Text Note: John 5:3b-4

Image:  Excerpt from John 5 in Codex Alexandrinus (see notes in margin on John 5:3-4)

I.  The issue:

The modern critical text, and translations based on it, omits the account of the angel stirring the water in John 5:3b-4.  The traditional text, and translations based on it, includes this passage.  Compare translations based on the traditional text (disputed portion in bold):

KJV John 5:3 In these lay a great multitude of impotent folk, of blind, halt, withered, waiting for the moving of the water. 4 For an angel went down at a certain season into the pool, and troubled the water: whosoever then first after the troubling of the water stepped in was made whole of whatsoever disease he had.

NKJV John 5:3 In these lay a great multitude of sick people, blind, lame, paralyzed, waiting for the moving of the water. 4 For an angel went down at a certain time into the pool and stirred up the water; then whoever stepped in first, after the stirring of the water, was made well of whatever disease he had.

II.  External Evidence:

Note:  We are dealing here with vv. 3b-4 together but there are some variations, noted below, between vv. 3b and 4 separately among manuscripts.

The traditional text (including vv. 3b-4) is supported in general by the following:

Greek witnesses:  A (though NA-28 indicates that v. 3b is missing in the original but it appears in a corrected hand), C (apparently includes in a corrected hand), K, L (apparently includes v.4,  but missing v. 3b), Gamma, Delta, Theta, Psi, 078, family 1, family 13, 565, 579, 700, 892, 1241, 1424, and the vast Majority tradition.

Versions:  The Vulgate and part of the Old Latin apparently support the inclusion of v. 3b, while the Old Latin (with minor variations) and the Clementine Vulgate support the inclusion of v. 4.  The traditional text is also supported by the Syriac Peshitta and the Syriac Harklean and, in part, by the Coptic Bohairic.

Church Fathers:  Of note is the fact that v. 4 is cited in the writings of the Church Father Tertullian (c. 220 AD).

The modern critical text (omitting vv. 3b-4) is supported by the following:

Greek witnesses:  p66, p75, Aleph, B, C (original hand), D (though it apparently includes v. 3b), T, W [supplement] (though it apparently includes v. 3b), 33 (though it apparently includes v. 3b).

Versions:  Individual Latin mss. f and l (though they apparently include v. 3b), Latin ms. q, the Stuttgart Vulgate (2007), the Curetonian Syriac, and the Coptic.

To help sort out some of the variations on the inclusion/exclusion of vv. 3b and 4, according to the NA-28, compare:

Include v. 3b but exclude v. 4	Exclude v. 3b but include v. 4
Greek codex D	Greek codex A
Greek codex W [supplement]	Greek codex L
Greek codex 33
Individual Latin ms. f
Individual Latin ms. l

Evaluative notes on external evidence:

First, it is obvious that there has been much textual activity around vv. 3b-4, indicating serious early controversy over their transmission.

Second, closer examination of the passage in the online version of Codex Alexandrinus (p. 45 recto, column 2, lines 13-14) indicates that the NA-28 apparatus notes may be somewhat misleading regarding vv. 3b-4.  Though some corrections to vv. 3-4 are included in the margin, these verses seem to be part of the original text of Codex A.  See this study of John 5:3-4 in Codex Alexandrinus.

Third, one might give weight to the fact that two papyri omit vv. 3b-4.  This should be tempered, however, by the following considerations:  (a) the recognition that the papyri evidence, in general, is limited, and it reflects traditions from only one general geographical area; (b) the weighing of the two individual papyri cited here.  Of p66, in The Story of the New Testament Text (SBL, 2010), Robert Hull notes, “The manuscript contains more than 400 singular readings, nearly half of them the result of carelessness in copying, and most of them corrected by the scribe himself” (p. 116).  Of p75, Hull notes “its text is remarkably similar to that of Codex Vaticanus; in fact, p75 and B are more closely related than any other NT manuscripts” p. 117).

Fourth, the conclusion that must be reached, in the end, is that the exclusion of vv. 3b-4, like so many other points of textual difference between the traditional and modern texts, rests primarily on the evidence of two codices:  Sinaiticus and Vaticanus.

III.  Internal Evidence:

In his Textual Commentary [Corrected Ed., 1975], Metzger treats v. 3b and v. 4 separately.

First, he notes that “a variety of witnesses” add v. 3b, speculating they do so, “perhaps in order to explain the reference in v. 7 to the troubling of the water” (p. 209 here and in other references below).  He then adds that the reading is lacking in “the oldest and best witnesses” (citing p 66, p75, Aleph, the original hand of A, and B) and that it includes “two non-Johannine words” [ekdechesthai and kinesis].  Two questions immediately arise for the reader:  (1) How do we know what the “oldest and best witnesses” are?  And (2) How do we know what is and is not within the limits of the Johannine vocabulary?

Second, with regard to v. 4, Metzger declares definitively that it is a “gloss” whose “secondary character” is clear in four ways:

(1) It is absent from “the earliest and best witnesses” (citing again p66, p75, Aleph, and B, etc.).

(2) He notes that there are “asterisks or obeli” that “mark the words as spurious in more than twenty Greek witnesses.”  Note:  The manuscripts he lists here, however, are all relatively late ones:  S, Delta, Pi, 047, 1079, 2174.  Could these marks indicate not that the text is “spurious” (for why then would the verse have been included?) but an acknowledgement of conflict in textual transmission?

(3) “The presence of non-Johannine words or expressions.”  He gives these examples:  kata karion, embaino [of going into the water], ekdechomai [expecting, awaiting], katechomai [to hold fast, to hold back], kinesis [movement], tarache [disturbance, stirring], and nosema [disease].  He adds that the last three three words appear only here in the NT.  Again, we must question how Metzger (or anyone else) is able to define so authoritatively the limits of Johannine vocabulary.  He also displays here circular reasoning.  For if v. 3b is considered authentic, one these words (kinesis) is definitely Johannine.   Is this conceivable for John?  Yes, it is.  Compare his limited use of the term “The Twelve” to refer to the twelve disciples in John 6:67, 70, 71; 20:24.

(4) He notes that since the passage is missing “in the earliest and best manuscripts” it “is sometimes difficult to make decisions among alternative readings.”  This seems, however, to be more of an expression of the difficulty of determining the eclectic modern critical text than an objection to the traditional text.

Edward F. Hills notes that the disputed passage is cited by Tertullian in a theological reference to baptism (see The King James Version Defended, pp. 145-146).  He quotes Tertullian as saying, “Having been washed in the water by the angel, we are prepared for the Holy Spirit.”  He also notes its citation in Didymus (c. 379 AD) and Chrysostom (c. 390 AD).  He notes:  “These writers, at least, appear firmly convinced that John 5:3b-4 was a genuine portion of the New Testament text.”   He adds that the text was also included in the Diatessaron by Tatian (c. 175 AD), “which also strengthens the evidence for its genuineness by attesting to its antiquity.”

How then did the text come to be omitted?  Hills cites a theory by Hilgenfeld and Steck:

These scholars point out that there was evidently some discussion of the Church during the 2^nd century concerning the existence of this miracle-working pool.  Certain early Christians seem to have been disturbed over the fact that such a pool was no longer to be found at Jerusalem.  Tertullian explained the absence of this pool by supposing that God had put an end to its curative powers in order to punish the Jews for their unbelief.  However, this answer did not satisfy everyone, and so various attempts were made to remove the difficulty through conjectural emendation.  In addition to those documents which omit the whole reading there are others which merely mark it for omission with asterisks and obels.

Hills also point out that the entire passage shows evidence of having been tampered with by “rationalistic scribes” noting as an example the fact that the spelling of the place name for the pool in v. 2 varies widely.   Compare:

Bethesda:  A, C, K, N, etc. (Majority reading)

Bethsaida:  p66 (corrected hand), p75, B, etc.

Bethsaidan:  p66 (original hand)

Belzestha:  D

Bethzatha:  Aleph, (L), 33, and the Old Latin (the reading adopted by the modern critical text)

Though Hills’ suggestion is worth consideration, the truth is that the reasons this passage came into dispute are now lost to us in the mists of the past.  One might speculate that it concerned disputes over the theology of angels (cf. Col 2:18; Rev 19:10; 22:9).  We will likely never know why the passage came into dispute.

In a commentary published in 1947 Edwyn Hoskyns concluded:

The passage is either a gloss added to explain v. 7, or it belonged to the original text of the gospel, and it was struck out in order to avoid giving support to popular pagan practices connected with sacred pools and streams…. (The Fourth Gospel [Faber and Faber, 1947]:  p. 265). 

Conclusion:

John 5:3b-4 clearly has ancient support.  It was known by Tertullian, appeared in ancient codices like Alexandrinus, and was adopted by the majority as the traditional reading.  Its absence is supported by the two major heavyweights of modern text criticism:  Sinaiticus and Vaticanus.  Though it is missing in two ancient papyri, one of those (p66) is notorious for it omissions, and the other (p75) apparently reflects the same stream as that represented by Vaticanus.

The arguments against the text by Metzger seem to rely on circular reasoning. He assumes that Sinaiticus and Vaticanus are “the earliest and best manuscripts” and then reasons that if the passage does not appear in those witnesses it cannot be original.  Likewise, he assumes that any less common vocabulary used in disputed passage must necessarily be “non-Johannine.”

Though there is no clear reason known to us as to why vv. 3b-4 might have been omitted, there is also no clear explanation as to why these words might have been added.  The ancient church clearly accepted 5:3b-4 as authentic, as did the Reformed Fathers.  One wonders if the passage’s exclusion in the modern critical text of the nineteenth century might not have been shaped by an Enlightenment influenced bias against the supernatural.  The comment on John 5:3-4 in The Orthodox Study Bible (based on the NKJV text of the Psalms and the NT) notes that these verses are “often omitted from modern English translations,” but adds, “The role of spiritual powers in the world must never be discounted” (p. 224).

I see no compelling reason to exclude John 5:3b-4 from consideration as part of the legitimate text of Scripture.