Word for word, thought for thought: what the labels mean

Every Bible publisher describes its translation with a label: word-for-word, thought-for-thought, or something in between. The labels are useful, but they are a publisher’s own description, not a measurement. This guide says where they come from, what they change on the page, and how to read them.

Where the labels come from

The modern terms were set out by the linguist Eugene Nida in Toward a Science of Translating (1964). He called one approach formal equivalence: keep the form of the original, its words and word order, as far as the receiving language allows. He called the other dynamic equivalence: produce in the reader the effect the original produced in its readers, even where that means changing the form. Nida and Jan de Waard later preferred the term functional equivalence for the second (From One Language to Another, 1986). “Word-for-word” and “thought-for-thought” are the everyday names for the same two ideas.

Publishers add labels of their own. Crossway calls the ESV “essentially literal”. Holman calls the CSB’s approach “optimal equivalence”. The NKJV’s preface calls its approach “complete equivalence”. Biblica says the NIV balances word-for-word and thought-for-thought. Tyndale describes the NLT as thought-for-thought. The spec sheet on each translation’s page records the publisher’s own words with a source and a fetched date.

A paraphrase restates the text freely in the translator’s own words. The Living Bible (1971) was a paraphrase of an earlier English version, the American Standard Version. The Message (Eugene Peterson, 2002) was made from the Greek and Hebrew by one author and is usually described as a paraphrase for its freedom with form.

What the choice changes on the page

Idioms. Hebrew and Greek have figures of speech that make no sense carried over literally. Amos 4:6 reads “cleanness of teeth” in the KJV and “empty stomachs” in the NIV. Both are rendering the same Hebrew; one keeps the picture, the other gives its meaning.

Set phrases. 1 Kings 2:10 reads “David slept with his fathers” in the ESV and “David rested with his ancestors” in the NIV. The first keeps the Hebrew expression for dying; the second says what it means and, in passing, settles a gender-language question too.

Word order and sentence length. Greek sentences in Paul’s letters can run for a paragraph. A formal translation keeps them long; a functional one breaks them into English-length sentences. This is much of what the readability meter here picks up.

Consistency. A formal translation tries to render the same original word with the same English word wherever it can, so a reader can trace a theme by vocabulary. A functional translation chooses the best English word for each context, which reads more naturally and makes tracing harder.

Measures, money and time. “A denarius” or “a day’s wage”. “The third hour” or “nine in the morning”. Formal translations keep the ancient unit and footnote it; functional ones convert.

Gender terms. Whether Greek adelphoi is “brothers” or “brothers and sisters” is partly a philosophy question and partly a separate dispute; the guide on gender language covers it.

No translation is purely one thing

Every translation makes functional choices somewhere, because a strictly word-for-word English text would be unreadable, and every translation keeps form somewhere, because the form often carries the meaning. The labels mark a tendency, not a category. Two translations with the same label can differ a great deal; two with different labels can agree in most verses. That is why the literal-to-idiomatic meter on each translation’s page comes from readers who used it and not from the publisher’s label. The meter shows where people felt it sits. Neither end is better, and we do not score accuracy.

How to use the labels

If you want to see what is there, including the awkward parts, go toward the formal end and expect to use footnotes. If you want to read long stretches without stopping, go toward the functional end and expect the translator to have made some decisions for you. If you want both, keep one of each; the comparison pages here put two translations side by side, fact for fact and meter for meter, without picking one. An interlinear or a reverse interlinear, which prints the original under or beside the English, is the tool for checking a single word, and it is not a reading Bible.

Sources

Eugene A. Nida, Toward a Science of Translating (Brill, 1964). Jan de Waard and Eugene A. Nida, From One Language to Another (Thomas Nelson, 1986). The publisher statements cited on each translation’s spec sheet here: Crossway (ESV), Holman Bible Publishers (CSB), Thomas Nelson (NKJV preface), Biblica (NIV translation philosophy), Tyndale House Publishers (NLT), NavPress (The Message). The verse comparisons above are quoted from the KJV, ESV and NIV texts.