13 March 2006


Don’t trust statistics. They can be manipulated. They can present a complex situation so simply that they deceive. When it gets down to specifics, they are just plain untrustworthy.

However, every once in a while there is one that grabs you, one that is so drastic that it makes a point even when you build in all the caveats. I recently discovered one of these using my trusty BibleWorks 6.

Total number of words in the standard editions of the Hebrew and Greek testaments: 447,996

Total number of words in the NASB, the self-proclaimed “most literal” translation in English: 775,306.

To me, it looks like for every four original words the NASB has added nearly three more. Why? They do it to make the translation intelligible. The funny thing is that Young’s Literal Translation (which is so literal as to yield incorrect and unintelligible English pervasively) has even more words than the NASB (786,937).

Rendering one English word for every original word does not seem to be a real possibility. Don’t get me wrong: inspiration extends to the very words of the original autographs. But those autographs (and the copies which are leftover today) were written in three foreign (and/or dead) languages. These languages cannot be drawn up into precise mathematical equivalence with English. Article usage, case usage, mood usage have different ranges in each of them. The words themselves have different ranges. If strict formal equivalence is so nearly impossible to attain, perhaps we should be nice to people who are not quite so caught up with linguistic formalities, people, for instance, who use the NIV (726,109 words).


Unknown said...

very interesting

Tim Barker said...

I appreciate the declaration ofthe impossibility of word-word translation. As we are dealing with human languages that God has chosen to use, why would we expect something that cannot occur in any other human language translation to happen with the Bible? I mean there is not 1:1 correspondence consistently for any human languages. I agree Pitts well said. Spend our time in the mother language and know it well, yet give an excellent rendering in the receptor language for this is afterall the purpose of a "translation."

Sarcastic aside: I guess the reason NIV takes less words to express the original is because of all the doctrines they leave out!!! :)

Tim Barker said...

Question: What do you think about trying to translate from mother to receptor language in the case of the original? e.g. greek participle must be rendered as an english participle.

Frank Sansone said...


I am sure you probably understand this, but there may be some that do not.

One of the reasons why there are so many more words in a literal English translation than there are in the corresponding Greek or Hebrew is due to the nature of the languages. In English, for instance, we generally have to supply the subject for verbs. Many other languages (including Greek) have verbs whose declension tell you information that our verbs do not.

Most of us are more familiar with this from Spanish, for instance:

vamos = we go

Although the verb is a form of "go", the "we" is not really an extra supplied word because it is actually part of the verb (due to the ending). Thus the one word is literally translated with two words.

Likewise in the Greek the same thing exists:

In Hebrews 12:1 the Greek word "trechōmen" is tranlated as "let us run" in the King James Version so as to give the sense of the 1st person plural (us) and the subjunctive mood (let us). Thus, one word is literally translated as three English words. (Of course Robertson indicates that if you wanted to give the full sense of the verb, you could even make it "let us keep on running" (which we turn it into five words).

Anyway, I just wanted to bring this us for anyone reading who sees your post and comes to the wrong conclusion that the literal translation are making up or adding a bunch of words. As I said, I am sure you probably already understand this.

In Christ,

Pastor Frank Sansone


Jeremy Pittsley said...

Good point, Pastor Frank. Making up words is not something most translators are in the business of doing.

The biggest reason for the difference is that English is a wordy language. Not only do we add helping verbs to render our verbs into different tenses and moods as you indicated, we also have both a definite and indefinite article for nouns. Neither Hebrew nor Greek use the definite article exactly the way English does, and neither language has an indefinite article. Aslo in translating both languages, we also "add" the word "of" to many of the nouns in the genitive case. After awhile these differences add up.

So here's my point: the languages are formally different. Why should we demand that translations be formally equivalent?

Jeremy Pittsley said...

I think most of the translations recognize a temporal participle when they see one and translate accordingly (at least some of the time). That would be one instance in which the Greek participle does not find a formally equivalent English particple.

The grammatically categories we use (e.g. participle) seem to be only roughly equivalent from language to language. That's why good English students still have to study Wallace when they get to second year Greek. If an English participle and a Greek participle had the same range of uses, everyone who paid attention English class would be all set. Because the languages are formally different, however, we have to learn what the Greek participle does that her English counterpart can't do (and vice versa). At least in those cases, it is imperative that we not use an English participle because the English participle can't do that. In those cases, it would be wrong outright to have a formal equivalent.

Unknown said...

Pitts, what you need to do is add up how many letters occur in the GNT compared with letters in the NASB/NIV. Whereas the English includes a helping verb (say, 3 letters) to show tense, the Greek simply adds seven letters (3 for tense and, let's say, 4 for person)!

Jeremy Pittsley said...

Oh! You're right! All this time I've been thinking about word-for-word. We need to look at letter-for-letter.
Unfortunately...BW6 did not bother to include that stat for any of my versions. Apparently, they're out of the loop.

Dave said...

Do you really believe that this is a logical argument or are you engaging in a little satire?

Jeremy Pittsley said...

There is some satire involved here. I am certain that many of those committed to a "word-for-word" translation readily acknowledge the wordiness of English that we have been talking about. Admitting the differences, they would probably want the English and the original to match as closely as possible in form.

The logical argument here, however overstated and unnuanced, was simply to point out how disparate "as closely as possible" (as represented by the NASB) still is.

On the other side, I think others committed to formal equivalency would be genuinely surprised to see how different Greek and Hebrew are from English. The numbers demonstrate that difference strikingly, if simplistically.

Jeremy Pittsley said...

If you are talking about the letter-for-letter thing that is completely satirical.