Bases and Models Revisited: The Importance of Using Different Types of Reference Translations

Bases and models

Some years ago, Harold Fehderau developed a basic theoretical outline for what subsequently became common practice in many Bible translation projects. He called the approach “the role of bases and models in Bible translations” in a seminal article in The Bible Translator in 1979. The starting point for this approach was the fact that most translation projects worldwide did not (and still do not) work directly with the source languages, but rather work(ed) with the biblical text by way of an intermediate reference translation (or translations) in one or more languages of wider communication within the local context.

Fehderau pointed out that most translators tend to use one intermediate translation almost exclusively. He further stated that if the translators depend on a formal intermediate translation, the comprehension level will not generally be as good as if the (usually quite literal) translation were being made from an intermediate translation that is more intelligible. If a functional intermediate translation is used, the comprehension level can normally be raised somewhat, according to Fehderau. He claimed, however, that the comprehension level can be raised much more by a judicious use of a formal translation in order to have a window on the original languages (a base), on the one hand, as well as a model for the style (register, naturalness, etc.) that the translators are aiming for, on the other. To quote Fehderau at some length (some of the language he uses is dated now, but the fundamental ideas are still sound),

Most translators understand the importance of modifying the form of the original language in order to preserve the most accurate meaning possible in the “receptor” language (the one into which the translation is being made). But it is still largely beyond the ability of all but the most experienced translators to carry out the kind of work required to find the component meaning of a word, for example, or to find the structure of a complex expression. As a result, many translators do not feel fully equipped to do adequate work.

During the course of a translation workshop or through other contact with Bible Society personnel, translators soon realize that there are modern language versions of the Bible available that measure up to the standards that are being taught. So, in order to make up for their own inability to apply the procedures for finding the basic meaning of the text, many translators adopt one of the versions as their translational base. What they then produce in their own language is often simply a literal version of such versions as Today’s English Version (Good News Bible), Français Courant (French), or Versión Popular (Spanish). The translation in the receptor language is usually quite a bit better than a translation done literally from a formal equivalence text (such as the Revised Standard Version or the King James Version). But the new translation is still quite wide of the mark. The fundamental mistake is that a text which should be considered simply as a model of good translation is used as a base from which to work. (Fehderau 1979, 401)

Yet in practice it is still the case that many translators and translation teams will use basically one translation with minimal reference to any others. This happens particularly when the translators’ understanding of the language(s) of wider communication is limited. The clear advantage is that translators will by definition be translating from the text they are best able to understand, which presumably gives them the best chance to represent the original text well. But there are pitfalls to this method, some of which Fehderau alluded to in his article (403–4). One such pitfall will now be examined, from a consultant visit to the Tojolabal translation project being carried out in southern Mexico.

“Sanctuary” and related terms in the Old Testament

The issue surfaced as we were using the Biblical Terms tool in the ParaTExt program to check certain related terms which had been generally translated santuario (sanctuary, holy place) in the Traducción en lenguaje actual (TLA), a functional equivalent translation in Spanish. We will return to the Biblical Terms tool later on, but this generic solution works quite well in many cases in Spanish, since the contexts within which the terms occur usually sufficiently make the distinctions to which the terms themselves specifically refer.

The list of terms we were checking is as follows:

house of God

house of Yahweh

tent of meeting

tabernacle

temple

holy place, sanctuary

These terms do not exactly comprise a natural class, but have been treated together because of the fact they were all most often represented by the same term (santuario) in the TLA translation, and the translators in this particular (Tojolabal) project followed TLA closely (thus creating a “daughter” translation of TLA).

It would be convenient if the terms in the biblical texts were as straightforward as they are portrayed in sets of Bible illustrations. Below are three of the major terms—tabernacle, tent of meeting, and temple—as interpreted in the Horace Knowles set of Bible illustrations (Figs. 1 and 2).

OPEN IN VIEWER

Figure 1.

Tabernacle / tent of meeting (LB00259).

OPEN IN VIEWER

Figure 2.

Solomon’s temple (HK00245).

However, in the biblical texts nearly all the terms in the list can have a variety of meanings and are used (at times seemingly interchangeably) to refer to various places within and without the larger structures of tabernacle and temple. For example, in various contexts “holy place” or “sanctuary” can be the tent of meeting itself inside the tabernacle, or the area inside the tent, or an area surrounding it, or the area behind the curtain inside the tent (see Appendix).

There are also many cases where “tent of meeting” and “tabernacle” seem to be used interchangeably. This may be due to synecdoche (whole for part or part for whole), or perhaps is simply an indication that the ancient authors did not always wish to make a distinction between the terms. And the term “tabernacle” may be simply a way of describing the tent of meeting. Thus Exod 40.29 (NRSV): “He set the altar of burnt offering at the entrance of the tabernacle of the tent of meeting.”

Finally, “house of God” and “house of Yahweh” seemingly were used variously to refer to tent, tabernacle, and temple (much like “holy place” or “sanctuary”) at various times in Israelite history. (There even seems to be a preference for certain of the terms “tent,” “holy place,” and “house of God/Yahweh” on the part of some biblical writers, e.g., the Chronicler. The complex relationship among the terms in the list deserves fuller treatment, but is not dealt with here.) Of the list of terms in the biblical texts, the Hebrew term for “temple” alone seems to be stable, referring either to a building in Jerusalem or a non-Israelite (“pagan”) house of worship.

“Sanctuary” and related terms in the Tojolabal translation project

In the Biblical Terms tool (in ParaTExt) for the Tojolabal project, the terms in the list above are given in Spanish, as in the following:

casa de Dios house of God

casa de Jehová house of Yahweh

tienda de encuentro tent of meeting

tabernáculo tabernacle

templo temple

lugar santo, santuario  holy place, sanctuary

The Tojolabal translators have variously used words for “house” (na’its, naj), “tent” (karpa), and “meeting place” (tsomjel) to render the biblical terms. In the case of the last term, the default meaning is “place where people meet together,” and is the common Tojolabal term for church. This seems to have been taken from the translators’ understanding of the church as congregation in the New Testament, and tsomjel has become a fixed term which they have extended to include (at times) references to various holy place(s) in the Old Testament as well. The problem is that (especially in terms of the tent of meeting, but also the tabernacle and temple as well) the meeting in the Old Testament text is not between/among the people (the “congregation” in the Old Testament), but is between a representative of the people (Moses, high priest) and God.

Now it may be the case that the term tsomjel is something like a dead metaphor and that the congregational aspect is no longer in focus. But it seems more likely that the limited knowledge of the OT cultural context has led the translators to use the only term they have in their language for a “holy place”—their church. If the latter is the case, the translation suggestion would be to choose something more generic like “holy place” (following their model translation more closely) or else to find different terms to distinguish some or all of the actual objects being referred to.

The problem is not merely that the translators followed a single model (the TLA) to the exclusion of other (including more formal or “base”) versions, but that, in seeing just one term nearly everywhere in their model, they naturally gravitated toward the term they were acquainted with in their context for a holy place—their church. Ironically, if they had followed their model more closely (literally), they may well have avoided the problem in this particular case.

But the better practice yet would be to follow Harold Fehderau’s advice and consult at least one base (formal) and one model (functional) reference version, along with other available exegetical helps. It should be noted, however, that this method takes some training as well for translators to understand that intermediate translations vary quite a bit from each other and for quite a few different reasons, and to be able to sort out the issues represented by these differences. Again, Fehderau is well aware of the limitations of his base/model approach:

This approach does not guarantee that the interpretation of the base text is the correct one in the translation, nor does it guarantee that the interpretation given in the text is the correct one. It is still necessary for a Biblical expert to work with the translation team to make sure the best text has been followed at all points.

This approach also does not guarantee that the language usage will be completely natural. The translation will have to be submitted for review to many other native speakers of the language, and subjected to testing procedures. The translators must also be aware of the patterns of usage in the receptor language that have to do with the meanings and structures of paragraphs and larger units in the language. It is not enough to only think in terms of words, phrases, and individual sentences as they translate. (Fehderau 1979, 414)

Amos 4.1 in the RV, DHH, and TLA: an example of a complex base–model analysis

Fehderau’s point of departure is to use one formal base and one functional model, but in actual experience the situation is often more complicated. Below is how the metaphorical phrase “cows of Bashan” (perhaps a fixed idiomatic expression in the Hebrew text) from Amos 4.1 is rendered in various versions. Reina-Valera (RV) is a formal translation, while the other three are functional.

RV 1960 “vacas de Basán”

DHH 1979 “aristócratas damas, flor y nata de Samaria”

DHH 1994 “vacas de Basán, damas de Samaria”

TLA “Ustedes están gordas como vacas de la región de Basán”

RV 1960, unsurprisingly, translates the expression literally, but the other versions vary quite a bit. DHH 1979 tries to translate the phrase with a rather different metaphor, flor y nata (flower and nectar), which conveys something like “cream of the crop” in English. But it adds a non-metaphorical reference, aristócratas damas (aristocratic ladies). TLA goes back to the “cows” metaphor. TLA also adds a non-metaphorical reference to the women of Samaria (the pronoun ustedes “you”) for clarity. DHH 1994 uses the phrase damas de Samaria (women of Samaria) for the same purpose but retains the “cows” metaphor as well.

Interestingly, the flor y nata expression only works in this context if read as sarcasm. Sarcasm is not so easy to convey in written texts, so it is perhaps not surprising that the subsequent edition of DHH went back to the original metaphor. It could well be argued that RV is more understandable (as long as the metaphor is not taken literally as a reference to cows) than the DHH 1979 edition, but it does require that the reference be clear. Thus TLA captures both aspects of idiom and reference, though some might argue that readers do not like everything to be explained to them.

The point here is that all Bible translation is caught in the middle of divided loyalties to the source texts versus modern-day audiences. (Anthony Pym is best known for bringing this idea forward in translation studies; see, for example, Pym 2012.) If one wants to achieve a formal similarity with the source text in Amos 4.1, cows will certainly be mentioned, and will not likely be explained. If one prefers to help the modern reader understand the meaning of the text, cows may or may not be mentioned, but the rich women of Samaria almost certainly will be. Having the binocular vision of both a formal base and a functional model reference text in view almost certainly will help the translator find his or her way along this treacherous divide, whether by some creative solution in the translation itself, or by means of footnotes.

Conclusion

After almost forty years, Fehderau’s base–model approach still has much to teach us. Some of us may have replaced terms like “receptor language” with “translating language,” and others may find the phrase “component meaning of a word” almost quaint. Furthermore, translators have made great gains in education and experience in the time since the article appeared in 1979, so that we would rarely (if ever) speak anymore of their relative “inability to apply the procedures for finding the basic meaning of the text.” Still, the problem of translators many times following only one intermediate version is still with us, and Fehderau reminds us there is a better way.