HistoricalChristian.Faith

John 1:1c — is the Word “God” or “a god”? (the anarthrous theos)

Type: supporting argument (off-timeline) · Assessment: sound — the interpretation that in kai theos ēn ho logos the anarthrous, pre-verbal theos is qualitative — “the Word had the nature of God,” fully deity, yet a person distinct from “the God” (ho theos) he is “with” — holds up as the near-consensus of Greek grammarians. The rival “a god” rendering (the Jehovah’s Witnesses’ New World Translation) rests on the fallacy that an anarthrous noun must be indefinite and is applied inconsistently even within John’s own prologue; the ancient subordinationist reading (the anarthrous theos marks a lesser deity) over-reads a distinction that guards the Word’s personhood, not his rank · weighed against the readings advanced by Arius’s heirs and the modern Arianizing translations.

The claim

ἐν ἀρχῇ ἦν ὁ λόγος, καὶ ὁ λόγος ἦν πρὸς τὸν θεόν, καὶ θεὸς ἦν ὁ λόγος.
“In the beginning was the Word, and the Word was with God, and the Word was God.” (John 1:1)

The crux is the third clause. Two features of the Greek do the work. First, the subject ho logos (“the Word”) carries the article, while the predicate theos (“God”) does not — it is anarthrous. Second, the anarthrous theos stands before the copula ēn (“was”), a fronted position. Three readings compete for that anarthrous theos: the definite (“the Word was God,” identifying the Word with deity), the indefinite (“the Word was a god” — one divine being among others, and by implication a lesser one), and the qualitative (“the Word was as to nature God” — sharing the divine essence). This page steelmans and rates the qualitative full-deity reading, weighing the indefinite and subordinationist rivals against it.¹

Who advances it

The grammatical distinction is not a modern discovery: Origen already built on it in the third century. On this verse he observes that Scripture uses ho theos (God with the article) for the Father, “the God” who is “Very God” (Autotheos), and the bare theos for the Word, who is God “by being with the Father” — and from that he drew a gradation, granting elsewhere that the Son might be called a “second God.” So the ancient subordinationists (and, sharpening it, Arius’s party) could read the anarthrous theos as marking a divinity real but derivative and lesser.² In the modern period the indefinite reading is pressed hardest by the Jehovah’s Witnesses, whose New World Translation renders the clause “and the Word was a god,” defending it in an appendix that appeals to the missing article. The orthodox qualitative reading is the settled position of twentieth-century Greek grammarians — E. C. Colwell, Philip B. Harner, Bruce Metzger, Daniel B. Wallace, and the standard reference grammars.³

The evidence for

The word order is the tell. In 1933 E. C. Colwell observed a regularity in the Greek New Testament: a definite predicate nominative regularly drops the article when it precedes the verb (it keeps the article when it follows). Since John’s theos is a pre-verbal predicate nominative, its lack of the article is exactly what the grammar predicts and gives no ground for reading it “a god.” That much is solid.⁴ But the stronger and more careful case is Harner’s. In a 1973 study he showed that anarthrous predicate nouns preceding the verb are, across the Gospels, primarily qualitative: they state the nature of the subject. On John 1:1 he concluded that the clause means “the Word had the same nature as theos” — the fronted, anarthrous theos stresses that the Word fully shares the divine nature, while its being anarthrous (not ho theos) keeps the Word from being identified with the Father. Wallace’s grammar reaches the same verdict: the construction is most likely qualitative, definiteness is possible, and the indefinite (“a god”) is the least probable option grammatically.⁵

The construction is a precision instrument, not a defect. John had other ways to say it and chose this one. Had he written ho theos ēn ho logos (“the Word was the God”), he would have identified the Word with the Father — the modalism (Sabellianism) the church also rejected. Had he placed an anarthrous theos after the verb, he could have suggested mere class-membership (“a god”). The fronted anarthrous theos threads between both errors: same nature as the Father, distinct person from him. That is precisely the doctrine the whole Nicene settlement would spend three centuries articulating.

Adversarial assessment

The strongest form of the “a god” case. Its best points are real and should be conceded. (1) Theos genuinely lacks the article; an anarthrous noun can be indefinite in Greek. (2) The Word is, in the same verse, “with God” (pros ton theon) — a distinction of persons that a flat “the Word was God” can blur. (3) The third-century Sahidic Coptic version renders the clause with its indefinite article, which the NWT cites as ancient support. A responsible reading has to account for the distinction the verse itself draws, and the qualitative reading does — it is the indefinite conclusion drawn from these premises that fails.

Anarthrous does not entail indefinite. This is the load-bearing error. Greek has no indefinite article, and countless anarthrous nouns are definite or qualitative, not indefinite — including, in John’s own prologue, the anarthrous theos of John 1:6 (“a man sent from God”) and 1:18, which the NWT itself renders “God,” not “a god.” Metzger pressed exactly this: the NWT’s appendix lists thirty-five Johannine passages where a predicate theos takes the article — but in every one the predicate follows the verb, where (by Colwell’s regularity) the article is expected; none is parallel to the pre-verbal 1:1c, so the list confirms the rule rather than the “a god” reading. The rendering, Metzger wrote, is “a frightful mistranslation,” and taken seriously would make its translators polytheists.⁶ The Coptic point cuts the other way, too: the Coptic indefinite article attaches to qualitative nouns as readily as to count-nouns, so “a god” over-translates a construction that in Coptic idiom signals “divine / of the class God,” not “one god among many.”⁷

Where the orthodox case must not over-reach. Symmetry requires flagging the opposite abuse. Colwell’s rule is frequently misused by apologists to “prove” theos is definite — but the rule runs one way only (definite pre-verbal predicates tend to be anarthrous); it does not follow that every anarthrous pre-verbal predicate is definite (that is the converse fallacy). So the definite rendering “the Word was the God” is not secured by Colwell, and would in any case slide toward modalism. The sound ground is Harner’s qualitative category, not a definiteness claim — and the honest verdict credits the qualitative reading, not an over-strong “Colwell proves it” shortcut.

The subordinationist over-reading. Origen’s observation about ho theos vs. theos is grammatically acute, but the gradation he (and later the Arians) built on it does not follow. The article’s absence marks that theos is functioning as a predicate describing the Word’s nature, and its presence on ho theos marks the subject/person (the Father) the Word is “with” — a distinction of reference and person, not of rank of deity. Nothing in the anarthrous form says the Word’s divinity is a lesser or derived kind; that is a theological inference imported into the grammar. Read qualitatively, the clause tells against Arius, not for him: whatever theos means when predicated of the Father, the same nature is predicated of the Word.

Assessment

Sound. On its best available case the qualitative full-deity reading is the mainstream, near-consensus verdict of Greek grammarians, and it fits the verse’s own architecture: the Word shares the divine nature (theos) yet is a distinct person from “the God” he is with (pros ton theon). Keep the two true points the rivals rest on — theos is indeed anarthrous, and the verse really does distinguish the Word from the Father — but the conclusions drawn from them fail. The indefinite “a god” is grammatically the weakest option: it treats anarthrous as indefinite (it is not), ignores the qualitative force of the fronted predicate, and is applied inconsistently against the translators’ own handling of other anarthrous theos in the same chapter. The subordinationist “lesser deity” reading turns a marker of personal distinction into a marker of ontological rank the grammar will not carry. Two cautions keep the finding honest: the orthodox side should not lean on Colwell to force a definite rendering (that is a converse fallacy, and would drift into modalism), and the qualitative reading is a claim about nature, not a proof-text substitute for the wider Johannine and canonical case for the Son’s deity. Rated on the grammar of the clause itself, the interpretation that the Word is fully God, distinct in person, holds.

Sources