1 Introduction

The translation of certain biblical religious key terms, such as “Logos,” “God” and “Holy Spirit,” into Chinese can hardly be satisfactory, because such Christian concepts do not exist in the Chinese language, or had a very different connotation from Chinese religious terminology. Ricoeur (2006), one of the important thinkers of philosophical and biblical hermeneutics, in the last years of his life dealt with the question of translation in lectures presented between 1997 and 2002 that are collected in his book On Translation (Hereafter OT). For Ricoeur, while translation can always accomplish something, it inevitably involves some losses in transferring meaning. Translation always includes “segments of untranslatability” (OT: 5). In this paper, I will show how Ricoeur’s translation theory and the study of Chinese Bible translation can shed light mutually on each other. I will illustrate how Ricoeur’s criticism of the aspiration of perfect translation, his call for “equivalence without identity” and “linguistic hospitality,” and his notion of “constructing comparables” can help us to analyze and reflect on the Chinese Bible translation. And the study of Chinese Bible translation can provide support for Ricoeur’s translation theory and show the paradox involved in Ricoeur’s notion of linguistic hospitality.

2 Ricoeur on Issues of Untranslatability and the Multiplicity of Languages

Ricoeur, in his On Translation, describes the difficulties of translation as a wager which can be “summarized in the term ‘test’ [épreuve] in the double sense of ‘ordeal’ [peine endurée] and ‘probation’” which is a testing period of the urge to translate (OT: 3). For Ricoeur, to translate is to serve two masters: foreign author and reader with desire for appropriation. It means that the process of translation always stands in the middle of tensions between “a vow of faithfulness and suspicion of betrayal” (OT: 4). The tension is due to the inescapable “segments of untranslatability” of languages caused by different syntaxes that are not equivalent, and by different semantic fields that cannot be superimposed on one another, idioms and intertextual references (OT: 5–6).

According to Ricoeur, the fundamental reason for the issue of partial untranslatability is the fact of the multiplicity of languages. The diversity of language operates at different levels including the phonetic, lexical, syntactic division, etc. Furthermore, there are different combinations of sense and reference in different languages (OT: 33). This suggests, following von Humboldt, “the idea of a radical heterogeneity that should render translation impossible a priori” (OT: 30). Although facing the difficulties of partial untranslatability, Ricoeur insists on the necessity of translation. Ricoeur offers an unusual interpretation of the story of Babel and argues that the story of Babel is not to tell the curse of God; rather it is “the non-judgmental acknowledgment of the original separation,” the fact of the multiplicity of languages and, more importantly, the desire and passion of translation (OT: 18–21).

According to Steiner (1975: 73–74), there are two opposing views of translation: universalist versus monadist view. Steiner considers Noam Chomsky as a universalist. Chomsky (1965: 3–59) argues that the human mind possesses certain universal linguistic capacity, that is, an internalized universal grammar common to all languages. By means of transformational operations, we can generate sentences at the surface structures from these linguistic deep structures. For Chomsky, the idea of universal grammar provides the foundation of translation. Although Ricoeur admits that some progress has been made through Chomskian generational grammars, he rejects the universalist view. Such universalist view assumes that the source and target text can be matched with one another through a third non-existent text. However, for Ricoeur, there is no such third text that can be taken as a perfect model for comparison. There exists no universal perfect language that can mediate between different languages (OT: 24). Ricoeur rather argues that “within a vast cultural area where the community identities, including linguistic, are themselves the product of long-lasting exchanges” (OT: 35). Equivalence is produced by the translation rather than presupposed by it.

Following Benveniste, Ricoeur stresses that the basic unit of meaningful language is the sentence rather than the word (OT: 30–1). In order to make sense of sentences, the translator has to work with the text, to take seriously the examples of texts and extended discourse, not just individual words or sentences. A text is the aggregation of cultural knowledge, “it is texts, not sentences, not words, that our texts try to translate. And texts in turn are part of cultural groups through which different visions of the world are expressed, visions which moreover can confront each other within the same elementary system of phonological, lexical, syntactic division, to the extent of making what one calls the national or the community culture a network of visions of the world in secret or open competition” (OT: 31). Translation should be based on hermeneutical analysis of cultures and worldviews of different nations, as he argues “the work of the translator does not move from the word to the sentence, to the text, to the cultural group, but conversely: absorbing vast interpretations of the spirit of a culture, the translator comes down again from the text, to the sentence and to the word” (OT: 31).

3 Intralingual Translation and Ontological Foundation of Translation

For Steiner (1975: 77), Edward Sapir and Benjamin Lee Whorf are monadists or linguistic relativists.Footnote 1 Monadists argue that languages are thought-systems carrying within themselves a unique worldview which inseparably incorporates syntax and semantics, words and meanings. Language shapes our thinking and our understanding of reality. Steiner (1975: 74) criticizes, “if the Humboldt–Sapir–Whorf hypothesis were right, if languages were monads with essentially discordant mappings of reality, how then could we communicate interlingually?” In other words, if languages are monads, they are ultimately untranslatable. However, translations obviously occur continuously. In the face of these two extremes of linguistic universalism and linguistic relativism, Ricoeur, following George Steiner, holds a position between these two. Although Ricoeur rejects the ideal of perfect translation, he also refuses the idea of total untranslatability (Kharmandar 2015, 74). Ricoeur stress that translation is still possible; a good translation can always achieve something, while it would inevitably involve certain loss of meaning; and there is no perfect translation (OT: 3).

Following Steiner, Ricoeur argues that there are two senses of translations. The first is an interlingual translation, the translation among different languages. The second is an intralingual translation, or translation within a language, “synonymous with the interpretation of any meaningful whole within the same speech community” (OT: 11). While the diversity of languages makes interlingual translation difficult, Ricoeur reminds us that there also exists the fact of the universality among these different languages, that is, all men speak by using language. Historically speaking, the diversity is not a disaster because people have always been engaged in the practice of translation (OT: 13). Thus, while we must acknowledge the fact of the diversity of languages, we also need to recognize the existence of a desire to translate which stems from the desire to go beyond constraint, to broaden the horizons of one’s own language, and to discover its own inherent but omitted potential. Translation is a task that must be done for the continuation of human action (OT: 19). The works of translation in the past also show that everyone has the ability to learn and to use others’ language (OT: 13). The desire of going beyond one’s own culture and translation and the history of interlingual translations not only show that human beings by nature tend to translate, but also show that human beings have the capability to translate. And, for Ricoeur, the ability to learn other languages and translation is related to our experience of intralingual translation, which involves our language’s ability to reflect upon itself, to express itself, and be considered as a language among other languages. The intralingual translation shows that “it is always possible to say the same thing in another way” (OT: 25). This ability to say the same thing in different ways is related to his idea of “an equivalence without identity” in his translation theory (OT: 22). This equivalence without identity links interlingual and intralingual translation together, showing that it is possible to translate in more than one way in interlingual translation; there is “something foreign in every other” (OT: 25), and thus, it leads to the significance of a dialogical nature of language (OT: 24). And thus, I would argue that Ricoeur’s explorations and linkage of interlingual and intralingual translation have provided an ontological foundation of translation. This also provides the condition for linguistic hospitality which is to acknowledge others’ languages and to be receptive to them, as discussed below.

4 The Limit of Translation: An Equivalence Without Identity

While we affirm that translation is possible, the case of untranslatability by Ricoeur or what Kearney (2019: 4) calls, “untranslatable kernel” appearing in translations also reminds us that the languages of the host and foreigners are never the same. With such discontent and mourning, Ricoeur calls for the abandonment of the dream of the perfect translation; what translation can do at best is approximation (OT: 8, 35), or what Ricoeur calls, “an equivalence without identity” that is an adequate equivalence, one “not founded on a demonstrable identity of meaning” (OT: 22). The only recourse to evaluate the adequacy of translation will depend on a critical reading of bilingual specialists who attempt to retranslate the work in question (OT: 7). As he states, “This equivalence can only be sought, worked at, supposed. And the only way of criticizing a translation—something we can always do—is to suggest another supposed, alleged, better or different one. And this, moreover, is what happens in the world of professional translators” (OT: 22). The assumption of “an equivalence without identity” is clearly shown in the phenomenon of ceaseless retranslation of the great classics of global culture such as the Bible, Shakespeare, and Homer in which we can see the urge of translation stimulated by the discontent about existing translations (OT: 7, 34).

In the past, according to Ricoeur, there were two ways in which the dream of creating a perfect universal language and perfect translation kept returning. The first was the Enlightenment ideal of establishing the complete library, by accumulation of the translation of all the works in all languages, from which all of the untranslatabilities would be eliminated. And there is “a rationality fully released from cultural constraints and community restrictions, this dream of omni-translation would try to fill the interlinguistic space of communication” (OT: 9). The second was the dream of a pure language, a kind of messianic expectation expressed by Walter Benjamin, by which perfect translation can be achieved without loss. However, for Ricoeur, the creation of a perfect universal language is impossible for two reasons. First, the creation of a lexical database of the universal language requires a total equivalence between symbols and references, and in a broader sense between language and a worldview of different cultures. However, there is no such consensus that could characterize such a universal language. Second, there is a gap between the universal language and different empirical natural languages that is insurmountable. As Foran (2018: 99) states, the ideal of universalism tied to an ahistorical view of language. It denies the various historical events that led to the multiplicities of languages. In the discussion of the history of Chinese Bible translation below, we will see how the meaning of religious terms varies according to different cultures and periods in history.

According to Venuti (2010: 6), there are two models of translation throughout history: instrumental and hermeneutical. The instrumental model, as advocated by St. Jerome, considers translation as conveying “an unchanging essence inherent in or produced by the source text, so that even if assimilated to the receiving language and culture that essence is transmitted intact.” It assumes the idea of an invariant, a perceived fixed meaning, contained in the source text; thus, for St. Jerome, a text could be rendered “word-for-word” and “sense-for-sense” translation. Venuti (2010, 6), instead, prefers the hermeneutical model which considers translation as conveying “one interpretation among other varying possibilities, each of which transforms the source text so as to reflect the receiving language and culture at a particular stage of development, in a specific social situation at a specific historical moment.” It assumes that the correspondence between source and translated text is partial and contingent because it is incomplete in re-creating the source text and the derived meaning is always one among other possible interpretations. Venuti (2010: 6–8, 24–8) prefers the hermeneutical model because languages and texts are created thickly mediated by historical, social and cultural determinants, and the hermeneutical model can expose diverse conditions of a translation, how these various determinations work in any translation and translators’ relation to their social-cultural values. It allows translators to apply different but equally valid interpretations. And it is obvious that Ricoeur’s translation theory belongs to the hermeneutical model which stands against the instrumental model aspiration of perfect translation.Footnote 2

Throughout the history of Bible translation in China, most translators would not accept the instrumental model because of the huge difference between the Bible and Chinese culture they faced. However, the aspiration of the perfect translation is still shown in a few cases of Chinese Bible translations, in particular the translation of certain religious special terms, but just expressed in the way of escapism. As Romano (2017: 179–181) states, in the face of translating certain “divine” terms, missionaries found themselves in a dilemma between translating them into certain similar Chinese philosophical and religious terms and offering a phonetic transcription. Both options have pros and cons. Translating into Chinese philosophical and religious terms would risk creating confusion, misinterpretations and inappropriate assimilations, while it is more easily understood by the general population. Although phonetic transcription seems to preserve fidelity, it will result in generating puzzles and awkward Chinese words. For instance, the Jesuits in China, during the Ming Dynasty (1368–1644), used the phonetic transcription of the Latin word Deus (Dousi 陡斯) to translate “God” into Chinese. And Jesuit priest Xiao Jingshan, 蕭靜山 (1855–1924) offered the phonetic transcription of the Latin verbum (wuerpeng 物尔朋) to translate “Logos.” However, such translations were incomprehensible and could hardly be popularized.

What is worse was the case of translating the term “Holy Spirit.” Initially, Catholic missionaries preferred to use a phonetic transcription in translating “Holy Spirit” from Latin because they were afraid that if the translation was not accurate then baptisms would be invalid. Thus, the translation found in a copy of Tianzhu Jiaoyao 天主教要 (Compendium of the Doctrine of the Lord of Heaven) by Father Pasquale D’ Elia S. J. (1890–1963) is Sibilitusisangketi 斯彼利土斯桑克提. However, such transliteration was found to be incomprehensible; to remember such sequence of meaningless characters was also too difficult. Thus, there were different formulas introduced for the translation of “Holy Spirit” later, such as Shengshen 聖神 (Holy God) suggested by Catholic, and Shengling 聖靈 suggested by Protestant missionaries as ling靈 means “spirit.”

Indeed, the motivation of such phonetic transcriptions in Chinese Bible translation was motivated by the aspiration of the perfect translation, but expressed as escapism. According to Longeway (1990: 1–2), escapism “is defined as the attempt to avoid awareness of aversive belief.” It attempts to draw us away from the unpleasant truth, to keep those unpleasant beliefs “out of consciousness, then, and, should they enter consciousness, to distract one from them or put them out of mind.” Distracting oneself with irrelevant concerns is one of the common strategies to force the unpleasant beliefs out of one’s consciousness. Translators were worried about offering a misleading translation by using a similar Chinese religious term; thus, they felt safe rather by offering a phonetic transcription. However, what such transliterations can offer is indeed meaningless to indigenous people. It just helps the reader to pronounce the term; it is just like phonetics; it is not a translation. Such avoidance of translation violates our desire for translation; it is ironically motivated by the fear of being incapable of offering the perfect translation. They were not willing to face the reality that they could not provide a perfect or satisfactory translation, and thus, they offered a phonetic transcription instead as if it could not be wrong. However, they might not have been aware that offering phonetic transcription instead of translation is indeed a kind of escapism.

I do not in principle reject all possibilities of phonetic transcriptions. This seems to be inevitable in the translation of names and certain concepts which do not exist or when one cannot find similarities in the culture of the target language, such as the translation of “hamburger” as 漢堡包, or 寿司 as Sushi. If the Bible is translated into a kind of language in which one cannot find any concept of a spiritual, transcendent, supreme God or deity, and people in that place believe in secular materialism, then phonetic transcriptions of these religious terms seem to be a plausible option. However, this is not the case for the Chinese community. Thus, I would argue that phonetic transcriptions in Chinese Bible translation were the avoidance of translation ironically motivated by the aspiration of the perfect translation expressed as a form of escapism. The contribution of Ricoeur’s philosophy is that it helps us to face the fact that there is no perfect translation—what we can do at best is approximation.

5 Linguistic Hospitality

By giving up the dream of perfect translation, Ricoeur argues for achieving “the happiness associated with translating” in “linguistic hospitality,” and “its scheme is definitely that of a correspondence without adequacy” (OT: 10). Ricoeur states that this is a “fragile condition” which admits that no verification can be done other than retranslation, to translate afresh after the translator (OT: 10; Kearney 2019: 3–4). And he further argues, “just as in the act of telling a story, we can translate differently, without hope of filling the gap between equivalence and total adequacy” (OT: 10). Through linguistic hospitality, one can experience “the pleasure of dwelling in the other’s language,” just as it is “the pleasure of receiving the foreign word at home, in one’s own welcoming house” (OT: 10).

Munday’s (2012: 254) observation is right; Ricoeur has presented translation as “an ethical problem—it risks betraying author and reader but it operates its practice of ‘linguistic hospitability,’ allowing the two texts to live side by side.” As Bottone (2011: 67) states, what matters to Ricoeur throughout the reflection of translation is not simply linguistic, but ethics. Instead of a correct method or principle of translation, Ricoeur emphasizes the correct ethical attitude of translation in encountering different languages and cultures. As Kearney argues, good translations, for Ricoeur, involve critical openness towards other languages. One must give up the sense of the self-sufficiency of the native language in order to accommodate the foreign language. Linguistic hospitality, in Kearney’s (2007: 151) words, “calls us to forgo the lure of omnipotence: the illusion of a total translation that would provide a perfect replica of the original. Instead, it asks us to respect the fact that the semantic and syntactic fields of two languages are not the same, nor exactly reducible the one to the other.” According to Kearney (2019: 1–2), translators are always facing the temptation of absorbing and reducing other languages into one’s native speech on the one hand, and surrendering one’s native language to the foreigner, on the other hand. Ricoeur’s linguistic hospitality is the middle position to overcome the temptation of the extremes of linguistic hegemony and humiliation. It also helps us to resist the temptation of the perfect translation because it asks us to “honor a dialectical balance between proximity (welcoming the stranger into our midst) and distance (acknowledging that something is always lost in translation: the other’s meanings can never be completely mine). A ‘hospitable’ translator is one who aims at approximate correspondences between tongues without ever assuming these to be final or adequate. Which is why translation is always an endless task” (Kearney 2019: 2).

6 Chinese Bible Translation and Ricoeur’s Idea of Linguistic Hospitality

In view of the history of Bible translation in China, I would argue that what most missionaries and translators of Bible translation have done could demonstrate Ricoeur’s ethics of linguistic hospitality. Unlike general translators, many of these missionaries and translators are not native Chinese. They risked their lives to go to China, learned their language, lived with them and lived like them. Matteo Ricci (1552–1610) even aligned himself with the Confucian intellectually elite literati and adopted their way of dressing. The motivation for them to translate the Bible is not profit-making or the pursuit of personal benefit; rather they risk their lives to come to China with a strong mission to preach Christian beliefs to the Chinese. Their motivation for translating the Bible into Chinese was to let the Bible become more popularized and widely accepted by Chinese. Thus, they put a lot of effort into learning Chinese culture and knowledge and expressing appreciation of them. A few of them, e.g. Matteo Ricci, Walter Henry Medhurst (1796–1857), Robert Morrison (1782–1834), and James Legge (1815–1897), even became respectful scholars in Chinese language and culture among Chinese intellectuals and officials. They not only contributed to the translation of the Bible into Chinese, but also translated Chinese classics into Latin and English, and composed certain writings about Chinese culture and compiled a Chinese-English dictionary. In their struggle to translate religious terms, we can see their proficiency of both the source and target languages and cultures. We can say that these efforts and achievements of Bible translations can be considered as one of the paradigm cases of linguistic hospitality in translation. Nevertheless, further studies of Chinese Bible translation in the following also show that translations in a culture containing diverse traditions inevitably involve certain paradoxes with linguistic hospitality.

6.1 Controversy in the Translation of Logos

Missionaries’ and translators’ effort and attitude of Bible translation were particularly shown in their debates and struggles of translating certain religious philosophical terms. For instance, regarding the translation of Logos, the earliest translation was suggested by Robert Morrison and Joshua Marshman (1768–1837) by using the term Yan 言 (word). However, Medhurst (1852: 58) criticized such translation as being weak and not expressive. He proposed to translate Logos with the term Dao 道, and it has been endorsed by Protestants. However, the use of indigenous philosophical terms is criticized for being in danger of inappropriate assimilations and the risk of syncretism.

According to Zhang 張子元 (1995: 443–49), a Chinese pastor and scholar, while there are certain analogies between Dao and Logos in the Bible, several incompatible elements can also be found. From the perspective of their attributes, Dao is an abstract principle and impersonal force operating in the universe. However, Logos in the Bible is a personal God. The concept of Dao is derived from philosophical reflection from observations of the world no matter whether by Daoists, Confucians or Legalists, while Logos is God who reveals himself actively by incarnation. From the perspective of creation, everything was generated or produced starting from Dao spontaneously and relentlessly without reason and purpose, while God created the world through Logos with love and the final goal of salvation. Thus, Zhang concludes that Dao and Logos are indeed very different concepts; not many similarities can be found between them. In order to avoid the risk of syncretism, Xiao Jingshan, in the Catholic translation, translated logos as Shengyan 聖言. Superficially, Shengyan 聖言 seems to be a better translation of Logos than Dao because it is a new Chinese term and it does not refer to other religious languages. However, despite the conceptual differences between Logos in the Bible and Dao, the translation in terms of Dao is very popular and seems to be more widely appreciated than Shengyan 聖言 in China. A survey by Romano (2016: 92–93) shows that Chinese Protestant respondents show greater appreciation of the use of Dao (87.7%) in translating Logos than Chinese Catholics do for Shengyan (80.3%). Furthermore, while no Chinese Protestant (0%) agrees that Shengyan is a better translation, around 12% of Chinese Catholics consider Dao to be a better translation. Indeed, I compiled a chronology of different Chinese Bible versions with the translation of John 1:1 (see Appendix). There are four interesting observations. First, the chronology involves 33 translations. There are four Catholic translations, one Russian Orthodox translation, 27 Protestant translations, and one recent secular translation by Feng Xiang 馮象. Second, while eight versions translate logos as Yan 言 or Shengyan 聖言, two versions translate them as Hua 話 (speech) and 23 versions translate them as Dao 道. Third, initially, both Catholic and Protestant translations (the first three translations) translated logos as Yan 言, but later while three Catholic translations, one Russian Orthodox translation, and one secular translation translated them as Shengyan 聖言, almost all the rest of the Protestant translations (23) translated them as Dao 道. Fourth, Feng (2010: xii–xiii) in his preface to New Testament translation comments that Logos should be translated as Yan 言. However, he does not think that translating it as Dao 道 is a careless mistake. Rather he thinks that it is probably a deliberate decision made by the Protestant missionaries based on certain doctrinal considerations. Based on the above observations, it seems to show that although literally speaking, the translation of Logos as Yan 言 seems to be more natural and a more faithful translation of the Latin verbum, it is an odd translation for general Chinese readers because Yan 言 generally means ordinary people’s talk; it never refers to something religious or transcendent. On the contrary, Dao 道 can refer to words, principles, paths, cosmopolitan and transcendent religious concepts (Lu 2016b: 341–344), and thus, most later translators, even though they are all Protestants, decided to translate it as Dao 道, except for two Protestants who translate it as Hua 話 (speech). In fact, Xie 謝扶雅 (1995), a Confucian Christian Scholar, also argues that Dao 道 is a better translation because the meaning of Dao 道 can include that of Yan 言, while the meaning of Yan 言 cannot cover that of Dao 道. Indeed, Confucian, Daoist and Legalist classics also use the word Dao 道 as the law of the nature, the source and principle of life, and the way of governance in the universe that is also linked with the idea of God, Heaven and the Creator (Koutsoumpos and Zhuang 2016: 216–220; Lu 2016a: 344–346; Guo 2014: 271). Although Feng Xiang thinks Yan 言 is a more appropriate translation, he is a nonreligious scholar, and he is not very familiar with Christian doctrines. Indeed, he admits that he has conducted no research on the missionary history of the late Qing Dynasty and the early Republic of China. Thus, I would argue that Feng may not be able to fully understand the religious concern regarding such translation. Finally, the reason for Dao being preferred over Shengyan Protestant Chinese Christians is probably because, first, unlike Catholics, they do not need to consider the translation of the Latin verbum, and second the term Dao is more familiar to Chinese. Chinese can easily grasp certain basic features of it such as transcendence, reason and speech from the term.

6.2 Controversy in the Translation of “God”

Controversy also happened in the translation of “God.” As discussed above, by rejecting the transliteration as Dousi 陡斯, Matteo Ricci (1552–1610) suggested translating God into indigenous philosophical and religious terms such as Tian 天 (Heaven) and Shangdi 上帝 (supreme ruler). Both terms denote a transcendent sovereign derived from the Shang and Zhou dynasties (1600–256 B.C.), also called the Pre-Qin Period, contained in Confucian Classics. Matteo Ricci believes that the presence of these two terms shows that the ancient Chinese had been inspired by knowledge of God which could be restored. For Ricci, the ancient Chinese during the pre-Qin period were already worshiping the supreme God; the difference between Shangdi and the European term for a personal God is terminological only (Liu 2014: 47).

Later, Jean Basset (1662–1707) used a Chinese religious term Shen 神 to translate “God.” However, the use of indigenous religious terms was controversial. For instance, the Franciscans and Dominicans rejected the use of local religious terms because they denied that Chinese had an idea of God or Shangdi which was eternal and had created the universe. Furthermore, the Chinese term Tian is a very vague concept covering a wide range of meaning and identified with impersonal force for ordering the universe (Sheppard 1955: 28). Gernet (1984: 210; see also Romano 2017: 171), a French sinologist, argues that to identify the Chinese notion of the Heaven with the God of the Bible is to merge notions that were incompatible which was the origin of confusion among both the Chinese and the missionaries. Indeed, the term shen 神 was usually associated with “ghosts and spirits” (guishen) (Pfister 2004a: 621). Shen means “god” but in the sense of “deity,” or a good spirit in contrast to gui 鬼, an evil spirit; thus, it can be in plural form. It is different from the Christian idea of God which is unique and exclusive. James Legge worries that using the term Shen would eventually evolve into teaching polytheism and so reduce the doctrine of the Trinity to an absurdity (Pfister 2004b: 190).

In 1583, a new term, Tianzhu 天主, was accidentally introduced by a Chinese convert (Sheppard 1955: 27). The missionaries thought that this term was appropriate to translate “God” because they thought that it had never been used in the Chinese Classics. However, missionaries did not know that the term was actually present in a classic historical book, Shiji 史記 and used by Buddhist writers to indicate an Indian god. Protestant translators on the contrary preferred to use the term Shangdi and Shen because Tianzhu was not a generic term and hence not appreciated by the Chinese, and it was used by the Catholic Church (Romano 2017: 172–174). In particular, Legge (1852: 23–59, 64, 129–31), a Scottish sinologist who had translated many Chinese Classical texts into English, rejects the use of Tianzhu because it would bring forth a strange God of whom Chinese ancestors had known nothing. Legge was famous in advocating the use of Shangdi because it was highly appreciated by the Chinese; he also believed that the ancient Chinese must have known the Christian God that was revealed not in philosophical texts, but through the sacrificial ritual ceremonies (Pfister 2004b: 188).

6.3 Advantages of the Generic Term

Indeed, there are five reasons why I prefer to use the generic term, such as Dao and Shangdi, to translate the religious term. First, such translation is more familiar, naturalized and understandable for Chinese; thus, it helps the popularization of Christian belief. As Dai (2015: 341–346) argues, a “naturalized translation” would generally be considered a good translation. It could not only bring new ideas to the target culture, but also eliminate barriers between different people and facilitates communication between them. Second, from today’s perspective, there is no need to worry about confusing Christian beliefs by using the generic term because most contemporary Chinese are not so philosophical; they do not really know the theory of Dao in Daoism in detail. They may know of the Jade Emperor (Yudi, 玉帝) because of the popular novel Journey to the West or Xiyou Ji 西游記, but few of them have ever heard of Shangdi from the Shang and Zhou dynasties. Nowadays, with the popularity of Christianity, when you talk about Shangdi, Chinese usually refer to the Christian God rather than the one from the Shang and Zhou dynasties. As Zhao Xiaoyang (2010: 175) states, “The Chinese word—Shangdi—underwent a fundamental shift of meaning, becoming gradually ‘Christianized’ and losing its original cultural and religious connotations.” When you Google search the term Shangdi 上帝 in Chinese, most of the websites it refers to are about Christianity.Footnote 3 Third, regarding the worry of misleading by translating Logos as Dao, with the elaboration of the nature of Logos in the first chapter of the Gospel of John, most readers can easily distinguish Christian Dao (Logos) from Daoist Dao. Thus, I think what is important is explanation and elaboration of the Christian belief by preachers. We should not rely on seeking a perfect translation in order to convey beliefs accurately.

Fourth, although Zhang is right that Dao and Logos in the Bible are very different concepts, the concept of Dao is actually similar to Logos in ancient Greek philosophy, such as Heraclitus and Stoicism. They both denote certain principles regulating the order of existence in the universe. Generally, biblical scholars find that there are both Hellenistic and Jewish backgrounds for the word Logos that St. John employed. The idea of logos was first found in Heraclitus (fifth century BC) as the unifying, rational principle holding together a world in perpetual flux. Later, Stoicism (third century BC onwards) considered Logos the unifying principle and the source of all things, and through the logoi spermatikoi all things come into being. Logos was also considered the natural law that people must follow. In the Septuagint (Greek Old Testament), Logos is used to translate Hebrew dāḇār (word), usually referring to the word of God that is self-revelatory, creative power, and wisdom. And Philo of Alexandria, the Hellenized Jewish philosopher, combined Greek philosophy and Jewish thought in his philosophy. From the Hebrew Scriptures, he got the subject of his writings. And from Platonism and Stoicism, he derives hermeneutic axioms to interpret the subject. Following Plato, Philo argues that there are two worlds, the ideal world of God and immortality and the world of physical phenomena. Sometimes in Philo, logos represents the word by which God created the world. At other times, it refers to the mediation between the ideal and the phenomenal world. Despite this Hellenistic-Jewish background, St. John decisively broke with these Greek concepts and goes beyond the Jewish perspective by affirming the personal pre-existence and incarnation of Logos by identifying Christ as the divine Logos, who created the world, who was the light of humans but was rejected, who became flesh and made people the children of God (Johnson 1992: 482; Donner 2000).

The above discussion shows that the meaning of Logos is not fixed or static; rather it is evolving through constant negotiation between different ideas and traditions. That St. John uses the concept of Logos to express Christian ideas is just a kind of appropriation similar to what other philosophers did at that time. And for Chinese bible translators, to use the term Dao is also similar to St. John himself borrowing the term Logos from ancient Greek philosophy and reinterpreting it in order to make the concept understandable to everyone regarding the idea of divine incarnation (Romano 2016: 83). There is no reason to reject such appropriation in Chinese translation while accepting St. John doing so.

And fifth, allowing such appropriation in translation, making it more popularized among Chinese, can enhance cross-cultural dialogue and mutual cultural enrichment. As Zhao (2010: 174) argues, “Rendering “Deus” or “God” into Chinese involves the deepest kind of cultural dialogue between two of the oldest civilizations on Earth. It means a repositioning of one’s own culture with that of another and brings to bear the widest range of cultural exchange, religious assimilation, and linguistic reproduction.” Indeed, using generic terms, instead of phonetic transcriptions, also demonstrates Ricoeur’s idea of linguistic hospitality by acknowledging Chinese religious languages and being receptive to them.

6.4 Further Reflection and Limitation of Translation and Linguistic Hospitality

However, the process of such translation involves certain tensions generated from Neo-Confucianism in the seventeenth century. Neo-Confucianism does not believe in a personal God; rather it interprets Tian as Taiji 太極 (the Great Ultimate) and li 理 (principle of nature). Neo-Confucianist understanding of Tian is a kind of transcendent nonpersonal principle. It is very different from the Christian understanding of God. Ricci’s translation was actually rejected by Niccolò Longobardo (1565–1655) who sided with the Neo-Confucianist interpretation of Tian and finally fueled controversies during the late seventeenth and early eighteenth centuries. Longobardo finds that Confucian scholar-officials at that time never agreed with Ricci unequivocally (Liu 2014: 49–51). However, Ricci also disagreed with Confucian scholar-officials. He criticized the association of Tian with the concepts of Taiji and li as just following the official interpretation established by founders of Neo-Confucianists, such as Zhu Xi (1130–1200), Zhang Zai (1020–1077), and the brothers Cheng Yi (1033–1107) and Cheng Hao (1032–1085), in the Song dynasty. And Ricci was aware of Daoist and Buddhist influences on Neo-Confucian metaphysics, which he denounced as corruptions (Liu 2014: 48; Chen 2016: 271).

Legge refers to Zhu Xi’s argument that “Tian means principle” and argues that such principle must be based on the existence of a divine governor which Neo-Confucianism fails to recognize (Chen 2016: 274–275). Legge, by exploring the Qing imperial rites related to sacrifices to Huangtian Shangdi (Supreme Lord dwelling in the sovereign heavens), discovered that Shangdi is not simply a “patterned principle” as suggested by Neo-Confucianism, but a monotheistic being who had powers of blessing and judgment over the world (Pfister 1999: 216). Indeed, apart from Ricci and Legge, certain contemporary Confucian scholars, such as Tu (2008: 122–134) and Shen (2013: 32), also consider Tian in pre-Qin Classics as a personal dialogical God. Thus, if Ricci’s and Legge’s analysis is right, I would argue that Ricci and Legge were sincere about the theistic affinity of pre-Qin Confucianism and Christianity; their genealogy is a kind of historical retrieval of the ancient Chinese idea of personal God. Chen (2016: 270–278) further argues that Ricci’s and Legge’s translation of Tian and Shangdi involved a kind of dialogical process of one reading the other, so as to accommodate the culture-specificity of the source to the target culture. Ricci’s and Legge’s translation had achieved a kind of intertextual coherence (fidelity or coherence between the source and target material) and intratextual coherence (the translator’s communication of his or her text with the target audience), showing an intertextual theology that richly merges Confucian hermeneutics with Christian thought through cross-cultural communication.

Basically, my main purpose here is not to arbitrate which translation is better. Through discussing controversies about the translation of religious terms, I argue for four points. First, the debates have given support to Ricoeur’s rejection of perfect translation and have illustrated Ricoeur’s theory that translation is a kind of hermeneutical dialogical activity. And I would argue that even if Chomsky’s theory of universal linguistic capacity is right, perfect translation is still impossible. This is because meaning of religious vocabulary of a given community is grounded in the shape of historical cultural religious beliefs and practices. In ancient China, where there was no Christian faith, it is impossible to find equivalent Chinese religious terms for achieving perfect translation. What translators can do at best is approximation. Rejecting the possible outcome of approximation would likely render translation unpopularized at best, and at worst incomprehensible, as discussed above.

Second, the debates show the struggles of missionaries and translators, their cautiousness and seriousness in choosing the appropriate words for translating important concepts of the Bible into Chinese. As discussed above, such attitudes and efforts show that Chinese Bible translation can be considered as one of the paradigm cases of linguistic hospitality in translations.

Third, the debates surrounding the translation of God also show that there exist certain paradoxes with, what Ricoeur says, linguistic hospitality. Let’s call it the paradox of linguistic hospitality. By “paradox,” one usually means a puzzling conclusion driven by a set of apparently incontrovertible premises or certain counter-intuitive reasoning. Unlike semantic or logical paradoxes, the paradox of linguistic hospitality does not logically, or in principle, involve contradictions; but empirically speaking, it most likely results in certain contradictory situations. The above analysis shows that Ricoeur’s idea of linguistic hospitality is based on certain hidden assumptions or conditions. It is feasible only if the target language of the translation involved belongs to a monolithic society, which is a closed society, not open to new ideas, and in which there is only one dominant idea, or at least there exists no conflicting ideas. In a pluralistic society in which there exist different competing and conflicting ideas, to show linguistic hospitality towards one idea would just indirectly express inhospitality towards its opposing views. However, if a society is monolithic and not open to new ideas, people in a closed society may not welcome the introduction of any translation of other new ideas or religious canons. Thus, it is really a paradox. In the case of Chinese Bible translation discussed above, while Ricci and Legge demonstrated linguistic hospitality towards Confucian metaphysics of the Pre-Qin Period in their use of Tian and Shangdi, they also asserted their criticism and inhospitality towards Neo-Confucianism. As Cawley (2013, 294–297) argues, Ricci’s translation has uncovered what the hegemonic tradition constructed by Neo-Confucian scholars has hidden or repressed from its own philosophical history. He had deconstructed God’s name(s) between traditions, opening up the name(s) of God to new Confucian possibilities. In other words, in a culture containing different traditions or schools of thought, offering linguistic hospitality towards one tradition may imply expressing inhospitality towards another tradition. The paradox of linguistic hospitality also demonstrates the limitation of Ricoeur’s idea of linguistic hospitality. It is impossible for people to show linguistic hospitality to all different kinds of cultures, traditions or ideas through the interpretation and translation done by them. What the translator must do, apart from being linguistically hospitable, is to carry out substantial studies of different traditions and to distinguish the substantive meanings of all these metaphysical/religious terms, just like Ricci and Legge did, as discussed above, so that the philosophical/religious affinity between the source and target languages can be found.

Fourth, the above cases show that reasons for such untranslatability are not simply because of the multiplicity of languages and the absence of prior meaning as mentioned above. It is also because there is ambiguity or polysemy with these religious terms in both the source and target language. The exact meaning of these religious terms varies depending on different schools of thought in China. Take the term Tian for example. Originally Tian (Heaven) in the Western Zhou dynasty (1045–1771 BC) was considered a transcendent creator and sustainer, the God on High. The first Zhou kings believed that their legitimacy as rulers was based on Tianming 天命 (mandate of Heaven); emperors of later dynasties, called Tianzi 天子 (son of Heaven), were supposed to follow the will of Tian and exhibit the virtue of ren (benevolence); and emperors were the only persons entitled to chair state rituals serving Tian. However, in the period of Spring and Autumn (771 to 476 BC), with the collapse of royal authority resulting from emperors’ presuming loss of virtue, Tian as the symbol of “benevolence” and “justice” developed into “fate.” Although Confucius still considered Tian as the transcendent creator and sustainer, he transformed the meaning of Tianming as “fate” into “mission.” And later, because of the influence of Daoism and Buddhism, Neo-Confucianism interpreted Tian as li 理 (principle). However, the Daoists, such as Laozi, reject the transcendent dimension of Tian, consider it as “sky” and reduce it to a purely “natural” level. For Daoism, it is Dao, rather than Tian, that represents the idea of transcendence (Fu 2003: 726–728). Thus, we can see that there are very different understandings of Tian between different schools of Confucianism and Daoism in China. Indeed, whether the attribute of Tian is personal or impersonal is also controversial among contemporary Confucians.

7 Constructing Comparables and Semantic Innovation

As discussed above, Ricoeur rejects the idea of perfect translation and the presumption of existence of a prior meaning. For Ricoeur, what translation can achieve is “an equivalence without identity” and such mystery of equivalence is solved by constructing equivalence rather than being presupposed by it. This leads to Ricoeur’s idea of “constructing comparables” (OT: 36) or “semantic innovation.” As Ricoeur states, “The construction of the comparable has even become the justification for a double betrayal insofar as the two incommensurable masters [author in his strangeness and reader in his desire for appropriation] are rendered commensurable through the translation-construction” (OT: 38). For Ricoeur, theory of meaning must cover the issue of novelty throughout the translation. Ricoeur argues that translation is the “construction of the comparable” which finally expresses itself in “the construction of a glossary” (OT: 37). As Ricoeur states, “Ordinary words that have not had a philosophical destiny and which, owing to the effect of translation, are removed from contexts of use and promoted to the rank of equivalents, those great equivalents without identity, whose antecedent reality we had presupposed, believing that it was hidden somewhere so to speak, and the translator would discover it” (OT: 37).

I think a few examples of Chinese Bible translation of religious terms are demonstrating Ricoeur’s idea of constructing comparables. For instance, with regard to the translation of Logos as Shengyan 聖言, God as Tianzhu 天主, Holy Spirit as Shengling 聖靈, etc., these are not original Chinese terms; they are neologism by combining two generic existing characters. Although they are new terms, they can be understood by combining the meanings of two familiar characters, just like “Holy Spirit” in English. It represents a compromise between indigenous philosophical religious terms (such as Tian) and a neutral but meaningless transliteration (such as Dousi) (Romano 2017: 185). Such combination of two Chinese characters led to what Ricoeur (2003: 350) calls “semantic innovation” which “results from founding a new semantic pertinence at the level of the metaphorical statement as a whole, the conceptual articulation proper to the speculative mode of discourse finds its condition of possibility in the semantic functioning of metaphorical utterance.”

It is precisely the contact and the cross-fertilization of these two distinct languages in translation that a novelty can be born. And this natality can be multiple as shown in different versions of translations of the Bible, not only Chinese translations, but also including Septuagint translations from Hebrew into Greek, St. Jerome’s translation into Latin, and the many different versions in different languages. As Kearney (2019: 4) argues, “With each rendition a new ‘semantic surplus’ is triggered by the creative collision of separate tongues—something mutually enhancing for both cultures.” For instance, in the Septuagint translation of Exodus 3:14 (“I am who I am”), “the Greek ontological notion of being (ontos on), understood as formal and material substance, is radically transformed by its encounter with the Hebrew notion of God’s becoming as historical and eschatological promise.” This semantic transformation also happened in Chinese in which the notion of Shangdi is also transformed by its encountering with the notion of God in the Bible, as discussed above.

Ricoeur also highlights the case of the biblical translation by Martin Luther: “Luther not only constructed a comparable in translating the Bible into German, in ‘germanizing’ it, as he dared to say, in the face of St Jerome’s Latin, but created the German language, as comparable to Latin, to the Greek of the Septuagint, and to the Hebrew of the Bible” (OT: 37). As Kearney (2007: 149) argues, Ricoeur’s translation theory indicates “the way in which some of the great translations of biblical and classical texts played formative roles in the development of both national and cultural identities…. In all these instances of inter-linguistic translation, the transmigration of one linguistic thesaurus into another was linked with modern ideas of human emancipation and change.” Indeed, the cultural transformation led by the achievements of Bible translations also happened in modern China. Shao (2010), in her studies of Chinese Bible translations in the period from the Late Qing to the Early Republican Period, argues that the wide variety of language and writing systems of Chinese Bible translations has offered the large Chinese population, including those with little literacy, with the possibility of reading the Bible. Church Romanization developed by Western missionaries has also influenced the movement of the Romanization of Chinese which became the foundation of pinyin. The Bible translations in Imperial Mandarin and local dialects have contributed to the concept of “consistency between speaking and writing” (yanwen yizhi 言文一致) which helped the generation of a “Chinese new literature movement” in the May Fourth Movement by involving Vernacular Chinese as the formation of the “newness” and the “Chineseness.” Many leading thinkers and writers of modern Chinese literature, such as Lu Xun, Shen Congwen, Guo Moruo, Ba Jin, Xu Zhimo, Lin Yutang, Zhou Zuoren, and Yu Dafu, have quoted the Chinese Bible, and have reviewed and introduced the Chinese Bible to Chinese people. Many of them have acknowledged that their writings and their thought have been inspired and influenced by the Chinese Bible (Jiang 2001: 301–305).

8 Conclusion

In this paper, I have shown how the debates of Chinese Bible translation can support Ricoeur’s hermeneutical dialogical translation theory and his criticism of the ideal of perfect translation. I also argue that a few phonetic transcriptions of religious terms in Chinese Bible translations were indeed driven by the aspiration for a perfect translation expressed as a form of escapism. By giving up the aspiration of perfect translation, what we can achieve in translating certain religious terms is “an equivalence without identity.” This is not only because of the existence of the multiplicity of languages, but also because of the problem of polysemy of philosophical religious terms whether in source or targeted languages. After reviewing the struggles, debates, and controversies of translating certain philosophical religious terms in the history of Chinese Bible translation, I argue that missionaries and translators have exemplified the paradigm of what Ricoeur calls “linguistic hospitality.” The debates also demonstrate that there exist certain paradoxes of linguistic hospitality when translation occurs in a culture containing diverse traditions. The works of the Chinese Bible translations in the past have not only brought to us a translated Bible which is comprehensible to Chinese readers, but also the construction of certain new Chinese terms and concepts that did not exist before. The results of Bible translations have also contributed to the movement of cultural renovation in modern China. These results have attested to Ricoeur’s idea of semantic cultural innovation.