical responses to them are couched in importantly different terms to those most familiar from similar discussions in Western philosophy. In place of appeals to truth, belief, and fallibility of the senses, action, discrimination, and difference of perspective play crucial roles. The aim of this dissertation is to explain why this should be so, and what consequences this had for the early Chinese understanding of knowledge.

In an attempt to answer these questions, I argue that, likely influenced by both facts about the classical Chinese language and key philosophical trends and interests of the time, discussions of knowledge by early Chinese thinkers generally referenced a broad notion of knowledge that was seen as being closely related to action. Linguistic factors also contributed to theorizing about knowledge focusing not on beliefs or other sentential structures, but rather on the drawing of action-guiding shi-fei distinctions, and the same shi-fei framework that was applied to perception was also applied to knowledge.

Language, understood most fundamentally in terms of an ability to distinguish shi-fei and apply names to things in the correct way, also played an important role in the pre‐Qin understanding of knowledge. On a linguistic level, knowledge corresponded to reliably correct language use, and rigid fa (法 standards, models) were seen as underwriting this by providing the standard of correctness. Just as these fa could be used to measure the correctness of individual terms, thinkers interested in the correctness of doctrines and speech in general attempted to apply the same idea to larger linguistic structures such as sentences, in the hope of finding fa for correct language use at a higher level. In doing so, they discovered facts about natural language use that could not be accounted for using the types of fa they considered.

Likely in part influenced by similar observations, others called into question the existence and uniqueness of standards in general and the adequacy of language in expressing knowledge. I argue that the prevailing positive view of knowledge ultimately gave rise to an interesting and nuanced form of skepticism grounded in a form of perspectivism. This skepticism does not merely have the negative consequence that we should question some of our knowledge commitments, but can also be used to suggest that – while still doubting – we can make practical use of our skepticism to improve our knowledge by considering a wider range of perspectives.

Posted in Philosophy | Comments Off

Chinese Text Project – New Wiki section

Update to the CTP:

The Wiki section of the site provides online browsing and full-text search for numerous texts not yet included in the textual database. Since some of these texts have not yet been adequately proofread, users are invited to help in the process of correcting these texts using a Wiki interface, and encouraged to upload historical Chinese texts not yet included. For more details, please see the instructions or browse the Wiki.

Posted in Digital Humanities | Comments Off

Chinese Text Project – Parallel passage database update

Update to the CTP:

Hundreds of thousands of new parallel passages have been added to the parallel passage database by means of an automated study, meaning that over half of all passages in the pre-Qin and Han section of the database now contain at least one parallel with some other text on the site (many of course containing far more). Parallels can be browsed by clicking the yellow parallel passage icon to the left of a paragraph, and can be searched using the Advanced search function.

In addition, the parallel passage display now provides a visual summary of text reuse within the selected paragraph. Parallel groups are indicated using shades of red; the stronger the shade of red a region is shaded, the more parallel groups it belongs to. Clicking on a shaded region jumps down to the corresponding list of parallels below. More details available here: ctext.org/tools/parallel-passages .

Posted in Digital Humanities | Comments Off

The Daoist Nazi Problem

The following is a guest post I originally made on the Warp Weft and Way group blog. Unfortunately the full text and discussion no longer seems to be available there; below is the post as it originally appeared.

The Daoist Nazi Problem

Suppose there is a person, or a group of people, committed to practicing what we can for convenience call a “Nazi Dao”: a Dao that, though practically successful from the perspective of its followers, involves commitment to some abhorrent practices that all “right-minded people” would condemn as exemplary immoral acts that should be universally condemned – “killing innocent babies for fun”, for example.

What can a Zhuangist – someone committed to a relativist position about differing practices and the nature of their justification, questioning of conventionally accepted values, and skeptical about certain kinds of knowledge – say about such a Dao? Can he condemn it? Is it a “bad” Dao, and if so in what sense? Or is it just as good a Dao as any other?

Before trying to imagine how a Zhuangist might try to respond, an obvious question to ask might be whether or not these are fair questions to ask of the text at all. Here it is worth pointing out that although the term “Nazi” itself is obviously alien to pre-Qin thought, the idea that a group of people might be committed to practicing what from our own perspective is not only a radically different but also distasteful, morally wrong and twisted set of practices while at the same time themselves seeing these practices as quite justifiable is certainly not – in fact, the Mohists used precisely this kind of thought experiment when arguing against taking conventional mores as a sound basis for morality (Mozi 39/25/75-78). The Zhuangzi itself explicitly questions intuitively plausible candidates for “universal” human values, such as the valuing of life over death or usefulness over uselessness, and seems more than willing to contemplate the possibility of radically different worldviews and the suggestion that the unexpected and unconventional may in fact be being unfairly maligned. In this sense, it does seem that there are legitimate questions here to answer – questions that in some ways are actually quite similar to those that the text explicitly considers itself.

The Zhuangzi, like the Mozi, doesn’t see convention as a useful basis for morality. But unlike many of its contemporaries, the Zhuangzi also emphasizes how differing schemes of commitments and conventions – different Daos – in some sense can only be justified internally; we cannot, as Confucians and Mohists might argue, hope to find some single standard or grounding that would allow us to distinguish what is universally right from wrong in an absolute sense. Justification for differing practices is ultimately relative to a particular Dao – a Confucian might give very good Confucian justification for upholding three-year mourning rituals, and a Mohist might give very good Mohist justification for the opposite stance – but in each case the justification offered ultimately presupposes many aspects of their own Dao. This relativist picture makes a direct response to the Nazi problem difficult for a Zhuangist if he is to avoid making universal claims from within his own Dao that make him appear just as dogmatic as a Mohist or Confucian.

A more serious difficulty in responding to the Nazi in a manner consistent with the Zhuangist position is that the Nazi himself has his own perspective – that of the Nazi Dao to which he is committed – that may very well be internally justifiable. The Nazi too may be able to give very good justification for his practices and actions from within the Dao to which he is committed. If we take seriously relativism of perspective, we can’t just rule out his perspective on the grounds that we find it abhorrent any more than we can argue that our own perspective is the uniquely correct one; many of our own practices might be abhorrent from alternative perspectives – our mass murder of insects with insecticide would surely be abhorrent from the insect perspective. It also doesn’t help to consider an overall natural perspective (i.e. the perspective of tian), because all of these perspectives, and all the actions of those taking them up, ultimately are natural, and nature provides no way of adjudicating between them that might tell us which are right and which are wrong.

One way of dealing with the problem is to simply accept this conclusion, and agree that the Zhuangzi is completely agnostic about competing Daos – no Daos are any better, worse, more or less justified or justifiable than any others. But this response is problematic: why then does the text seem so clearly critical of certain perspectives and Daos? If they are ultimately all equally valid and equally good, why bother criticizing the Mohists and Confucians? Surely we also want to be able to say that there is a sense in which the Nazi Dao might be worse than a Mohist or Confucian Dao, or worse than the Dao of a Daoist sage. It also seems counter to other values that we want to ascribe to the text that it would view these as equally acceptable Daos; a philosophy that genuinely is equally at ease with killing innocent babies for fun as free and easy wandering would seem to be taking moral relativism a little too far.

In a forthcoming paper, I argue that the Zhuangzi’s skepticism about knowledge does not merely have the negative consequence that we should question some of our knowledge commitments, but also suggests that – while still doubting – we can improve our knowledge by considering a wider range of perspectives. Knowledge commitments that hold up from a wider range of perspectives are in some sense preferable to those that fail to do so – though this can never make them absolutely justified or guarantee that they are universally correct.

I want to suggest that what I see as this positive aspect of Zhuangist epistemology can provide the Zhuangist with a consistent and meaningful response in the Nazi case. This response is inevitably not as strong as outright condemnation of the Nazi Dao as bad, or an endorsement of Daos that condemn the Nazi’s actions as good. Rather, the response focuses on two questions that the Zhuangist is better placed to respond to: firstly, as a Zhuangist, can I follow a Nazi Dao, and secondly, can I in any way condemn a Nazi Dao follower. Both of these questions, I argue, can be responded to using a perspectival account of knowledge.

On my account, a wise and knowledgeable person for the Zhuangzi is someone who, unlike the petty Confucian or Mohist debater each of whom claims that he is right and his opponent wrong, or the small cicada or giant bird each of whom knows that his kind of flying is the best kind, prefers knowledge commitments that remain valid across a wide range of perspectives, including those he does not occupy himself. On this reading, one serious problem with the Nazi’s commitments is their high degree of contingency upon perspective. “Killing innocent babies for fun” is acceptable from the Nazi’s own perspective, but not from many other perspectives, and in particular not from those of the babies themselves, from those of their parents, other members of society, and so on. The Zhuangist criticism of the Nazi is not so much that he is wrong but that he is stupid: a wise Zhuangist would no more follow a Nazi Dao than he would become a fervent and ideologically committed Confucian or Mohist insisting that he knows the uniquely right way for all to follow. For the same reason, he would condemn the Nazi practitioner as a fool who failed to see other important perspectives on his situation. He might – echoing a technique we see repeatedly in anecdotes in the Zhuangzi – point out perspectives from which the Nazi’s commitments fail to hold even from within his own Dao: would the Nazi still think killing innocent babies for fun was acceptable with respect to his own babies, or with respect to himself as an infant? Rather than offering a moral criticism, the Zhuangist can challenge the Nazi with an epistemic and cognitive criticism of his commitments and actions.

This criticism of the Nazi is naturally weaker than a direct moral condemnation of his actions, and also leaves open the possibility that the Zhuangist criticism may itself be open to attack. Perhaps like Zhuangzi who, as he stalks the bird in the forest hunting the mantis that preys upon the cicada, loses sight of his own vulnerability as he himself is pursued by the gamekeeper, we have applied a technique in criticizing the Nazi that someone else may use to criticize us. There is – as always – the possibility that we will some day meet with a “great awakening” in which we will come to see that our past commitments were in fact wrong, and that we too had failed to appreciate other important perspectives on our own situation, and so responded to it in a way we subsequently might come to see was mistaken. Nonetheless, in taking into account a wider range of perspectives than the Nazi did, though we cannot be sure that we are right, we can at least articulate one thing that is wrong with him and his Dao.

Posted in Philosophy | Comments Off

Zhuangzi, perspectives, and greater knowledge

This paper appeared in Philosophy East and West 65:3 (July 2015), accepted April 2012.

Zhuangzi, perspectives, and greater knowledge

Abstract

Although the text of the Zhuangzi appears to present prima facie skeptical arguments, there has been much debate as to the nature of its skeptical stance, and even whether or not its stance is substantively skeptical at all. In this paper I attempt to engage with both the skeptical aspects of the text and its positive agenda, by firstly accepting that the Zhuangzi takes a substantive skeptical stance, but also arguing that in doing so the text provides a positive account of how to improve our epistemic position – an account which might be a motivating factor of the Zhuangist ethical stance. My argument will focus on Zhuangist attitudes to different types of knowledge, specifically what the text refers to as “lesser knowledge (小知 xiao zhi)” and “greater knowledge (大知 da zhi)”, and the relationship between the two. I will attempt to show that, far from promoting “epistemological nihilism” as has been claimed by some, the Zhuangist stance is actually that of a “positive skeptic” who can offer wide-ranging practical advice on how to improve our own epistemic situation, while at the same time warning us of the ultimate limits of what we can come to know.

Download the full paper.

Posted in Philosophy | Comments Off

Chinese Text Project – Word clouds

Update to the CTP:

Word clouds have been added for all texts on the site. These show a graphical representation of unexpectedly frequent words and characters in a particular text, with more frequent and/or unusual words in larger fonts. To display the word cloud for a particular text, click on the icon shown on its contents page under “Media”.

Posted in Chinese, Digital Humanities | Comments Off