The Greek Text behind the English Versions

For a true textual critic, the text of a modern translation of the Bible is of little real significance. It is based exclusively on materials the critic has made available, and almost never represents an advance in the theory of the text. Also, it is rarely as literal as, say, the Vulgate, so even if it were textually significant, it would not testify to the reading of some variants. But, since moderns are often interested in the text underlying their translations, the following table summarizes the text used to create the New Testament of various modern English translations. (If someone wants to send me an equivalent list for German or another language, I will happily add it, but I'm not in position to create such a list myself.)

Except for the pre-King James versions, I freely confess that I have include only those translations I have on my shelf. With the recent explosion in Bible translations, I can't possibly acquire every version in print, and to be honest, I don't have the slightest urge to try. Many of these translations strike me as frankly silly; the only translations I ever really use, these days, are the New Revised Standard Version (my basic everyday Bible), the Revised English Version (when I want something with real stylistic merit and don't care about literal accuracy), the King James Version (solely for research into folklore based on it), and my own literal translation.

For those with very little background in the types of texts, I have also included a "star rating," giving each version from 1 to 10 stars (★). This is, of course, personal opinion, and in fact I do not give any version more than eight stars (because I don't consider any texts to be entirely properly done). As a data point, I consider a version based on the Textus Receptus to be worth three stars; those based on other languages will be worth less. Others will disagree quite violently with my ratings, of course. And, it must be stressed, a version can have a good text and be a bad rendering, or have a bad text and be a good rendering. The ratings are based solely on the quality of the Greek text of the New Testament. (So, for example, the Challoner translation is much better, simply as an English version, than the Rheims translation; it is in good English and renders the Latin carefully. But both Rheims and Challoner are based on the Vulgate, and on poor Vulgate manuscripts at that; in both cases, the text is very bad. There is political influence -- indeed, we should probably say "sucking-up" -- in the Coverdale version; of the two earliest editions, one is dedicated to Queen Anne Boleyn, but after her fall, a new edition was hastily produced with a mention of Queen Jane Seymour. But the text was still what it was.)

Note that the dates refer to the publication date of the New Testament; the Old Testament and Apocrypha may have been published at other times, and some (like Tyndale) published the New Testament in multiple parts over several years. This is the best date I can come up with for a complete version. If there are two dates, they generally refer to the first edition of the work and the final edition.

VersionDateTextual Source and DescriptionRating
Wycliffe1384, 1388
(Purvey revision)
(no reference to the Greek; made from Vulgate MSS)
Tyndale1525, 1534Textus Receptus (Erasmus), with influence from Luther★★★
Coverdale1535Based mostly on Tyndale plus the Vulgate, Luther's German, and some other Latin references★★
"Thomas Matthew"
(John Rogers)
1537 A minimal revision of Tyndale, still based on the Textus Receptus★★★
Taverner1539 Revised for style with reference to the Greek, but based mostly on Tyndale, "Thomas Matthew," and the Vulgate★★★
The Great Bible1539 A cleaned-up version of Tyndale executed by Coverdale★★
Becke1549 Although sometimes listed as a separate translation, this is really Matthew's Bible with a new set of notes by Edmund Becke. The underlying Greek of course is the same as Matthew's (and hence Tyndale's).★★★
Geneva Bible1557, 1560 Textus Receptus (Stephanus, plus probably information from Beza); first English Bible to note textual variants, though it rarely departed the TR (and the textual notes were pretty well buried among the extremely tendentious and controversial notes of other sorts). During the many editions of this translation, the notes were sometimes modified, but the text remained largely constant. (There was a 1576 revision by Tomson which made a few modifications based on Beza's Latin translation of the Greek; obviously this can't have helped the text much.)★★★
Bishop's Bible1568 Compiled by Matthew Parker, supposedly with the assistance of other scholars, but there really seems to have been relatively little original scholarship; the result was pedestrian and never popular. The Greek was, once again, the Textus Receptus, with influence from Great and Geneva Bibles★★★
Rheims/Douai1582, 1610Catholic translation from Vulgate manuscripts. The title page makes clear both its editorial goals and its thoroughly bad text: "The Holie Bible Faithfully Translated into English, Out Of the Authentical Latin. Diligently conferred with the Hebrew, Greeke, and other Editions in diverse languages. With Arguments of the Bookes, and Chapters: Annotations: Tables: and other helpes, for better understanding of the text: for discoverie of Corruptions in some late translations: and for clearing Controversies in Relgion. (In this one case, I can't help but add: If the translators wanted readers to understand the text, far better to translate it into actual ENGLISH, rather than literal word-for-word nonsense in which they often printed the Latin word minus its ending, and forget their stupid tendential notes.)
King James
1611 Textus Receptus (primarily Stephanus and Beza), but with influence from Erasmus via Tyndale (who was the source for much of the wording)★★★
1749Clementine Vulgate
Revised Version
1881 Text created by vote based on arguments from F. H. A. Scrivener (arguing for a text close to the Textus Receptus) and F. J. A. Hort (arguing for a text like Westcott-Hort). The result is a strange compromise with a mix of Byzantine and Alexandrian readings. The Greek text was reconstructed by Palmer and made the base of Souter's critical edition, though Palmer's production isn't really a very good representation of the text the revisors probably intended.★★★★★
American Standard Version1901 A slightly modified variant on the English Revised Version, with a few differences in text but overall much the same strengths and weaknesses★★★★★
Moffatt1913, 1935Von Soden, with a few small changes. Almost all of the few marginal notes are on textual topics★★★★★
Goodspeed1923Westcott & Hort, though Goodspeed seems to have occasionally adopted readings from the margin rather than the text.★★★★★★★
CCD 1941 rev
1941Although reference is regularly made to the Greek in the notes at points where the Greek differs from the Clementine Vulgate, and quite a few readings lacking in the Greek are placed in square brackets (with or without footnotes), the Vulgate is regularly followed, so we must still consider this based on the Vulgate. Note, e.g., that it actually includes 1 John 5:7-8, although in brackets. Still, the attention to the Greek makes this a significant advance over the previous Catholic versions.★★
Revised Standard Version1946, 1971 The translators expected to produce an eclectic text; the starting point was, very loosely, the 17th edition of the Nestle text. The result was not as expected. The notes on the text by Frederick C. Grant in An Introduction to the Revised Standard Version of the New Testament, p. 41, confess, "[I]t is really extraordinary how often, with the fuller apparatus of variant readings at our disposal, and with the eclectic principle now more widely accepted, we have concurred in following Westcott and Hort. Not that we agreed in advance in favor of Hort -- quite the contrary, there was no such unanimity." The scholars who produced the 1971 edition backed away from this slightly, however, notably in putting the longer ending of Mark back in the text. In the table below, the second edition of the RSV is used as the basis for the list of readings.★★★★★★★
Phillips1947-1962 The introduction claims to be based on the "best available Greek text" (which at the time would mean either Westcott-Hort, von Soden, or Nestle), but Phillips seems in fact to have been somewhat eclectic (and, in a few cases, to have paraphrased his way out of the situation. My favorite is Ephesians 1:1, where the text reads "to all faithful Christians at Ephesus (and other places where this letter is read)"!) ★★★★
New English Bible1961, 1970 The translators started from a Nestle edition, but determined their own text -- and they have admitted that the resultant text was not edited according to one particular principle, but was often heavily influenced by the particular scholar who proposed the first draft of each individual section. As a result, it has a tendency to lurch between a very Westcott-and-Hort type text and one heavily influenced by the "Western" witnesses. The reconstructed text (with commentary on variants found in the margin) was published by Tasker, but given the inconsistency of the compilation process and the fact that the NEB is a rather loose translation, this should not be considered a very reliable text. ★★★★★
Jerusalem Bible1966, 1968 A peculiar edition, supposedly taken from the "ancient languages" but with marginal material, and suggestions for translation, based on the French La Bible de Jérusalem. The result is obviously eclectic, with more Vulgate influence than is probably desirable (though much less than in the earlier Catholic translations; on the whole it is closer to the Greek than the Vulgate), and rather erratic. If there is any sort of critical principle in here, I truly cannot find it.★★★★
New American (first edition)1970 The first Catholic Bible to be made directly from the original languages, there are perhaps some teething pains in the first edition. (E.g. the verse numbering is strange -- where a verse is omitted, the previous verse is often split to cover it. So John 5:3b becomes 5:4, and Romans 16:23b becomes 16:24.) The text is mostly modern, but the Vulgate still exercises influence. Some other textual peculiarities may be hidden under the loose rendering.★★★★★
New American (revised edition)1986 The text of this revised edition has been largely adjusted to match the UBS edition -- but with peculiarities. For example, it includes both the longer and shorter endings of Mark, but with a footnote declaring that the church has found the longer ending canonical. A strange way to approach the matter, which I have observed in the catalog of major variants below by printing (GNT) when the text agrees with GNT but the margin has a comment disagreeing with this.★★★★★★★
New American Standard1960- This is a continuously updated version, so it is hard to give it an exact date; the most recent copyright date in the edition I'm using is 1977. (It's an astonishing edition, which doesn't even have the word "Bible" on the cover. The major headline is "Good News America, God Loves You." It literally just showed up on my doorstep -- and left me rather afraid of seeking a more normal copy of the NAS.) The text is strange at best. It claims to be based on Nestle, but most of the major additions in the Byzantine text are included. In brackets, to be sure, or with footnotes, but a casual reader is unlikely to understand what that means.★★★★
New International1970-1978, 1984 Even though published well after the release of the UBS text, the initial edition is eclectic, with no clear editing criterion that I can see. It does not strike me as a particularly original or valuable text. It is perhaps worth noting that, though the translators claim "no sectarian bias," the introduction refers to the Bible text as possessing "infallibility." Although I can't point to a particular reading or rendering which reveals that, the whole text feels that way, somehow -- as if they're trying to be good modern scholars of the text, but their hearts aren't in it.
I have based my textual tests on the 1978 edition, though there was a slightly revised edition released in 1984 (and more recently a "Today's New International Version" and other side branches).
New King James1979, 1982 The introduction makes the astonishing claim that "a growing number of scholars now regard the Received Text as far more reliable than previously thought." This, of course, is not true; scholars such as Hodges, Farstad, and Robinson regard the Byzantine Text as valuable, but the Byzantine Text is not the Received Text. The introduction does not say which version of the Textus Receptus is used, but it is noteworthy that it does include the long form of 1 John 5:7-8 -- the only modern version translated from the Greek to include this clearly extracanonical passage. Although I have to give this translation's text three stars on the grounds that I gave every other version based on the Textus Receptus three stars, in terms of the knowledge of the translators, it deserves lower than any other translation given here, including those not taken from the Greek.★★★
New Jerusalem1985 Claims to be translated from "the Greek text as established in modern times by critical work," but even a casual inspection shows that this is not the UBS3/4 text, nor Westcott and Hort; it doesn't look like any of the Nestle versions either. Although it claims to be translated from the Greek, there seems to be a lot of French and Latin influence, and the resulting text is eclectic and erratic. Overall, it still seems to be quite close to the original Jerusalem Bible.★★★★
Revised English1989 The introduction says that the UBS4 text was a "major point of reference," but examination shows that it is still an eclectic text with many similarities to its immediate predecessor, the New English Bible.★★★★★★
New Revised Standard1989UBS4 with minor modifications★★★★★★★★
New Living Translation1996 The text claims to be translated from UBS4 and NA27. The original Living Bible was not based on anything; it wasn't even a translation, properly speaking -- the compiler had no Greek, and just assembled the readings he liked from many different versions. This version does have reference to the original, but I would still call it a paraphrase, and a very free one; at many places, it hardly matters what the Greek reads.★★★★★★
Today's New International Version2001, 2005 A revision of the New International Version, with some change in translation philosophy (it seems to me, though this is not stated in the introduction very clearly: The new version seems freer than the old -- frankly, it's reached what I would consider the point of paraphrase -- and it also seems to have less King James influence). The text is stated to be "an eclectic one based on the latest editions of [Nestle-Aland and UBS]." Although this could be read as referring either to the TNIV or the UBS text as eclectic, it seems to refer to TNIV -- which is closer to UBS4 than was the NIV, but still differs in some places. The handling of the notes can sometimes be rather peculiar -- e.g. long sections such as Mark 16:9-20 and John 7:53-8:11 are in the text but printed in italics. There is no sign of the shorter ending of Mark. In general, though the text is closer to the UBS text, the feeling is still that of the original NIV, with all that implies. To be blunt, I'll take a volume with a real UBS4 text, such as the NRSV, any day.
Not that I'll have much choice. The original NIV is the preferred translation of modern conservative Christians (which is why it is the best-selling modern translation). This revision drew so much criticism (apparently for its use of gender-neutral language) that the translation is being withdrawn. I imagine it will still be sold for the moment, but it is a dead end. The next revision of NIV will be based on the original NIV, not the TNIV. One has to suspect that, since the translation will be de-modernized, it may also revert to a less modern text.

To give a very slight feeling for the versions cited, I have taken a selection of variant readings, all significant enough that we can without doubt know what text underlies the English translation (or so I thought until I started reading some of the wild paraphrases in some translations!), and marked how each of the post-King James translations reads it. In the list below, "GNT" represents the reading of the United Bible Societies text; "Byz" represents the Byzantine text (as usually found in the King James); I also occasionally cite "West" for Western text, "Alex" for Alexandrian, "WH" for the Westcott-Hort text, and VgCL for the Clementine Vulgate. (N.B. GNT and Alex are assumed to agree unless otherwise specified.)

The list of readings follows the table below.

1 Co
1 Jo
Challoner (CCD 1941 revision) GNTGNTGNTByzGNTGNTByzByzByzByzByzGNTGNTVgCLGNT
Phillips (1962) GNTGNTByzByzWHWestByzGNTGNTGNTByz(both)GNTGNTGNT
Revised Standard (1971) GNTAlexByzByzWHWestGNTGNTGNTGNTByzAlexGNTGNTGNT
New American Standard (1977) ByzGNTByzByzWHGNTByzByzByzByzByzGNTGNTGNTGNT
New King James (1982) ByzGNTByzByzByzGNTByzVgCLByzByzByzGNTByzVgCLGNT

Readings Examined

Matt 6:13: add "for yours is the kingdom and the might until the age of ages" Byz; omit GNT West

Matt 12:47: add "Someone said to him, 'Your mother.... seeking to speak to you" Byz West [GNT]; omit Alex

Mark 6:22 substitute "the daughter of Herodias" Byz; "his daughter Herodias" GNT

Mark 16:9-20 add Byz, West, [[GNT]]; omit Alex. NOTE: For this variant, I consider a version to agree with Byz if it includes 16:9-20 with no clear indication in the text that they are questionable. (A footnote is not enough to make them questionable.) I consider it to agree with GNT if, like the NRSV and Moffatt, it includes both the longer and shorter endings or if, like the first version of RSV, it omits the ending altogether.

Luke 8:43 add "and had spent all her living on healers" Byz [GNT]; omit WH

Luke 24:40 add "And having said this, he showed them hands and feet" Alex Byz UBS; omit West WH

John 5:3b-4 add 5:3+5:4 Byz VgCL; add 5:3b only West Vulgmss; add 5:4 only MSS; omit both verses GNT

Acts 8:37 add verse "If you believe with all your heart.... the son of God" VgCL; omit GNT

Acts 15:34 add verse "But it seemed good to Silas to stay there" Byz (West) VgCL; omit GNT

Rom 16:24 add verse "The grace of our Lord Jesus Christ is with you all" Byz West VgCL; omit GNT

1Co 13:3 substitute "my body to be burned" Byz West; "my body in order to boast" GNT

Eph 1:1 add "in Ephesus" Byz West [GNT]; omit Alex

Col 3:13 subst "just as Christ has forgiven you" Byz; "just as the Lord forgiven you" Alex West GNT Vulg

1Jo 5:7-8 add "the Father... three on earth" VgCL; omit Alex Byz GNT

Rev 15:6 substitute "clothed in pure bright linen" Byz GNT VgCL; "clothed in pure bright stone Alex

Addendum: The English Text of the King James Bible

Those interested in textual criticism may note with some irony that the text of the King James Bible is itself uncertain. There is no original printer's copy. And, it appears, there are two first editions. It has been suggested that the initial printing called for 20,000 copies. This was apparently more than a single print house could produce in one run. As a result, there were two impressions of this first edition -- and they aren't the same.

Should you chance to come across one of the first editions, the impressions can be told apart by looking at Ruth 3:15. In one impression, the verse reads "SHE went into the citie" the other has "HE went into the citie." Not too surprisingly, they are therefore called the "He" and "She" Bibles. (Interestingly, although "He" is the correct translation, it is said that the majority of later KJV editions print "She." Both of my King James editions are guilty of this, e.g.)

The first major attempt to clean up the printing errors in the King James Bible was made in 1638. This Cambridge edition became the standard for many years to come. It did not, however, entirely eliminate errors from subsequent printings. Several editions became famous for some rather strange errors:

Of no particular textual note, but a fascinating example of how printers could get in trouble, is the "Leda Bible." Like many books of the period, it had illustrations -- in this case, illustrations borrowed from an edition of Ovid's Metamorphoses. One of these, which was used in Hebrews, was an illustration of Zeus, ahem, "visiting" Leda while in the form of a swan. (This would, of course, result in the birth of Apollo.) The result, not too surprisingly, was a lot of protests!

Incidentally, there are "named Bibles" other than the King James. Some of the stranger ones:

The Bug Bible.
Coverdale's edition. So called because Psalm 41:5 is translated "Thou shalt not nede to be afrayed for eny bugges by night." The word Coverdale rendered "bugs" should be translated as "terror" or something similar.
The Place-Maker's Bible.
An edition of the Geneva Bible which has "Blessed are the placemakers" instead of "Blessed are the peacemakers" in Matthew.
The Treacle Bible.
The Bishop's Bible. So-called because it has Jeremiah 8:22 refer to "treacle in Gilead," where most modern editions read "balm in Gilead." (Coverdale also translated as "treacle," which may have inspired the Bishop's Bible reading.)