It is easier for a hippopotamus to…

I recently returned from Africa, where I was working with a translation of the Gospel of Luke into a language that has had no previous Bible translation and a culture that has had very little contact with Christianity. I was not responsible for producing the translation into this language, but I was responsible for evaluating the translation. This was a very isolated language group, geographically and culturally. But the people were not what I would consider primitive. They are sophisticated in their own way. The traditional language and culture provided some key language for the translation that I would not have expected, including words for “altar,” “priest,” “miracle,” “holy,” “spirit,” “disciple,” “righteous,” “grace,” “savior,” and even “synagogue” (literally, their word for a meeting house).

As to be expected, there were some translation challenges when it came to certain terms for flora and fauna and geography. Though there are sheep and cows, this group has no donkeys or camels, and no words for them. It is possible to say “east” (the side where the sun rises) and “west” (the side where the sun sets), but no simple way to say “north” or “south.” Some concepts in the Bibe have to be translated as a phrase, such as “people mouth of God” for “prophets” and “woman death of man” for “widow.” (I believe these phrases come off sounding better in this language than they do in English.) It is just a fact of translation that you cannot always expect to have a matching target language word for every source language word, but that doesn’t render translation impossible.

I was fascinated to find out that in this language group, people ride cows. And their translation of Jesus riding into Jerusalem had him riding in on a cow. Interesting! Unfortunately, this was not historically accurate. I would only resort to borrowing a word if there is no other good option, because if you are borrowing words, you aren’t translating. However, in this case, we borrowed a word for “donkey” to say what Jesus rode into Jerusalem. The story of the Good Samaritan still has the Samaritan putting the injured man on a cow to take him somewhere where he can be fixed up. Some English translations like the NIV, CEV and NLT have “donkey” there, but the Greek has a more generic word.

This brings us to the verse in Luke that reads, in this language, “It is more easy for a hippo to pass in the hole of a needle than a rich person to accept that God can be king over him.” This is the English backtranslation of Luke 18:25. Interesting! Is this legitimate, or, for the sake of accuracy, do you have to insist that a word for “camel” be borrowed into the language to translate this verse? I have a hard time saying that the translation is not accurate and legitimate. I kind of like it, really. Now, obviously, if you were looking for a match for the Greek word κάμηλος, this target language word backtranslated as “hippo” wouldn’t seem to be a good match. But if you widen your perspective a bit, and don’t just look at words but rather at meanings in context, then in this particular context, a target language word for “hippo” is arguably a good translation of Greek κάμηλος.

In Bible translation, as in any kind of translation, there are norms that govern acceptable behavior. The norms don’t answer the question of what is and is not legitimate translation, which is very elusive to try to answer, but rather what is and is not considered acceptable in a community of practice. Granted, there are different subgroups, and not all Bible translators adhere to the same set of norms. But one norm in Bible translation that is widely–though not necessarily universally–accepted is that it is possible to take a little more liberty in translating an idiom, metaphor, proverb or parable, because the meaning of those language units is more than just the sum of the parts. I would argue that, for a language group that knows about hippos but not about camels, and based on testing with representatives of the target audience, it might be more accurately meaningful to translate Luke 18:25 using a target language word that corresponds to our English “hippo” than to try to find some way to use a word that corresponds to our English word “camel” that is not naturally a part of that language.

In which I ask if there’s any value to conveying morphosyntax

There are many things people to use describe translations: literal, formal, functional, dynamic, idiomatic, figurative, literary, interpretative, accurate, thought-for-though, word-for-word, relevant, paraphrase.

Most of these suck. Most of them are almost entirely useless in my opinion. They get so misused and everyone uses them in their own subtly different way.

Instead I think it’s much better to ask what a translation is attempting to convey from its source. It might try to convey the meaning (semantics) of the source. It might try to convey the purpose of the author (pragmatics, broadly.)

When people talk about a literal, formal, non-interpretative or word-for-word translation, they usually mean that it attempts to convey the morphology and syntax of the source into the target language. So my question to BBB’s readers is: is there any value in conveying morphosyntax? If you believe there is, put your best case forward and convince me!

First things first

Hi! Welcome to the new year. For those who don’t know me, I’m Dannii, an Australian linguistics student. I’ve guest posted here once before. But for my first official post on this first day of a new year, I thought what could be better than to write about the word first? Specifically that favourite verse of many: “But seek first the kingdom of God and His righteousness, and all these things will be added to you.” (Matthew 6:33, ESV)

Rich Rhodes blogged previously about this verse, focusing on seek‘s formal usage in modern English and its unsuitability for this very ordinary Greek verse. Today we’ll look at first, which unfortunately does not mean what the ESV team, and many other translation teams, thinks it means.

We all understand the basic meaning of first, but to define it with a more obscure word, it’s all about primacy. There are many contexts where first is used, but two of the most common are time and prominence.

First is very often used to talk and compare things or events in time. We’ll say we liked the first movie more than its sequel, or that the first person to finish the race will get the prize. We start driving in first gear and there are a whole lot of churches named “First …” (though I think that’s an American thing.)

But first is also used about things which have no special place in time but are prominent for other reasons. Mathematical first principles are generally neither the principles first discovered, nor the first principles taught to maths students, but are instead the foundational principles that everything else is build upon. When safety is called a first priority it doesn’t mean that after safety is achieved we move on to other priorities, but that at all times safety must be practised.

It’s this second meaning that applies to Matthew 6:33. Jesus does not mean that seeking God’s kingdom is our first goal after which he will give us others, but that at all times we must be focused on the kingdom. But there’s a catch. All the examples of first I just gave are adjectives but in our verse it is an adverb!

Adjectives and adverbs are very similar: they’re both modifiers and many words can be used as both parts of speech. First is one of these. But… the senses each part of speech allows are limited. I don’t know how it was in Englishes past, but modern standard English usually only allows the adverb first to have the sense of time. You can check this yourself with the Corpus of Contemporary American English. I admit I didn’t check all 25671 times when first follows a verb, but I think it’s quite clear that aside from the idiom first and foremost, first in this context has a temporal meaning.

What does this mean for Bible translation? Well quite simply that the ESV has it wrong, as do many other translations. Of the three English Bible translations in progress, only the CEB gets this right. I already submitted a revision suggestion for the NIV, I guess I now have to do that for the ISV too (which is even worse than the ESV at this verse.) I just hope no one will get confused and think that seeking God’s kingdom is a completable goal based on this mistranslated verse.

But this is only a single verse. I have no idea how many other times mistakes like this have been made throughout the rest of the Bible. And more generally, it shows a major flaw with the word-for-word principle. That first may match well the meanings of the corresponding words in the original languages isn’t enough. The full range of meanings that the adjective has are almost irrelevant when the adverb has a limited subset of them. Word-for-word cannot be the dominant translation principle as contexts matter too much.

The Exceptional Translations

We talk a lot about word-for-word translation. In a previous posting, Wayne points out “[a] truly word-for-word translation would be an interlinear translation,” So, it seems to me, no translation would then be truly word-for-word. Wayne goes on to offer three possible explanations for what ESV proponents might mean by the phrase as they compare their preferred translation with those translations felt to not be word-for-word:

  1. There is greater concordance of words…
  2. There is a higher degree of formal equivalence…
  3. There is an attempt to translate each word of the original biblical text with some word or words in English.

I think the last is the most likely. And, to be more specific, I think the term word-for-word, as it is used by its proponents, refers to the translation method. I think it has less to do with concordance and/or formal equivalence. These first two explanations appear to me to form metrics which measure the word-for-word quality of the translation. The later, however, is the kernel.

I think this would explain why ESV’s translation of Hebrews, when compared to other NT books, is less conformable to the word-for-word norm. Hebrews doesn’t measure up as well to the “matching a word in the original with a word in the translation” metric. Even though it is still thought of as word-for-word, it evidences exceptions to the method.

Wayne’s use of comparative terms (greater concordance, higher degree) in his assessment of what the ESV team means by word-for-word translation also indicates there are exceptions to the word-for-word rule when this method is followed.  The word attempt in the last explanation clearly shows that exceptions are required.

And this brings us closer to the point of this post.  I think there is a deeper reason word-for-word proponents allow for, even require, exceptions to their otherwise word-for-word method.  And that’s my question:  When following a word-for-word method to translations, why are there exceptions?

Whenever we discuss a translation choice according to a word-for-word perspective, we invariably run head-first into the list of exceptions. Word order is an obvious problem. So, we have to handle that exception. Metaphors sometimes don’t work. Though, thankfully, many Biblical metaphors are quite basic to human life and they tend to work in Western, civilized cultures.  But it’s still true that as the translation process progresses, each metaphor is considered for its value within the destination culture—that is, should an exception be made?  Within the word-for-word perspective there’s hesitancy to translate an original word with more than one destination word. Doing so would also be an exception. However, it is necessary to do so in many cases; so, the exception is allowed. Translating with less words is also frowned on—it is exceptional. John 3:27 presents a common case. The KJV and NKJV (as well as other word-for-word translations such as NASB and ASV) render it “John answered and said…” The ESV allows (rightfully, I think) the exception, ”John answered,” rendering with two words what is four in the original. Puns and other purposeful ambiguities are not even thought of as exceptional. They’re just not handled at all. Admittedly, these communication devices are very difficult to handle by any translation method. They lay outside the purview of the current science of translation. In fact, all these exceptions seem to reside outside of the linguistic theory which supports word-for-word translation.


Why the exceptions? Why are they even viewed as exceptions? Why doesn’t the theory supporting the word-for-word translation methodology explain these practical, common artifacts of the text? And do so, not as exceptions, but as part and parcel of how the language works?

William James said:

Round about the accredited and orderly facts of every science there ever floats a sort of dust-cloud of exceptional observations, of occurrences minute and irregular and seldom met with, which it always proves more easy to ignore than to attend to… Anyone will renovate his science who will steadily look after the irregular phenomena, and when science is renewed, its new formulas often have more of the voice of the exceptions in them than of what were supposed to be the rules.

I suggest the reason for the exceptions is because the linguistic theory underlying the word-for-word translation method is inherently broken. A renewed translation method will incorporate the exceptional into the norm as, in fact, normal. As I see it, the theory supporting word-for-word translation does not have the explanatory power to incorporate the exceptions into the norm.  It has to handle them as exceptions.  And worse, while William James refers to exceptions “minute and seldom seen,” the exceptions met with in word-for-word Bible translation are significant and frequent.

The problem is: the normal, communicating human being doesn’t view as exceptional the kinds of exceptions required by a word-for-word method. Language simply doesn’t work that way. They think of these artifacts as just normal and natural.  That is, a language provides to the language user a rather coherent system.  And the language user processes a text written in that language in a rather unexceptional way–the user doesn’t have to start, stop, and stutter their way through a text.  This is not to say that authors can’t use the language in creative ways.  It’s to say that the creative ways don’t require exceptional ways of processing them.

But, word-for-word translation method does view them as exceptional?  So, why doesn’t the method sync-up with the normal way language works?  Why are word-for-word translations so exceptional?