Where the world comes to study the Bible

The Text and Grammar of John 1.18

February 25, 2004

In Bart Ehrman’s provocative book, The Orthodox Corruption of Scripture: The Effect of Early Christological Controversies on the Text of the New Testament (Oxford, 1993), the author devotes five pages to a treatment of John 1.18 (78-82). Inter alia, he says this:

The more common expedient for those who opt for [oJ] monogenhV" qeov", but who recognize that its rendering as “the unique God” is virtually impossible in a Johannine context, is to understand the adjective substantivally, and to construe the entire second half of John 1:18 as a series of appositions, so that rather than reading “the unique God who is in the bosom of the Father,” the text should be rendered “the unique one, who is also God, who is in the bosom of the Father.” There is something attractive about the proposal. It explains what the text might have meant to a Johannine reader and thereby allows for the text of the generally superior textual witnesses. Nonetheless, the solution is entirely implausible.

…. It is true that monogenhv" can elsewhere be used as a substantive (= the unique one, as in v. 14); all adjectives can. But the proponents of this view have failed to consider that it is never used in this way when it is immediately followed by a noun that agrees with it in gender, number, and case. Indeed one must here press the syntactical point: when is an adjective ever used substantivally when it immediately precedes a noun of the same inflection? No Greek reader would construe such a construction as a string of substantives, and no Greek writer would create such an inconcinnity. To the best of my knowledge, no one has cited anything analogous outside of this passage.

The result is that taking the term monogenhV" qeov" as two substantives standing in apposition makes for a nearly impossible syntax, whereas construing their relationship as adjective-noun creates an impossible sense.1

Ehrman thus suggests that it is nearly impossible and completely unattested for an adjective followed immediately by a noun that agrees in gender, number, and case, to be a substantival adjective. His argument assumes that monogenhv" cannot normally be substantival, even though it is so used in v 14—as he admits. There are many critiques that could be made of his argument, but chief among them is this: his absolutizing of the grammatical situation is incorrect. His challenge (“no one has cited anything analogous outside of this passage”) is here taken up. There are, indeed, examples in which an adjective that is juxtaposed to a noun of the same grammatical concord is not functioning adjectivally but substantivally.2

The following are texts that show Ehrman’s view to be incorrect. I have only looked at a few NT books (John, the corpus Paulinum, James, 1-2 Peter, Jude). The texts are of two kinds: first, those in which a similar semantic situation as Ehrman sees in John 1.18 occurs, though it is not exactly the same (these will be explained below); and second, those in which exactly what Ehrman calls for occurs but the adjective is not modifying the noun. (Asterisked items are the clearest examples that invalidate Ehrman’s absolute rule.)

Category 1: Similar Structural Parallel

Rom 10:19: prw'to" Mwu>sh'" levgei (“at first, Moses says”): Although the adjective is not substantival, it is adverbial, and thus fits into the ‘impossible’ category that Ehrman says does not exist. The proper form would have been prw'ton (since accusative neuter adjectives can regularly function adverbially).

1 Cor 5:10: toi'" pleonevktai" kaiV a{rpaxin h] eijdwlolavtrai" (“with coveters and swindlers or idolaters”): Again, not exactly a parallel, but a noun followed by a substantival adjective followed by a noun (though all are joined by conjunctions) comes awfully close.

1 Cor 6:9: ou[te moicoiV ou[te malakoi; ou[te ajrsenokoi'tai (“neither adulterers nor effeminate nor homosexuals”): Same construction virtually as 1 Cor 5:10.

1 Cor 9:6: h] movno" ejgwV (“or [do] only I [and Barnabbas]...): here the adjective is functioning adverbially, similar to Rom 10:19.

1 Cor 12:29: mh; pavnte" ajpovstoloi… mhV pavnte" profh'tai… mhV pavnte" didavskaloi… mhV pavnte" dunavmei"… (“All are not apostles, are they? All are not prophets, are they? All are not teachers, are they? All are not [workers of] miracles, are they?”): The pronominal adjective pavnte" is functioning substantivally (though this is common for pronominal adjectives) and the noun following is functioning as a predicate nominative. These are good parallels, but are not as helpful as pure adjectives would be. However, the final word, dunavmei", though a noun that can mean ‘ruler,’ is here used in the sense of ‘miracle [-worker].’ This last clause may be unusual grammar, but it seems to be irrelevant to the situation at hand.

1 Cor 14:24: ti" a[pisto" h] ijdiwvth" (“a certain unbeliever or ungifted person”): this one with a conjunction between the two.

2 Cor 3:3: ouj mevlani ajllaV pneuvmati (“not with black [ink] but with the Spirit”): this one with a conjunction between the two (but cf. 2 Cor 9:7 et alii where a conjunction [postpositive gavr] separates the adjective from the noun).

Eph 5:5: pa'" povrno" h] ajkavqarto" h] pleonevkth": see 1 Cor 5.10.

*Col 1:2: toi'" ejn Kolossai'" aJgivoi" kaiV pistoi'" ajdelfoi'" (“to the saints and faithful brothers”) comes as close as any text to meeting the requirements that Ehrman says are impossible. The parallel is not exact, but an adjective in the construction ARTICLE + ADJECTIVE + kaiv + ADJECTIVE + NOUN. See discussion of this text in my dissertation.

1 Tim 1:13: blavsfhmon kaiV diwvkthn (with ‘and’ between) (but 1 Tim 2:5 illustrates that both gavr and kaiv can intervene between a modifying adjective and its noun [ei|" gaVr qeov", ei|" kai; mesivth" qeou' kai; ajnqrwvpwn])

**Heb 9:24: ouj gaVr eij" ceiropoivhta eijsh'lqen a{gia Cristov", ajntivtupa tw'n ajlhqinw'n (“For Christ did not enter a holy place, a copy of the true...”) Although “Christ” separates the adjective from the noun, this fits well the semantics that Ehrman says are impossible. The reason it does is precisely because a{gia is seen as substantival in the context. Does not John 1:14 do that for John 1:18?

Category 2: Identical Structural Parallel

John 6:70:kaiV ejx uJmw'n ei|" diavbolov" ejstin. Here diavbolo" is functioning as a noun, even though it is an adjective. And ei|", the pronominal adjective, is the subject related to diavbolo", the predicate nominative.

*Rom 1.30: katalavlou" qeostugei'" uJbristaV" uJperhfavnou" ajlazovna", ejfeuretaV" kakw'n, goneu'sin ajpeiqei'" (“slanderers, haters of God, insolent, arrogant, boastful, inventors of evil, disobedient to parents”—true adjectives in italics)

**Gal 3:9: tw/' pistw/' jAbraavm (“with Abraham, the believer” as the NASB has it; NRSV has “Abraham who believed”; NIV has “Abraham, the man of faith”). Regardless of how it is translated, here is an adjective wedged between an article and a noun that is functioning substantivally, in apposition to the noun.

*Eph 2:20: o[nto" ajkrogwniaivou aujtou' Cristou' jIhsou' (“Christ Jesus himself being the chief cornerstone”): although ajkrogwniaivo" is an adjective, it seems to be functioning substantivally here (though it could possibly be a predicate adjective, I suppose, as a predicate genitive). LSJ lists this as an adjective; LN lists it as a noun. It may thus be similar to monogenhv" in its development.

*1 Tim 1:9: dikaivw/ novmo" ouj kei'tai, ajnovmoi" deV kaiV ajnupotavktoi", ajsebevsi kaiV aJmartwloi'", ajnosivoi" kaiV bebhvloi", patrolw/vai" kaiV mhtrolw/vai", ajndrofovnoi" (law is not made for a righteous man, but for those who are lawless and rebellious, for the ungodly and sinners, for the unholy and profane, for those who kill their fathers or mothers, for murderers [adjectives in italics]): this text clearly shows that Ehrman has overstated his case, for bebhvloi" does not modifypatrolw/vai" but instead is substantival, as are the five previous descriptive terms.

2 Tim 3:2: e[sontai ga;r oiJ a[nqrwpoi fivlautoi filavrguroi ajlazovne" uJperhvfanoi blavsfhmoi, goneu'sin ajpeiqei'", ajcavristoi ajnovsioi (“For men will be lovers of self, lovers of money, boastful, arrogant, revilers, disobedient to parents, ungrateful, unholy, ...”) Adjectives are italicized. Similar to 1 Tim 1:9; although it could be said that the adjectives are adjectival, they are not modifying the noun in question, but, like it, refer back to the subject mentioned earlier.

Titus 1:10: EijsiVn gaVr polloiV ?kaiV? ajnupovtaktoi, mataiolovgoi kaiV frenapavtai, mavlista oiJ ejk th'" peritomh'" (similar to 1 Tim 1:9; 2 Tim 3:2).

*1 Pet 1:1:ejklektoi'" parepidhvmoi" (“the elect, sojourners”): This text is variously interpreted, but our point is simply that it could fit either scheme for John 1.18. It thus qualifies for texts of which Ehrman says “no one has cited anything analogous outside of this passage.”

***2 Pet 2:5: ejfeivsato ajllaV o[gdoon Nw'e dikaiosuvnh" khvruka (“did not spare [the world], but [preserved] an eighth, Noah, a preacher of righteousness”). The adjective ‘eighth’ stands in apposition to Noah; otherwise, if it modified Noah, the force would be ‘an eighth Noah’ as though there were seven other Noahs!

Added to my examples are those that a doctoral student at Dallas Seminary, Stratton Ladewig, has culled from the NT: Luke 14.13; 18.11; Acts 2.5. As well, he has found several inexact parallels. See his Th.M. thesis, “An Examination of the Orthodoxy of the Variants in Light of Bart Ehrman’s The Orthodox Corruption of Scripture,” Dallas Seminary, 2000.

In light of these examples, we can thus respond directly the question that Ehrman poses: “when is an adjective ever used substantivally when it immediately precedes a noun of the same inflection?” His remark that “No Greek reader would construe such a construction as a string of substantives, and no Greek writer would create such an inconcinnity” is simply not borne out by the evidence. Keep in mind that we have only looked at a sampling of the NT, yet Ehrman suggests that “no Greek writer would create such an inconcinnity.” Yet if Peter, Paul, Luke, and John create such expressions, this internal argument against the reading monogenhV" qeov" is null and void.

It now becomes a matter of asking whether there are sufficient contextual clues that monogenhv" is in fact functioning substantivally. Ehrman has already provided both of them: (1) in John, it is unthinkable that the Word could become the unique God in 1.18 (in which he alone, and not the Father, is claimed to have divine status) only to have that status removed repeatedly throughout the rest of the Gospel. Thus, assuming that monogenhV" qeov" is authentic, we are in fact driven to the sense that Ehrman regards as grammatically implausible but contextually necessary: “the unique one, himself God…” (2) that monogenhv" is already used in v 14 as a substantive3 becomes the strongest contextual argument for seeing its substantival function repeated four verses later. Immediately after Ehrman admits that this adjective can be used substantivally and is so used in v 14 that way, he makes his grammatical argument which is intended to lay the gauntlet down or to shut the coffin lid (choose your cliché on the force of the connection with v 14. But if the grammatical argument won’t cut it, then the substantival use of monogenhv" in v 14 should stand as an important contextual clue. Indeed, in light of the well-worn usage in biblical Greek, we would almost expect monogenhv" to be used substantivally and with the implication of sonship in 1.18.

In conclusion, the internal arguments against monogenhV" qeov" in John 1.18 simply are not sufficient to overturn the strong external evidence in its favor. We have not here dealt with the external evidence, as that is considered to be the strong suit in the argument for the authenticity of this reading. But if the internal evidence is actually on its side as well, then what is to prevent us from rendering this reading in our English translations—and rendering it something like “the unique Son, himself God…”? The NET Bible, as well as other modern translations that adopt the reading monogenhV" qeov", stand vindicated in the face of this syntactical argument.


1 Ehrman, Orthodox Corruption, 81.

2 Another criticism is that Ehrman has too hastily asserted that monogenhv" cannot have the implied force of “unique son” as in “the unique Son, who is God” (ibid., 80-81):

“The difficulty with this view is that there is nothing about the word monogenhv" itself that suggests it. Outside of the New Testament the term simply means ‘one of a kind’ or ‘unique,’ and does so with reference any range of animate or inanimate objects. Therefore, recourse must be made to its usage within the New Testament. Here proponents of the view argue that in situ the word implies ‘sonship,’ for it always occurs (in the New Testament) either in explicit conjunction with uiJov" or in a context where a uiJov" is named and then described as monogenhv" (Luke 9:38, John 1:14, Heb 11:17). Nonetheless, as suggestive as the argument may appear, it contains the seeds of its own refutation: if the word monogenhv" is understood to mean 'a unique son,' one wonders why it is typically put in attribution to uiJov", an attribution that then creates an unusual kind of redundancy ("the unique-son son"). Given the fact that neither the etymology of the word nor its general usage suggests any such meaning, this solution seems to involve a case of special pleading.

The problem with this assertion is threefold: (1) If in the three texts listed above monogenhv" does, in fact, have both a substantival force and involves the implication of sonship, then to argue that this could be the case in John 1.18 is not an instance of special pleading because there is already clear testimony within the NT to this force. (2) Ehrman's argument rests on going outside of biblical Greek for the normative meaning of a term within the Bible. But since in the NT—as well as patristic Greek (see next footnote) and the LXX (cf. Judg 11.34 where the adjective is used prior to the noun that speaks of Jephthah's daughter; Tobit 3.15 is similar; cf. also Tobit 8.17)—monogenhv" often both bears the connotation of 'son' and is used absolutely (i.e., substantivally) to argue for a secular force within the Bible looks like special pleading! (3) To argue that an implied lexical force becomes “an unusual kind of redundancy” when the implication is brought out explicitly in the text requires much more nuancing before it can be applied as any kind of normative principle: on its face, and in application to the case in hand, it strikes me as almost wildly untrue. In grammar and lexeme, the NT is filled with examples in which the ebb and flow of implicit and explicit meaning intertwine with one another. To take but one example from the grammatical side: eijsevrcomai eij" is a generally hellenistic expression in which the increased redundancy (by the doubling of the preposition) gets the point across. It is found over 80 times in the NT, yet it does not mean “come-into into”! Yet, it means the same thing as e[rcomai eij", a phrase that occurs over 70 times in the NT. English examples readily come to mind as well: In colloquial speech, we often hear “foot pedal” (is there any other kind of pedal besides one for the feet?).

3 A quick look at Lampe’s Patristic Greek Lexicon also reveals that the substantival function of this adjective was commonplace: p. 881, def. 7, the term is used absolutely (i.e., substantivally) in a host of patristic writers.

Related Topics: Textual Criticism