PARADOXES of TRANSLATION by EA nida

APRIL 1991) UBS TRENDS IN TRANSLATING in the International Federation of Translators, the American Translators Association, and university programs in a number of countries When the United Bible Societies began, fully 90l1Jo of Bible translations in the Third World were being made by missionaries with the help of informants or translation helpers Now in 90l1Jo of the projects the translators are nationals, and missionaries have become the resource persons During the last forty years there has been a considerable expansion in the use of different disciplines as a means of highlighting problems and providing more satisfactory answers to difficulties of functional equivalence During the first few years the focus was on the contribution which linguistics could make to the problems of equivalence, but it was soon evident that sociolinguistics (the science of language in use) could provide somewhat more crucial information about the role of language in society This led inevitably to the use of communication theory, with its emphasis on source, message, receptor, feedback, noise, and setting All of this resulted in a concern for the implications of sociosemiotics as providing a broader framework for the understanding of equivalence, especially in the area of rhetoric and style Sociosemiotics has in turn focused attention on the indeterminacy of language; as Edward Sapir used to say, "All languages leak." It is in this area of the unpredictability of creativity in figurative meaning that language exhibits its greatest impact Such matters become especially crucial for Bible translators since fully 90l1Jo of the key vocabulary of the Scripture is figurative EUGENE A NIDA THE PARADOXES OF TRANSLATION Dr Nida is Consultant to the UBS Subcommittee on Translation No human activity seems so paradoxical as translating As T.A Richards (1953) has said, translating is "probably the most complex type of event yet produced in the evolution of the cosmos," and yet it is also completely natural for almost any bilingual (Harris and Sherwood, 1978) and especially so for children, particularly before they have gone to school and have learned about nouns, verbs, and prepositions Despite the fact that tens of thousands of pages of material are being translated every day throughout the world and the results of such translations "make the world go round," some translation theorists have argued in all seriousness that translation is impossible, although these same scholars seem to be quite content to have their books translated and their royalties increased (Giittinger, 1963) Some translators think that faithfulness to the original text requires them to drag the reader back to the original by torturing the receptor (or THE BIBLE TRANSLATOR (VOL 42 NO 2A target) language in such a way as to reveal how different, strange, and unique the original text really is Wherever possible they insist on wordfor-word renderings, and they compensate for any loss of meaning by numerous notes to make sense of the text Other translators, however, insist that priority must be given to meaning and that a translation should not disguise the intention of the original, but reveal it Any important formal features of the original which cannot be adequately reflected in the translation can then be dealt with in marginal notes These contradictory, and even paradoxical, views of translating are nowhere better illustrated than in the rendering of religious texts, where word-worship may satisfy the spirit, but can numb the mind Islam assumes the radical position, namely, that the Koran should not be translated at all Those who truly wish to worship Allah must so with the very words of the Koranic text Some Christian confessions have at various times accorded translations an almost koranic status and have insisted on using Latin, Old Syriac, and Old Church Slavonic far beyond the time that these translations were meaningful to people in general No doubt, for some worshipers the mystery of faith has been strengthened by the mystery of the language, but for most people the unintelligible words have been equated with an irrelevant message, and the "good news" has become no news Despite the fact that the essence of religious language is to reveal what has been hidden in the past, some translators are reluctant to put the message of the New Testament into the same kind of "common language" as occurs in Koine Greek The Gospels were not written in an academic form of Hellenistic Greek, but in the widespread form of Koine, which overlapped between the literary and colloquial levels Some translators seem fearful of making the Scriptures as meaningful in a modern language as they were in Greek or Hebrew Certain words seem to have acquired such an aura of holiness and some grammatical constructions appear to be so sanctified by tradition as to be untouchable, even though most readers and hearers either not understand or misunderstand the divine revelation Some years ago I was lecturing to a group of Bible translators in Hungary about principles and procedures of translation, but there was strong resistance to the idea that translating means communicating Members of the New Testament committee felt certain that to adopt principles of functional equivalence (translating a text into the language of a receptor culture in such a way as to approximate the impact and appeal of the original text in the source culture) would rob the sacred text of its linguistic and sacred uniqueness Fortunately, the Secretary of the Communist Party of the National University was present for the discussions, and he came to my rescue by insisting that all Hungarians ought to be able to read the message of the New Testament in a form of language which would not only be fully intelligible but also aesthetically pleasing by exploiting the rich literary resources of Hungarian APRIL 1991) PARADOXES OF TRANSLATING The Paradoxes of Language The paradoxes or translating should not, however, come as a surprise, since they are all derived from the paradoxes of language itself Different languages seem to be so utterly diverse in sounds, words, and grammar, that statements by linguists to the effect that all languages are at least 90070 alike appear to be nonsense to most people Linguists admit that each language has its own distinct set of sounds, but whether there are a dozen or sixty different phonemes (distinctive sets of contrasting sounds), they are all systematically organized in very much the same way They are not randomly thrown together out of the hundreds which the human voice can make, but they are organized in contrasting sets, as Trubetskoy (1939) and his colleagues in Prague so effectively discovered and described Words of different languages may be quite distinct, but all words in all languages are constructed from minimal units (called "morphemes") in a strictly limited number of ways, that is, by compounding, affixation, reduplication, substitution, and reduction And though the grammar of one language never matches precisely the grammar of another, they all have propositions of activity, identity and class and employ ways of restricting the meaning of words by qualifiers, quantifiers, and expressions of time and space Then they bind the whole together into hierarchical structures of binary parts by rules of government and agreement Even on the level of rhetoric (or discourse structure) all languages employ figurative expressions and, as will be noted below in a later section, they use the same discourse building blocks to form the scenarios and schemata of narrative, descriptive, expository, and dialogic texts When studied only in terms of surface phenomena, languages are very different, but their underlying structures and their principles of organization are essentially similar and universal This organizational similarity of languages, as against surface diversity, is not only one of the major paradoxes of language, but also one of the basic reasons why translation is possible In fact, in all the work that translation consultants of the United Bible Societies have done in helping translators on several hundred languages, there has not been one case of a biblical expression which completely defied translation into another language What can be said in one language can be said in another, but not always with the same degree of efficiency For example, many languages not employ comparative forms of adjectives followed by the compared standard, e.g., Bill is taller than John But they can and make comparisons by saying, "Bill is tall; John is not tall." The form may be different, but the meaning is the same Another paradox of languages is the fact that they seem to be so rigid and rule-governed, with long lists of grammatical rules and welldocumented dictionaries specifying what words should mean and how they are to be used, but one soon finds that some of the best writers constantly break the rules and dictionaries are usually about a generation behind the times For centuries English poetry had to have rhyme or it was not regarded as poetry, but now rhymed poetry is often considered to be THE BIBLE TRANSLATOR (VOL 42 NO 2A artificial, naive, and even childish For most people one of the most unexpected features of language is the fact that the transmitting of verbal information is a relatively limited function of language The most common use of language is for thinking (the cognitive function), and another major role is for self-expression (the expressive function), e.g cursing, shouting for joy, writing poetry, playing games, and singing love songs and lullabies Language is also used for giving commands (the imperative function), for changing someone's status (the performative function) in sentencing a criminal, in marrying a couple, or in pronouncing absolution Language is also used to change the emotive state of receptors (the emotive function), in which demagogues and politicians have always been particularly skilled, since they are trained to make maximally favorable impressions by communicating a minimal number of promises And finally, there is the interpersonal function of communication, by which people establish and maintain interpersonal relations by the proper use of at least five different registers of language (Joos, 1962): ritual, formal, informal, casual, and intimate Paradoxically, it is this interpersonal function of language which creates so many difficulties in translating the Scriptures into some languages, e.g Japanese with its five different levels of first-person singular reference, Korean with three levels of politeness, and Thai with politeness levels involving both grammar and lexicon Many Africans are shocked and surprised at the manner in which literal translations render the disciples' questions and statements made to Jesus The blunt boldness of the biblical texts makes the disciples appear both discourteous and uncouth Perhaps the most paradoxical aspect of language is the fact that with a limited number of significant sounds, a restricted number of grammatical constructions, and a relatively limited inventory of words (usually not more than 50,000 in the active vocabulary of the average university graduate), it is possible to speak and write relevantly about the infinite number of objects, events, features, and relations in the real world, which is made up of unique phenomena Languages have to be the way they are, since we could not efficiently manage hundreds of verbal sounds They would be too close together in acoustic space and thus misleading We couldn't possess a separate word for every distinct object in the universe, nor could we employ a grammar which had a different construction for every possible relation between entities, actions, states, and features The codes of languages have to be sufficiently rule-governed to be easily internalized and practically automatic-in fact, we sometimes speak before we think But life is constantly changing, even as Heraclitus insisted, and languages must grow and change with culture, which they largely reflect and of which they are a part Special Problems in Translating Biblical Texts Translating the works of a living author is generally much easier than translating a biblical text If a translator is in doubt about what a presentday author meant, it is often possible to write a letter or telephone The APRIL 1991) PARADOXES OF TRANSLATING Bible translator however, is separated from the biblical text by 2,000 to 3,000 years, and the temporal gap only highlights the cultural gap But in comparison with the typical European or North American, most West African villagers can more easily "see and feel" the meaning of the Old Testament text when they read about tribes, sacrifice, ordeals, stoning, holy wars, places of refuge, levirate marriage, swearing by the genitals, idol worship, the sacred box of the covenant, ritual uncleanness, demons, witches, and prophets in ecstatic trances The words of Jesus are usually clear enough, but much of what Paul wrote sounds like a very different world-which it is Not only is the culture gap a large one, but the Old Testament contains linguistic problems which add serious complications, since almost ten percent of the vocabulary occurs only once These hapax legomena are often tantalizingly difficult to analyze, and there is so little help from cognate languages, and such help that exists can be misleading, even as cognates in European languages constantly remind us, e.g., English knight and German Knecht, English demand and French demander, English assist and Spanish asistir Another problem is the wide variety of literary genres in the Scriptures, e.g., parables, riddles, allegories, laws, proverbs, legends, history, dramatic biography, wisdom literature, letters (personal and general), apocalyptic, genealogies, anthologies, poetry (philosophical, didactic, lyric, ritual), sermons, and prayers At the same time there are vast differences in the levels and qualities of language Compare, for example, the stark simplicity of Mark with the elaborate Greek of First Peter; or the pedestrian Hebrew of the Chronicles in contrast with the sublime language of Job The interpretation of the Greek and Hebrew texts is complicated by the long and often obscure history of production, redaction, and canon formation And it is not easy to determine where literary criticism leaves off and textual criticism begins Furthermore, different ecclesial confessions differ widely in their approaches to meaning Is the correct meaning to be found in the Ugaritic parallels to the Psalms, or in the ways in which the Psalms were used in the temple worship by the believing community, or in the manner in which present-day churches have found spiritual strength and renewal? Since Christendom has for the most part insisted on the uniqueness of the historical revelation contained in the New Testament, there has generally been a keen desire to go back to the closest documentary equivalence of the original autographs This has meant long and arduous work on thousands of manuscripts and fragments in an effort to ascertain the earliest and most authentic form of the text This concern is reflected in large measure in the United Bible Societies' sponsorship of the Greek New Testament text with some 1,400 significant differences of meaning in diverse textual traditions But these are relatively few in comparison with more than 5,000 passages in the Hebrew Old Testament text on which translators need guidance The results of a lO-year study by an 10 THE BIBLE TRANSLATOR (VOL 42 NO 2A interconfessional committee of Old Testament scholars is just now being published in a series of five large volumes But decisions about canon and text cannot be adequately answered without considering certain important issues For whom is the revision or new translation being prepared? Is it for well educated people? Or is it for those with limited educational backgrounds and very restricted knowledge of the Scriptures? Furthermore, what are the expectations of future readers as to the kind of text regarded as appropriate? And how is such a text to be used? Is it for liturgical worship, for personal reading, or for catechetical instruction? Since communication always moves in the direction of the audience which must understand and appreciate a translated text, one must take seriously the issue of the receptors One girl who first read Today's English Version, exclaimed to her mother that it could not be the Bible because she could understand it Unfortunately, the idea that the Scriptures are not to be understood is far more common than most people would like to believe But there are still other questions which need to be asked By whom should a new translation or revision be undertaken? What should be their qualifications in biblical languages, pastoral concerns, stylistic proficiency in the receptor language, and capacity to work together with others as a team? And finally, how is the translation to be tested for adequacy? Will there be reviewers and consultants to go over the text and suggest improvements? Will sample tentative publications be put out to test general reactions? Will specialists employ various testing procedures, e.g Cloze technique, evaluation of difficulties in oral reading, analysis of hearers' responses about content, and studies of facial gestures for interest indices In order to find answers to these questions and some of their paradoxical implications, we need to ask an even more difficult question, "What is translating?" What is Translating? Typical answers to this fundamental question about the nature of translating have traditionally described it as a "skill and an art." Some people have claimed that it is no more than "interlingual communication," and still others would like to regard it as a "science." In order to understand the essential issues involved in such differences of opinion, it is first necessary to distinguish clearly between the actual process of translating and the study of the process and its results The actual process of translating involves both skill and art if it is to be done well But the critical analysis of the process and the resulting texts can be regarded as a science in the broad sense of the term, since any human activity can be studied scientifically with clearly defined methods and rigorous criteria for determining comparisons and contrasts The actual process of translating is perhaps best described as a technology which employs the insights and principles of a number of branches of behavioral science in order to accomplish its goal of effective interlingual communication Translating may be regarded as a type of language engineering which draws upon linguistics, cultural anthropology, APRIL 1991) PARADOXES OF TRANSLATING 11 communication theory, sociolinguistics, psychology, sociology, and sociosemiotics to build bridges for communication between societies In fact, bridge-building may be a very close analogy to translating Those who build bridges across rivers must draw on various branches of physics, mathematics, and aesthetic design if the results of their efforts are to stand up to the load requirements and to satisfy the aesthetic tastes of those who want something better than most railroad bridges Similarly, translators must employ the structural principles of all the closely related sciences dealing with human behavior, as well as be concerned with the artistic factors of design, especially if a translation is to be more than a clumsy caricature of the original Because of the enormous differences in literary genres, types of content, diversities in linguistic categories, standards of equivalence, and aesthetic tastes, no one has produced a comprehensive theory of translating which has won wide acceptance, although a number of scholars have made important contributions to our understanding of many problems and solutions These will be noted at various points in the subsequent sections, but for general treatments of the history and principles of translating the following bibliographical references should be helpful: Brower (1959), Mounin (1963), Nida (1964), and Wilss (1988) Although one may view the process of translating from a number of different perspectives, e.g author, transmission of the text, history of interpretation, content of the message, capacities of the translator, differences and similarities between source and receptor languages, receptors, ways in which translations are used, and the effect of a translation on a developing literature in the receptor language, there are four major perspectives: philological, linguistic, communicative, and sociosemiotic, which include most of the diverse theories and sub-theories which people have proposed as ways to explain the nature of the translation process Only by examining these major perspectives can one appreciate fully the built-in paradoxes of translating and of theories about translation These four different perspectives should not, however, be regarded as competitive or contradictory, but rather as complementary and supplementary They do, however, reflect a historical development which parallels the shift of attention in communication from the intent of the source to the interpretation by receptors The Philological Perspective The philological perspective on translation in western Europe goes back primarily to some of the seminal observations by such persons as Cicero, Horace, Augustine, and Jerome, whose principal concerns were the correct rendering of Greek texts into Latin Jerome's translation of the Vulgate text of the New Testament was especially important in that it broke away from the earlier Old Latin tradition of literal rendering of the Greek text, but his boldness caused him no end of trouble In eastern Europe the most influential development was the pioneering work of Cyril and Methodius, who together with colleagues not only translated the Bible in the ninth 12 THE BIBLE TRANSLATOR (VOL 42 NO 2A century into what is called "Church Slavonic," but also developed a new alphabet, largely based on Greek Several centuries earlier, however, outstanding translations of the Scriptures had already been made in the Armenian and Georgian languages At the time of the Renaissance some of the most important early contributions to principles of translation were made by Martin Luther (1530) and Etienne Dolet (1540), and in the western world the issue of this philological perspective is still very much alive, as evidenced by the contributions of such persons as Cary and Jumpelt (1963), Octavio Paz (1971), George Steiner (1975), and John Felstiner (1980) The philological perspective on translating has continually focused attention on faithfulness to the intent of the author of a text, and criteria about the adequacy of a translation have almost always been stated in terms of the degree to which a translation reflects both the content and the form of the original, with some people arguing that, when necessary, form should be sacrificed for the sake of the content, and others claiming that content should be sacrificed for the sake of preserving the form The result has been swings of fashion from one extreme to another, and Bible translations have been caught in the middle Compare, for example, the following translations of John 1.1 made during this century: American Standard Version (1901) In the beginning was the Word, and the Word was with God, and the Word was God James Moffatt (1922) The Logos existed in the very beginning, the Logos was with God, the Logos was divine J B Phillips (1960) At the beginning God expressed himself That personal expression, that word, was with God, and was God Today's English Version (1966 edition) From the very beginning, when God was, the Word also was; where God was, the Word was with him; what God was, the Word also was New English Bible (1961) When all things began, the Word already was The Word dwelt with God, and what God was, the Word was Living Bible (1971) Before anything else existed, there was Christ, with God He has always been alive and is himself God Discussions of translating based on the philological perspective have tended to be too heavily subjective and almost wholly lacking in definite APRIL 1991) PARADOXES OF TRANSLATING 13 criteria by which meaningful discussion could proceed More recently, a quite new approach has been adopted by the journal Translation Review, published by the University of Texas at Dallas on behalf of the American Literary Translators' Association In this journal theoretical discussion has been largely abandoned, but the experience of outstanding literary translators is creatively explored largely through personal interviews The composite experience of outstanding professionals in the field provides a much more valuable source of insight on philological issues than all the handbooks of do's and don'ts and all the heavy tomes dedicated to the pros and cons of various translations The Linguistic Perspective While the philological approach to translating has focused on the form and history of particular texts, the linguistic approach has concentrated on the respective languages, and what the philological perspective may lack in providing methodological principles and objective data for testing, the linguistic perspective supplies in providing quite new approaches to the meanings of lexical units (words and idioms), the functions of grammatical constructions, and the organization of texts (also spoken of as "discourse structure," "rhetoric," and "poetics") Words and Idioms Perhaps the most important insight to be gained from the study of the meaning of words and idioms is that the lexical units are not like labels on a map, but are designations of areas in the cultural terrain of a society's experience Furthermore, these words cover the entire cultural realm and in fact in most instances they cover it several layers deep, with highly generic words such as thing, matter, object, event, action, feature, and state in contrast with many specific words such as collie, mosquito, housefly, oak, rose, chimpanzees, dance, and gyrate, and various semantically intermediate lexical items in between, as in dog, insect, tree, plant, animal, move about Trying to define the area of experience covered by a term means considering the areas of meaning of other closely related terms For example, how thick does a string have to be before it is called a cord, and how thick and heavy does a cord have to be before it is called a rope? People not differ appreciably in the case of typical tokens, but the margins between such types as string, cord, and rope are often uncertain and fuzzy In order to define the meanings of words, it is essential to study them in semantic domains, that is, with other words relating to the same areas of experience For example, in Hellenistic Greek there are several different terms for population centers: polis, komopotis, kame, and agros, often translated as 'city,' 'town,' 'village,' and 'hamlet,' respectively, but size was not the only distinctive feature There were other relevant features, e.g legal status, control of surrounding territory, existence of markets, and possibly a wall or fortification 14 THE BIBLE TRANSLATOR (VOL 42 NO 2A Most semantically related sets of meanings not constitute such obvious clusters with relatively clear distinguishing features Many of the relationships are overlapping (partial synonymity), e.g., agapao 'to love' and phileo 'to be fond of.' Others are included, e.g., katakrino 'to condemn' and dikaioo 'to acquit' are included within the range of krino 'to judge' Some meanings are reversives, e.g., deo 'to bind' and luo 'to loose,' and others are contrastive, e.g., agathos 'good' and kakos 'bad,' while still others involve different complementary procedures, e.g., agorazo 'to buy' and poleo 'to sell.' Some sets of meanings involve merely series e.g days of the week and numbers For the study of different types of semantic structures, see Semantic Fields and Lexical Structures (Lehrer, 1974) and Componential Analysis of Meaning (Nida, 1975) In contrast with these sets of closely related meanings, the different meanings of a single word are usually quite far apart in semantic space Note, for example, the different meanings of the Greek term pneuma: 'Holy Spirit,' 'spirit,' 'evil spirit,' 'ghost,' 'inner being of a person,' 'way of thinking,' 'wind,' and 'breath.' Sometimes the different meanings of a word form chains of related meanings, but more often they constitute (1) galaxies, with a central meaning and a number of peripheral meanings dispersed in a number of different directions, or (2) constellations, groups of meanings without any central meaning to which they are all related In dealing with the different meanings of a term, it is essential to distinguish between meaning (also spoken of as "analytical meaning") and reference (contextual meaning") The Greek term aiteo means 'to ask for something on the basis of presumed need,' but in some contexts it is used of asking from God, and hence can be translated into English as 'to pray' (e.g John 16.26) One may say, therefore, that the specific reference in a text (contextual meaning) such as John 16.26 is 'pray' but the meaning (analytical meaning) is 'to ask for.' It is often helpful to distinguish clearly between four different types of words: (1) those with unique referents, e.g proper names and titles (when used within specific contexts), (2) those with multiple referents, socalled "common words," e.g., boy, tree, road, house, run, eat, before, under, good, lazy, (3) those which refer only to contextually identifiable entities, actions, features, etc., and called "pronouns," e.g., this, it, he, and (4) markers, which only serve to mark other words or phrases, e.g English to, which marks infinitives, and Greek kai, which in the Gospel of Mark often simply marks the beginning of a new sentence In Luke the phrase kai egeneto 'and it happened' often marks a new section in the discourse Lexical semantics is concerned with two quite different kinds of structures: (1) single words and easily isolatable affixes, although often morphologically quite complex, and (2) idioms, combinations of words whose combined meaning cannot be directly derived from the meanings of the parts, e.g 'heap coals of fire on the head' (Romans 12.20), 'mouth to mouth' (2 John 12), 'male opening the womb' (Luke 2.23) These idioms have their own internal grammatical structure, but as far as their meaning APRIL 1991) PARADOXES OF TRANSLATING 15 is concerned, they must be treated as single units Some words may appear to have quite simple meanings and yet be semantically quite complex, since they belong to more than one semantic domain For example, the English term dancer belongs to the class of entities, since it refers to a person, but also to the class of actions, since it refers to the activity of dancing Such a distinction becomes specially important in certain grammatical constructions For example, in the statement she is a good dancer, the term good refers specifically to the component of activity of dancing, and not to the component of being a person Some single words may have extremely complex internal semantic structures For example, the English word heir may be said to mean 'a person who has or will receive something of significant value, presumably at the death of the one making the gift.' In some languages a term seeming to correspond to 'heir' has as an obligatory feature the death of the donor, so that a literal translation of heirs of God would imply that God had died The most serious problem encountered by anyone trying to determine the meaning of Greek or Hebrew biblical terms is the seemingly impossible number of alternative glosses For example, the Greek term charis has been translated into English in different passages as 'grace,' 'blessing,' 'generous service,' 'privilege,' 'gift,' 'credit,' 'special task,' 'grateful,' 'thanks,' 'favor,' 'generosity,' 'goodwill,' 'gracious work,' 'service of love,' 'respect,' 'love,' 'presence,' 'pleasing manner,' 'gracious,' 'beautiful,' and 'eloquent.' These are all legitimate references (contextual meanings), but they are not all analytical meanings A study of the distinctive features of this set of contextual meanings indicates that there are essentially only six meanings: (1) a favorable attitude toward someone or something (e.g Luke 2.52 and Acts 2.47); (2) the act of showing kindness to someone, with the implication of a gracious attitude on the part of the one showing such kindness (2 Corinthians 8.7); (3) that which is given generously and freely (1 Corinthians 16.3); (4) a state of thankfulness and gratitude (1 Timothy 1.12); (5) an expression of thanks (also a possible meaning of Timothy 1.12); and (6) attractiveness in form and/or manner (Luke 4.22) These meanings of charis are unusually close, and they illustrate a very serious problem in translation, namely, the difficulties posed by presuppositions and inferences Meaning 2, the act of showing kindness, has a presuppositional feature of a favorable attitude, otherwise, no act would take place At the same time, Meaning is an inference of Meaning 2, in that an act of kindness implies a result, either some tangible gift or a state of well-being in the benefitted party In some contexts it is quite simple to decide whether the focus is on the attitude, the act, or the resulting state, but often it is not possible to so Furthermore, in speaking about the grace of God it may not be necessary to specify a particular aspect of charis, despite the desire of some theologians to make a distinction All three aspects may be involved and in various proportions Splitting theological hairs may be both arduous and misleading The multiple meanings and fuzzy semantic boundaries of lexical units 16 THE BIBLE TRANSLATOR (VOL 42 NO 2A can be very frustrating for the lexicologist, but it is the price that must be paid for the intriguing capacity of language for growth, figurative usage, calculated vagueness, double entendres, meaning more than is said, and saying more than is meant Instead of the meanings of words fitting the cultural experience like a well organized mosaic, they are much more like multilayered bubbles floating on a moving current Syntax Some linguists have thought that syntax is the primary carrier of meaning, since it is so closely related to propositional logic They have proposed that the words can simply be plugged in whenever they are necessary as the various transformations generate sentences But this simply does not work, since the meanings of words and syntactic constructions are intimately bound together Perhaps the best way of dealing with this overlapping of semantic and syntactic types of structures is to recognize the five semantically distinctive classes of lexical units: entities, activities, features (including states), relations, and markers, since these are crucial in indicating the way in which important syntactic relations penetrate the lexical level For example, in the phrase his' 0ld trusted? servant", his is a pronominal substitute for an entity (E) The adjective old is a feature (F), trusted is semantically complex and consists of a feature (F), but at the same time of an activity (A), namely, that someone trusted somebody Finally, servant (like dancer cited above) is also complex and consists of an activity (A) and an entity (E) Semantically this phrase his old trusted servant is not simply a noun preceded by three attributives The relations are much more complex, and in a number of languages one should not translate this phrase literally In the first place, the relation between his and servant is reciprocal: He hires a servant, and the servant serves him In other words EI (his) commands A4 E4 (servant), but E4 does A4 (i.e., serves) E1 (his) The adjective old (F2) may be a feature of E4 (the entity of servant), meaning that the servant is old But F2 may also refer to the activity of serving, as taking place formerly and/or for a considerable period of time The p3 component of trusted is a feature of E4 (servant), since it designates a quality of personality, but at the same time it is derived from the fact that EI (his) has learned to trust E4 (servant) Two of the most tantalizingly difficult syntactic constructions of the Bible are the construct constructions in Hebrew and the genitive constructions in Greek So many of these constructions, which become in English two nouns combined by of, seem to be similar, while in reality their semantic relations are quite different In the phrases "your labor of love and endurance of hope" (l Thessalonians 1.3) the relations between love and labor and between hope and endurance are quite different In your labor of love, you labor because of your love, or your labor shows that you love But in your endurance of hope, your hope endures In "a hearer of forgetfulness" (James 1.25) someone hears and promptly forgets, APRIL 1991) 17 PARADOXES OF TRANSLATING and in "obedience of faith" (Romans 1.5) some from all the nations believe and obey In both "the peace of God" (Philippians 4.7) and "the God of peace" (Philippians 4.9) it is God who causes peace, but in "the glory of God" (2 Corinthians 4.15) it is God who is glorious, and in "the righteousness of God" (Romans 3.21) it is what God does in declaring or making righteous In the New Testament there are at least twenty-five different relations marked by the genitive construction Romans 1.5 illustrates in a very effective manner the implications of an approach to meaning which combines syntactic and lexical elements: di'1 houEs whom= Christ elabomen' A+Es A = to experience Es=we (epistolary I) kharirr' A+F A=act F = kind attitude of E S2 apostolen'' eis7 A+E R A = activity purpose E = of an apostle hupakoenf A to obey pisteos? A to believe ethnesin-? E nations hyper'" tou" R F on behalf of definite R intermediate agent pasin!' tois'? F F all definite kaiS R apposition en'? R within, among onomatos'" autou'? Ev Es name = person E s 17 = E S2 (figurative = E S17 ) In the above analysis of Romans 1.5 the superscript numbers only identify the succession of words, and the subscript letters and signs have the following values: A activity E entity Es entity represented by a pronominal or affixal substitute Ev a verbal entity F feature R relation + indicates more than one relevant semantic class Before describing the various syntactic relations, it may be necessary to point out certain features of some of the lexical units For example, elabomen is essentially a pseudo-active periphrasis for the passive meaning 'to experience,' so that the personal ending -men (literally 'we' but the epistolary equivalent of'!') is the experiencer of the act in kharin 18 THE BIBLE TRANSLATOR (VOL 42 NO 2A Similarly, apostolen may be interpreted as 'being an apostle' rather than as 'the activity of an apostle,' but the focus seems to be more on the role than on the status The various constructions in Romans 1.5 may be stated in terms of the most evident relations between agents, experiencers, features, and activities: E S2 (Christ) causes E S3 (me) to undergo A3 (an experience) E S2 (Christ) does A4 + F4 (the act of kindness) for the benefit of E S3 (me) E S2 (Christ) causes E S3 (me) to perform A6 +E6 (activity of an apostle) A6 + E6 (activity of an apostle) belongs to the class of or is the result of A4 = F4 (an act of kindness, a privilege) RIO - EI3 (in all the nations) qualifies an implicit reference to people as part of E13 RI - A6 + E6 happens (R7) so that EI3 (people of the nations) will A9 (trust) and A8 (obey), with the persons trusted and obeyed being unspecified, but either Christ or God E S3 (I) A + E6 (work of an apostle) for the sake of Ev 16 = E Sl7 (Christ) One may also interpret the relation between A and A9 as A9 qualifying A8, in other words, obedience based on trust, but this is only a nominal transform of the underlying verb expressions involving trusting and obeying On the basis of the contructions in 1-17, one may legitimately render this clause in English as "through whom I experienced the privilege of doing the work of an apostle for the sake of Christ in order that people in all nations may trust and obey God." Linguistics is by no means the answer to all the problems of exegesis, but it may serve as a useful tool in carefully defining many of the subtle and often obscure relations In other words, linguistic techniques may help in making clearer and more precise the interpretive alternatives Discourse Structure In contrast with the lexicon, in which the relations are between lexical units and referents, and the grammar, in which the relations are between words and between groups of words, the discourse structure consists of more or less elaborate patterns-what might even be called in some cases "verbal tapestries" with numerous formal and semantic processes employed to produce both impact and appeal Whereas the number of different types of word formations and syntactic constructions are relatively quite limited in order for languages to become more or less automatic, the varieties of discourse structures are practically unlimited The result is an amazing variety of discourse types or genres, often described in terms of scenarios, schemata, and frames But there are many hybrid types with seemingly contradictory and arbitrary features These are, however, an expression of the creativity of language usage and have been effectively described by APRIL 1991) PARADOXES OF TRANSLATING 19 a number of persons, e.g Jakobson (1970), Grimes (1972), and Traugott and Pratt (1980) Despite the enormous variety of discourse types, there are a limited number of primary and secondary organizational principles The primary principles involve organization based on (1) time e.g personal accounts, stories, sets of instructions, lists of procedures, etc., (2) space, e.g descriptions and certain mythological accounts, (3) class, e.g additive (and) alternative (or), subtractive (except, but), and duplicative (apposition, tautology) The secondary principles, which are structural elaborations of the primary principles, involve organization based on (1) consequence, e.g., condition, concession, purpose, result, cause-effect, basis-assumption, (2) rank, e.g taxonomic (generic/specific), degree (ascending or descending series based on value, power, size, cruciality, etc.), and (3) dialogic, e.g positive/negative, question/answer, similarity/contrast, old/new formation, intertextuality (parodies, quotations, allusions), and streams of consciousness Many translators regard these organizational principles of discourse structure as being irrelevant to the task of translating, since one is obliged to reproduce what is in the source language But in many instances a text will be seriously misunderstood, if a translator does not introduce certain modifications This is particularly true in the case of flashbacks, which introduce information about a prior event Unless these are clearly marked or the order of clauses and sentences altered, receptors may completely misunderstand what is happening In Greek the relations of consequence between clauses and sentences is almost always clearly marked In fact, certain stylistic canons required that every sentence begin with a conjunction (usually in postposition), but some of these conjunctions in Greek are so semantically empty that they simply mark syntactic units and are often better left "untranslated," in which case their real function is more accurately reflected in a language such as English But in Hebrew the tendency is to employ considerable parataxis and to leave to the context an understanding of the semantic relations between sentences and clauses Accordingly, in translating the Old Testament it is often necessary to add conjunctions Although Greek employs conjunctions extensively to mark sentence and clause units, it generally does not indicate the logical relations of participial constructions to the rest of a sentence or clause As a result a translator must often introduce appropriate conjunctions Preference for direct or indirect discourse may also pose a problem, especially in those cases in which there may be several levels of direct quotation The shifting of some of these into indirect quotations may be essential if readers are to understand who is talking to whom Perhaps the greatest obstacle to effective handling of discourse structure in translation is the tendency to translate clause for clause or sentence for sentence The significant unit for translating is actually 20 THE BIBLE TRANSLATOR (VOL 42 NO 2A the paragraph, not the sentence, since it is the paragraph which should represent a meaningful unit In some languages the principle of organization according to rank is to proceed from the less important to the more important, so that the final position is preeminent But other languages may employ an opposite system, and this may require considerable adjustments in order As a means of achieving maximum impact and appeal, there are a number of formal and semantic processes or devices which languages employ These devices carry significance largely because they involve shifts in expectancies, that is to say, they are not the normal ways of doing things with words As a result they have greater impact because of their unexpectedness The principal formal devices involve order of words, repetition, measured units (e.g poetic lines), embedding (parenthetical statements and quotations), condensations (e.g the succinctness of gnomic utterances), and parallelism In many languages repetition seems to be a perfectly natural way to emphasize something, but in some languages the repetition of words and especially of propositions (as in 'liturgical poetry) is interpreted as an insult to the intelligence of readers or hearers, since it implies that the people could not understand and therefore had to be told the same thing a second time But translators are faced with even more problems because of the various semantic devices used to create impact and appeal These devices consist primarily of figurative language, plays on the meanings of words, paradox, irony, overstatement (hyperbole), understatement (litotes), euphemisms, rhetorical questions, and imperatives for conditionals The figurative statement "I will give you cleanness of teeth" (Amos 4.6) is generally not understood as a reference to famine, and rhetorical questions, which are really intended as emphatic declarations, are frequently interpreted as revealing the stupidity of the writer or speaker who is so uninformed as to have to ask for such information When imperatives are used for conditionals, the situation becomes even more difficult, as in the literal rendering of Psalm 4.4 and its quotation in Ephesians 4.26, "Be angry and not sin." This is clearly not an exhortation to be angry, but a condition, "Even if you get angry, not sin." In many instances the figurative significance of something is almost entirely overlooked For example, in Amos the repeated references to "three" and then "four" transgressions is simply a figurative way of indicating continuous transgressions In the Scriptures the number forty is symbolic of a considerable amount, and seven often implies perfection But in many instances the symbolism of numbers is lost when units of measurement are changed from one system to another For example, in Revelation 21.16-17 the change from 12,000 furlongs and 144 cubits to 1,500 miles and 216 feet obscures completely the significant number 12 and its multiples, so important in speaking about the twelve tribes, the twelve disciples and the twelve gates In addition to all the different formal and semantic rhetorical devices APRIL 1991) PARADOXES OF TRANSLATING 21 for increasing the impact and appeal of a text, a translator must also reckon with the issue of what the Roman rhetoricians called varietas, including not only the variety of devices, but their distribution in a text and their frequency In translating biblical poetry, it is frequently necessary to change a figurative expression into a nonfigurative one since the biblical expression depends for its meaning on some local custom or event But to lose too many of these expressions may so change the character of the poetry as to make it sound like prose It is possible, therefore, to compensate somewhat for the loss of some figurative expressions by changing nonfigurative language in the source text into effective figurative speech in the receptor language Most creative translators this very thing The Communicative Perspective Despite all that can be learned from both the philological and linguistic perspectives on translating, there is still much to be gained from viewing translating as a communicative process involving source, message, receptors feedback, setting, noise, and media A translator may think that the process of translating ends with the submission of a manuscript to the publisher, but in fact the process does not end until readers have correctly comprehended the message Success in the process of translating cannot be judged primarily by verbal correspondences, grammatical similarities, and stylistic equivalences, but by the extent to which intended and qualified readers can understand and appreciate both the form and the content One important source of insight concerning the communicative aspects of language have come from sociolinguists studying language as used by a society to accomplish its cooperative goals The work of Labov (1972) and Hymes (1974) have been especially important in this regard But another important source has been the rediscovery of Vygotsky's important views (1939) about the mental development of children and the role of language within the society, as a code developed by human society, transmitted by society, and learned within the context of society In order to justice to the communicative significance of a text, a translator needs to understand and appreciate those features of form and content which figure prominently in the original effectiveness and its corresponding impact and appeal to a receptor audience Those features which involve primarily formal elements are cohesion, proportion (balance between different sections), and prominence (relative degrees of emphasis), and those which involve primarily matters of content are relevance (important for receptors), coherence (the extent to which the content of the text fits the real or imagined world), and wholeness (a matter of both unity and completeness) Several features, however, involve both form and content: clarity, appropriateness (in terms of both setting and themes), and novelty (the extent of unexpectedness) Most of these features of form and content are largely determined by the source text, but a good many accommodations must be made in a 22 THE BIBLE TRANSLATOR (VOL 42 NO 2A translation if it is to communicate effectively The original text normally fits the world-view and value system of the source culture, but a literal translation may be radically misunderstood because of a very different world-view in a receptor culture The Sadduccees' argument with Jesus about the resurrection based on the account of a woman who gave birth to no child although she was married to seven brothers one after the other makes very good sense in the biblical culture practicing levirate marriage But in most African societies the question about "Whose wife will she be?" is simply absurd In the African world-view any woman who had apparently caused the death of seven husbands would surely be a witch and no one would want her under any circumstances Footnotes to explain the differences in cultural presuppositions are essential In many societies the poetic format of a discourse carries a great deal of meaning, so that in some cases the content of the book of Job gains great importance by being printed in poetic lines, which expressly validate the truth of the content But some translators of Job into English have been reluctant to use a poetic format, since they believe that as poetry the text would not be interpreted as literally true They argue that if something is in a poetic format, then it never really happened And when some socalled "primitive peoples" discover in Genesis that the names of Adam and Eve have symbolic values and when they read about a snake speaking, they immediately interpret the passage as symbolic and figurative and are surprised when outsiders insist that the words must be understood in a literal and factual sense So many problems arise in communication because people are unaware of the paralinguistic and extralinguistic signs and symbols which accompany a verbal message In oral communication the various paralinguistic features are quite obvious, e.g., quality of voice, high pitch, unusual pronunciation, foreign accent, hyper-correctness of enunciation, loudness, and rapidity of utterance And there are almost always some extralinguistic features, e.g., gestures, degree of eye contact, body stance, and distance between participants But in written communication there are also paralinguistic features, e.g., style and size of type, format, extent of orthographic correctness, and number and types of footnotes (too many scare people off), as well as extralinguistic ones, e.g., quality of paper, type of publication (e.g., popular magazine or academic journal), the reputation of the publisher, and the quality of the binding Increasingly the reputation of the translator becomes an added extralinguistic feature in communication For many people the most paradoxical aspect of this new perspective about communication is the fact that the words are really only a limited part of any communication Traditionally, the focus has always been on words The meaning of grammatical constructions has been taken for granted, and the impact of the rhetorical devices have largely escaped the average reader's notice When readers finally recognize the multiplicity of codes and the relative weakness of words in convincing a reader of the truth of what is said, they are often astonished to realize how these covert aspects of APRIL 1991) PARADOXES OF TRANSLATING 23 language have so extensively influenced their judgements about a speaker's sincerity, commitment to truth, breadth of learning, ethnic background, concern for other people, and personal attractiveness The Sociosemiotic Perspective The significance of a sociosemiotic perspective on translating results from the fact that semiotics is the most comprehensive discipline dealing with codes of all types, from the biological code of the DNA to the chemical codes of hormone messages, to the song codes of birds and the chatter codes of monkeys, to the most complex code of all, human language Language can, of course, be viewed as a cognitive construct: Saussure's langue in contrast with parole or Chomsky's competence in contrast with performance, as a system to explain how an ideal speaker-hearer would potentially use language But such approaches only explain the bare bones of the system and fail to provide the kind of insight which is necessary to capture language in action Translators need to appreciate those aspects of language which no rigid formal system can ever capture, e.g poetic symbolism, mixed metaphors, Freudian slips, associative meanings, the impact of rhetorical patterns, and verbal creativity They also need to be able to read between the lines, to evaluate what is not said, and to sense what a writer is trying to say as he or she stumbles around to find just the right expression To capture language in both its sociological and aesthetic roles, a translator first needs a realistic epistemology, such as Charles Peirce's (1934) concept of the sign, the referent, and the interpretant based on the system of signs and the interpreter's dialogic interaction with a text Essentially this means that signs are only defined by other signs in each language code No sign, therefore, can be interpreted in isolation from all the other signs within the same semantic domain, and the semantic range of each domain is determined in turn by other contiguous domains This is precisely why the Greek-English Lexicon (Louw and Nida, 1988), published especially for Bible translators, is based on semantic domains The fact that signs are always defined by other signs means that no sign can ever be defined in a complete and absolute way This may be discouraging to some people, but it is a welcome antidote to dogmatism, a release from intellectual rigidity, and a basis for conceptual creativity Peirce's concept of three levels of meaning is particularly relevant for the translator-exegete, since it makes possible insightful discussion For example, in the biblical expression "justified by his blood" (Romans 5.9), the literal meaning of the physiological substance "blood" would be a case of what Peirce called "firstness." But in this biblical context what is relevant is not the physical substance but the metonymic character of blood as a figurative reference to Christ's sacrificial death (Peirce's "secondness") And in the context of the eucharist blood becomes a symbol of participation and identification between the worshiper and the one who is worshiped (a case of "thirdness") This concept of semiotic levels also applies to larger units of a text 24 THE BIBLE TRANSLATOR (VOL 42 NO 2A For example, the firstness in the Story of the Prodigal Son, which should be called the "Story of a Father and Two Sons," is the actual account of the father and his two sons But there is an obvious "secondness" about this parable, since the three persons parallel the three participants in the setting described by Luke for this parable Jesus is speaking to a crowd consisting of so-called "sinners," who had rejected the institutional religion of Judaism and had become social outcasts But these persons "heard Jesus gladly" and had returned from the "far country" to the celebration of full acceptance by Jesus But the Pharisees who were also listening could never be reconciled to the grace of God as revealed in Jesus At the same time they had nothing but disdain for these religious outcasts who had become the followers of Jesus This second level of understanding of the parable would constitute Peirce's "secondness." So the triad of the father and the two sons and the triad of Jesus, the religious outcasts, and the self-righteous Pharisees form an essentially triadic symbol of God, repentant sinners, and religious legalists, who have never been able to accept God's offer of righteousness based on covenant trust Jakobson (1970, 1972) has dealt with a number of the semiotic aspects of meaning, especially in his treatment of metaphor and metonymy, which he likened to imitative and contagious magic respectively, and to certain parallels which he found in different forms of aphasia Some very useful treatments of the sociosemiotic implications of verbal signs can be found in Baron (1981), Krampen (1979), Merrell (1980), Eco (1979), and particularly for translators, in the astute observations of Halliday in Language as Social Semiotic (1978) Even within these four approaches to the processes and problems of translating there are certain inherent paradoxes In the philological approach so much attention has been focused on the form and history of texts that the readers who are to receive, interpret, and presumably appreciate such texts have too often been largely overlooked When some philologists have sought to redress this oversight, so much emphasis has been placed on "the role of the reader" that the text has often been divorced from its context and given an "autonomous existence." On the other hand those approaching the problems of translating from the linguistic perspective have so often been concerned with the structure of the respective languages that they have dismissed the aesthetic elements of style and discourse as being irrelevant or only marginal But those who have seen translation primarily through the lenses of communication theory have tended to downgrade the significance of the distinctive stylistic features of the text and the context of the source text At the same time the numerous new insights coming from sociosemiotics have been so varied and significant that many translators have been overwhelmed by the realization of all that they have overlooked in the source text, the linguistic media, and the readers' set of values and presuppositions about reality APRIL 1991) PARADOXES OF TRANSLATING 25 The Strangest Paradoxes It may seem incredibly strange that those who so vigorously claim to follow the one who is "the way, the truth, and the life" and who revealed so uniquely the grace of God and the truth about the Father, should seemingly be more concerned about the form of the verbal revelation than about its content Even when the revisions of Scriptures are more faithful to the best textual evidence and are more easily understood, they are often strongly denounced and even violently rejected, with copies burned or buriedalthough fortunately Bible translators are no longer subject to the same treatment Perhaps such reactions should not be regarded as strange, for as Jesus said, "those who have drunk old wine not want new wine" (Luke 5.39) Paradoxically, new translations are usually more readily accepted than revisions One reason for this is that people not know just where to begin to criticize a new text But what seems even more anomalous is that translations so often succeed more on the basis of publicity and marketing than on the basis of quality of language or truth of content Some people in this scientific age find it almost unbelievable that a book written between 2,000 and 3,000 years ago is even now being translated and revised in more than 1,000 languages and that each year at least one billion Bibles and publications containing selections from the Bible are distributed throughout the world in response to an increasing demand for a book which uniquely combines a message of realism and hope But the greatest paradox of all is how even technically inadequate translations-and no translation is ever perfect-have proved to be an immeasurable blessing to countless numbers of people In a truly paradoxical way the Spirit of God succeeds in translating the written Word into the lives of those who have learned to listen with reverence to the message about the living Word Bibliography American Standard Version of the Bible 1901 New York: Thomas Nelson and Sons Baron, Naomi S 1981 Speech, Writing, and Sign: A Functional View of Linguistic Representation Bloomington: Indiana University Press The Bible: A New Translation 1922 James Moffatt, translator New York: Harper and Brothers Publishers Brower, Reuben A (ed.) 1959 On Translation Cambridge: Harvard University Press Cary, E., and R.E Jumpelt (eds.) 1963 Quality in Translation New York: Macmillan Catford, J.C 1965 A Linguistic Theory of Translation London: Oxford University Press de Waard, Jan and Eugene A Nida 1987 From One Language to Another Nashville, Tenn.: Thomas Nelson 26 THE BIBLE TRANSLATOR (VOL 42 NO 2A Dolet, Etienne 1540 'La maniere de bien traduire d'une langue en une autre,' in Edmund Cary, 1955, 'Etienne Dolet.' Babel 1.17-20 Eco, Umberto 1979 A Theory of Semiotics Bloomington, Indiana: Indiana University Press Felstiner, John 1980 Translating Neruda Stanford, California: Stanford University Press Grimes, Joseph E 1972 The Thread of Discourse Ithaca, NY: Department of Modern Languages and Linguistics, Cornell University Guttinger, Fritz 1963 Zielsprache Zurich: Manasse Verlag Halliday, M.A.K 1978 Language as Social Semiotic Baltimore, Maryland: University Park Press Harris, Brian, and Bianca Sherwood 1978 'Translating as an innate skill,' in David Gerver and H Wallace Sinaiko, eds., Language Interpretation and Communication, 155-170 New York: Plenum Press Hymes, Dell 1974 Foundations in Sociolinguistics: An Ethnographic Approach Philadelphia: University of Pennsylvania Press Jakobson, Roman 1970 Main Trends in the Science of Language New York: Harper and Row _ 1972 'Verbal communication.' Scientific American 227.7281 Joos, Martin 1962 'The Five Clocks.' International Journal ofAmerican Linguistics 28, No.2, Part V Krampen, Martin 1979 'Profusion of signs without confusion.' Ars Semiotica 2.327-359 Labov, William 1972 Sociolinguistic Patterns Philadelphia: University of Pennsylvania Press Ladmiral, Jean-Rene 1972 La traduction Paris: Didier and Larousse Lehrer, Adrienne 1974 Semantic Fields and Lexical Structures Amsterdam: North-Holland Publishing Company The Living Bible 1971 Kenneth Taylor, translator Wheaton, Illinois: Tyndale House Publishers, Inc Louw, Johannes P and Eugene A Nida 1988 Greek-English Lexicon of the New Testament, Based on Semantic Domains New York: United Bible Societies Luther, Martin 1530 Ein Sendbrief von Dolmetschen Werke Malone, Joseph L 1988 The Science of Linguistics in the Art of Translation Albany, NY: State University of New York Press Merrell, Floyd 1980 'Of Metaphor and Metonymy.' Semiotica 31.289307 _ 1963 Les problemes theoriques de la traduction Paris: Gallimard Mounin, Georges 1976 Linguistique et traduction Bruxelles: Dessart et Mardaga The New English Bible 1970 Oxford and Cambridge: Oxford University Press, Cambridge Unversity Press The New Testament in Modern English 1960 J.B Phillips, translator London: Geoffrey Bles APRIL 1991) PARADOXES OF TRANSLATING 27 Nida, Eugene A 1964 Toward a Science of Translating Leiden: E.J Brill 1975 Componential Analysis of Meaning The Hague: Mouton Paz, Octavio 1971 Traducion, literatura y literalidad Barcelona: Ediciones Tusquets Peirce, Charles 1934 Collected Papers Cambridge: Harvard University Press Richards, T.A 1953 'Toward a theory of translating,' in Arthur F Wright, ed., Studies in Chinese Thought, 247-262 Chicago: University of Chicago Press Steiner, George 1975 After Babel: Aspects ofLanguage and Translation London: Oxford University Press Today's English Version of the New Testament 1966 New York: American Bible Society Traugott, Elizabeth Closs, and Mary Louise Pratt 1980 Linguistics for Students of Literature New York and London: Harcourt Brace Jovanovich Trubetskoy, Nikolai Sergieevich 1939 Principes de Phonologie Paris: Klinksieck Vygotsky, L.S 1939 'Thought and Speech.' Psychiatry 2.29-52 Wilss, Wolfram 1988 Kognition und Ubersetzen: Zu Theorie und Praxis der menschlichen und der maschinellen Ubersetzung, Tiibingen: Max Niemeyer Verlag MILDRED LARSON THE SUMMER INSTITUTE OF LINGUISTICS AND TRANSLATION The author is Vice-President of Personnel with the Summer Institute of Linguistics During the past 50 years the Summer Institute of Linguistics (SIL) has been one of the institutions which has made a major contribution to Scripture translation in the minority languages of the world I will first trace the theoretical developments within SIL and then turn to the more practical matters of the translation task-the programs themselves and, finally, mention some trends and projects Developing a theory of translation Forty years ago, when I first attended the summer training course of SIL at the University of Oklahoma, the only book on translation which we were assigned to read was Eugene Nida's 1947 book called Bible Translation Surely no one will deny that the development of translation theory within SIL has its roots deep in the soil of Nida's version of translation theory  ... multiplicity of codes and the relative weakness of words in convincing a reader of the truth of what is said, they are often astonished to realize how these covert aspects of APRIL 1991) PARADOXES OF TRANSLATING... many societies the poetic format of a discourse carries a great deal of meaning, so that in some cases the content of the book of Job gains great importance by being printed in poetic lines,... obligatory feature the death of the donor, so that a literal translation of heirs of God would imply that God had died The most serious problem encountered by anyone trying to determine the meaning of