The Feigned Ignorance of the Existence of Other Translations

In my previous post I mentioned a correlation between the lack of variety in available Bible translation approaches and the fact that most Bibles are intended to stand alone (as a ‘one-size-fits-all’ style, that is, as producing Bibles to be used for both study and liturgy, by a wide audience). This is especially true of large committee Bible translation projects where the goal tends to be the out-dated idea of providing a new ‘standard’ translation that would monopolize the market of readers.

This means a missed opportunity for new Bibles to pursue the variety of translation methods (since they are all competing for the ‘word group’ and ‘clause’ level translation method, see previous post).

Today I’m following up the discussion of the implied correlation between the lack of variety and the ever-increasing number of new Bible translations.

It is no secret that the number of Bible versions created in the English language is vast—the number is in the hundreds (or thousands if one counts different editions). So it is strange that new Bible translations appearing in English usually overlook the obvious fact that many of their intended readers will own and read more than one translation.

Many Bibles appearing in English are produced as though they were intended to be read in ignorance of other translations/versions. I suppose it is partly a consequence of marketing (compare how TV networks rarely like to acknowledge the existence of other TV networks in their broadcasts). Such ‘ignorance’ is obviously pretend.

In the real world many readers (especially Bible college students, pastors, religious educators, and Bible study groups) are not so ignorant and so they do not forego reading their Bibles in various English translations. Granted many readers do gravitate to one Bible to use as as their primary Bible for practical reasons, but this does not obviate comparative use of more than one version at various times.

So what difference does this ‘feigned ignorance’ make for Bible readers? I suggest that it not only reproduces an ignorant reading experience (leaving it to the more inquisitive readers to go search out for themselves alternative translations) but that it is partly responsible for the never-ending over-production of new Bible translations in English.

We have already noted that multiple translations are a necessary phenomenon due to the nature of translation. This should be clear from the previous blog post. Here’s how Julio Trebolle Barrera stated it (italics added):

The biblical text is loaded with such richness of meaning in its smallest details that any translation can offer no more that [sic] a few of the many aspects which philological, historical and theological study find in the text. In effect, it is really always necessary to provide several alternative translations. This is one of the reasons why the student of the Bible cannot forego study of the original languages.

The Jewish Bible and the Christian Bible: An Introduction to the History of the Bible (Trans. Wilfred G. E. Watson; Leiden; New York; Koln: Brill, 1997), 126.

Here we will focus on the necessity for English readers of having access to multiple translations.

The point is widely acknowledged by those who study translations (and taken for granted by students of translation studies). For example, Tommy Wasserman concludes that

multiple translations that complement each other are necessary. There is certainly a great need for translations with different levels of language and with varying degree of interpretation, aimed at different addressees and purposes; the translation for the experienced reader should not necessarily be the same as for the first time reader; likewise, those who read the Bible to prepare sermons could need another kind of translation than those who read it as a book for culture. . . . The ideal would be a broad range of translations that could, so to speak, build a bridge into the world of the Biblical authors. . . . idiomatic translation . . . needs to be complemented with the literal translation, not least in order to enrich and amplify the target language and bridge the gaps between our modern world and the Biblical one.

“Theoretical Orientations In Bible Translation: AComparative Analysis of Two English-language Bible Versions” (2001)

Similarly, Philip Sumpter (in his article “English BibleTranslations: An Overview of Current Versions”) found that “None of the authors [translation theorists] cited in this article, each involved in translation projects themselves, claim that only one translation or translation type is sufficient.” (italics added)

So it is refreshing when a translation stops behaving like a TV network and acknowledges the mutual existence of other contemporary translations. This was a welcome feature of the NETBible when it appeared (online 1996; in print 2005) which does occasionally make reference to other versions in the footnotes when discussing an ambiguous passage. The most extreme endeavour in this vein would probably be the Comprehensive New Testament (2008) which provides references to twenty other commonly used English versions for every New Testament verse!

Whilst we should not expect that every new version can or should go to such extremes (and the CNT has its own agenda in doing so), we might at least expect that more Bibles are translated with the knowledge that their translations exist in mutual dialogue with other translations, and to make this fact more transparent to readers.

The myth that a newer Bible will be the version to end all versions (or the one and only Bible translation to be read by its faithful readers) in reality does little to reduce the outpouring number of new versions. In fact it would seem to hinder the proliferation of new Bibles because it fails to properly address the need for necessary complementary translation types that theorists have identified. This in turn is reflected in the production of Bible versions which inherently aim to supplant and compete with rather than supplement and complement.

Ultimately, there remains an ongoing need for newer Bible translations. What is needed, however, is for such translations to improve their awareness of what they are and how exactly they might be achieving what they are setting out to achieve in relation to other translation efforts, by paying attention to what translation theorists and other experts have known and studied about the translation enterprise.

I will follow up the matter with some examples soon.

Truth and Myth in the Literal-Free Diagram (part 4)

I’ve been way too busy lately to blog. But I’m finally now able to add a ‘part 4’ that I’ve been wanting to add since the last post! The current series is because I felt it was important to have something on the internet that critiques the traditional ‘diagram of translation theory’ (sometimes I've called it 'literal-to-free' diagram and sometimes 'literal-to-paraphrase' diagram). In other words something brief that will broaden the horizons of those who will undoubtedly encounter its (mis)use. 

Part 4 concerns the ‘size’ of a ‘unit’ being translated.

The ‘unit size’ being translated may theoretically be quite large or quite small. But, for whatever reason (and there lies a thesis or two!), in almost all English Bible versions the chosen unit size is very similar.  

So in theory a translator could chose to translate a whole paragraph at once (as a single large ‘unit’ that tries to relay the larger ‘point being made’, namely the various rhetorical and social implications being implied) or at the other extreme could chose simply to translate word-by-word, or perhaps at an even smaller scale to translate all the ‘morphological’ elements within a single word (for example the parts of a ‘compound word’, including prefixes and affixes -- we discussed this point briefly in part 1 for Mk 1:1 for εὐαγγελίον eu-angelion (literally “evangel” but traditionally either “gospel” or “good news”, though more effectively “proclamation [of salvation]”). 

It’s probably no surprise that our old friend the ‘literal-to-free diagram’ again does not really help to display the differences. That is, it does not cope well with displaying the different unit sizes employed by each translation. The fault this time lies not so much with the diagram but with the lack of variety of Bibles that translate ‘above’ the size of a ‘sentence’ (that is, larger than clause-level). As a rule, almost every English version only translates somewhere between the ‘phrase level’ (or ‘word group’) and the ‘sentence level’ (or ‘clause complex’).

Putting this observation alongside the traditional diagram produces something like the following:

This time what the diagrams does help to demonstrate is that there is hardly any difference between the alleged ‘extremes’ using the traditional literal-to-free diagram! We’ve had to shrink the diagram down to squeeze between only three or four ‘levels’.

Essentially this is just what Stanley Porter has argued (and whence the above table derives) in:

Stanley E. Porter, “Assessing Translation Theory: Beyond Literal and Dynamic Equivalence.” Pages 117–45 in Translating the New Testament: Text, Translation, Theology. Edited by Stanley E. Porter and Mark J. Boda. Grand Rapids: Eerdmans, 2009. 

The reason for the lack of variety will probably have something to do with the fact that most English Bible versions are designed to stand alone (not many are deliberately made to be used primarily in conjunction in another existing translation). This means that there is an inherent pressure to find a compromise between the potential extremes (as there would presumably not be something else to gauge such an extreme against for comparative purposes). 

However, the more English versions that are produced makes such an assumption seem more and more ridiculous (anyway, that’s another tangent we could easily follow up…no doubt another time).

I’ll resist saying that’s all for this series this time (on the literal-to-free diagram) but I really would like to move onto other matters…

In all, I think that the diagram still has some utility if the issues raised here (including parts 1, 2, & 3) are also taken into account.

Truth and Myth in the Literal-to-Paraphrase Diagram (part 3)

Now that I've critiqued what such diagrams claim to display, by looking at what is “literal” (part 1) and at “relative positioning”(part 2), it’s appropriate to mention some things that the literal-to-free diagram does not account for. 

For now let’s discuss: “reading level” and “register.”

“Reading level” is to do with the range of vocabulary (and complex sentences) that a Bible version employs, and therefore it may have an “age level” suitability (identified as a school-grade reading level). These recommended “reading levels” will not be agreed upon by all. I tend toward the more optimistic side of the Australian-based reading levels.

The NIV, for example, is suited for about a grade 6 competence reading level which is lower than the NRSV (approx 10th grade). But differences in reading level are not always very noticeable, so Mk 5:25-28:

NRSV: Now there was a woman who had been suffering from hemorrhages for twelve years. She had endured much under many physicians, and had spent all that she had; and she was no better, but rather grew worse. She had heard about Jesus, and came up behind him in the crowd and touched his cloak, for she said, “If I but touch his clothes, I will be made well.”

NIV: And a woman was there who had been subject to bleeding for twelve years. She had suffered a great deal under the care of many doctors and had spent all she had, yet instead of getting better she grew worse. When she heard about Jesus, she came up behind him in the crowd and touched his cloak, because she thought, “If I just touch his clothes, I will be healed.”

Actually both NRSV and NIV have simplified this single long sentence in similar ways, by breaking it into three sentences to make it easier to read. “For she said” (NRSV) also belongs to a more formal “register” than NIV’s “because she said.” 

(NOTE "Formal register" is nothing to do with "formal/literal method of translation"--"formal register" has a very different meaning. "Register” refers to the kind of language suited to a particular purpose or setting. The kind of language spoken among friends tends to be much less formal than the kind of language one is expected to use in a high-school essay which is still not as formal as that of a PhD thesis. So "formal" as in "not casual" or "not informal." 

Sometimes, a more advanced reading-level corresponds to a more “formal register.” Compare, however, Eugene Petersen’s The Message (at the furthest extreme on the right of the diagram) and the Good News Bible (fourth version from the right-hand side). The Message was written in contemporary American English for adult readers, using highly idiomatic English phrases that try to engage modern readers with a more “casual” tone. By contrast the Good News Bible (GNB) was made for low-level readers, especially non-native English speakers (whose first language is not English). What is interesting about this comparison is that even though these two are positioned closely together on Bible version diagrams, they are quite different in both register and reading level.

The GNB was written in the kind of English “commonly used by people most of the time” and uses a moderately formal register (not colloquial style) and so it tries to avoid informal language (slang and colloquialisms). It also tries to avoid “big words” (theological jargon). It is generally aimed at a moderately low reading level (about grade 4 or 5 [Australian grades]). Likewise its replacement version, the CEV (made in 1995), was aimed at quite a low reading level (about grade 3 or 4) so that reading aloud could be done easily and fluently. By contrast, The Message requires a much higher reading level (grade 6+) and deliberately uses informal (American) language.

Yet the literal-to-paraphrase diagram does not display any of these factors. Simply adding reading levels to the diagram would appear rather jumbled:

By now it should be becoming clear that arranging English Bible versions on a two-dimensional scale does not do justice to the complexity of Bible translation philosophy. Such diagrams are more useful for exposing them for what they fail to show.

I could go on critiquing such diagrams . . . For example, the degree of “literalness” means different things to different translators. For example, the KJV is usually considered to be quite “literal” (that is, relatively closer to the left-hand extreme). However, it lacks one of the most common components of "formal/literal" methods--the use of “word consistency” whereby the same source words are consistently represented in the target text by the same corresponding words in English. Instead, the KJV translators tried to avoid the overuse of stereotypical correspondences so as help make the translation more readable (by employing a variety of English words for the same Hebrew or Greek word).

So that’s about it folks for the literal-to-free diagram!

Oh, here’s one more picture showing the lack of correspondence between word counts and alleged “literalness” (care of epistlesofthomas based on stats from a paper by Karen H. Jobes “Bible Translation as Bilingual Quotation.”):

But stay tuned for plenty more snippets on Bible translation theory . . .

 

Truth and Myth in the Literal-Free Diagram (part 2)

Part 1 of this series discussed a basic notion within the alleged “translation philosophy scale”, namely “literal” translation (at the left extreme, see part 1 for two example diagrams).

In part 2, I will discuss briefly another basic notion of the diagram, namely “placement” (relative positioning) of certain English Bible versions (within the literal-to-free diagram).

“Placement” or “positioning” is what appears to give such diagrams their basic utility. It enables people to make quick judgments about various well-known Bible versions by positioning them across an alleged “translation spectrum” in order to do two things:

1.  to note how close (or how far) each version sits to the extremes (“literal” on the left side or “free” at the right side), as well as 

1.  to observe where a particular version sits relative to other versions.

The key problem, however, is that in reality not every verse of a particular Bible version will satisfactorily maintain its “position” according to the alleged scale.

For example, if we take the two most “polarized” versions (in the traditional diagram)—NASB and The Message—we would expect that every verse would always be radically different in each. And for most verses their relative positioning appears accurate, that is, in most verses the NASB translates much more formally (“literally”) than The Message. But we don’t always find the alleged extremes we would expect. That is, when we look at a list of names, such as the names in Lk 6:14, we find that The Message is barely much less literal or more “freely paraphrased” (“though-for-thought”) than NASB in that particular verse.

So for all the other versions not at the extremities of the diagram we find that not every verse of a particular version will maintain its alleged position in relation to another version. So, for example, the NRSV will not always more literal (less free) than the NLT for every verse (even though the diagram places them at some distance apart). If we compare Prov 16:3–9 in the NLT with the NRSV it is difficult to say which version is always more literal. So for verse 8, for example: Better a small (amount) with righteousness than great (amount) without justice the NRSV is hardly much more literal than NLT:

NRSV: Better is a little with righteousness
    than large income with injustice
NLT: Better to have little, with godliness,
    than to be rich and dishonest.

Whilst NLT uses “to be rich” to correspond to the phrase “great produce”, the NRSV has “large income.” Both are trying to relay the overall implied sense of “to be wealthy” or “to have riches.”

In all, the notion of relative positioning (and therefore the entire diagram) can be misleading, particularly as it suggests that translation philosophy is simply a two-dimensional enterprise, as though a “literal-to-free” scale can satisfactory explain the translation philosophy of all Bible versions.

The main reason why no version will maintain its alleged relative position for every verse is that translation is not primarily done with an eye on keeping any alleged relative position on the diagram! The translation philosophy in most contemporary Bible versions is primarily set by a mix of other agendas.

I will briefly discuss two of these in part 3, namely “reading level” and “register.”

Truth and Myth in the Literal-Paraphrase Diagram

Presently, as the topic is too big for one post, I am beginning to blog a short series of posts dealing with the truth about the categorization of Bible versions as more or less “literal.”

Within popular Bible translation theory and method it is perhaps the most widely perpetuated myth, namely the artificial categorization of translation philosophy according to a degree of “literalness,” by creating a polarized “literal-to-paraphrase” diagram (with “interlinear” at one extreme and free “paraphrase” at the other extreme).

The following two diagrams (both originally from Zondervan I think) whereby “literal” is contrasted it with free “paraphrase”:

Like many myths, such diagrams are not without merit and do contain elements of truth.

The first post in this series begins by discussing where the diagram seems to succeed in saying something true. 

Part 1: What is "Literal"?

There is some truth to the notion that some versions are generally more or less “literal” than others. And at the left extreme (Interlinear and NASB) we can honestly say that the NASB does fall relatively close (closer than the others in the diagram) to an “interlinear” approach to translation (interlinear being a method that preserves the original words and simply supplies English words directly below every Greek or Hebrew word).

But what exactly counts as “literal” is problematic. The notion of something being “literal” is a subjective judgement. The complexity of language means that there are potentially infinite ways different people might choose to represent speech from one language more or less “literally” in another language. 

Are, for example, any of the following successfully “literal” translations of the first verse in Matthew (without footnotes)...?

The book of the genealogy of Jesus Christ, the son of David, the son of Abraham: (NASB 1977)

The record of the genealogy of Jesus the Messiah, the son of David, the son of Abraham: (NASB 1995)

Most “literally” Mt 1:1 is composed of the Greek words:

Βίβλος  γενέσεως  Ἰησοῦ  χριστοῦ  υἱοῦ  Δαυὶδ  υἱοῦ  Ἀβραάμ

which might be “literally” represented by replacing each word with an English word, that is, by using an “interlinear” method of representation:

Book  of-record  of-Jesus  Anointed  son-of  David  son-of  Abraham

 Already we can see that the notion of “literal” quickly degenerates into a provisional, “makeshift” representation even though I have already begun here to make decisions about which English words to capitalize, and to add other punctuation (which is needed in English but not in ancient Greek) to make it look more like a translation and less like an interlinear. I have also attached “of” using hyphens (for genitive cases) in order to indicate that the number of words chosen does not match the number of Greek words. So I have still not sufficiently managed to “translate” into English the sense of the entire Greek verse (grammatically and/or semantically). Granted, our verse has no verb and is not really a complete clause (sentence). Still, we are hovering at the boundary between translation and interlinear.

“Book” is, however, somewhat anachronistic here in that “books” (in the modern “book” format) did not really exist yet … scrolls and wax tablets existed … so “book” here would have to mean something like “scroll” or “document” or “record,” which is why many Bible versions avoid the term “book” in Mt 1:1. “Book” seemed like a suitable “literal” equivalent for “biblos” probably because of the etymological histories and usages of both words where similar enough (“biblos” [papyrus-bark] and “book” [beech-bark] extended by both to be used for creating a record by means of writing: apparently the English word “book” originally meant “writing tablet” and is probably etymologically connected with “beech” as in the name of the beech-tree, OED 2nd ed., 1989). 

But is the etymology of “book” really relevant here? Not really, we don’t expect readers to read words etymologically! In fact if we want a “literal” translation to work we are hoping that readers can automatically recognise that the chosen English words take on meanings based on the source culture, not the target culture. An example like “book” might not be too unsatisfactory so long as the reader can resist imposing more modern ideas of “book.”

But why stop at “book”? Why not go more literal and translate “biblos” as “papyrus”? Unfortunately it seems one can always find ways to go more “literal” when it comes to representing individual words … but at what cost to the sense of sentences? A “literal  translation” still has to deal with “sense” of the words and (preferably) of the sense of words used in combination. Thus “biblos Yesou,” as a phrase, determines that “papyrus of-Jesus” is insufficient as a translation (sense-wise), and so we see that “more literal” begins to become nonsensical and inaccurate (since the medium of writing material is probably the wrong sense to take). Hence more “literal” eventually leads to a nonsensical interlinear. We cannot adequately relay meaning by simply providing fragmented pieces (interlinear glosses).

And what about transliteration? Almost all Bible versions will choose to transliterate certain words (such as with χριστοῦ as “Christ” in Mt 1:1). Transliteration primarily intends to carry over the phonetic sounds of the source (here Greek) and is presumably for when the target language (English) lacks a good translation “equivalent” (another common example is using the word baptize for βαπτίζω, rather than using a translation such as “wash” or “clean” or “cleanse”).

Does transliteration count as more “literal” or less “literal”? If transliteration were to be considered (always?) more literal then it is interesting that no Bible versions have chosen the word “genesis” to “translate” (transliterate) γένεσις in Mt 1:1 - doing so would give us

(The) book of genesis of-Jesus Anointed Son of David Son of Abraham

and readers might then notice not only that the first verse is functioning as a title for the whole Gospel of Matthew as well as a description of the first two “historical” chapters (a “genealogy” and birth “origins story”) but would also note the probable allusion being made to the same formulaic expression βίβλος γενέσεως (book of-genesis) used several times in the early chapters of the book of Genesis in the Greek OT, which translated the Hebrew phrase sefer toledot (stories of family generations/genealogies). So transliteration might sometimes be considered as more successfully literal (as “genesis” is here), but not necessarily always.

Finally, I must mention compound words. Let's look at the first verse in Mark (with an interlinear):

Ἀρχὴ τοῦ εὐαγγελίου Ἰησοῦ χριστοῦ [υἱοῦ θεοῦ]
beginning of-the gospel of-Jesus Anointed [Son of-God]

The word I’m interested in here is εὐαγγελίου (which might be transliterated as “evangel” but no Bible versions I know transliterate this compound word as evangel). By compound word I mean that it has two (or more) parts which can (in theory) each be translated separately. Many Bibles consequently translate τοῦ εὐαγγελίου as “of the Good News” (or “of the good news”). Should we always split up compound words like εὐαγγελίου? Probably not. Bible versions do not have any strict method for always translating separately the parts of Greek compound words with separate English words. In fact most compound Greek words should probably not be split up, just as most compound English words are considered as whole words. 

The question here for εὐαγγέλιον is: Is “Good News” more “literal” than “Proclamation”? Perhaps a more literal translation for the “eu” prefix is “well” or “wellness.” But perhaps more literal would, in fact, be to translate the “eu” prefix with its intended meaning within the Gospel of Mark, namely as “salvation” or “victory” or “peace” given that the word belongs within the “semantic domain” of heralds, kings and kingdoms. . . .

Conclusion 

In this first post (in my series dealing with truth and myth in the “literal-to-paraphrase diagram”) I have briefly discussed “literal” (at the left extreme) by looking at two simple verses (Mt 1:1 and Mk 1:1) neither of which contained difficult grammar or strange speech idioms. Even here we found we could always find ways of going more “literal,” but at the cost of losing meaning. Since translation is about relaying meaning, when more literal leads to a disregard for meaning, it can no longer be considered translation but merely an interlinear representation. On this much our diagrams may actually be in agreement with the above points.  

Unfortunately the “diagrams” above are not very clear about the significance of the boundary between interlinear and a “literal translation” like NASB. They might seem to suggest that “interlinear” is simply an extreme form of literal translation whereas in reality an interlinear is not really a translation if it only translates the broken pieces. Some Bible versions attempt to position themselves very closely to the boundary between “interlinear” and “translation” probably in the hope that they will provide more insight into individual words by playing chicken with an interlinear approach. But more “literal” does not necessarily lead to more “accuracy of meaning,” especially as “literal” ultimately depends on interpretation of words in combination. When it comes to the translation of grammar of clauses (sentences), the issue of what might count as more literal is not a simple cut and dry issue.  

Transliteration is sometimes more successfully “literal” (as is “genesis” in Mt 1:1) but transliteration is not usually considered to count as translation (even though used by most Bible versions!). Similarly, compound words do not usually require splitting into separate words but this is again not a clear cut issue. Likewise, interlinear “piecemeal glosses” are not really translations when they ignore grammatical combinations of phrases and clauses.

So in conclusion, since interlinears do not translate the meaningful combination of words, an “interlinear translation” is really an oxymoron (self contradictory). And to further complicate things we could talk about “machine translation” (translation performed by a computer) where an attempt at translation (to reproduce meaningful grammar) has been made but inadvertently falls short as unsuccessful, ultimately landing us back in the realm of the nonsensical, piecemeal interlinear. 

In the next post (part 2) I plan to discuss some issues regarding the “placement” (relative positioning) of certain Bible versions on the “mythical diagram.”