*Editor’s Note: As more installments of this series are released, they will eventually be linked together. 

A Worthy Inclusion? The Johannine Comma of 1 John 5:7–8

Part 2: Arguments for its Exclusion from the External Evidence

To get a general overview of the manuscript situation of the Comma, I’ll begin with a longer quote by Daniel Wallace.

Up until a few weeks ago [in 2010], scholars knew of only eight Greek NT MSS that had the comma either in the text (four MSS) or in the margins (four MSS). The earliest textual reading … is found in a 14^th century MS. The marginal readings are found in one 10^th century MS, one 11^th century MS, and then three later MSS. But the marginal notes are all written by a later hand. If you’ve been counting, you’ll notice that there are nine GNT MSS that have the comma. I discovered the ninth one at the Bayerische Staatsbibliothek, Munich, in July 2010. This is an 11^th century MS, codex 177.[1] Its early Gregory number indicates that it has been known for a long, long time. Yet the marginal reading of the comma has gone unnoticed. I did not see it in the catalogs of the BSB, and it’s not listed in Nestle-Aland²⁷, which does list the other eight MSS. The marginal gloss was added no earlier than the second half of the sixteenth century, and probably a century or two later.

As bleak as this sounds for its inclusion, it gets worse.

Elijah Hixson wrote a very helpful overview of the 10 manuscripts, which I will greatly rely upon in this post. Included in the link are pictures of all 10 manuscripts containing the Comma. Hixson explained:

There are 10 Greek manuscripts that have the CJ, but only three of them have it in the same form as in Stephanus’ 1550 edition and Scrivener’s edition reprinted by the TBS—these three are 221^marg, 2318 and 2473… sometimes an eleventh Greek manuscript is cited. GA 635 is sometimes cited as having the CJ in the margin, but it does not…. GA 177 is fun. The manuscript dates to the 11th century, but the CJ dates to 1785.

Now I will dispute his claim that there are 3 manuscripts that are identical to the TR, as Stephanus’ 1550 edition and Scrivener’s 1881 edition are not perfectly identical. But we’ll save that for a later installment (see part 6).

Comparison of Readings

With these two quotes above, both Wallace and Hixson capture well the scant external evidence for the Comma’s inclusion. It is an alarming point that while there are only 10 manuscripts that have a form of the Comma, all 10 are very late, many written after the first printed TR edition. And more shockingly, among the 10, not one single reading is like another![2] They are all singular readings of the Comma. If one were to conclude that the Comma should be included, further textual criticism would have to be done to determine which of the 10 manuscripts is the most accurate. In other words, which version of the Comma is original?

See a comparison of all 10 with differences being underlined:

Scrivener (1881)

ὅτι τρεῖς εἰσιν οἱ μαρτυροῦντες ἐν τῷ οὐρανῷ, ὁ πατήρ, ὁ λόγος, καὶ τὸ Ἅγιον Πνεῦμα·
καὶ οὗτοι οἱ τρεῖς ἕν εἰσι. καὶ τρεῖς εἰσὶν οἱ μαρτυροῦντες ἐν τῇ γῇ, τὸ Πνεῦμα,
καὶ τὸ ὕδωρ, καὶ τὸ αἷμα· καὶ οἱ τρεῖς εἰς τὸ ἕν εἰσιν.*

GA 629 (1362): A Greek-Latin diglot

ὅτι τρεῖς εἰσιν οἱ μαρτυροῦντες ἀπὸ τοὺ οὐρανοὺ, _ πατήρ, _ λόγος, καὶ _ Πνεῦμα Ἅγιον·
καὶ οὗτοι οἱ τρεῖς εἰς τὸ ἕν εἰσι. καὶ τρεῖς εἰσὶν οἱ μαρτυροῦντες απο της γης, τὸ Πνεῦμα,
__ τὸ ὕδωρ, καὶ τὸ αἷμα. ________________.

GA 61 (XVI): Codex Montfortianus, perhaps the manuscript that influenced Erasmus

ὅτι τρεῖς εἰσιν οἱ μαρτυροῦντες ἐν τῷ οὐρανῷ, _ πατήρ, _ λόγος, καὶ _ Πνεῦμα Ἅγιον·
καὶ οὗτοι οἱ τρεῖς ἕν εἰσι. καὶ τρεῖς εἰσὶν οἱ μαρτυροῦντες ἐν τῇ γῇ, _ Πνεῦμα,
καὶ _ ὕδωρ, καὶ _ αἷμα. ________________.

GA 918 (XVI): Most likely copied from Erasmus’s third edition (1522) as it is an exact match.

ὅτι τρεῖς εἰσιν οἱ μαρτυροῦντες ἐν τῷ οὐρανῷ, _ πατήρ, _ λόγος, καὶ _ Πνεῦμα Ἅγιον·
καὶ οὗτοι οἱ τρεῖς ἕν εἰσι. καὶ τρεῖς εἰσὶν οἱ μαρτυροῦντες ἐν τῇ γῇ, _ Πνεῦμα,
καὶ _ ὕδωρ, καὶ _ αἷμα· καὶ οἱ τρεῖς εἰς τὸ ἕν εἰσιν.

GA 2473 (1634): Nearly identical to Scrivener, except missing the final nu at the end.

ὅτι τρεῖς εἰσιν οἱ μαρτυροῦντες ἐν τῷ οὐρανῷ, ὁ πατήρ, ὁ λόγος, καὶ τὸ Ἅγιον Πνεῦμα·
καὶ οὗτοι οἱ τρεῖς ἕν εἰσι. καὶ τρεῖς εἰσὶν οἱ μαρτυροῦντες ἐν τῇ γῇ, τὸ Πνεῦμα,
καὶ τὸ ὕδωρ, καὶ τὸ αἷμα· καὶ οἱ τρεῖς εἰς τὸ ἕν εἰσι_.

GA 2318 (1700s): Identical to Scrivener’s TR, but produced well past the KJV and the Elzevirs 1633 TR.

ὅτι τρεῖς εἰσιν οἱ μαρτυροῦντες ἐν τῷ οὐρανῷ, ὁ πατήρ, ὁ λόγος, καὶ τὸ Ἅγιον Πνεῦμα·
καὶ οὗτοι οἱ τρεῖς ἕν εἰσι. καὶ τρεῖς εἰσὶν οἱ μαρτυροῦντες ἐν τῇ γῇ, τὸ Πνεῦμα,
καὶ τὸ ὕδωρ, καὶ τὸ αἷμα· καὶ οἱ τρεῖς εἰς τὸ ἕν εἰσιν.*

GA 429^marg (1600s; marginal text in block quotation): Copied from Erasmus^3rd (1522).

ὅτι τρεῖς εἰσιν οἱ μαρτυροῦντες [ἐν τῷ οὐρανῷ, _ πατήρ, _ λόγος, καὶ _ Πνεῦμα Ἅγιον·
καὶ οὗτοι οἱ τρεῖς ἕν εἰσι. καὶ τρεῖς εἰσὶν οἱ μαρτυροῦντες ἐν τῇ γῇ], τὸ Πνεῦμα,
καὶ τὸ ὕδωρ, καὶ τὸ αἷμα· καὶ οἱ τρεῖς εἰς τὸ ἕν εἰσιν.

GA 177^marg (1785; marginal text in block quotation)

ὅτι τρεῖς εἰσιν οἱ μαρτυροῦντες [ἐν _ οὐρανῷ, _ πατήρ, _ λόγος, καὶ _ Πνεῦμα Ἅγιον·
καὶ ___ οἱ τρεῖς εἰς τὸ ἕν εἰσιν. ὅτι τρεῖς εἰσὶν οἱ μαρτυροῦντες,] τὸ Πνεῦμα,
καὶ τὸ ὕδωρ, καὶ τὸ αἷμα· καὶ οἱ τρεῖς εἰς τὸ ἕν εἰσιν.

GA 221^marg  (after 1854; marginal text in block quotation)

The youngest manuscript with the latest marginal note (after 1850!). How this marginal reading was understood is not easy to piece together, as a new page (marked “|” ) at the middle of v. 8. Likely, the original hand of 221 wrote the standard text, which excludes the comma, and the marginal reading was added much later as an alternate reading. It was introduced with the expression “οὗτως ἐν ἄλλῳ” meaning “thus in another.” From where, we are not told. The final phrase of the marginal reading is “καὶ τὰ λοιπά,” which means “and the rest” or etcetera. The generous way to read 221 and the much later marginal note would be as follows. It originally had the standard readin,g excluding the Comma. However, the Comma was inserted at the bottom right margin of the text where the page broke between “and the water, and the blood” and “and these three.” The “and the rest” at the end of the marginal reading is to signal where to pick back up in the text, just after “and the Spirit” (the second reference to “spirit”). If this is the case, then it is almost identical to Scrivener and perfectly in line with Stephanus^3rd (1550) and manuscript 2473. However, such a recreation requires omitting a large part of the original hand at the start of verse 7 (“ὅτι τρεῖς εἰσιν οἱ μαρτυροῦντες τὸ Πνεῦμα,”). Further, the “καὶ τὰ λοιπά” at the end of the note gives it a unique reading in the most technical sense. Finally, there is no mark signaling where the marginal insertion should be located.

ὅτι τρεῖς εἰσιν οἱ μαρτυροῦντες τὸ Πνεῦμα, καὶ τὸ ὕδωρ, καὶ τὸ αἷμα· |
[ὅτι τρεῖς εἰσιν οἱ μαρτυροῦντες ἐν τῷ οὐρανῷ, ὁ πατήρ, ὁ λόγος, καὶ τὸ Ἅγιον Πνεῦμα·
καὶ οὗτοι οἱ τρεῖς ἕν εἰσι_. καὶ τρεῖς εἰσὶν οἱ μαρτυροῦντες ἐν τῇ γῇ, τὸ Πνευμα,
καὶ τὰ λοιπά] καὶ οἱ τρεῖς εἰς τὸ ἕν εἰσιν.

GA 88^marg  (marginal text in block quotation):

Likely, the note is post-Erasmus. The microfilm is very difficult to read in the main text; however, the margin is clear enough. There is a repetition in the margin of the final phrase “τὸ Πνεῦμα, καὶ τὸ ὕδωρ”.

ὅτι τρεῖς εἰσιν οἱ μαρτυροῦντες [ἐν τῷ οὐρανῷ, ὁ πατήρ, ὁ λόγος, καὶ τὸ Ἅγιον Πνεῦμα·
καὶ οὗτοι οἱ τρεῖς ἕν εἰσι. καὶ τρεῖς εἰσὶν οἱ μαρτυροῦντες ἐν τῇ γῇ, τὸ Πνεῦμα,
καὶ τὸ ὕδωρ] τὸ Πνεῦμα, καὶ τὸ ὕδωρ, καὶ τὸ αἷμα· καὶ οἱ τρεῖς εἰς τὸ ἕν εἰσιν.

GA 636^marg  (marginal text in block quotation):

A 15th-century manuscript with a post-Erasmus era note.

ὅτι τρεῖς εἰσιν οἱ μαρτυροῦντες [ἐν τῷ οὐρανῷ, _ πατήρ, _ λόγος, καὶ τὸ Πνεῦμα Ἅγιον·
καὶ ___ οἱ τρεῖς ἕν εἰσι. καὶ τρεῖς εἰσὶν οἱ μαρτυροῦντες ἐν τῇ γῇ], τὸ Πνεῦμα,
καὶ τὸ ὕδωρ, καὶ τὸ αἷμα· καὶ οἱ τρεῖς εἰς τὸ ἕν εἰσιν.

A Latin Source

A closer look at each of the 10 manuscripts reveals the potential source of the Comma. There is a pattern of these readings appearing similar to the Latin versions that include it. And there is also a papist incentive to match it as well. Westcott said,

The words [of the CJ] are not found in any ancient version except the Latin; and it was not found (a) in the Old Latin in its early form (Tert Cypr Aug), or (b) in the Vulgate as issued by Jerome or (c) as revised by Alcuin. On the other hand, the gloss is found from the 6^th century in Latin Fathers; and it is found also in two copies which give an old Latin text, in some early copies of the Vulgate and in the great mass of the later copies and in the Clementine text.[3]

This led some, including Erasmus, to believe that Jerome (according to the pseudo-Jerome prologue to the canonical epistles) was the culprit behind its insertion into the Latin tradition.[4]

For those who accept the inclusion of the Comma, many of them recognize its Latin source. Even Edward Hills, a supporter of the KJV and the inclusion of the Comma, claimed that it dropped out of the Greek manuscript tradition, only to be preserved in the Latin versions until such time as Erasmus could restore it.[5]

What are these Latin influences? GA 629 is the earliest extant manuscript to include the Comma in the text rather than in the margin. It is a 14^th century Greek-Latin diglot housed at the Vatican. This also reads close to the Latin rather than as the TR renders it:

Scrivener:

ὅτι τρεῖς εἰσιν οἱ μαρτυροῦντες ἐν τῷ οὐρανῷ, ὁ πατήρ, ὁ λόγος, καὶ τὸ Ἅγιον Πνεῦμα·
καὶ οὗτοι οἱ τρεῖς ἕν εἰσι. καὶ τρεῖς εἰσὶν οἱ μαρτυροῦντες ἐν τῇ γῇ, τὸ Πνεῦμα,
καὶ τὸ ὕδωρ, καὶ τὸ αἷμα· καὶ οἱ τρεῖς εἰς τὸ ἕν εἰσιν.

GA 629:

ὅτι τρεῖς εἰσιν οἱ μαρτυροῦντες ἀπὸ τοὺ οὐρανοὺ, _ πατήρ, _ λόγος, καὶ _ Πνεῦμα Ἅγιον·
καὶ οὗτοι οἱ τρεῖς εἰς τὸ ἕν εἰσι. καὶ τρεῖς εἰσὶν οἱ μαρτυροῦντες απο της γης, τὸ Πνεῦμα,
__ τὸ ὕδωρ, καὶ τὸ αἷμα. ________________.

Clementine Vulgate:

Quoniam tres sunt, qui testimonium dant in cælo : Pater, Verbum, et Spiritus Sanctus :
et hi tres unum sunt. Et tres sunt, qui testimonium dant in terra : spiritus,
et aqua, et sanguis : et hi tres unum sunt.

What is most significant is how well the Greek text matches the Clementine Vulgate, including the order of “Holy Spirit” and leaving off the articles for the “heavenly witnesses,” as Latin has no article and therefore did not render any in that portion. In 629, the Greek article was, however, included for the three earthly witnesses as that is how the original Greek renders it. This adherence to the Latin at the “heavenly witnesses” but not the “earthly witnesses” is telling as 629 is a diglot (containing the Latin and the Greek).[6] It is also housed in the Vatican. Perhaps there was an incentive to have the Greek text match the official version of the Roman Catholic Church, the Clementine Latin Vulgate, in order to legitimize both the Comma and the Latin tradition over against the original Greek and Hebrew. This is not farfetched when considering our next manuscript: Codex Montfortianus.

It is widely believed that GA 61, called Codex Montfortianus, was used by Erasmus to update his 3^rd edition of the GNT (1522). His initial two editions excluded the Comma. His 3^rd edition includes a version of it and is different from his 4^th and 5^th editions as well as Scrivener’s rendering. Likely that is because manuscript 61 is so unique.[7]

Scrivener:

ὅτι τρεῖς εἰσιν οἱ μαρτυροῦντες ἐν τῷ οὐρανῷ, ὁ πατήρ, ὁ λόγος, καὶ τὸ Ἅγιον Πνεῦμα·
καὶ οὗτοι οἱ τρεῖς ἕν εἰσι. καὶ τρεῖς εἰσὶν οἱ μαρτυροῦντες ἐν τῇ γῇ, τὸ Πνεῦμα,
καὶ τὸ ὕδωρ, καὶ τὸ αἷμα· καὶ οἱ τρεῖς εἰς τὸ ἕν εἰσιν.

GA 61:

ὅτι τρεῖς εἰσιν οἱ μαρτυροῦντες ἐν τῷ οὐρανῷ, _ πατήρ, _ λόγος, καὶ _ Πνεῦμα Ἅγιον·
καὶ οὗτοι οἱ τρεῖς ἕν εἰσι. καὶ τρεῖς εἰσὶν οἱ μαρτυροῦντες ἐν τῇ γῇ, _ Πνεῦμα,
__ _ ὕδωρ, καὶ _ αἷμα. ________________.

Clementine Vulgate:

Quoniam tres sunt, qui testimonium dant in cælo : Pater, Verbum, et Spiritus Sanctus :
et hi tres unum sunt. Et tres sunt, qui testimonium dant in terra : spiritus,
et aqua, et sanguis : et hi tres unum sunt.

Manuscript 61 was most likely produced by a Franciscan monk, thus by a papist with motivation to match it to the Latin version. Most scathingly, Hixson pointed out:

61 is a copy of 326 in the Catholic Epistles. GA 326 doesn’t have the CJ, and there are a number of places where 61 diverges from the text of 326 to ‘Latinise’ the manuscript an[d] introduce Vulgate-derived readings. Of course, that’s more what you would expect from a Western, Latin tradition making Greek manuscripts. Another piece of evidence that 61 likely reflects a translation from Latin into Greek is that it also lacks definite articles here.

In this case, both sets of three witnesses are lacking the article, making it even closer to the Vulgate and thus more “Latinized.”

GA 221 is the oldest manuscript on our list of ten, dating to the 10^th century. Yet ironically, it has the youngest marginal note of the Comma. In 1854, the catalog reporting manuscripts included 221 and cited it as not containing the Comma! It is assumed that this includes marginal notes as well, though we cannot be certain. Therefore, it is reasonable to conclude that it was added after the catalog report of 1854. The handwriting is so vastly different, this cannot be disputed.

There is good reason to believe that all five of the marginal readings arose after the TR, and so are not nearly as relevant as the number of manuscripts reported. For example, GA 429 gives the Comma reading in the margin and also cites many such marginal notes as from Erasmus. Other such manuscripts come after Erasmus’s 3^rd edition and likely were based on it. This is true for GA 2318, the only manuscript of the 10 to match perfectly Scrivener’s text. Scholars date it around the 1700’s, long after the printed editions of the GNT are widely disseminated. Several of the manuscripts have Roman Catholic origins as well, which again identifies it with the Latin tradition that includes the Comma. The papist motivation for its inclusion is probably to match the ancient Greek text with the Latin when making new Greek manuscripts.

In truth, we cannot ascertain the motivation of the Romanist scribes, but it does seem suggestive that those manuscripts show Latinizing influences. And given that Spanish inquisitors were investigating Erasmus, accusing him to be an Arian for leaving the Comma out in his first two editions, we can imagine there was papal pressure to include it in both printed and hand-written editions.[8] Galiza and Reeve helpfully explained, “Other [inquisitors] were adamant in their position that the doctrinal usage of the comma by the church conferred canonicity (authority) to this passage, despite the lack of manuscript tradition. The major assumption of this later argument was that whatever the church transmitted [in this case, in the Latin version] was the correct text. Any variation [in the Greek manuscript tradition] was seen as a deviation from orthodoxy.”[9] In essence, when the Council of Trent declared the Latin Vulgate to be the authentical edition of Scripture (cf. WCF/2LCF 1.8), this forced the papists to accept the inclusion of the Comma and the need for Greek manuscripts to support it.

Summary of the External Evidence

When it comes to the Johannine Comma, the external evidence is stacked against its inclusion. If we ascertain the best reading based by sheer numbers, 98% of the extant Greek manuscript tradition exclude it. If we ascertain a variant by weighing the manuscripts, the 10 that include it are all very late (many later than printed editions of the GNT) and considered of very minor significance. If we ascertained by Occam’s razor, not one of the 10 readings agree perfectly with another, making these each a singular reading and therefore the least likely to be original. If we ascertain by a transmissional history, the evidence suggest that the Comma arose in the Latin tradition and found its way into a small number of late Greek manuscripts, many of whom were composed by Roman Catholics. But the vast majority of the church (geographically and chronologically) did not operate with the Comma as if it were in their Bibles (see part 3).

If we were to let the external criteria be our deciding factor, then we would have no choice but to reject the Comma as original. And it is for this reason that those who appeal to its authenticity do so on other grounds: primarily upon the internal evidence but also its place throughout church history. In part 3, we’ll take up the historical claim, and then in part 4 the internal evidence.

[1] See also his article on this Ms GA 177: https://www.csntm.org/2010/07/02/the-comma-johanneum-in-an-overlooked-manuscript/

[2] Depending on how one takes 221^marg, it could be argued that it does perfectly match 2473. But this is not so in a technical sense. See the comparisons below.

[3] B. F. Westcott, The Epistles of St. John, Second Edition: The Greek Text with Notes and Essays, Second Ed. (Cambridge & Londong: MacMillan and Co., 1886), 202.

[4] Grantley McDonald, “Raising the ghost of Arius: Erasmus, the Johannine comma and religious difference in early modern Europe,” PhD dissertation, Leiden University, 2011: 126.

[5] Edward F. Hills, The King James Version Defended (Des Moines: The Christian Research Press, 1984), 210.

[6] View here: https://digi.vatlib.it/view/MSS_Ott.gr.298 page 105v. That Latin portion is difficult to transcribe as it has been worn away.

[7] View here: https://digitalcollections.tcd.ie/concern/works/000004722?locale=en page 439r.

[8] Grantley Macdonald, “Erasmus and the Johannine Comma (I John 5.7–8),” The Bible Translator 67.1 (2016): 49–50; David M. Whitford, “Yielding to the Prejudices of His Times: Erasmus and the Comma Johanneum,” Church History and Religious Culture 95, no. 1 (2015): 19–40.

[9] Rodrigo Galiza and John Reeve, “The Johannine Comma (1 John 5:7–8): The Status of Its Textual History and Theological Usage in English, Greek, and Latin,” Andrews University Seminary Studies 56, no. 1 (2018): 67.

Timothy Decker

Dr. Timothy Decker is one of the pastors of Trinity Reformed Baptist Church of Roanoke, VA, having joined them in 2018. He holds a B.A. and M.A. biblical studies from Carolina University (formerly Piedmont International University), a Th.M. in New Testament from Southeastern Baptist Theological Seminary, and a Ph.D. in biblical studies from Capital Seminary and Graduate School. In his dissertation research, he examined the style of biblical Hebrew poetry in the New Testament. He has presented various papers at academic society meetings and authored numerous articles in several different scholarly journals. He is a member of ETS and IBR. When he is not reading or researching, he enjoys spending time with his wife and four children.

 

Courses taught at CBTSeminary: Elementary Greek I, Elementary Greek II