Comparing Translations Header

Counting The Words

Written by Ronald J. Gordon Published: April, 1997 ~ Last Updated: February, 2016 ©
This document may be reproduced for non-profit or educational purposes only, with the
provisions that this document remain intact and full acknowledgement be given to the author.

his brief instrument numbers the Greek and English words that are contained in three separate passages, of three separate translations following three major disciplines: literal, equivalent, and paraphrase. Knowledge of word totals equips the student with one more tool for better understanding the product of translation. It principally addresses questions such as: “How many extra words do translators add? How many words are omitted? How many words are changed in some way? How many words are free of influence?” By their very nature, each translation discipline represents both a mechanical and cultural approach to scripture. Three different translations were arbitrarily chosen to represent each discipline: King James Version (literal), New International Version (equivalent), and Living Bible (paraphrase). Many translations fall into each of these three categories, and since only three were considered, the Summary totals may have slightly different results if other versions would have been used. We make no attempt to produce an exhaustive inquiry over a wide selection of translations, for the results may predictably be similar according to discipline. Hopefully, the selection of these three will accommodate the modest expectations of the reader, for popularity of a translation by discipline was our guiding principle. The King James was selected because it enjoys wide acceptance as one of the most literal translations of the Bible, and in many circles is still considered to be the Bible. The NIV was selected because it likewise enjoys broad acceptance as certainly one of the most representative of the dynamic equivalent. The Living Bible was also chosen for its immense popularity among younger readers or students who enjoy a simpler approach to interpretation.

Since two different Greek texts are represented in these versions and we desire to compare apples with apples, each version was counted accordingly with its underlying source text. Theodore Beza's 5th edition was used to evaluate the King James and the United Bibles Societies 3rd edition was used to evaluate both the NIV and the Living Bible. Words were counted from three chapters representing three different styles of communication. Matthew 1 is a very rigid chapter, which allows very little imagination on the part of the translator. The basic textual structure is primarily “so and so was the father of so and so.” Philippians 1 is a very expressive chapter wherein Apostle Paul exhibits an increase in vocabulary, and this happily opens the door for a much wider latitude in word selection. 3 John 1 contains many picture words which do not easily lend themselves exclusively to one English word. In this passage, the translator will experience his greatest temptation toward subjectively.

Four results were anticipated: (1) English words added that do not literally appear in Greek, (2) Greek words that were omitted, (3) translated words having a different English rendering, and (4) words translated free of bias or influence. Here is the methodology and procedure that was employed. Photocopies of both Greek texts and all three translations were made for each of the scriptural passages. Fluorescent highlighter felt-tip pens in three colors were used to mark the photocopied texts. One color represented English words that did not appear in Greek (additions). A second color represented Greek words that did not clearly appear in English (omissions). A third color identified translated words that received a slightly different meaning (modification). Naturally, the remaining uncolored words, by default, would be free of influence but each was still checked just to make sure. Finally, after departing from the project for several days, the exercise was completed again, from scratch, using new photo-copies of both Greek texts and each translation for each scriptural passage, along with the same highlighter colors. Finally, both word counts were then cross-referenced with each other. Out of more than a thousand words, there were eight places where this writer had changed his mind during the second exercise, usually involving a decision between addition or modification. Lastly, these eight discrepancies were resolved.

This instrument is a static representation, for it does not presume the necessity of adding or omitting words. Generally it is often essential for translators to add or omit at least a few words in order to maintain an intelligent or rhythmic sense. For example, in Colossians 1:12 there are five definite articles: THE Father, THE partakers, THE inheritance, THE saints, and THE light. Would it not be redundant to include more than two or three? Conversely, in 1 Corinthians 2:16 there are no definite articles, but the textual sense requires their inclusion. Would it be sensible to translate: A mind to render, “We have A mind of Christ?” Additionally, as is true of most European languages, word endings declare person and number. The English phrase, “he has been running” would be contained in only one Greek word, thus proper conjugation of verb tense almost requires additional words in English. However, some modern translations routinely add words that can hardly be considered anything but guesswork. For example, in Matthew 24:41, Greek texts state “two women grinding at the mill.” It does not say what kind of a mill. The sensational phrase loving NIV could not resist the temptation to assert that it was a HAND mill. Modification for no apparent reason also seems high on the list of many recent translations, and the NIV also modifies “wounded” in Acts 19:16 into BLEEDING, a distinctively different meaning. Wounded means that an injury or bruise of some type has occurred. It does not necessarily suggest a piercing of the skin. Professional boxers have been wounded for many rounds with no obvious sign of bleeding. We offer a static representation because to do otherwise would require an enormous interpretation on our part. An opportunity which we now give to you.

Matthew 1

Style is rigid and diminishes imaginative interpretation

         Greek  English  +/-             Add         Omit      Modifiy         Free
    KJV    444      473   6.5%       16  3.4%      8 1.7%      34  7.2%    423 89.4%
    NIV    436      508  16.5%      113 22.2%     59 11.6%    106 20.9%    289 56.9%
    LB     436      524  20.2%      176 33.6%     68 13.0%     98 18.7%    250 47.7%

Philippians 1

Style is expressive and encouranges imaginative interpretation

         Greek  English  +/-             Add         Omit      Modifiy         Free
    KJV    499      632  26.7%       38  6.0%     24  3.8%     19  3.0%    575 90.9%
    NIV    502      674  34.3%      105 15.6%     83 12.3%    137 20.3%    432 64.1%
    LB     502      866  72.5%      358 41.3%    223 25.8%    261 30.1%    247 28.5%

3 John 1

Style is compressed and forces imaginative interpretation

         Greek  English  +/-             Add         Omit      Modifiy         Free
    KJV    219      296  35.2%        22  7.4%     6  2.0%     26  8.8%    248 83.8%
    NIV    219      330  50.7%        71 21.5%    21  6.4%     53 16.1%    206 62.4%
    LB     219      395  80.4%       170 43.0%    63 15.9%     95 24.1%    130 32.9%


         Greek  English +/-              Add         Omit      Modifiy         Free
    KJV  1,162    1,401 20.6%        76  5.4%     38  2.7%     79  5.6%  1,246 88.9%
    NIV  1,157    1,512 30.7%       289 19.1%    163 10.8%    296 19.6%    927 61.3%
    LB   1,157    1,785 54.3%       704 39.4%    354 19.8%    454 25.4%    627 35.1%

There are perhaps many conclusions that can be drawn from this instrument but we recognize principally two: The more literal the Greek text the less expressive will be the English translation with the greatest number of words that are free of bias or influence, and conversely, the more expressive the underlying Greek text, the more subjective the English translation with the fewest number of words that are free of influence. The selection from Matthew produced high counts of free words because the text was inflexible, whereas the passage from 3 John was the most expressive and the number of free words significantly declined.

Word count according to translation discipline was predictable. The literal King James experienced a 20.6% increase in English words to render its underlying Greek text, and the paraphrased Living Bible incurred a 54.3% increase in the number of words. The former allowed 88.9% of its Greek text to pass through translation while the latter permitted only 35.1%. Different people will no doubt come to different conclusions regarding our exercise but this last point will remain without substantial challenge.

“Study to show thyself approved unto God, a workman that needeth not to be ashamed,
rightly dividing the word of truth.”
2 Timothy 2:15