On translating the divine name.

Cunningham, David S.

[Greek Text Omitted]

- Acts 2:6

Contemporary translation theory has seized upon the story of Babel as its touchstone. We live "after Babel," says one contemporary theorist;(2) another believes that the story of Babel "can provide an epigraph for all discussions of translation."(3) But the Church understands the legacy of Babel as having been profoundly altered by the event of Pentecost. No longer are languages confused and the people scattered; rather, each person hears the message of the gospel in his or her own native tongue. But this is not a reversion to the era before Babel, in which "the whole earth had one language and the same words" (Gen 11:1). The multiplicity of language remains, but the confusion and failed communication, the legacy of Babel, has been decisively overcome through the power of the Holy Spirit.

Thus, in its very constituting event, the Church acknowledged that its message could be heard "in translation": that the differences between one's own native tongue and the Ur-text of revelation would not stand as a barrier to the proclamation of the gospel. In contradistinction to Islam (and to some versions of Judaism), the Holy Scriptures of the Christian faith are available not only in their original languages, but also in various vernaculars.

However, the early emergence of Greek as the common tongue of Christian theology, and the maintenance of linguistic univocity in the West through its replacement by Latin, helped to mask the essential translatability of the Christian witness. A similar phenomenon has evolved in the Anglophone world, not only because of the pervasive influence (due to aesthetic, literary, and political factors) of the Book of Common Prayer and the King James Bible, but also because of the increasing hegemony of English as the new lingua franca in the international exchange of money, commodities, and ideas. We who speak English as our native tongue are not often cognizant of the problems of translation, if only because we are so rarely forced to grapple with them. Indeed, many seem scandalized that translation is necessary; and even when so convinced, they expect it to operate straightforwardly and without controversy.

It was not always thus. The theory and practice of translation was of urgent concern to Christian thinkers of another time and place; it was, in fact, intimately bound to theology, as witnessed by the work of translation-theorists-cum-theologians such as Jerome, Augustine, and Luther. But in more recent theology, matters of translation have typically been marginalized. Neither the significance nor the perplexities of translation are admitted, except by the denizens of relatively isolated enclaves - professional translators, for example, or scholars of modern and classical languages.

In this article, I propose that Christian theologians who ignore the enormous complexities of the theory and practice of translation do so at their peril; and, more specifically, that their nearly universal willingness to do so, at least in the English-speaking world, has oversimplified and polarized theological arguments which might otherwise become more nuanced. The adverse effects of ignoring translation theory can be witnessed above all in the protracted current debate concerning the naming of God.

Status quaestionis de divino nomine

The mention of "the divine name" in the contemporary Christian context evokes a wide range of issues, including the doctrine of the Trinity, and even the very nature of theology as "talk about God." Here I can offer no more than a thumbnail sketch of these discussions. To put the matter simply, some theologians have argued that the language that is traditionally used most frequently to refer to God (such as the word Father or the phrase Father, Son, and Holy Spirit) is inadequate and should be replaced, perhaps with a variety of alternative names. The reasons for this claim range from complex theological critiques of traditional trinitarian theology, to concerns about the effects of this language on the victims of abusive relationships, to arguments about pluralism and aesthetic viability.

In response, others have argued that the traditional language provides Christians with the only appropriate way to name God, or (in a less extreme version of the argument), the best way. Again, varying arguments are offered on behalf of this stance: references to the text of Matthew 28:19, coupled with the thoroughgoing presence of "Father/Son" language in the New Testament and in the early traditions of the Church; the ecumenical acceptance of the formula (one of the few elements upon which the fractured denominations have, for the most part, agreed); and the failure of any and all proposed alternative formulae to attain true equivalence with the traditional one, with accompanying theological arguments for the necessity of such equivalence.

My purpose in the present article is not to evaluate these arguments,(4) but rather to note their polarized contentiousness and their almost complete failure to grapple with the problem of translation. On the first point, any sampling of the current literature will indicate the degree to which absolute (and frequently hostile) claims are made on both sides of the argument.(5) Indeed, especially in matters of baptism, this issue has become something of a line drawn in the sand. Some of the would-be reformers have indicated, with regret, that they will leave the Church if made to endure the exclusive use of the traditional language; and some of their opponents have insisted so firmly on its retention as to encourage them, at least implicitly, to do precisely that.

One would expect, however, that arguments in Christian theology concerning the acceptance or rejection of particular English words would make some mention, however fleeting, of the problem of translation. For example, in another active debate over trinitarian language, concerning whether or not to retain the language of person, discussions typically begin by considering that word's relationship to the word which it attempts to render, namely the Greek word [Greek Text Omitted], as well as the related [Greek Text Omitted] and the Latin words substantia and persona.(6) But in discussions about our naming of God, no one seems to be very interested in the fact that the traditional formula, "Father, Son, and Holy Spirit," is but one of very many actual (and possible) translations of the phrase [Greek Text Omitted], gleaned from a part of the penultimate verse of Matthew.

This is not to say that the question of translation is completely absent from the debate. For instance, the question of whether Jesus actually addressed God with the Aramaic form [Hebrew Text Omitted], and if so, what this would mean, is very frequently a topic of discussion; however, the interlocutors' various positions on this question do not divide along the same lines as their views on the revisability of the traditional language.(7) But the larger questions of translation - and the theological implications of acknowledging that "Father, Son, and Holy Spirit" is a translation - do not merit even an index entry in books that focus on this topic, let alone a major section or a chapter heading. There are a few exceptions to this rule, which will be noted in due course; but they are rare. And the reason for this is clear. The question of translation makes matters so much more complicated that it threatens to blur, or even to efface, the lines that have been so carefully drawn in the sand.

[Greek Text Omitted]

Our first concern will be to seek some clarity about what constitutes a "name," and whether God can be said to have one (or more). Aristotle tells us that "there are no names by nature";(8) we must designate names by convention, and these designations will affect our stance toward that which is named. A different stance will be adopted by those who believe that a particular word or phrase constitutes the one and only divine name (or is at least primus inter pares) than by those who consider the same phrase to constitute a revisable liturgical rubric.

As Shakespeare reminds us, there is much "in a name"; but he was not the first. "The Lord God formed [Hebrew Text Omitted] of dust from the [Hebrew Text Omitted]" (Gen 2:7): by wordplay and definition, the name is not "just a name." It is given semantic weight; the author of the text attempts to endow it with meaning. Similarly, of [Hebrew Text Omitted] (Moses) it is said, [Hebrew Text Omitted], "I drew him out" (Exod 2:10); and [Hebrew Text Omitted] (Isaac) is conceived in [Hebrew Text Omitted], "laughter" (Gen 17: 17-19).

The name of God is given meaning as well; moreover, God would appear to have more than one name. [Hebrew Text Omitted] and other specifications of [Hebrew Text Omitted] are frequently employed nominally of God; [Hebrew Text Omitted] is the name of God revealed to Moses. According to the gospel accounts, when Jesus speaks to God, he uses the name [Greek Text Omitted], but also [Greek Text Omitted] and [Greek Text Omitted], and, quite frequently, [Greek Text Omitted]. This last-mentioned name would appear, at first, to be a common noun - or, more precisely, an attributive definite description.(9) But it is also a proper name, in that it signifies a specific individual, rather than a class or category; its particular aptness to its subject is coincidental. "A proper name is proper just insofar as it is used independently of aptness to the one named, but it need not therefore lack such aptness."(10) Derrida offers the example of Babel, which has, "as a proper name, the function of a common noun."(11)

Moreover, [Greek Text Omitted] can be used as a proper name (and not just as a common noun) even when it is used without qualification. If I use the word queen in ordinary speech ("she had always wanted to be queen"), I use it as a common noun - referring to a category of monarchs (real and potential, literal and figurative). If I want to specify a single person, I may need to qualify the term: "Elizabeth is the Queen of Great Britain." But when the context admits of no confusion, I can drop the qualification; the word then becomes a proper name, referring to precisely one person. A British subject, using the word without qualification ("The Queen has spoken") uses it as a proper name. In the New Testament, [Greek Text Omitted] is typically used in this way. While a phrase such as [Greek Text Omitted] sometimes appears on the lips of Jesus, he is more often described as saying [Greek Text Omitted] alone. It would surely have been clear (to Jews, at least, and to anyone with a knowledge of Judaism) that he was not referring to one member of a category containing many entities that could be so named. Rather, he is using [Greek Text Omitted] as a proper name, pointing to precisely one person. The common (and often unqualified) New Testament use of [Greek Text Omitted] may be partly due to the influence of the Septuagint.

Thus, we have discovered many biblical names for God, even restricting ourselves to proper names. If we included metaphorical descriptions as well, our list would be much longer. One might argue that this multiplicity of names is overshadowed by the centrality (for Israel, at least) of the Tetragrammaton. Yet this name would eventually be unspoken, thereby prompting even more alternatives. Clearly, some biblical authors are especially endeared to certain names; but any claim that one of these is "the one-and-only scripturally authorized name of God" simply cannot be sustained.

For example, some modern theologians have claimed that [Greek Text Omitted] is exclusively and absolutely "the" biblical name for God. Others have argued that it is the only biblically justified name for the first person of the Trinity. But both these claims come to grief on a number of factors:

(1) Many biblical personages, including Jesus, seem to have employed other names, some of them quite frequently.

(2) In the Gospels, no one but Jesus - not even the disciples - ever refers to God as [Greek Text Omitted].

(3) John uses [Greek Text Omitted] with such disproportionate frequency that one would be hard pressed not to attribute it to the evangelist's stylistic predilections (even when admitting the theological significance of the final form of the canonical texts).

(4) Paul is far more fond of [Greek Text Omitted] than [Greek Text Omitted], even when referring to what later generations would call "the first person of the Trinity."

(5) In general, the most common New Testament appellation for God is clearly [Greek Text Omitted], occurring around four times as frequently as [Greek Text Omitted].

Nor will it do to claim that [Greek Text Omitted] is merely the name of God-in-general, and therefore not comparable to [Greek Text Omitted], which names one of the hypostases. As Karl Rahner demonstrated in his exhaustive study, in the New Testament, [Greek Text Omitted] typically signifies what the Church would later define dogmatically as the first person of the Trinity.(12) Thus, even if the name of the first person can, by synechdoche, name the Trinity, the word [Greek Text Omitted] has at least as much scriptural warrant for this role as does the word [Greek Text Omitted].

Or perhaps more. When Paul uses the word [Greek Text Omitted], he does not use it as a proper name; instead, he always qualifies it in some way. In 54 occurrences of the word (in the 13 letters self-designated as Paul's), 12 refer to human fathers. Of the 42 others, the majority (about 34) pair the word with [Greek Text Omitted] (in combinations such as [Greek Text Omitted] and [Greek Text Omitted]), and/or connect it explicitly with Christ. The remaining occurrences contextualize the word in some way, in order to make it clear that Paul is referring to the God of Israel and of Jesus - by quoting the Old Testament, or referring to God's work of creation, or pairing it with the word [Greek Text Omitted]. In other words, while Paul seems willing to allow [Greek Text Omitted] to stand on its own as a proper name, [Greek Text Omitted] was still seen as enough of a common noun to require constant qualification. Because many may be called [Greek Text Omitted], Paul makes his reference explicit: [Greek Text Omitted] (Rom 15:6). This is not to say that Paul does not consider [Greek Text Omitted] an important word; however, it is certainly not alone.

Other theologians have claimed that [Greek Text Omitted] is the one true and definitive name of God.(13) It is difficult to make this claim on the basis of Scripture alone; after all, these words appear as a cluster only once, and many other names are employed much more frequently. Admittedly, the verse which actually employs this phrase also uses the word [Greek Text Omitted], which has misled some commentators into assuming that we are here given access to "the" name of God.(14) In Matthew 28:19, Jesus tells his followers to make disciples of all nations, baptizing them [Greek Text Omitted]. The formula is repeated in Didache 7. But we certainly cannot claim that this is the one-and-only name of God simply on the basis of [Greek Text Omitted]. Doing so would require us to interpret Matthew 28:19 as follows: "this name, which I am about to pronounce, is the name of God; it supersedes every other name that anyone has ever used." Surely [Greek Text Omitted] cannot be asked to bear this much weight. After all, our ordinary use of the English phrase in the name of is not about designating a name; it is most commonly an invocation of authority ("Stop in the name of the law!"). In the biblical text, the phrase is often read as meaning "in reference to" or "in thinking of" (compare Matt 18:20, [Greek Text Omitted] - "two or three are gathered in my name").(15) It can also mean "dedicated to," which is the sense that many readers give it in this verse, though it could as easily mean "by the authority of." In any case, readers of this verse will bring to it particular theologies of baptism which will certainly influence their readings of the text.

Thus, the appearance of the word [Greek Text Omitted] in this verse does not somehow magically designate the phrase which follows it as "the" name of God. Moreover, the verse was apparently not always taken to require that the phrase be recited at baptisms. Indeed, in the earliest era of the Church, baptism apparently did not include the phrase in Matthew 28:19. Baptism was either "in the name of Jesus" (Acts 2:38, 8:16, 10:48, 19:5, and possibly 22:16), or by means of three questions, asking the candidate to express belief in each of the divine persons, individually named - questions which were later elaborated into protocreeds. Each of these questions, upon being answered affirmatively by the candidate, was followed by an act of baptism.(16)

I have thus far restricted my analysis to the language of Scripture. When we turn to the later tradition, we find a movement in two different directions at once: on the one hand, we can observe a general acceptance of [Greek Text Omitted], [Greek Text Omitted], and [Greek Text Omitted] as naming the three hypostases of the Trinity, along with a general acceptance of some conjoining of the three to name God - although the precise way in which these three should be linked together (with prepositions, to form a formula) varies according to the context in which the words are to be used.(17) Yet despite this early enthusiasm for a comparatively narrow range of "names" for God, we simultaneously witness a phenomenal multiplication of such names, the locus classicus of which may be found in the pseudo-Dionysian treatise On the Divine Names.(18) The variety of divine names seems to be one of the implications of the via negativa: God cannot be named, and is therefore the God of every name.

Negative theology has become a touchstone for writers Who seek to argue against the exclusive use of the traditional language for God.(19) If it is true, as the Damascene says, that "the deity, being incomprehensible, is also assuredly nameless,"(20) then clearly it would amount to something like a denial of the mystery of God to claim that God has but one name. But while the via negativa claims that God is beyond all names, it does not make the same claim for the individual hypostases of the Trinity. As Thomas Hopko notes,

God is said to be essentially beyond being, divinity, paternity, sonship, spirithood, goodness, wisdom, power, and so on. But God is never said to be hypo-statically beyond Father, Son, and Holy Spirit. For God is supraessential and even nonessential. But God is not suprahypostatic or nonhypostatic, supra-personal or nonpersonal.(21)

From this perspective, negative theology cannot, by itself, be decisive as an argument against the naming of the individual trinitarian hypostases as Father, Son, and Holy Spirit. But it is simply a non sequitur that "the names of Father, Son, and Holy Spirit for the three divine hypostases are never changed or amended," as Hopko states in the next sentence of his text.(22) The reasons for this will become clear in the next section.

To summarize this investigation into "the very idea of the name of God": while some theologians may continue to argue that one or another name is ultimate, this appears a very difficult claim to justify. In Scripture, the preeminent names for God, measured by sheer quantity, are [Hebrew Text Omitted] and [Hebrew Text Omitted] in the Old Testament and [Greek Text Omitted] in the New; beyond this, a variety of names is clearly the norm. In the later tradition, the via negativa seems to assure that there will be no unanimity. While [Greek Text Omitted], [Greek Text Omitted], and [Greek Text Omitted] certainly retained a certain pride of place, their use was not uniform. Thomas Aquinas, for example, argues that qui est, the Vulgate's translation of [Hebrew Text Omitted], is the most appropriate (maxime proprium) name for God.(23) One is thereby forced to register grave doubts as to whether it is possible to speak coherently of "the" name of God in the Christian tradition.

I hasten to add, however, that my argument in the present essay does not depend upon the acceptance of this conclusion. For even if we grant, for the sake of argument, that either the word [Greek Text Omitted], or the phrase [Greek Text Omitted] is the (only, or preeminent) name of God, we must still wrestle with the thorny issue of translation.

Das Unbehagen in der Ubersetzung

Some theorists argue that translation is, in principle, impossible. On these accounts, a language creates its own discursive universe, hermetically sealed and inaccessible to outsiders. I do not wish to advocate such a view. Nevertheless, there is a particular class of words which, by definition, cannot be translated: pure proper names. The untranslatability of these words results from their lack of the "fulcrum" that is needed to move a word from one language into another. Most people typically assume that this fulcrum is provided by the "meaning" or "definition" of the word.(24) For example: if bread in English refers to a food made from flour which is kneaded, shaped, and baked, then we look for the German word which refers to the same sort of thing, and find Brot. But if we have a pure proper name (that is, one which is used to refer uniquely to one entity), we will not "find" a different corresponding word in another language which similarly refers. "A proper name as such remains forever untranslatable, a fact that may lead one to conclude that it does not strictly belong, for the same reason as the other words, to the language."(25) Pure proper names are transliterated - either in a strong sense. (Cyrillic characters into Roman ones, for example), or in a weak sense, when certain letters are modified within similar orthography (as when the English name John becomes Johannes, Jean, Juan, Ian, Ivan, or Giovanni).

Certainly, some words which appear to be used as proper names may be translated, but only to the extent that they are not pure proper names (that is, they are not used to refer to only one entity). They can be translated if they are "given semantically,"(26) that is, if they can be given one or more "meanings" which can allow them to be moved from one language to another. Thus some Native American names are translated by means of the common elements of nature upon which they are metaphorically based: Little Feather, Night Horse. But these are not pure proper names, because they do not refer to just one entity. In fact, they gain their poetic effect from their metaphorical richness. Pure proper names, by definition, have no semantic equivalents.

But in addition to the pure proper name and the semantically given name, there is a third, hybrid variety. Such names are translated on some occasions and transliterated on others; or, the reader/hearer is offered both a transliteration (e.g. "Isaac") and a translation ("laughter"). Many of the names of persons in Western society fall into this category; they have some semantic reference (the sort of thing one finds in a "name-your-baby" book), but because the name is rarely used except in reference to a person, this semantic value is rarely noticed. The name "Monica" may mean "advisor" according to some deep etymology; but one rarely hears "She's been a monica to me all my life." By contrast, in English, the phrase "little feather" is used more often to refer to the plumage of birds than to a human person.

Among the names in this third, "hybrid" category, we find names for God. The Tetragrammaton, for example, often appears as YHWH in languages which use a Roman lettering system, but may also appear as [Greek Text Omitted] or qui est or "I am who I am." Similarly, the Greek vocative [Greek Text Omitted] found its way into one part of the Latin Mass, but various inflections of the word dominus were more commonly employed. This "hybrid" nature of the divine name should provide some forewarning of the minefields we are likely to encounter when undertaking the project of its translation. Not that this project can be avoided altogether; after all, Christianity understands itself as a translatable discourse, and thus does not restrict the possibility of revelation to its "original" language. But this poses some theological difficulties, since translations are, by their very nature, imperfect and transient.

In the New Revised Standard Version of the Bible, for example, the Hebrew [Hebrew Text Omitted] is sometimes "humankind"; the Greek [Greek Text Omitted] is often "brothers and sisters." These are new translations, inspired by new sensibilities. But such interpretive concerns are hardly new to the process of translation; while the most recent translations have attempted to avoid unnecessary masculine specificity, earlier translations actually added it. One particularly glaring example is the King James version of Numbers 11:12, wherein Moses complains that he has been asked to care for the people "as a nursing father beareth the sucking child." Nursing father? Even Luther was willing to write Amme ("wetnurse"), which the context almost seems to demand. But lest the Authorized Version be considered a mere throwback to bygone days, note that even the New American Bible cannot abide Moses as wetnurse: "like a foster father carrying an infant," it says, thereby completely obscuring the fact that the child is breastfeeding. In this case, as in many others, translators are unlikely to reach quick agreement as to what the most "natural" sense of a word might be.

What implications might this have for the question at hand? If Matthew 28:19 is the unique scriptural revelation of the divine name, and if this name should be translated rather than transliterated (two unresolved issues), then we still have to make choices about the most appropriate translations of [Greek Text Omitted], [Greek Text Omitted], and [Greek Text Omitted]. Competent users of a language may certainly disagree about the most "obvious" translation of a particular word, and have good reasons for doing so. For the sake of intellectual honesty, we must disabuse ourselves of the notion that even the simplest of words can be painlessly and unproblematically carried over into a new language. Thus, even if [Greek Text Omitted] is "the" divine name, it does not therefore follow that "Father, Son, and Holy Spirit" is "the" divine name. In support of this claim, I offer five arguments.

Languages

The first argument is based on the sheer multiplicity of languages and their differing nuances. As translation theorists are quick to point out, even the most basic and banal phrases cannot be carried across into different languages without remainder. To translate is to alter, to interpret, to transform. This is due primarily to the differently nuanced associations called up by ostensibly equivalent words in different languages. On this matter, allow me to quote George Steiner at length; he makes the point clearly, even in the case of the "simplest" words.

Though they deny it, phrase-books and primers are full of immediate deeps. Literally: J'aime la natation (from Collins French Phrase Book, 1962). Word-for-word: 'I love natation', which is mildly lunatic though, predictably, Sir Thomas Browne used the word in 1646. 'I like to go swimming' (omitting the nasty problem of differential strengths in aimer and 'like'). "Swimming" turns up in Beowulf; the root is Indo-European swem, meaning to be in general motion, in a sense still functional in Welsh and Lithuanian. Nager is very different: through Old French and Provencal there is a clear link to navigare, to what is "nautical" in the governance and progress of a ship. The phrase-book offers: je veux aller a la piscine. "Swimming-pool" is not wholly piscine. The latter is a Roman fish-pond; like nager it encodes the disciplined artifice, the interposition before spontaneous motion, of the classical order. "I want to go . . ." / je veux aller . . . "Want" is ultimately Old Norse for "lack," "need," the felt register of deprivations. The sense "to desire" comes only fifth among the rubrics which follow on the word in the OED. Vouloir is of that great family of words, derived from the Sanskrit root var, signifying volition, focused intent, the advance of "will" (its cognate). The phrase-book is uneasily aware of the profound difference. "I want should not be translated by je veux. In French this is a very strong form, and when used to express a wish creates the unfortunate impression of giving a blunt and peremptory order rather than of making a polite request." But the matter is not basically one of differing forces of demand. "Want" as Shakespeare almost invariably adumbrates, speaks out of concavity, out of absence and need. In French this zone of meaning would be circumscribed by besoin, manque, and carence. But j'ai besoin d'aller nager is instantaneously off-pitch or obscurely therapeutic.(27)

If we have so much difficulty guaranteeing that discussions about a trip to the beach will be equivalent in English and French, should we not be somewhat skeptical that we have so easily happened upon the final, unrevisable, perfectly equivalent translation of the divine name?

The depths of difference among languages led Walter Benjamin to argue that the notions of "fidelity" and "license" in translation, which had hitherto so thoroughly governed the enterprise, were "no longer serviceable."

What can fidelity really do for the rendering of meaning? Fidelity in the translation of individual words can almost never fully reproduce the meaning they have in the original. For sense in its poetic significance is not limited to meaning, but derives from the connotations conveyed by the word chosen to express it. We say of words that they have emotional connotations. A literal rendering of the syntax completely demolishes the theory of reproduction of meaning and is a direct threat to comprehensibility.(28)

I will return to Benjamin presently. For now, perhaps it will suffice to make two observations. First, we need to be cautious about appealing to Greek patristic arguments concerning the use of a particular divine name, whether [Greek Text Omitted] alone or the formula derived from Matthew 28: 19. The move from Koine Greek to the Greek of the Patristic era, while not utterly unproblematic, was much less complicated than is the move from Greek to English.

Second, we should note that those who claim, without linguistic nuance, that "Father, Son, and Holy Spirit" is "the" name of God must be mistaken, at least insofar as English is not the only language in which God's name may be spoken. This objection might be thought trivial if the divine name were merely transliterated, since that process is governed by widely accepted conventions within a given target language (p for [Pi], a for [Alpha], and so on).(29) But because we have chosen to regard this name not as a pure proper name, but as one given semantically (and therefore translatable), its appearance in the garb of a variety of languages is manifestly not a trivial manner: to translate is to interpret, and one never translates without remainder. There must be multiple names for God, for there are many languages - among which there can never be exact replication, but always interpretation.

Words

The second argument concerns the particular Greek words [Greek Text Omitted] and [Greek Text Omitted], which themselves do not have narrow ranges of meaning. In addition to offering the English word father as a definition, Liddell and Scott remind us that Zeus is called [Greek Text Omitted], and that the word is also used as "a respectful mode of addressing elderly persons."(30) Metaphorically it can mean the parent of anything; in this sense it is similar to the Latin auctor. The plural [Greek Text Omitted] can mean both parents, or, in the Christian context, can mean all deceased Christians - suggesting that the notion of maleness is not so firmly ensconced in the phoneme [Greek Text Omitted]-as some would assume. The person who instructs a novice is also [Greek Text Omitted](31) (compare the German Doktorvater).

Needless to say, any attempt to render this vast range of meanings into another language is bound to fail at some level. The English "father" is troublesome even on the first and most "obvious" definition, in that it is probably not even the most commonly used term to refer to a male parent; and, in current American English at any rate, often connotes some degree of formality (though this obviously varies by time, space, family structure, and social class). In French, pere is much more common and certainly less formal. Its close relationship to a word such as compere (fellow, comrade) makes it very different from father. German is even more problematic, given the recent historical manifestations of Vaterland and its cognates. This is not to say that these other meanings are necessarily called up in every invocation of father or pere or Vater, but only that the semantic registers are different in different languages. Steiner's phrase-book examples, quoted at length above, remind us that words are never "the same" in another language.

The range of [Greek Text Omitted] has been narrower, but hardly univocal. It, too, has a role in the master-apprentice relationship: it refers to the pupil or follower. Arndt and Gingrich also offer "those who are bound to a personality by close, non-material ties; it is this personality that has promoted the relationship and given it its character"(32) - a description which seems to be supported, by later dogmatic theology, in the case of Jesus. And this is to say nothing of the far-ranging resonance that the term would have for readers of the Gospels. It might call up a number of related phrases, including [Greek Text Omitted], [Greek Text Omitted], and of course [Greek Text Omitted], with their enormous ranges of reference.

English son and German Sohn are, if anything, even more restricted in their range of meanings, but French fils offers something quite quite different. Because of its close orthographic connection to the feminine fille, and to other words referring to parent-child relationships, such as filial(le), filiation, and filleul(e), its gender does not stand out nearly as starkly as do its English and German "equivalents."

Again this argument does not endorse an alternative translation, nor does it even suggest that some or all of the aforementioned meanings of [Greek Text Omitted] and [Greek Text Omitted] ought to be displayed when translating these words. Neither does it suggest that other references within the New Testament and in later ecclesial usage might not push us very strongly in the direction of the traditional English translation. I simply wish to make the (comparatively limited) claim that, given the very different semantic ranges of these two words in the two languages, we ought to accept all translations tentatively and provisionally, and should therefore not make extraordinary claims of unrevisability on their behalf. As Walter Benjamin admits, "all translation is only a somewhat provisional way of coming to terms with the foreignness of languages. An instant and final rather than a temporary and provisional solution of this foreignness remains out of reach."(33)

This is manifestly not to argue that languages are isolated monads, forever insulated from one another by the principle of untranslatability. Translation is a process, and can always be continued by further explanation, dialogue, and conversation. Apparent failures may simply be, in Stephen Fowl's words, "contingent difficulties of translation at any point in time."(34) Benjamin, however, remained beholden to "equivalence" in translation, accepting that some future translator might approach, at least asymptotically, the desideratum of true equivalence.(35) But the very idea of "equivalence" requires certain assumptions about words and meanings - assumptions which may not be wholly adequate.

Meanings

I took note earlier of the apparently "common-sense" claim that translation takes place via a common "meaning." It is now time to return to this claim, and to note its insufficiency. This conventional view of semantics assumes that words can be associated with nonlinguistic "meanings," which are related to their linguistic bearers rather as Platonic ideals are related to the phenomena that participate in them. On this model, translation operates in two steps: first, from the concrete word in one language to the disembodied "meaning"; and then from this "meaning" back to a concrete word in another language.

The appeal of this approach to meaning is obvious enough. For one thing, it seems to help explain how we acquire language: by association of a word with an object, a signifier with a signified. This association typically takes place by precise and detailed description, or by ostensive definition. This is how Augustine describes his own acquisition of language.(36) It seems obvious to us, because we use this method to teach children their first words. And in using this approach in my earlier example, I deliberately chose very common words (bread/Brot), learned early in life by native speakers of both the languages in question.

But this is not the end of the story, as Wittgenstein points out at the beginning of the Philosophical Investigations. After quoting Augustine's account at length (in Latin!), he comments: "If you describe the learning of language in this way you are, I believe, thinking primarily of nouns like 'table', 'chair', 'bread', and of people's names, and only secondarily of the names of certain actions and properties; and of the remaining kinds of word as something that will take care of itself."(37) In the more advanced stages of language use, Wittgenstein argues, we do not constantly refer back to charts and tables which encode these ostensive definitions; we move directly from language to action, without conscious reference to the "meaning" of the words. Thus the notion that words refer to particular objects is "appropriate, but only for this narrowly circumscribed region, not for the whole of what you were claiming to describe."(38) Then follows Wittgenstein's famous discussion of a language consisting only of the words block, pillar, slab, and beam, and the very many different ways in which these apparently simple words are used. As linguistic activity becomes more complicated, it becomes increasingly obvious that words are not invariably associated with meanings.(39) Nor is lack of such association a merely theoretical insight: the utter failure of machine translation to produce anything more than the most rudimentary sentences in another language has helped to certify Wittgenstein's skepticism.(40)

If, as Wittgenstein claims, "the meaning of a word is its use in the language,"(41) then word-for-word translation via the vehicle of "meaning" is either an unachievable ideal or, worse, a mask for hegemonic discourse. And of course, the "meanings" which animate word-for-word translations can themselves be expressed only in linguistic terms. This is a very basic and generally accepted insight of hermeneutics; however, it has rarely been rigorously applied to the problem of translation. To claim that one has arrived at the final, definitive, ultimate translation, one would have to have direct access to a single, fixed and final "meaning" behind the word - which simply does not exist.(42)

Of course, to note that words are not tied to a single meaning is not to claim that they are meaningless. Quite the contrary: they typically call forth multiple meanings, which must then be negotiated. Indeed, all words are potentially meaningful - even nonsense words ("'Twas brillig, / and the slithy toves . . ."). They require only a reader or hearer willing to construct a possible meaning for them. This does not deny that the author of the words may have intended something, perhaps even one single thing, in employing them; but this is not necessarily the meaning, and certainly not the only one, that will be constructed by the readers or hearers. "Meanings arise out of, and invariably revolve around, contextual uses. Words do not 'have' meanings in any objective sense, in the so-called null context, in abstract isolation from real speech-use situations. Neither the dictionary nor the thesaurus nor any other formalization of semantic fields holds sway over the volatility of actual speech use."(43) Or, as Mikhail Bakhtin argues,

No living word relates to its object in a singular way: between the word and its object, between the word and the speaking subject, there exists an elastic environment of other, alien words about the same object, the same theme, and this is an environment that it is often difficult to penetrate. It is precisely in the process of living interaction with this specific environment that the word may be individualized and given stylistic shape. . . . The living utterance, having taken meaning and shape at a particular historical moment in a socially specific environment, cannot fail to brush up against thousands of living dialogic threads, woven by socio-ideological consciousness around the given object of an utterance; it cannot fail to become an active participant in social dialogue.(44)

And since all words can be attributed multiple meanings, their translation cannot be a simplistic process of invoking another word, in another language, which has "the same meaning." Bakhtin's comments here are further applied to the theory and practice of translation by Douglas Robinson:

Since words do not really belong to anyone, since they aren't "property" that can be allotted or stolen or trespassed upon, but float freely in the dialogical public domain, there can be no pure or perfect or ideal distinctions between texts, and thus no pure or perfect or ideal correspondences between them either. . . . There is no way of establishing the objective "equivalence" between texts, or between receptor responses to texts. Artificial boundaries can be set up and jealously maintained, but dialogized words flow back and forth across any such boundaries and render them thus politically and historically contingent.(45)

Again, this argument does not claim that words have no meaning, nor that meaning cannot sometimes, even frequently, be communicated through language. But there can be no automatic and perfect correspondence of meanings between words in different languages.

Audiences

My fourth argument makes a rhetorical turn: the translator always writes for a particular audience, the members of which must be kept in mind as work on the translation proceeds. One cannot rest content with dictionary definitions and "common sense" assumptions about the connections between two languages; one must always bear in mind how the language of the translation will be received by those who are most likely to read it (or hear it read). Obviously, one can never include all possible readers and hearers in this group. Any two readers may cling tenaciously to mutually exclusive definitions of a particular word, and the translator can never convey the same meaning (or range of meanings) to both these readers without employing different words. Occasionally one may have recourse to interpolation or footnotes, in order to clarify a matter on which the audience is likely to be divided; thus, the NRSV puts "brothers and sisters" in the text, but assures readers that the translators still retain possession of their Greek Grammars by placing "Gk brothers" in a footnote in every case. But this option is distracting at best, and at worst has the effect of undermining the chosen translation by hinting that it is not "really" appropriate.

More commonly, the translator simply excludes from his or her audience those who cannot avoid a connotation that was not intended. Thus, if the members of the NRSV committee had chosen to continue the practice of past English versions by translating Paul's vocative use of [Greek Text Omitted] as "brethren" or "brothers," they would be knowingly eliciting false judgments on the part of many readers, who would assume that a text addressed to "brothers" does not apply to women. Yet most theologians would claim that the historical addressees of phrases such as [Greek Text Omitted], [Greek Text Omitted] (NRSV: "I appeal to you, therefore, brothers and sisters") included all members, male and female, of the Christian community to which Paul was writing. Apparently, the committee felt that the number of persons likely to read "brothers" as exclusive had grown sufficiently large to justify the change.

This does not imply a judgment on the "correctness" of these readings. Nor does it mean that the translation problem here could not be explained to the willing listener. One could point out that Paul frequently seems to imply the existence of female Christians, and that he does not seem to exclude them when addressing the community as [Greek Text Omitted]. And in any case, the feminine singular [Greek Text Omitted] would be assumed to be included in the masculine plural, in the same way that mixed groups are often referred to in the masculine in a number of modern languages (in French, if the group is all women, it is elles, but if it is either all men or a mixed group, it is ils). But providing such lengthy explanations is functionally equivalent to translating the word as "brothers and sisters." In both cases, one assumes that, if Paul were speaking American English in the 1990s, he wouldn't say "brothers," because many members of his audience would assume, incorrectly (but reasonably, given current usage), that he was only talking to the men.

And in fact, what we might call the "new translation" solution to the problem of [Greek Text Omitted] has certain advantages over the "explanation" solution. Specifically, it has the advantage of easy repetition: each time the Greek text has [Greek Text Omitted], the English text simply has "brothers and sisters" (or another inclusive translation), without repeating or reinforcing the lengthy explanation offered in the previous paragraph. Moreover, the first solution does not exclude the use of the second one as well: the rationale for the new translation is highlighted by means of the explanation. The two solutions would, in fact, mutually strengthen one another. Lastly, the new translation helps to prevent the implicit exclusivity of earlier linguistic forms from becoming a "stone of stumbling" in the evangelization of those who live in a linguistic world in which grammatically masculine words such as "brothers," "men," and "he" are assumed to refer only to males. This is manifestly not to say that Christianity can abide no scandal; but one must not let the most important scandal - that of the Cross - be eclipsed by those "stumbling blocks" that Christians are urged to eradicate (Matthew 18:6).

By analogy, when translating the divine name, one must be attentive not only to what one considers to be the most "natural" or "common-sense" rendering, but also to how the translation will be received. Again, one cannot make space for every eccentric hearer, but one must also recognize the gravity of being "misheard" or "misread" by a large number of people. If one continues to argue that "Father" and "Son" are the exclusive English translations for "the" names of the first two persons of the Trinity, one must also admit that, in the United States in the 1990s, this will be heard by a substantial number of people as indicating some degree of maleness in God. In the attempt to defuse this misunderstanding, one may quote Athanasius, Hilary, and Gregory of Nazianzus to the contrary, not to mention a large number of modern commentators.(46) But these declarations can hardly outweigh the effect of the overwhelming usage of masculine language, masculine metaphors, and masculine personal pronouns in reference to God - not to mention art, architecture, and, in general, "the effective-history of the father symbol in Christianity, which grew hardened and fixed in alliance with patriarchal rule, thus imprisoning rather than releasing the good news it was originally intended to convey."(47) Those who are bombarded with such language and imagery, consistently and exclusively, both in theory and in practice, might be forgiven for assuming that biological sex, or at least gender, apparently must be predicated of God, and that God's sex (or gender) must be male.

And this is especially a problem in the English-speaking world, due to some grammatical oddities of the language.(48) Consider the counterexample of French, in which people are quite accustomed to referring to a wide variety of entities, from God to human beings to inanimate objects, with grammatically gendered nouns. If I point to the desk and name its color in French, I will say "il est noir"; the literal translation is "he is black," but the presence of gender in this sentence doesn't have the same jarring effect on the French speaker as it does on the English speaker. Indeed, in many languages, people are quite accustomed to appropriating gendered language to items with no gender. English is different: gendered language is almost exclusively for persons. (The primary exceptions are anthropomorphized entities - those which seem to have something resembling human "personalities," such as our cars, boats, and pets.) We use gendered language primarily of persons, and "all the persons we know are either male or female"(49) - all persons, that is, except the three persons of the Trinity. This would be acceptable if these words were taken to apply to God metaphorically, since a certain degree of dissociation is always necessary for metaphors. But those who argue most strenuously for a single form of the divine name also insist that these words are not metaphors, but names. So as English speakers, we are asked to assume that the words "father," "son," and "he" will refer ("literally") to a biologically male being in every instance except when they are used to speak of God. This demands a feat of linguistic differentiation that is impossible for all but the most semantically conscientious speakers of the language.(50) And even they may be deluded; as recent psychoanalytic criticism suggests, language affects our psyche and our construction of gender more deeply than we realize.(51)

Despite these factors, some theologians still argue that the use of feminine language and imagery for God "introduces sexuality" into the concept of the divine, whereas masculine language does not.(52) This claim, too, is oblivious to the problems of translation in general, and to the specific hazards of translation into English in particular. In the original cultural-linguistic settings in which the Bible was written, it may have been possible to refer to God with grammatically masculine pronouns and not thereby introduce sexuality into the divine. But when the biblical narrative is translated into contemporary English, the new audience is unlikely to read or hear the stories as gender-neutral. In English, sexuality is introduced every bit as much by "he" as by "she," and to claim otherwise is either sheer ideology or nostalgia for the days when one could say "man" or "men" and not be assumed to be referring only to males. The "generic masculine" has fallen out of approved usage in American English. From the style sheets of academic journals to the Weekly Reader, from television news to the corporate meeting room, "men" means "male" - and so do "he," "him," and "his."(53) Some translators and theologians may not be happy with this development, but they will not change it by mere force of will.

It is true, of course, that masculine language about God may work in the opposite direction as well. Instead of provoking its hearers to imagine that God is to be understood on the model of the human fathers and sons they have known, it may instead become a call to conversion: human beings must model their own fatherhood and sonship on that of God.(54) Unfortunately, the language does not seem to have been employed this way very often in the history of the tradition; with luck, it will find a place on the agenda of future Christian catechesis. But while this possibility clearly presents a warrant for retaining the translations "father" and "son," it does not invalidate the claim that other translations might also be quite appropriate, alongside the traditional ones. Indeed, if we wish to increase the likelihood that audiences will understand God-language as a critique of human relationships (rather than as a divine imprimatur on male privilege), we should actively seek out alternative translations of the divine name, for their very multiplicity would discourage the hearer from understanding God only under a single genus or species.

In any case, to many people, the qualities that some theologians have assumed are evoked by the use of father are opaque at best, and often even misleading. If we had followed Jesus' command to "call no one your father on earth" (Matt. 23:9), we might be able to apply this word to God without engendering manifest confusion. But having ignored that injunction on so widespread a basis, we are now paying the price. At the very least, we must become aware that for many Christians, the traditional names of the first two persons of the Trinity are associated with male sexuality. If we wish to correct this interpretation, only two routes appear to be open to us: explanation or retranslation. We cannot simply declare that there is no problem, and then repeat the. old translation without explanatory nuance.

If one chooses the explanatory approach, then one must be aware that every time the old translation is repeated, it must once again be explained; failing to do so will invite the more common semantic associations of words such as "father" and "son" to swiftly regain the audience's attention. If one chooses retranslation, one must, of course, still debate the relative success of any alternative; for example, the early experimental formula "Creator, Redeemer, Sustainer" has been brought under cogent critique from a number of different theological perspectives. However, this formula is something of a straw figure in the current debate; despite all the intellectual energy expended against it, no one seems much in favor of it!(55) In any case, we should much prefer such specific discussions of the relative merits of various new translations over the claim that one can continue to use gendered language and not thereby imply sexual difference in God. I have already attempted to indicate the reasons for preferring retranslation to explanation; and in this regard, my fifth argument will be decisive.

Bodies

My advocacy of the "audience's perspective" on translation has thus far focused on intellectual categories: what the text "means," or better, how it is "heard" by the audience. However, words are received not just by the mind, but by the whole body. This point is made powerfully in an important recent contribution to translation theory by Douglas M. Robinson. We do not just "think" words; we feel words. And we

most typically guide our choice of words when we speak (and our interpretation of words when others speak) emotionally, by recourse not to an abstract cognitive system of rules but to what feels right. . . . We often have a gut-level sense that a word is wrong, off-base, inadequate, incorrect, or else perfect, exactly right for what we have in mind to say - and yet could not, if pressed, provide a dictionary definition for it, let alone analyze its semantic field. . . . We also feel words in the tactile sense - we can feel assaulted or bludgeoned by words. . . . Words can also caress, soothe, placate.(56)

Of course, how our bodies react to words will vary according to person, context, and language. But this does not mean that the assignment of meaning is simply random, according to how one feels on a particular day; for we often seem, at least, to understand one other. This is due primarily to the "interpretive communities" that grow up around particular uses of language.(57) These communities develop and control the range of possible interpretations; thus, even though we all might potentially "feel" differently about a particular word or phrase, our range of possible responses is controlled through the communities in which we participate. "Meaning and its interpretation are motivated and guided by feeling, or, more broadly, by body or somatic response; but that guidance is both contextually and personally variable (the flexibility and uniqueness of the individual speaking subject) and ideologically controlled (the shaping force of the speech community)."(58)

Robinson's description of the effects of words on our bodies - what he calls the ideosomatics of language - helps us understand the wide variety of interpretive responses to certain words, and also the tenaciousness with which we sometimes cling to particular interpretations. We have been "programmed" to respond, bodily, to certain words in certain ways; thus, certain words "feel" right while others "feel" wrong. "This is also why native speakers of a language can argue forever over the connotations of a word (in a poem, say); each has personal associations that awaken ideosomatic responses, and each has been programmed to objectify (reify, externalize) somatic response as textual property. Each responds to the poem slightly differently, and each wants to believe that his or her response is the true or correct one."(59)

All of this has enormous implications for translation, which must produce language that is not only mentally appropriate for the particular audience, but which "feels" right, i.e., evokes a somatic response in the target language which is akin to that evoked in the very different context of the source language. As Robinson argues, translation is never a "literal" or "mechanical" affair, but always involves figuration and ethical judgment. The translator employs tropes, and must decide which are the most important ones to employ in particular cases. Are we looking to reproduce the sound and rhythm of the original? We might employ a certain kind of metonymy. Do we think that a certain subset of the text in question is the key to the whole structure? Then we might effect a synechdochal translation, allowing this "key" to govern our interpretation of the whole. Do we want to remind our readers that this is only a translation - indeed, that any translation of the original is high treason? Use irony. Do we have an insight into the author's meaning that hasn't come through in the text? Exaggerate. Do we want to emphasize the text's distance from us? Archaize. Its nearness? Modernize.

Robinson's point is not so much that we should use these tropes when translating; rather, we do use them, but typically do not admit it. Because we claim to strive toward equivalence, we believe that tropes (so often considered "mere ornament") would violate the supposedly "logically exact science" of translation. But just as postmodern literary theory has forced us to recognize the ubiquity of rhetorical devices in all genres, so has translation theory forced us to recognize that the tropes have always played a major role in the process of rendering one language into another.

The translator's ethical judgments are also important. We place a great deal of trust in translators: first, in what they (and their patrons) choose to translate into our language, and secondly, in the way they do so. If we have no access to the source language, we are literally at the translator's mercy. Much will depend upon the translator's reaction to the text: a text that seems (to the translator) to be advancing the human condition may appear, in the target language, to have a great deal more gusto and persuasiveness than one which the translator considers malevolent.

If this seems to place a great deal of power in the hands of the translator, then Robinson has made his point. He wants us to recognize that translators wield enormous power. (He also wants translators themselves to recognize it, and stop pretending that they are instrumental technicians for whom a particular set of inputs must always produce identical outputs.) But I believe that Robinson's strong claims on behalf of translation are quite legitimate, and especially so for Christian theologians.

Christianity employs a complex and unique understanding of the significance of the body. Indeed, the Christian faith cut its teeth by differentiating itself from those mystery religions which employed stark dualisms of spirit and matter. Christianity adopted a more organic and unified understanding of the body, recognizing a heuristic distinction between the physical and the psychological, but also recognizing that the distinction was not an ultimate one.

Alas, this claim has often been overlooked or misunderstood; and the history of Christian thought is replete with language that can be read as advocating a starkly dualistic theological anthropology. Paul, Augustine, Thomas, and Luther are all frequently quoted by their cultured despisers in ways that make them sound no more enamored of the fleshly material "stuff" of human existence than were Valentinus and Basilides. One such despiser is Robinson himself, who sees Augustine and Luther as having altogether suppressed the corporality of human existence in general and of translation in particular; consequently, a persistent and most unfortunate anti-Christian polemic runs throughout Robinson's book. But it is also clear that Robinson does not clearly understand the faith he criticizes; for the fleshly reality of the Incarnation, so thoroughly affirmed in the historical creeds of the Church, will not allow the bodily existence of human creatures to be ignored. Indeed, Robinson's own project would be strengthened were he to recognize how thoroughly it is substantiated by any properly Christian theology of the body. Christians believe in the true Incarnation of the word, the bodily dwelling of God on earth - not just a divine, ethereal flesh, not a merely apparent body, but a true human body. And the significance of the body is reiterated in the Christian doctrine of creation, in the sacraments, and in its understanding of the end of human life and the end of the world: Christians believe in "the resurrection of the body."

So bodies are important to Christians, and the effects that words have upon bodies are also important. In certain circumstances, our words may need to challenge, annoy, or even accuse others. Ultimately, though, our words should not maim and injure; they should heal. Christianity cannot abide a gnostic flight from the effects that words have upon the body, or upon the body of Christ. If the translations have been employed that cause injury to that body, or to the individual bodies which collectively constitute it, then we should pause to consider whether we have properly discerned the inspiration of the Spirit. God's self-revelation does not occur "in ways that harm us. As Karl Barth noted, 'Brutal grace is not the grace of the true and living God.'"(60) In the realm of secret gnosis there might be room for a claim that certain words hold the magic power to invoke God, and that their effects on particular bodies are irrelevant. But Christianity cannot make that claim.

Il nome della rosa

Can we possibly conceive of what it might mean to retranslate the divine name, a name which has found such a clear and common presence in English-language Christian liturgy and devotion? Many of those who claim that there can never be any alternative to "Father, Son, and Holy Spirit" have suggested that the hard-won ecumenical agreement about baptism in the triune name would be horribly undone were any substitutions to be allowed. Of course, this evades the whole question of translation, because many names are already allowed; we baptize in the vernacular, not according to the Greek text. But even if we confine ourselves, for the moment, to English, can we possibly consider changing so venerable a translation?

Not only can we consider it; we already have done so. The translation of the name has already changed, in English. Until sometime in the middle of the 20th century, the form in common use was not "Father, Son, and Holy Spirit," but "Father, Son, and Holy Ghost." Those who believe this to be a "minor" change are obviously not familiar with the intensity of argument between, for example, devotees of the King James version of the Bible and those committed to more recent translations; or between "Rite I" and "Rite II" Episcopalians, for whom the difference between the 1928 and 1979 Prayer Books is a gulf become a chasm. The evolutionary change in the translation of the divine name came about because the word "ghost" was no longer considered an adequate translation of [Greek Text Omitted]. Perhaps the word ghost had evolved out of its previous connotations, and was now most commonly used to refer to the enduring presence of a dead human being (especially one that liked to haunt its former habitat), or, more problematically, to a faint or false image, such as appears in various forms of photography. The word ghost no longer had the connotations that Christians were trying to evoke when originally employing the word. So they chose a different word. Not everyone agreed with the new choice; and indeed, the two translations now coexist among English-speaking Christians.

The circumstance of father is very similar. On the one hand, the word carries a high degree of positive cultural capital (protector, breadwinner, powerful authority) just as "ghost" retains certain "pneumatic" qualities. But just as the exclusive use of ghost may tend to conjure up weak television images and bad horror films, so the exclusive use of father may produce, for many people, images of abuse, laziness, tyranny, and anger. Like the word ghost, the word father does not always have the effect it should have on the Body of Christ, and especially on the individual bodies within it. This is why we must allow for the possibility of retranslating the divine name.(61)

As I noted in the first section of this essay, those who have argued against any such alteration have weighted their cause with some very heavy freight, arguing about the potential invalidity of baptisms and the failure to identify the same God as Jesus did. One writer has even claimed that "the triune God has named himself, and he likes his name."(62) (Really? In what language does he like it? And was he mad when we changed his last name from Ghost to Spirit?) Surely a bit more theological humility is in order when Christians speculate upon the "most appropriate" name of God. In addition to the many problems of translation, we have gained no consensus as to whether we can even speak of "the" name of God, and if so, what that name might be.

In any case, we would do well not to assume that theologians must be the first line of defense for the protection of God and of God's name. By all means, let us argue about the appropriateness (theological, grammatical, and aesthetic) of various names of God. Let us prioritize them; let us counsel the most appropriate spheres of their invocation. But let us not draw lines in the sand over one particular translation of one divine name - even if that name is held, by some at least, to be primus inter pares. For even the most consistent use of a particular name cannot, simply by its employment, guarantee so much as the existence of the one who is so named - let alone assure an accurate or adequate description thereof, and even less, a proper attitude of reverence toward the one who is, in Meister Eckhart's words, innominabile et omninominabile.

In another world it may be otherwise, but here below, when we wish to refer to God, we must use names. Let us not forget that names are merely names. Names you will have always with you; you will not thereby have God. Stat rosa pristina nomine, nomina nuda tenemus.

With much help and encouragement from Margaret Adam and Phil Kenneson. Thanks also to A. K. M. Adam, Fritz Bauerschmidt, Steve Fowl, Reinhard Hutter, Ed Krentz, Carol LaHurd, David Landry, Marianne Meye Thompson, and David Penchansky.

2 George Steiner, After Babel: Aspects of Language and Translation (New York: Oxford University, 1975).

3 Jacques Derrida, The Ear of the Other: Otobiography, Transference, Translation, ed. Christie McDonald, trans. Peggy Kamuf (New York: Schocken, 1985; reprint, Lincoln: University of Nebraska, 1988) 100.

4 Though that, too, is a task that needs to be undertaken; a brief and very preliminary sketch is offered by Ted Peters, "The Battle Over Trinitarian Language," Dialog 30 (1991) 44-49.

5 As two examples, see Ruth C. Duck, Gender and the Name of God: The Trinitarian Baptismal Formula (New York: Pilgrim, 1991); and most of the essays in Alvin F. Kimel, Jr., ed., Speaking the Christian God: The Holy Trinity and the Challenge of Feminism (Grand Rapids: Eerdmans, 1992).

6 This is true whether or not the writer favors retaining "person." A case in favor is offered by Lawrence B. Porter, O.P., "On Keeping 'Persons' in the Trinity: A Linguistic Approach to Trinitarian Thought," TS 41 (1980) 530-48. A case against is presented by Nicholas Lash, Believing Three Ways in God: A Reading of the Apostles' Creed (Notre Dame: University of Notre Dame, 1993).

7 Duck, e.g., argues for a revision of the baptismal formula but finds no support for this position in Jesus' use of [Hebrew Text Omitted] (Gender and the Name of God 59-72).

8 Aristotle, On Interpretation 2.16a28; text in The Categories, On Interpretation, and the Prior Analytics, Loeb Classical Library (London: Heinemann; Cambridge: Harvard University, 1938) 116.

9 For a useful discussion of the various ways in which language refers, see Christian J. Barrigar, "Protecting God: The Lexical Formation of Trinitarian Language," Modern Theology 7 (1991) 299-310.

10 Robert W. Jenson, The Triune Identity: God according to the Gospel (Philadelphia: Fortress, 1982) 18. This sentence, which in its context is used to refer to "Father, Son, and Holy Spirit," applies equally to the New Testament use of [Greek Text Omitted].

11 Jacques Derrida, "Des Tours de Babel," trans. Joseph F. Graham, in Difference in Translation, ed. Joseph F. Graham (Ithaca, N.Y.: Cornell University, 1985), reprinted in Semeia 54 (1992) 8.

12 Karl Rahner, "Theos in the New Testament," in Theological Investigations 1, trans. Cornelius Ernst, O.P. (London: Darton, Longman, and Todd, 1965) 78-148.

13 So, frequently, the contributors to Speaking the Christian God.

14 As, e.g., in Alvin F. Kimel, Jr., "The God Who Likes His Name: Holy Trinity, Feminism, and the Language of Faith," in Kimel, ed., Speaking the Christian God 190.

15 For some other possible meanings, see William F. Arndt and F. Wilbur Gingrich, eds., A Greek-English Lexicon of the New Testament and Other early Christian Literature, 2d ed. (Chicago and London: University of Chicago, 1979) s.v. [Greek Text Omitted]. My use of this lexicon (along with Liddell and Scott, to which I will refer later) is obviously problematic, since these works were clearly influenced by some of the very forms of 19th- and 20th-century theology and translation theory which this essay seeks to call into question. I attempt to counter this influence with the following methodological safeguard: I invoke their authority not to restrict the number of meanings of a particular word, but to illustrate some of their wide variation. In this sense I am operating somewhat against the grain of standard lexicography of the Bauer/Arndt/Gingrich/Liddell/Scott variety.

16 J. N. D. Kelly, Early Christian Creeds, 3d ed. (London: Longman Group, 1972) 30-52.

17 As is clear in St. Basil the Great, On the Holy Spirit (Crestwood, N.Y.: St. Vladimir's Seminary, 1980).

18 English text in Pseudo-Dionysius, The Complete Works, trans. Colm Luibheid, Classics of Western Spirituality (New York: Paulist, 1987) 47-131.

19 Elizabeth A. Johnson, "The Incomprehensibility of God and the Image of God Male and Female," TS 45 (1984) 441-65.

20 On the Orthodox Faith 1.12, cited in Thomas Hopko, "Apophatic Theology and the Naming of God in Eastern Orthodox Tradition," in Kimel, ed., Speaking the Christian God 157.

21 Hopko, "Apophatic Theology" 160.

22 Ibid.

23 Summa theologiae 1, q. 13, a. 11.

24 Shortly, I will turn to the problems inherent in this assumption; for the moment, however, it provides a useful heuristic.

25 Derrida, "Des Tours de Babel" 8.

26 Robert W. Jenson, "A Quick Correction," Dialog 30 (1991) 247.

27 Steiner, After Babel 303-4.

28 Walter Benjamin, "The Task of the Translator" (1923), in Illuminations, ed. Hannah Arendt, trans. Harry Zohn (New York: Schocken, 1969) 78.

29 Of course, transliteration is also interpretation, since resonances associated with the particular features of particular orthography are lost in the process. Consider, for example, the transliteration of the Tetragrammaton into YHWH or, more pointedly (no pun intended), Yahweh.

30 Henry George Liddell and Robert Scott, eds., An Intermediate Greek-English Lexicon (Oxford: Clarendon, 1889) s.v. [Greek Text Omitted]. For my own methodological hesitation about the use of lexicons, see note 15 supra.

31 Arndt and Gingrich, A Greek-English Lexicon s.v. [Greek Text Omitted].

32 Ibid. s.v. [Greek Text Omitted].

33 Benjamin, "The Task of the Translator" 75.

34 Stephen E. Fowl, "Could Horace Talk with the Hebrews? Translatability and Moral Disagreement in MacIntyre and Stout," Journal of Religious Ethics 19 (1991) 5.

35 Benjamin, "The Task of the Translator" 70-72.

36 St. Augustine of Hippo, Confessions 1.8 (13); English text trans. Henry Chadwick (Oxford: Oxford University, 1992) 10-11. Of course, Augustine was not attempting to promulgate a full-scale theory of language here; compare his more nuanced discussion in De doctrina Christiana. Nevertheless, he provides a nice example of the "commonsense" approach.

37 Ludwig Wittgenstein, Philosophical Investigations, trans. G. E. M. Anscombe, 3d ed. (New York: Macmillan, 1953) sec. 1, p. 2.

38 Ibid. sec. 3, p. 3.

39 Anyone unconvinced by Wittgenstein's discussion of this matter should read or see Tom Stoppard's play Dogg's Hamlet, Cahoot's Macbeth (New York: Samuel French, 1980), in which Wittgenstein's four-word language is used as a starting point to illustrate the essential separability of words from meanings (and, in fact, to teach the audience an entirely new language by connecting English words to different meanings).

40 For a discussion of problems that develop in machine translation of material even so banal as weather reports, see Douglas Robinson, The Translator's Turn (Baltimore: Johns Hopkins University, 1991) 23-29.

41 Wittgenstein, Philosophical Investigations, sec. 43, p. 20.

42 Indeed, the multiplicity of meanings in the original is often necessarily obscured in translation. For a detailed example, see Jacques Derrida, "Plato's Pharmacy," in Dissemination, trans. Barbara Johnson (Chicago: University of Chicago, 1981) 65-171.

43 Robinson, The Translator's Turn 8.

44 Mikhail M. Bakhtin, The Dialogic Imagination: Four Essays, ed. Michael Holquist, trans. Caryl Emerson and Michael Holquist (Austin: University of Texas, 1981) 276.

45 Robinson, The Translator's Turn 105.

46 Some of the loci classici include Athanasius, De Synodis 42; Gregory of Nazianzus, Oration 31.7, and Hilary of Poitiers, On the Trinity 1.18.

47 Elizabeth A. Johnson, She Who Is: The Mystery of God in Feminist Theological Discourse (New York: Crossroad, 1992) 82.

48 See Dennis Baron, Grammar and Gender (New Haven and London: Yale University, 1986).

49 Johnson, "The Incomprehensibility of God" 460.

50 Some empirical evidence for the indirect ways in which people process gender-specific words that are supposedly being used "generically" (such as "he" and "man") may be found in the chapter by Mary Crawford and Roger Chaffin, "The Reader's Construction of Meaning: Cognitive Research on Gender and Comprehension," in Gender and Reading: Essays on Readers, Texts, and Contexts, ed. Elizabeth A. Flynn and Patrocinio P. Schweickart (Baltimore: Johns Hopkins University, 1986) 3-30.

51 See, for three examples among many in the work of Luce Irigaray, "Women's Discourse and Men's Discourse," and "Linguistic Sexes and Genders," both in je, tu, nous: Toward a Culture of Difference, trans. Alison Martin (New York: Routledge, 1993) 29-36; "Divine Women," in Sexes and Genealogies, trans. Gillian C. Gill (New York: Columbia University, 1993) 57-72.

52 As does, e.g., Elizabeth Achtemeier, "Exchanging God for 'No Gods': A Discussion of Female Language for God," in Kimel, ed., Speaking the Christian God 4. The sentiment is echoed by several other of the contributors to the volume.

53 See, e.g., Baron, Grammar and Gender 137-51.

54 The case for such a construal of the language is ably argued by Ellen T. Charry, though her point seems directed primarily at those who would banish masculine language altogether (see "Is Christianity Good for Us?" in Reclaiming Faith: Essays on Orthodoxy in the Episcopal Church and the Baltimore Declaration, ed. Ephraim Radner and George R. Sumner, with a Foreword by George A. Lindbeck [Grand Rapids: Eerdmans, 1993] 225-246).

55 As noted by Peters, "Battle" 48 n. 13.

56 Robinson, The Translator's Turn 5.

57 See Stanley Fish, Is There a Text in This Class? The Authority of Interpretive Communities (Cambridge: Harvard University, 1980).

58 Robinson, The Translator's Turn 10.

59 Ibid. 15.

60 Charry, "Is Christianity Good for Us?" 227.

61 In my future work, I hope to explore some of the issues that might be at stake in this process. Some initial suggestions have been sketched by Gail Ramshaw, God Beyond Gender: Feminist Christian God-Language (Minneapolis: Fortress, 1995).

62 Alvin F. Kimel, Jr., "The God Who Likes His Name," in Kimel, ed., Speaking the Christian God 188.