“Book of life” or “Tree of life” in Revelation 22:19?

As seen, the differences between the later edition of the Textus Receptus and the Byzantine readings are not that much greater than the differences between Codex Sinaiticus and the Byzantine. The amount of differences is what is to be expected between two texts of Revelation, a book with the highest variant per manuscript number ratio. It is amazing to see how close the Textus Receptus is to the text of Greek manuscripts if Erasmus relied solely on the Vulgate. The closeness would cause one to think that Erasmus did consult other sources besides the Vulgate, as he did with Revelation 22:20 (inserting “αμην ναι ερχου”).

Furthermore, Textus Receptus editions published after Erasmus’ 1516 edition appear to be based on Greek manuscripts since the revisions bring the Textus Receptus closer to the Byzantine Majority Text. This is a crucial point which demolishes the accusation that the last 6 verses of Revelation in the King James Version were not based on any Greek manuscript. How could the Textus Receptus have gone through the following revisions if the editors did not have access to any Greek manuscript of the passage?

Verse 18: Erasmus may have back-translated the Vulgate’s “si quis” as “εἴ τις”. However, the Textus Receptus is later revised to reflect the Byzantine/Majority Text (2000):

• • “προφητείας βιβλίου τούτου. εἴτις” (Erasmus 1516)

  • “προφητείας τοῦ βιβλίου τούτου, ἐάν τις” (Stephanus 1550)

  • “προφητείας τοῦ βιβλίου τούτου, ἐάν τις” (Byzantine/Majority Text (2000))

Verse 19: Erasmus again may have back-translated the Vulgate’s “si quis” as “εἴ τις”. However, the Textus Receptus is later revised to reflect the Byzantine/Majority Text (2000):

• “καὶ εἴτις” (Erasmus 1516)

• “καὶ ἐάν τις” (Stephanus 1550)

• “καὶ ἐάν τις” (Byzantine/Majority Text (2000))

Verse 19: Erasmus may have back-translated the Vulgate’s “et de civitate sancta” as “καὶ πόλεως ἁγίας”. There is no preposition “out” in the Latin so Erasmus’ text does not have “ἐκ”. The Textus Receptus is later revised to reflect the Byzantine/Majority Text (2000):

• • “καὶ πόλεως ἁγίας,” (Eramus 1516)

  • “καὶ ἐκ τῆς πόλεως τῆς ἁγίας” (Stephanus 1550)

  • “καὶ ἐκ τῆς πόλεως τῆς ἁγίας ” (Byzantine/Majority Text (2000))

The Textus Receptus has been revised to reflect the Byzantine Majority Text where the editors so wished. Therefore the more recent Textus Receptus editions as we have now are not blindly back-translated texts. Where the editors kept Erasmus’ readings, the editors may have had Greek manuscripts with those readings. This possibility is demonstrated below.

Manuscript support

The Textus Receptus editors may have refused to conform the following Erasmian readings to the Byzantine Majority Text because the Greek manuscript support for these readings were known to the editors. The following Erasmian readings have Greek manuscript support:

• Verse 16: καὶ (Alexandrinus, 5th century)

• Verse 17: καὶ (218, 13th century)

• Verse 17: λαμβανέτω (2324 and 1894)

• Verse 18: omission of τω (110, 12th century)

• Verse 18: ἐπιθήσει (Alexandrinus, 5th century)

• Verse 19: βίβλου (1957, 15th century)

• Verse 21: ἡμῶν (2067, 15th century)

• Verse 21: ὑμῶν (2050 has “( )μῶν”, 12th century)

Taking the above into consideration, the following is the Stephanus 1550 Textus Receptus with readings with Greek manuscript support indicated by underlines. We see that only a handful of readings are not supported by any extant Greek manuscript.

Since the Textus Receptus editors appeared to have access to Greek manuscripts, they may have had Greek manuscripts to support all the readings that appear in the last 6 verses of the Textus Receptus. Perhaps these manuscripts with these readings have become lost. However, despite the remaining Erasmian readings not having any extant Greek manuscript support, the differences between these Erasmian readings and the Nestle-Aland readings do not affect the sense in an English translation (except for two words). For these words of inconsequential differences, other variants identified by Herman Hoskier in Concerning The Text Of The Apocalypse, vol. 2 (1929) are listed in order to show that the witnesses against the Textus Receptus readings are not in unison (though some are obvious misspellings):

Verse 16

• TR: ὀρθρινός

• NA: πρωινος

  • Both KJV and ESV have “morning”

Verse 17

• TR: ἐλθέ

• NA: ερχου

  • Both KJV and ESV have “come”

• TR: ἐλθέτω

• NA: ερχεσθω

  • Both KJV and ESV have “come”

• TR: λαμβανέτω

• NA: λαβετω

  • Both KJV and ESV have “take”

  • Hoskier lists the following variants:

    • λαμβανέτω is supported by 2324 and 1894, though their dates are uncertain.

    • λαμβειν: 12 manuscripts.

• TR: τὸ ὕδωρ

• NA: ὕδωρ

  • Both KJV and ESV have “the water”

Verse 18

• TR: συμμαρτυροῦμαι

• NA: μαρτυρω εγω

  • Both KJV and NASB have “I testify”

  • Hoskier lists the following variants:

    • μαρτυρομαι εγω: 17 manuscripts.

• TR: πρὸς ταῦτα

• NA: επ αυτα

  • Both KJV and ESV have “to them”

• TR: ἐπιτιθῇ

• NA: επιθη

  • Both KJV and ASV have “shall add”

  • Hoskier lists the following variants:

    • επιθησει: 6 manuscripts.

    • επιθησι: 1 manuscript.

    • επιθηση: 12 manuscripts.

    • επιθει: 8 manuscripts.

Verse 19

• TR: ἀφαιρῇ

• NA: αφελη

  • Both KJV and ASV have “shall take away”

  • Hoskier lists the following variants:

    • αφελειται: 1 manuscript.

    • αφεληται: 9 manuscripts.

    • αφελοι: 1 manuscript.

    • εφελη: 1 manuscript.

    • εφειλη: 1 manuscript.

    • αφελει: 10 manuscripts.

• TR: ἀφαιρήσει

• NA: αφελει

  • Both KJV and ASV have “shall take away”

  • Hoskier lists the following variants:

    • αφελι: 1 manuscript.

    • αφελη: 7 manuscripts.

    • αφελοι: 80 manuscripts.

    • αφελαι: 15 manuscripts.

The following is the Stephanus 1550 Textus Receptus with readings with Greek manuscript support indicated again by underlines. Blue shading indicates words of inconsequential differences in an English translation:

Stephanus 1550

16 … ὁ ἀστὴρ ὁ λαμπρὸς καὶ ὀρθρινός 17 Καὶ τὸ πνεῦμα καὶ ἡ νύμφη λέγουσιν, Ἐλθε, καὶ ὁ ἀκούων εἰπάτω, Ἐλθε, καὶ ὁ διψῶν ἐλθέτω· καὶ ὁ θέλων λαμβανέτω τὸ ὕδωρ ζωῆς δωρεάν 18 Συμμαρτυροῦμαι γὰρ παντὶ ἀκούοντι τοὺς λόγους τῆς προφητείας τοῦ βιβλίου τούτου, ἐάν τις ἐπιτιθῇ πρὸς ταῦτα, ἐπιθήσει ὁ θεὸς ἐπ’ αὐτὸν τὰς πληγὰς τὰς γεγραμμένας ἐν βιβλίῳ τούτῳ 19 καὶ ἐάν τις ἀφαιρῇ ἀπὸ τῶν λόγων βίβλου τῆς προφητείας ταύτης ἀφαιρήσει ὁ θεὸς τὸ μέρος αὐτοῦ ἀπὸ βίβλου τῆς ζωῆς καὶ ἐκ τῆς πόλεως τῆς ἁγίας καὶ τῶν γεγραμμένων ἐν βιβλίῳ τούτῳ 20 Λέγει ὁ μαρτυρῶν ταῦτα, Ναί ἔρχομαι ταχύ. Ἀμήν, ναί, ἔρχου, κύριε Ἰησοῦ 21 Ἡ χάρις τοῦ κυρίου ἡμῶν Ἰησοῦ Χριστοῦ μετὰ πάντων ὑμῶν. Ἀμήν

The only remaining words not supported by extant Greek manuscripts which affect the sense in an English translation are the following 2 words indicated above by gray shading:

• Verse 18: γὰρ (for)

• Verse 19: καὶ (and)

Even if Erasmus back-translated these verses from the Vulgate for the 1516 edition, subsequent Textus Receptus editions made revisions based on Greek manuscripts and most of the remaining unique Erasmian readings have Greek manuscript support or carry the same sense as Nestle-Aland readings when translated into English. The remaining 2 unsupported words are a conjunction and a particle. These are minor words that are often supplied anyway in an idiomatic translation regardless of manuscript support. The only significant Vulgate-based variants in the last 6 verses of Revelation are “book of life” at verse 19 and “you all” at verse 21. Nonetheless, “you all” is arguably supported by the 12th century minuscule 2050 which has “( )μῶν” (a one letter difference from ὑμῶν). Even considering these two differences, the fact that Erasmus used the Vulgate for these verses for his 1516 edition is used by critics to grossly exaggerate the unreliability of the King James Version. This is even assuming that the Vulgate readings are spurious. It may well be, however, that the Vulgate has preserved these verses better than the extant Greek manuscripts.

“Book of life” is found in the Vulgate as early as in Codex Fuldensis of the 6th century. The value of Vulgate readings cannot be dismissed. The Vulgate preserves several passages that are found in the Sinaiticus but were later lost in the majority of Byzantine copies: e.g. “raise the dead” (Matthew 10:8), “the Jews” (John 3:25), “Church of God” (Acts 20:28), Doxology (Romans 16:25-27). Moreover, even the NIV, ESV and NASB prefer several Latin Vulgate readings over readings that are preserved in the originally inspired language. The Old Testament was written in Hebrew. However, the NIV in Genesis 4:8 adds the line, “Let us go out to the field” from the Vulgate even though the line does not exist in the Hebrew. The NIV, ESV and NASB in 1 Chronicles 4:13 add “and Meonothai” from the Vulgate despite its nonexistence in the Hebrew. The NIV, ESV and NASB in 2 Chronicles 15:8 add “Azariah the son of” from the Vulgate despite its nonexistence in the Hebrew. The NIV, ESV and NASB acknowledge that the Vulgate can at times preserve readings that are lost in the originally inspired language text. Please go to the page, Aren’t some Textus Receptus readings based on little or no Greek manuscript evidence? for more on this issue.

The reading “book of life” appears to have been common in the Latin stream from early times. The text of Ambrose (4th century) in De Paradiso, Book One, c. 12 reads as the Textus Receptus:

“Nam si Joannes hoc judicavit de suis scriptis: Si quis apposuerit, inquit, ad hoc, adjiciet in illum Deus plagas, quae scriptae sunt in libro isto: et qui dempserit de verbis his prophetiae hujus, delebit Deus partem illius de libro vitae” (Migne, Patrologia Latina, vol. 14, p. 320)

Ambrose also appears to allude to the reading of “book of life” in On the Holy Spirit. In expressing concern about a passage removed by Arians, Ambrose says that the act of removing the words from Scripture has caused those Arians themselves to be “blotted out” from the “book of life”. He writes in Book III, chapter 10 of On the Holy Spirit:

“Nor does the Scripture in this place alone bear witness to the qeoths ,that is, the Godhead of the Holy Spirit; but also the Lord Himself said in the Gospel: “The Spirit is God.” Which passage you, Arians, so expressly testify to be said concerning the Spirit, that you remove it from your copies, and would that it were from yours and not also from those of the Church! For at the time when Auxentius had seized the Church of Milan with the arms and forces of impious unbelief, the Church of Sirmium was attacked by Valens and Ursatius, when their priests [i.e. bishops] failed in faith; this falsehood and sacrilege of yours was found in the ecclesiastical books. And it may chance that you did the same in the past.

And you have indeed been able to blot out the letters, but could not remove the faith. That erasure betrayed you more, that erasure condemned you more; and you were not able to obliterate the truth, but that erasure blotted out your names from the book of life.” (English translation at www.newadvent.org)

Only Revelation 22:19 explicitly establishes the causality between removing God’s word and the remover himself being removed from the book of life.

The text of Primasius of Hadrumetum (6th century) also has “book of life”. Commentariorum super Apocalypsim, which drew from the Revelation commentary by Tyconius (4th century), reads:

“Si quis addiderit ad haec, adjiciet Deus super cum plagas scriptas in libro hoc. Et si quis dempserit, adimt Dominus partem ejus de libro vitae, et de civitate sancta, et de his quae scripta in libro hoc.” (Migne, Patrologia Latina, vol. 68, p. 934)

The text of Haymo of Halberstadt (9th century) also has “book of life”.

“Et si quis diminuerit de verbis hujus libri prophetiae vel sensum, vel etiam verba contra suam conscientiam, sicut Arius fecit ex Evangelio, ut praedictum est, auferet Deus partem ejus de libro vitae, id est in memoria prescientie Dei non scribetur.” (Migne, Patrologia Latina, vol. 117, p. 1220)

It has been argued that “tree of life” became “book of life” due to an intra-Latin accidental alteration. The basis of this argument is that “tree” in Latin, “ligno”, and “book” in Latin, “libro” are different by just two letters. The words in Greek, however, are also very similar. The similarities may be less apparent in computer font Greek. Hence the words below are constructed from the Greek script of Codex Sinaiticus:

βιβλου (book)

ξυλου (tree)

The two words in Greek share three of the same consecutive letters. The Beta of βιβλου and the Xi of ξυλου could look very similar with the angled diagonal strokes. If the Iota became assimilated into the second Beta of βιβλου due to poor writing, the amalgamated script could appear to be one letter. The words in Latin share three same letters whereas the words in Greek share three consecutive letters. The chances for confusion appear to be even for Latin and Greek.

There are reasons to believe that Latin witnesses of Revelation could be more accurate than Greek witnesses. The first reason being that the book of Revelation was canonized first by the Latin Church whereas the Greek Church took until 397 AD at the Council of Carthage. While influential Greek fathers in the 4th century such as Chrysostom and Gregory of Nazianzus were still hesitant to canonize Revelation, Christians ministering in the Latin West in the 2nd century – Justin Martyr, Irenaeus, and Tertullian – recognized its canonicity early on. The Muratorian Canon, the oldest known canon that includes Revelation, is a Latin canon. Codex Vaticanus, a Greek codex, does not even have Revelation. Latin commentaries on Revelation by Victorinus and Tyconius existed by the fourth century, but the earliest known Greek commentary on Revelation is by Andreas of Caesarea in the seventh century. Thus the reading “book of life” is the reading of Christians who considered Revelation to be inspired from early times. The reading “tree of life” is the reading of Christians who initially did not consider those words to be inspired. Moreover, the Greek text of Revelation was corrupted early and extensively. Proof of this is given on this KJV Today article on Revelation 16:5 [CLICK HERE]. Revelation 22:19 may well be one of these early corruptions that were never corrected in the Greek speaking Church.