Jesus as Θεός (God): A Textual Examination

Editor’s Note: This paper was originally given at the Evangelical Theological Society’s southwestern regional meeting, held at Southwestern Baptist Seminary on March 23, 2007. Brian was one of my interns for the 2006-07 school year at Dallas Seminary. He did an outstanding job in presenting the case that the original New Testament certainly affirmed the deity of Christ.

Daniel B. Wallace

Note: For a less academic treatment on this article go to Jesus as “God”: Scriptural Fact or Scribal Fantasy? Also, A final form of this work will be included in a new series of books coming out by Kregel Publications on the Text and Canon of the NT, edited by Daniel B. Wallace (the first volume is due out during the spring of 2010, titled "Revisiting the Corruption of the NT: Manuscript, Patristic, and Apocryphal Evidence)."

From Aland to Zuntz, every major NT scholar has explored the canon of the NT for texts that call Jesus θεός.² While this may seem like a painless pursuit with plenty of “proof-passages,” several stumbling blocks quickly emerge.³

No author of a synoptic gospel explicitly ascribes the title θεός to Jesus.⁴ Jesus never uses the term θεός for himself.⁵ No sermon in the Book of Acts attributes the title θεός to Jesus.⁶ No extant Christian confession(s)⁷ of Jesus as θεός exists earlier than the late 50s.⁸ Prior to the fourth-century Arian controversy, noticeably few Greek MSS attest to such “Jesus-θεός” passages.⁹ And possibly the biggest problem for NT Christology regarding this topic is that textual variants exist in every potential passage where Jesus is explicitly referred to as θεός.¹⁰ This plethora of issues may provoke one to repeat, for different reasons, what a Gnostic document once confessed about Jesus, “Whether a god or an angel or what I should call him, I do not know.”¹¹

On the other hand, “It was the Christians’ habit on a fixed day to assemble before daylight and recite by turns a form of words to Christ as God,” Pliny the Younger wrote in a letter to Emperor Trajan about Christians.¹² “We must think about Christ as we think about God,” the author of 2 Clement opens his homily. “I bid you farewell always in our God Jesus Christ,” concludes Ignatius in his letter to Polycarp.¹³ “True God from true God,” the first ecumenical council ultimately dogmatized concerning Jesus.¹⁴

When, then, did this boldness to call Jesus θεός begin?¹⁵

CONDENSED EXAMINATION

Although this work will examine the textual certainty of every potential NT ascription of θεός to Jesus,¹⁶ 10 of the possible 17 passages will be dismissed up front for the following reasons:¹⁷

(1) Romans 9.5 involves a punctuation issue “which our earliest manuscripts do not answer.”¹⁸ Moreover, even if the absence of any discernable type of standardized punctuation cannot be definitively traced back to the earliest Greek NT MSS, “the presence of punctuation in Greek manuscripts, as well as in versional and patristic sources, cannot be regarded as more than the reflection of current exegetical understanding of the meaning of the passage.”¹⁹

(2) Colossians 2.2. Although this verse contains fifteen variants,²⁰ the issue focuses on syntax rather than the textual pedigree and is therefore outside the scope of this investigation. The same holds true for Matt 1.23,²¹ John 17.3,²² Eph 5.5,²³ 2 Thess 1.12,²⁴ 1 Tim 3.16;²⁵ Titus 2.13;²⁶ 1 John 5.20,²⁷ and Jude 4.²⁸ This leaves seven texts warranting extended examination.

EXTENDED EXAMINATION

New Testament Passages

John 1:1²⁹

According to Aland’s Kurzgefasste Liste, the Gospel of John has more papyrus fragments than any other book of the NT.³⁰ Surprisingly, though, neither the UBSGNT4 nor the NA27 list any variants for John 1.1c. Only three major published NT Greek texts even list textual variants in their apparatus: Tischendorf, Merk, von Soden (with 100% unanimity as to its Ausgangstext: καὶ θεὸς ἦν ὁ λόγος).³¹ No textual debates on John 1.1c exist in any standard work on Jesus-θεός passages, and until fourteen years ago³² NT textual critics were unanimous in their certainty of John 1:1c. This scholarly agreement continues today even though one textual critic, Bart Ehrman, stated his reluctance to dismiss a single eighth-century Alexandrian manuscript, L.³³ To Ehrman, an articular θεός gives him the “distinct impression” that the Orthodox party changed it due to the Arian controversies.³⁴ In other words, Ehrman points out that an articular θεός possibly makes this otherwise implicit identification (Jesus as simply divine) an explicit one (God himself).³⁵

Although the most probable understanding of the anarthrous θεός is qualitative (the Word has the same nature as God),³⁶ three points concern us here textually. First, both P75 and Codex B attest to the absence of the article in John 1.1c. This is significant since “[t]hese MSS seem to represent a ‘relatively pure’ form of preservation of a ‘relatively pure’ line of descent from the original text.”³⁷ Kenneth W. Clark concludes, “it is our judgment that P75 appears to have the best textual character in the third century.”³⁸ Likewise, Ehrman concurs, “[a]mong all the witnesses, P75 is generally understood to be the strongest.”³⁹ Thus, this evidence significantly strengthens our initial external examination in favor of an anarthrous θεός.

Second, only two MSS contain an articular θεός (L W^s): καὶ ὁ θεὸς ἦν ὁ λόγος.⁴⁰ In addition, these two MSS are late (eighth century)⁴¹ and have never produced a reading that has found acceptance into the base text of the NA27 or UBSGNT4 without the support of better and earlier MSS. In fact, regarding Regius (L), “the article with θεός in John 1.1c represents the only sensical variant involving a single letter in all (53) of this scribe’s singular readings. . . . The best explanation for the addition of the article is the sloppy scribal behavior evident in every aspect of this manuscript [i.e., the Gospel of John portion of Regius].”⁴² As for W^s:

First, there is no evidence to establish a direct relationship between these two eighth-century manuscripts. As a result, the article with θεός in John 1:1c found in both would appear to be isolated corruptions that are not dependent upon each other. Second, alignment of Codex L and W^s never merits the “original” text according to NA27 without support from other key MSS (a, B, C, D, P66, P75). Third, there are no known instances where W^s combined with a single other witness can be found as the accepted text of NA27. Therefore, the inclusion of W^s as a sub-singular reading in John 1:1c does not negate the significance of the scribal behavior in Codex L and the combination of the two possesses insufficient testimony to consider the reading καὶ ὁ θεὸς ἦν ὁ λόγος to be a plausible original.⁴³

This scant evidence, at best, struggles to gain any viability in going back to the Ausgangstext. In addition, it is highly improbable that this was a deliberate corruption by the Orthodox Church five centuries after the Arian controversy.

Third, Sahidic Coptic MSS,⁴⁴ usually considered decent representatives of the Alexandrian form of text,⁴⁵ offer an intriguing clue to the textual certainty in John 1.1c. In short, Sahidic has both an indefinite and definite article (whereas Koine Greek only has a definite article). What gives this fact significance is that John 1.1c has the indefinite article in Sahidic (and Bohairic) MSS:⁴⁶ ⲁⲅⲱ ⲛⲉⲩⲛⲟⲩⲧⲉ ⲡⲉ ⲡϣⲁϫⲉ.⁴⁷ It should come as no surprise, then, that the occurrence of the indefinite article (ⲟⲩ, which has contracted) before “God” (ⲛⲟⲩⲧⲉ) in this passage suggests that the Coptic translator was looking at a Greek Vorlage with an anarthrous θεός. In other words, the fact that θεός was translated into Sahidic (and Bohairic) as an indefinite noun strongly suggests that the translator was translating a Greek text without the article.

To flesh this out a little more, Horner translates John 1.1c into English as follows: “. . . and [a] God was the Word.”⁴⁸ The apparatus, however, states, “Square brackets imply words used by the Coptic and not required by the English, while curved brackets supply words which are necessary to the English idiom.”⁴⁹ Unlike English, the Sahidic indefinite article is used with abstract nouns (e.g., truth, love, hate) and nouns of substance (e.g., water, bread, meat).⁵⁰ An example of this can be seen in Horner’s translation of John 19.34b (where there are no Greek articles, καὶ ἐξῆλθεν εὐθὺς αἷμα καὶ ὕδωρ): “. . . and immediately came out [a] blood and [a] water.”⁵¹ None of the words in brackets are necessary in English but are still noted by Horner due the presence of the indefinite article in the Coptic MSS.

Circling back to the textual assessment above, the question we must now answer is, “Did Coptic translators uniformly translate the nominative singular θεός?” To answer this, I examined every occurrence of the nominative singular θεός in every potential Johannine writing (i.e., John; 1 John; 2 John; 3 John; Rev).⁵² My examination revealed that John 1.1c was the only time the nominative singular θεός (articular or anarthrous) was translated with a Coptic indefinite article (see tables 1-3 below). Putting this in further perspective, of the five NT books I examined, there were only four other anarthrous uses of θεός examine (if one includes the textual variant in Rev 21.3).⁵³ The difference still, however, is that whatever one understands the Coptic translator to have done with the other four potential instances (assuming their Vorlage contained them), John 1.1c is the only text we can be certain that the Coptic translator was in fact looking at a Vorlage that contained an anarthrous θεός (i.e., no evidence to the contrary exists to-date). As stated above, only two late eighth century MSS contain an articular θεός, and both the Sahidic and Bohairic versions were composed prior to then. In other words, until (or unless) new evidence is discovered to the contrary, it is highly probable that the Coptic translator(s) were looking at a Greek Vorlage with an anarthrous θεός as reflected with the only Coptic indefinite article with a nominative singular θεός in the five NT books mentioned above.

Table 1. Coptic Translation of θεός in the Gospel of John

Table 2. Coptic Translation of θεός in the Johannine Epistles⁵⁴

Table 3. Coptic Translation of θεός in Revelation

This short summary shows that it is highly improbable that the Coptic translator was translating a Greek Vorlage containing an anarthrous θεός.

Internally (and syntactically), the absence of the article does not necessarily deny the full deity of Jesus. “Neither in LXX Greek nor in secular Greek,” Harris explains, “is a firm or a fine distinction drawn between the articular and the anarthrous θεός. This judgment is confirmed, as far as Hellenistic Greek writings contemporaneous with the NT are concerned, by Meecham, who cites specific examples from the Epistle to Diognetus.”⁵⁵ More specifically, “The term θεός appears in some form 83 times. Of these 63 are articular and 20 anarthrous. Still, it is highly improbable that the Fourth Evangelist intends any consistent distinction to be drawn between θεός and ὁ θεός.”⁵⁶

At any rate, the scholarly consensus is correct, then, that the text is certain and every viable MS ascribes the title θεός to Jesus. For that reason, I will press on to John 1.18.

John 1:18

At least 13 variant readings,⁵⁷ of which three are viable, exist in John 1.18.⁵⁸ All the variants, however, divide into two distinct groups either reading υἱός or θεός.⁵⁹ If the latter is chosen, the final decision ultimately depends on the presence or absence of the article:

μονογενὴς θεός	P66 a* B C* L S* 423 Diatessaron^Arabic syr^{p, h(mg)} geo² Apostolic Constitutions Arius^{acc. to Epiphanius} Basil Clement^grlatCyril^1/4 Didymus Epiphanius Greory-Nyssa Heracleon Hilary Irenaeus^{lat 1/3}Jerome Origen^{gr 2/4} Pseudo-Ignatius Ptolemy Synesius^{acc. to Epiphanius}Theodotus^{acc. to Clement} Valentinians^{acc. to Irenaeus and Clement}
ὁ μονογενὴς θεός	P75 a¹ D 33 cop^{sa, bo} Basil^1/2Clement^2/3Clement^{from Theodotus 1/2}Cyril^2/4 Epiphanius Eusebius^3/7Gregory-Nyssa Origen^{gr 2/4}Serapion^1/2
ὁ μονογενὴς υἱός	A C³ Δ G Q K Π T X Y W^supp 063 0141 0211 ¦¹¦¹³ 1 13 22 24 63 68 69 79 106 114 118 124 131 138 152 154 157 158 160 165 168 173 178 180 185 191 205 209 213 220 222 228 245 265 268 270 280 295 333 345 346 348 352^c357 370 377 382 389 391 397 401 423 430 472 482 489 508 513 515 537 543 544 555 557 565 579 589 597 649 679 683 700 709 713 716 720 726 731 732 733 736 740 744 747 775 787 788 792 799 807 809 821 826 827 828 829 833 841 851 863 865 873 874 878 883 884 888 889 891 892 899 904 931 968 969 979 982 983 989 992 994 1006 1009 1010 1014 1021 1026 1029 1038 1043 1071 1079 1085 1087 1093 1113 1118 1128 1187 1188 1195 1200 1216 1230 1241 1242 1243 1253 1292 1342 1344 1365 1424 1505 1546 1646 2148 Byz [E F G H] Lect it^{a, aur, b, c, e, f, ff(2), l} vg syr^{c, h, pal} arm eth geo^l slav Alexander Ambrose^10/11 Ambrosiaster Athanasius Augustine Basil^1/2 Caesarius Irenaeus^{lat 1/3} Irenaeus^{lat 2/3} Clement^{from Theodotus 1/2} Clement^1/3 Clement^2/5 Cyril^1/4 Chrysostom Hippolytus Origen^{lat 1/2} Letter of Hymenaeus Eustathius Eusebius^4/7 Serapion^1/2 Gregory-Nazianzus Proclus Theodoret John-Damascus Tertullian Hegemonius Victorinus-Rome Hilary^5/7 Ps-Priscillian Faustinus Fulgentius Gregory-Elvira Phoebadius Jerome Varimadum Letter of Hymenaeus Nonnus Synesius Titus of Bostra Victorinus of Rome

Let us first exam the external evidence.⁶⁰

Θεός is attested in the best Alexandrian majuscule (B) and in the earliest available MSS (P66 P75).⁶¹ The significance of this is that if the Alexandrian witnesses for υἱός (e.g., T Δ Ψ 892 1241) cannot reasonably go back to the Alexandrian archetype its attestation therein is almost a moot point.⁶² Ehrman rightly concludes that the semi-recent discovery of P66 and P75 did “very little (in this instance) to change the character of the documentary alignment. . .[and] done nothing to change the picture.”⁶³ Granted, no scholar to my knowledge argues against this fact. Unfortunately, though, these two MSS continue to persuade certain scholars (particularly in evangelical circles) that θεός is now the superior reading. For example, Kostenberger and Swain recently concluded, “With the acquisition of P66 and P75, both of which read monogenēs theos, the preponderance of the evidence now leans in the direction of the latter reading [monogenēs theos].”⁶⁴ This evidence, albeit strong, has not really changed the picture. That is why scholars who opt for υἱός consistently point out the apparent isolation of θεός in the Alexandrian form of text. In fact, Ehrman argues that because “virtually every other representative of every other textual grouping—Western, Caesarean, Byzantine—attests to υἱός,” θεός does not “fare well at all.”⁶⁵ Let’s assess, then, the remaining textual groupings Ehrman mentioned since no one, to my knowledge, is arguing against θεός going back to the Alexandrian archetype.

Three main issues require comment concerning the Western tradition. One, the quality of the Western MS supporting θεός (a)⁶⁶ is comparatively greater than all Alexandrian MSS supporting υἱός. Meaning, unlike the overwhelming improbability of υἱός going back to the Alexandrian archetype, θεός does have a viable possibility of doing so in the Western tradition. Two, a is the earliest Western MS containing this passage (also strengthening its possible connection with the Western archetype). In the least, this demonstrates that θεός is not isolated in the Alexandrian form of text with weak attestation elsewhere. Third, although υἱός has relatively stronger support in the Western form of text (e.g., W^s it vg syr^c Irenaeus), one could still argue that “in the early period [pre-180] there was no textual tradition in the West that was not shared with the East.”⁶⁷ In other words, “the origin of the ‘Western’ text lies anywhere but in the direction its name would suggest.”⁶⁸ Moreover, Ehrman concludes, “[a]bove all, it is significant in saying something about the transmission of the so-called ‘Western’ text of the Fourth Gospel. To be sure, we have not uncovered any evidence of a consolidated form of this text that could match the carefully controlled tradition of Alexandria.”⁶⁹ The possible implication of this, then, is that even with the majority of MSS attesting υἱός in the Western form of text it does not necessarily add a lot of textual weight to its authenticity (especially without a or stronger Alexandrian support).

But what about the Caesarean textual grouping in order to strengthen the argument in support for υἱός? Again, the overwhelming majority read υἱός (Θ, 565, 579, 700, f¹, f¹³, geo¹). This, however, is problematic for at least two reasons. First, more recent nomenclature moves away from this label (Caesarean) since strong argumentation exists against it being a fourth form of text.⁷⁰ Admittedly, some merit might exist in using the label Caesarean with the result that further geographical distribution might be exposed. This leads me to point two. Assuming Caesarean does reveal further geographical distribution, θεός is attested in it, albeit scarcely (geo²). “Like the Armenian version, it [Georgian] is an important witness to the Caesarean type of text. Among the oldest known Gospel manuscripts are the Adysh manuscript of A.D. 897, the Opiza manuscript of 913, and the Tbet’ manuscript of 995. In most apparatus critici, the Adysh manuscript is cited as Geo¹ and the testimony of the other two, as Geo².”⁷¹ The evidence shows that once again θεός is present outside an exclusively Alexandrian tradition with a viable witness to an Archetype (increasing its geographical and genealogical solidarity).

Regarding the Latin and Syriac traditions (aligning with the “Western” type of text), υἱός occurs most frequently, with θεός still present in some Syriac MSS (syr^h(mg) syr^p).⁷² At first glance, this scant evidence seems irrelevant. What impresses us here, though, is that θεός is attested again outside the Alexandrian tradition (e.g., the Peshitta [syr^p] in the Gospels is close to the Byzantine type of text and was “transmitted with remarkable fidelity” and syr^h(mg) is close to the “Western” type of text).⁷³ At the same time, θεός is the exclusive reading in both the Arabic and Coptic traditions.⁷⁴ θεός, then, is also attested in one of the earliest versions of the NT where υἱός is completely absent (the Coptic versions).

Turning now to the Church Fathers, Ehrman emphasizes the early date of υἱός by listing three specific Church Fathers “who were writing before our earliest surviving manuscripts were produced” (Irenaeus, Clement, and Tertullian).⁷⁵ Regrettably, he does this without acknowledging any Church Father supporting θεός around the same period (or P66). I, therefore, will equally list three here: Irenaeus, Clement, and Eusebius. One may quickly notice that the same names appear on both sides of the debate. This redundancy, though, reveals the fact that many Fathers (both Greek and Latin) use υἱός as well as θεός in their writings at John 1.18. My point is that their are many names that could be used to support either reading. In fact, here are three more: Basil, Cyril, and Origen. At the risk of sounding repetitive, θεός shows up again outside the Alexandrian tradition (e.g., early Latin Fathers in the Gospels are Western witnesses)⁷⁶ with relatively strong textual weight (per Ehrman’s argument).

At least two more issues, though, are critical regarding the Church Fathers. First, McReynolds warns us that any reference to ὁ μονογενὴς υἱός by a Church Father is unsubstantiated unless it specifically denotes John 1.18. The citation or allusion alone could equally apply to any of the other passages in John (1.14; 3.16) or in the NT (Luke 7.12; Heb 11.17; 1 John 4.9) where μονογενής refers to the “Son.” On the other hand, the same problem does not apply to μονογενὴς θεός since it occurs nowhere else. Thus, one can be sure that John 1.18 is in view if μονογενὴς θεός, with or without the article, is read (e.g., Arius, Basil, Clement, Cyril, Didymus, Epiphanius, Eusebius, Gregory-Nyssa, Heracleon, Hilary, Irenaeus, Jerome, Origen, Ps-Ignatius, Ptolemy, Serapion, Synesius, Tatian, Theodotus, Valentinius). McReynolds concludes “that patristic evidence for various readings needs to be used much more carefully, and with a full view of the context of the Father being quoted.”⁷⁷

Second, the reading μονογενὴς θεός is not an anti-Arian polemic. Arians did not balk at giving the title θεός to Jesus.⁷⁸ In fact, Arius supports the reading θεός here (according to Epiphanius)⁷⁹ and even called Jesus “God” in a letter he wrote to Eusebius bishop of Nicomedia, “But what do we say and think? What have we taught and what do we teach? That the Son is not unbegotten or a portion of the unbegotten in any manner or from any substratum, but that by the will and counsel of the Father he subsisted before times and ages, full of grace and truth, God, only-begotten, unchangeable.”⁸⁰ If this is true, it throws into doubt that an orthodox scribe would change the text away from Arius if θεός bolsters the complete deity of Christ. Even if the reverse is true (Epiphanius’s testimony is wrong and/or Arius never wrote that letter), one would have to assume that each scribe that changed υἱός to θεός knew about the Arian controversy and knew how to change the text to the higher Christology (which would be many given the MS evidence listed above). Even then, the evidence shows inconsistency in their alleged corruption(s) given John 1.1 and 20.28. On top of all that, it would also have to be shown that all the textual evidence originated during or subsequent to this Arian controversy (which it does not). One might still argue, though, that there only needed to be one extremely early scribe who generated θεός. The real question would then become, “How early?” To answer this objection, the evidence reveals that earlier MSS (in fact, the earliest) attest to θεός (and well before the Arian controversy). This indicates that the objection would remain highly speculative and against the clearer testimony of earlier and better MSS. In other words, the earliest and best MSS heighten the argument away from the allegation that this is an orthodox corruption (as well as the fact that both sides of this Christological controversy use/quote θεός).⁸¹

Two other plausible reasons might explain the mainstream survival of υἱός. One, “Son” may have prevailed as the easier (more predictable) reading before the composition of most extant versions. In support of this, “Son” has universal agreement in later copies with no observable evidence of scribes to alter it. Two, given the external arguments above, even though θεός has wide geographical distribution it remains comparatively weak to υἱός in other non-Alexandrian forms of text. A probable explanation is that θεός is by far the more difficult reading theologically, statistically, and stylistically, which generally produces various textual variants (see internal discussion below).

In sum, externally, both readings enjoy wide geographical distribution, even though υἱός is relatively stronger in non-Alexandrian forms of text. Both readings co-existed in the second century, although weightier MSS support θεός.⁸² As a whole, then, I believe θεός is more probable due to the quality, antiquity, and transmissional history of the witnesses listed above. Nevertheless, this external evidence alone does not make θεός the exclusive heir to the throne. Let us now turn to the internal evidence.

Several internal observations initially seem convincing in support of υἱός. For starters, statistically, μονογενής refers to the “Son” elsewhere in John (1.14; 3.16) and in the NT (Luke 7.12; Heb 11.17; 1 John 4.9).“The only occasion in the NT where μονογενής is not used of an ‘only son’,” Harris observes, “is Luke 8:42, where it qualifies θυγάτηρ.”⁸³ Stylistically, the reading “Son” is more natural with the mention of “God” earlier in the verse as well as the mention of “Father” later in the verse. Otherwise, why would “God” be repeated twice and how could God reside in the bosom of another God (“the Father”)? Theologically, the NT rarely calls Jesus θεός, making the reading almost too difficult. All of these observations seem to point one in the direction of an original reading of ὁ μονογενὴς υἱός.⁸⁴

In response, the offense of using θεός probably drove a scribe to the less offensive Christology of υἱός, which comports well with the scribal tendency to simplify the text (substituting “God” for “Son” is highly improbable, perhaps best explaining the absence of θεός in later Greek MSS). Even more, μονογενὴς θεός is never used elsewhere.⁸⁵ One must ask, then, why here and only here do we have the textual variant μονογενὴς θεός (with or without the article)? My answer, given this scenario alone, is that θεός best explains the rise of the other variants.

Stylistically, θεός closes the inclusio begun in 1.1c; also possibly providing a parallel with 20.28 (the Gospel as a whole). Perhaps the intention was to shock the reader. If this phrase occurred frequently then the author may have failed in achieving his desired result. The reference “who is in the bosom of the Father” is an anthropomorphic metaphor for intimacy and fellowship.⁸⁶ In other words, it is an idiom for closeness and does not truly affect either reading. Lastly, the author of John’s Gospel has a penchant for varying Christological designations (cf., e.g., 1.49; 4.42; 6.69; 9.38; 11.27; 20.16).

Another internal argument sometimes given, a scribe could have easily erred since only one Greek majuscule letter differentiates “Son” from “God”: =u=-s or =q=-s. One problem with this option, however, is that υἱός was not one of the original (or earliest) nomina sacra.⁸⁷ At the same time, though, θεός (q=-s) was one of the four earliest (i.e., Ἰησοῦς, Χριστός, κύριος, and θεός) and most consistently rendered nomina sacra from the second century onward.⁸⁸ To state this differently, although this option is not impossible, it is highly improbable given the transmissional evidence we have.

What variant, then, best explains the rise of the others? I believe that the subtle meaning of the two words in their original apposition, μονογενὴς θεός, caused an early misconception. Thus, an article was assigned to the original reading, now ὁ μονογενὴς θεός, as early as P75, a, and cop^{bo, sa}. Ironically, this change wound up alleviating nothing and was inconsistent with other Johannine and NT usage. Accordingly, the next stage of evolution changed “God” to “Son”: ὁ μονογενὴς υἱός. Finally, although a few other variants arose which either combined the two readings (ὁ μονογενὴς υἱὸς θεός)⁸⁹ or simply omitted both (ὁ μονογενής),⁹⁰ ὁ μονογενὴς υἱός became the majority reading with no viable evidence of change in later Greek MSS.⁹¹

In retrospect, I conclude that μονογενὴς θεός is the best reading given all the evidence we have internally and externally. As a result, it is highly probable that the text of John 1.18 calls Jesus θεός.⁹²

John 20:28

Far beyond the confession of Nathanael in John 1.49 (“Rabbi, you are the Son of God; you are the kingof Israel!”),⁹³ the Gospel of John ends with the fullest Christological confession of faith in the entire Gospel: ἀπεκρίθη Θωμᾶς καὶ εἶπεν αὐτῷ· ὁ κύριός μου καὶ ὁ θεός μου (“Thomas answered and said to him, ‘My Lord and my God’”).⁹⁴ While this paper does not seek to demonstrate or articulate the sense in which Jesus was understood to be θεός,⁹⁵ the aim again is to find out whether a textual analysis will reveal a particular degree of textual certainty that this title was even ascribed to him.

Externally, a single fifth-century Western manuscript D (05) omits the second article in this verse, thus rendering θεός μου instead of ὁ θεός μου.⁹⁶ While this changes nothing contextually,⁹⁷ D is arguably one of the most important Western MSS textually. “When D supports the early tradition the manuscript has a genuine significance, but it (as well as its precursors and followers) should be examined most carefully when it opposes the early tradition.”⁹⁸ In this case, however, the latter is true. Furthermore, it is an eccentric MS and regularly drops the article.⁹⁹ Yet even if D is original and the second article is absent, this verse grammatically falls under the criteria of Granville Sharp’s Rule: ὁ κύριός μου καὶ θεός μου. “In native Greek constructions (i.e., not translation Greek), when a single article modifies two substantives connected by καί (thus, article-substantive-καί-substantive), when both substantives are (1) singular (both grammatically and semantically), (2) personal, (3) and common nouns (not proper names or ordinals), they have the same referent.”¹⁰⁰ In other words, if D is correct and there is no article before θεός, both “Lord” and “God” in this verse explicitly refer to Jesus because of this grammatical construction (cf. also 2 Pet 1.1 below). Thus, Granville Sharp’s Rule makes the phrase even more explicit and leaves “no wiggle room for doubt.”¹⁰¹

John 20.28, no matter which variant or MS one chooses, is categorically secure for referring to Jesus as θεός.¹⁰²

Acts 20:28

Acts 20.28 involves two distinct textual problems of which at least nine variants (seven and two respectively) exist. For convenience’ sake, the viable options are as follows:

τὴν ἐκκλησίαν τοῦ θεοῦ	a B H M S V W Θ 056 0142 4 104 218 257 312 314 322 383 424 454 459 614 621 629 917 1175 1409 1495 1505 1522 1611 1758 1831 2138 2147 2298 2412 2495 l60 l592 l598 l603 l1021 l1439 cop^bo vg it^{ar, c, dem, ph, ro, w}syr^{h, p} geo Ambrose Athanasius Basil Chrysostom Cyril-Alexandria Epiphanius
τὴν ἐκκλησίαν τοῦ κυρίου	P74 A C* D E S T Ψ 13 33 36 40 81 94 104 181 206 209 307 337 429 431 436 453 522 610 630 623 945 1678 1739 1829 1891 2344 2464 l164 l599 arm cop^sa it^{d, e, gig, p} syr^hmg
τοῦ αἵματος τοῦ ἰδίου	Ambrosiaster Didymus^{dub, lat} Irenaeus^lat Jerome Lucifer Pelagius Theodoret^1/2
	P41 P74 a^* A B C* D E Ψ 33 36 69 181 307 326 453 610 945 1175 1611 1678 1739 1837 1891 2464 l60 arm geo syr^hgr Cyril Theodoret
τοῦ ἰδίου αἵματος	H L P 049 056 0142 1 88 104 226 323 330 440 547 614 618 927 1241 1243 1245 1270 1409 1505 1646 1828 1854 2147 2344 2412 2492 2495 Byz Lect slav Athanasius Chrysostom Didymus^{dub vid}

With the external evidence geographically and genealogically proportionate on the first variant (the other five readings lack sufficient external support, are obvious conflations, or both),¹⁰³ the only thing a textual critic can do is appeal to the internal evidence. Yet this too is equally balanced.¹⁰⁴ Of course, it must be noted that “church of the Lord” is absent from the NT and Apostolic Fathers while “church of God” occurs 11 other times in the NT (1 Cor 1.2; 10.32; 11.16, 22; 15.9; 2 Cor 1.1; Gal 1.3; 1 Thess 2.14; 2 Thess 1.4; 1 Tim 3.5, 15) and 12 times in the Apostolic Fathers (1 Clem. 1.1; Ign. Eph. 17.1; Ign. Trall. 2.3; 12.1; Ign. Phld. 1.1; 10:1; Ign. Smyrn. 1.1; Pol. Phil. 1.1; Mart. Pol. 1.1; Herm. Sim. 18.2, 3, 4).

What then shall we say? Most scholars accept θεοῦ as original not merely because of its difficulty but also because of their confidence that the second variant reads τοῦ αἵματος τοῦ ἰδίου (“the blood of his own [Son]” or “his own blood”). Transmissionally, Lars Aejmelaeus proposes an actual literary dependence of Acts 20.28 on 1 Thess 5.9-10 and Eph 1.7. This is in keeping with his overall thesis that Pauline allusions in Acts are invariably due to Luke’s knowledge of the Pauline letters.¹⁰⁵ In addition, τοῦ αἵματος τοῦ ἰδίου is undeniably superior externally (P41 P74 a* A B C* D E Ψ 33 1739 geo syr). Its strength also rests on the logic that it is the harder reading and best explains the rise of the others. “That God suffered was acceptable language,” Harnack notes, “before criticism required some refinement of the conviction that God (or God’s Son) had become man and died on the cross.”¹⁰⁶ In addition, τοῦ αἵματος τοῦ ἰδίου received an “A” rating by the UBSGNT Editorial Committee¹⁰⁷ and all major published NT Greek texts are unanimous (e.g., NA27, Tischendorf, Bover, Merk, von Soden, Westcott and Hort, Vogels, and Weiss).

I suggest that the text originally read θεοῦ.¹⁰⁸ This reading quickly changed to κυρίου because of the difficulty in reconciling it with αἵματος τοῦ ἰδίου (which I also accept as original).¹⁰⁹ According to this conclusion, then, my theory seems verifiable and reinforced by the combination of variants in the witnesses. Here are a few examples of such combinations (in order of my proposed transmissional history):

1. Witnesses that read both originals (θεοῦ and αἵματος τοῦ ἰδίου): a* B 1175 l60

2. Witnesses that changed θεοῦ to κυρίου because of the second original
(αἵματος τοῦ ἰδίου): P74 A C* D E Ψ 33 453 945 1739 1891 36 181 307 610 1678 arm Theodoret

3. Witnesses that kept θεοῦ because of second non-original (ἰδίου αἵματος): H 056 104 614 1409 1505 2412 2495 Athanasius Chrysostom

4. Witnesses that support both non-originals (κυρίου and ἰδίου αἵματος): 2344 Didymus

To summarize, then, the variants that best explain the rise of the others are θεοῦ and αἵματος τοῦ ἰδίου. If Acts 20.28, therefore, equates Jesus with θεός it must do so on other “non-textual” grounds.¹¹⁰

Galatians 2:20¹¹¹

The original text of Gal 2.20, according to Metzger, Ehrman, and others, must have read, ἐν πίστει ζῶ τῇ τοῦ υἱοῦ τοῦ θεοῦ τοῦ ἀγαπήσαντός με (“faith in the Son of God who loved me”). In fact, all major published NT Greek texts contain this reading with the exception of Bover (who reads θεοῦ καὶ Χριστοῦ). The UBSGNT committee continues their support and certainty of it. In fact, the committee agreed to increase their rating from a “B” (found in the 3rd edition) to an “A” (found in the 4th edition). Additionally, the authors of the text-critical notes in the New English Translation, with different arguments (e.g., progressive revelation),¹¹² came to the same textual conclusion.¹¹³ Yet after considering the internal and external evidence (as well as the arguments from many secondary sources), I still think several stones have been left unturned and discourse left unsaid regarding the reading θεοῦ καὶ Χριστοῦ. Therefore, although I ultimately accept the reading υἱοῦ τοῦ θεοῦ, I will give several reasons why I am reluctant to give it an “A” rating or exclude it among the list of passages potentially proclaiming Jesus as θεός.

The four noted variants for this passage, in no particular order, are:

τοῦ θεοῦ	1985
τοῦ θεοῦ τοῦ υἱοῦ	330
τοῦ υἱοῦ τοῦ θεοῦ	A A C D² H K L P S T V Ψ 056 075 0151 0278 6 33 69 81 88 104 131 205 209 226 256 263 323 326 365 424 436 440 459 460 489 517 547 614 618 796 910 927 945 999 1175 1241 1242 1243 1245 1270 1315 1319 1352 1424 1448 1505 1573 1611 1646 1734 1735 1738 1739 1827 1836 1837 1852 1854 1874 1881 1891 1912 1962 1982 2125 2127 2147 2200 2400 2412 2464 2495 2815 it^{ar, f, r} cop^{sa, bo}syr^{h, p} Lect vg arm eth geo slav Ambrosiaster Augustine Chrysostom Clement Cyril Didymus^dub Jerome Marcion^{acc to Adamantius} Pelagius Severian Theodoret Varimadum
θεοῦ καὶ Χριστοῦ	P46 B D* G it^{(b), d, g} Marius Victorinus-Rome Pelagius

Externally, although the majority of witnesses favor τοῦ υἱοῦ τοῦ θεοῦ, the two oldest MSS support θεοῦ καὶ Χριστοῦ (P46 B) along with several other important witnesses (D* G it^{(b), d, g} Marius Victorinus-Rome Pelagius). Furthermore, along with its early “proto-Alexandrian” support (P46 B),¹¹⁴ a strong group of Western witnesses concur (D* F G it^{(b), d, g}Victorinus-Rome). This variant, then, is relatively early and possesses agreement between good Western and Alexandrian witnesses (though it does not attest in the Byzantine, Caesarean, or secondary Alexandrian form of text).

Next, two main internal arguments against this reading exist: (1) Paul nowhere else expressly speaks of God as the object of a Christian’s faith and (2) during the copying process a scribe’s eye probably passed over the first article to the second so that only τοῦ θεοῦ was written (as in MS 330).¹¹⁵ In response to the former, God is the object of a believer’s faith in Rom 4.24. Moo writes, “It is typical for Paul to designate God as the one who raised Jesus from the dead (cf. 8.11; 10.9; 1Cor 6.14; 15.15; 2 Cor 4.14), but it is somewhat unusual for him to designate God himself as the object of Christian faith. Undoubtedly he does so here [Rom 4.24] to bring Christian faith into the closest possible relationship to Abraham’s faith.”¹¹⁶ As to the latter, that theory best explains only one of the four noted variants above, τοῦ θεοῦ, not all of them.

Furthermore, θεοῦ καὶ Χριστοῦ does find some syntactical parallel in the corpus Paulinum: 1 Tim 5.21 and 2 Tim 4.1 (cf. also 1 Tim 6.13).¹¹⁷ Beyond this, “Son of God” is the easier reading and possibly explains why a scribe preferred it. It is also possible that there is a contextual harmonization of v. 19 (“live to God”) and v. 20 (“Christ lives in me”), keeping with the Western tradition and Pauline theology.¹¹⁸ Of course, textually speaking, harmonization seems to be more literal than conceptual.

Externally and internally, several issues still need more clarification and resolution as I have attempted to reveal. Though I sought to confront and consider most of them, I still opted for the traditional reading υἱοῦ τοῦ θεοῦ as the best of all probable scenarios. At the same time, I am reluctant to give this reading an “A” rating (even though the UBSGNT4 committee did) and dismiss it from this Jesus-θεός study.

Hebrews 1:8¹¹⁹

Two main interconnected textual issues exist in Heb 1.8 that possibly resolve the broader grammatical dilemma of how to interpret and translate ὁ θεός in vv. 8 and 9.¹²⁰

The first textual variant involves the presence or absence of τοῦ αἰῶνος after εἰς τὸν αἰῶνα:

ὁ θρόνος σου ὁ θεὸς εἰς τὸν αἰῶνα τοῦ αἰῶνος

ὁ θρόνος σου ὁ θεὸς εἰς τὸν αἰῶνα

Externally, the absence of τοῦ αἰῶνος is significantly inferior with only a small handful of concentrated MSS omitting it (B 33 t vg^ms). Although it is true that scribes often expanded readings, it is not the situation here for several reasons. First, τοῦ αἰῶνος is a direct quotation from both the LXX (Ps 44.7) and MT (Ps 45.7). Second, this reading is supported by some of the best and earliest MSS (only a few omit it: B 33 t vg^ms). Third, almost every time du <lwu occurs in the Psalms according to the MT the LXX translates it with τοῦ αἰῶνος (e.g., Ps 10.16; 45.7; 48.15; 52.10; 104.5). Even when this is not the case (e.g., Ps 20.5), it maintains a resemblance (εἰς αἰῶνα αἰῶνος).

Putting it another way, if one accepts the shorter Greek rendering of the OT quotation in Heb 1.8 (simply by εἰς τὸν αἰῶνα), and does not include τοῦ αἰῶνος, it goes against all the ancient versions of the OT. This variant’s potential implication for our study, though not directly determinative on the Jesus-θεός issue, is to establish all possible links to a Vorlage, best understand the grammatical structure, and assess every possible textual alignment (i.e., character count) in various MSS. In this case, its OT reference (or even Vorlage) was probably the LXX.

The second main textual issue in 1.8 is whether the last word in the verse should read αὐτοῦ or σου (i.e., πρὸς δὲ τὸν υἱόν· ὁ θρόνος σου ὁ θεὸς εἰς τὸν αἰῶνα τοῦ αἰῶνος, καὶ ἡ ῥάβδος τῆς εὐθύτητος ῥάβδος τῆς βασιλείας [σου/αὐτοῦ?]). The outcome, simply put, will help determine whether ὁ θεός is a nominative for vocative (if σου) or subject-predicate nominative (S-PN) construction (if αὐτοῦ):

1. Nominative for vocative = “Your throne, O God, is forever and ever, and a righteous scepter is the scepter of your kingdom”

2. S-PN = “God is your throne [or, Your throne is God] forever and ever, and a righteous scepter is the scepter of his [i.e., God’s] kingdom”¹²¹

Internally, whereas they are both grammatically possible,¹²² only the first resonates with the central theme of the section and book (i.e., the exalted Christ). Ehrman believes, however, that the orthodox party corrupted this text because of their “need to differentiate Christ from God.”¹²³ He concludes by saying, “. . . we are now dealing not with a corruption of the original text but with a corruption of a corruption.”¹²⁴ What I think Ehrman may be missing is that the author of Hebrews stands in the exegetical tradition of the quoted Psalm. “That Jewish exegetes regularly understood the text as an address is clear,” Attridge points out, “both from the Targum and from the revision of the LXX by Aquila.”¹²⁵ Little doubt remains, then, that the LXX translator construed it so, suggesting that ὁ θεός in Heb 1.8 points to Jesus’ essential unity with God while preserving his functional subordination (see ὁ θεός σου in v. 9).¹²⁶ “It is not impossible that the uniform testimony of the ancient versions in support of the vocative may reflect a messianic re-reading which stresses the transcendence of the King – Messiah,” Harris writes, “but it is at least equally possible that all these versions testify to the most natural way of construing <yh!Oa, whether they understood the word in reference to the Messiah, or, as Mulder believes (Psalm 45 48), to God.”¹²⁷

Caragounis summarizes several other salient points regarding the use of the nominative for vocative: (1) it occurred very early in classical Greek, (2) it originally applied to deities, (3) it was more frequent in poetry than prose, (4) it gave greater emphasis, and (5) its usage increased substantially in the NT from classical Greek.¹²⁸ In addition, after probing the centuries (from ancient to modern times) for the use of the nominative ὁ θεός in lieu of the vocative, he concludes, “the articular nominative ὁ θεός when used as vocative has a more exalted, a more distanced, tone belonging to a more formal, solemn and elevated diction. . .[and h]ardly a more solemn or dignified context could be imagined than the one in which this address is placed [Heb 1:8].”¹²⁹ In other words, all of these observations together strengthen the internal probability of understanding ὁ θεός as a nominative for vocative and thus supporting the more natural reading σου.

With all that in mind, there is also a μέν … δέ construction in vv. 7-8. Wallace feels that the nominative for vocative syntax adequately handles this construction; the predicate nominative does not. “Specifically, if we read v 8 as ‘your throne is God’ the δέ loses its adversative force, for such a statement could also be made of the angels, viz., that God reigns over them.”¹³⁰ To sum this up another way, if one holds to the predicate nominative view then there is no clear distinction between the angels (subordinate; ephemeral; servants) and Christ (superior; eternal; deity).

Lastly, various translators handle the preposition πρός differently throughout this pericope (namely; 1.7, 8, 13). Several translations translate it “of” (e.g., ESV, NAS, NET, RSV), some “to” (e.g., KJV, NJB, NLT), and still others “about” (e.g., CSB, NIV); with varying combinations throughout all three instances. However, the translations with “of” or “about” reflect a “misconstrual of the citation as a word about [of] the Son, not to him.”¹³¹ In other words, πρός in vv. 8 and 13 “must be translated ‘to’.”¹³² This pertains to our present internal investigation because it strengthens the μέν … δέ discussion above towards a nominative for vocative translation. I agree with Attridge, then, that “the variant ‘his’ was probably occasioned by the ambiguity of the preposition used to introduce the citations and the failure to construe the whole citation as an address.”¹³³

Externally, albeit both well attested in the Alexandrian tradition, I believe the pronoun σου has more impressive weight and variety than αὐτοῦ.¹³⁴ Here is a snapshot of the witnesses supporting each:

αὐτοῦ	P46 a B H S
σου	A D F K L P Ψ 075 0121 0150 0243 0278 6 33 81 104 256 263 326 365 424 436 459 1175 1241 1319 1739 1852 1881 1912 1962 2127 2200 2464 arm Byz cop^{sa, bo, fay}geo Lect it^{ar, b, comp, d, t, v} slav syr^{h, p, pal(ms)} vg Chrysostom Cyril Gregory-Nyssa Jerome Theodoret

Outside the Alexandrian tradition (primary, P46 a B; secondary, H), αὐτοῦ is almost non-existent to-date (one majuscule, S). And of these six MSS, they are only present in one class of NT witnesses (Greek MSS), two categories of Greek MSS (Papyri and Majuscules), and four centuries (i.e., P46 [III] a [IV] B [IV] H [VI] S [949]). On the other hand, σου is ubiquitous. Every possible geographic area—Alexandrian (e.g., primary, 1739; secondary, 0243), Western (e.g., D), Byzantine (e.g., K), Caesarean (e.g., geo), and other important MSS (e.g., Ψ)—and every century from the third to the 14th century (e.g., cop^sa [III] Gregory-Nyssa [IV] A [V] D [VI] syr^h [616] Ψ [VIII] 33 [IX] 1739 [X] 424 [XI] 365 [XII] 2200 [XIV]) contains at least one witness to σου. What is more, σου is present in every class of NT witnesses (Greek MSS, ancient translations into other languages, and quotations by early ecclesiastical writers) and every category of Greek MSS (Papyri, Majuscules, and Minuscules).

One more external issue requires a response. Ehrman remarks, “It is interesting to observe that the same MSS that evidence corruption in Hebrew 1:8 do so in John 1:18 as well, one of the other passages.”¹³⁵ First, while this brief statement is basically correct, he leaves the reader with a distorted view of scribal activity and transmissional history. Indeed, many examples of the reverse exist. I will briefly list five examples from the MSS he used numerous times regarding our present topic:

I. P66

a. Corrupted text according to Ehrman (i.e., calls Jesus θεός): Gal 2.20.

b. Texts that support Ehrman’s reading (i.e., does not call Jesus θεός): Heb 1.8.

II. A (01)

a. Corrupted texts according to Ehrman (i.e., calls Jesus θεός): John 1.18, 20.28.

b. Texts that support Ehrman’s reading (i.e., does not call Jesus θεός): Acts 20.28; Gal 2.20; Heb 1.8; 2 Pet 1.1.

III. L (019)

a. Corrupted texts according to Ehrman (i.e., calls Jesus θεός): John 1.18, 20.28.

b. Text that supports Ehrman’s reading (i.e., does not call Jesus θεός): John 1.1.

IV. L (020)

a. Corrupted texts according to Ehrman (i.e., calls Jesus θεός): Heb 1.8; Jude 4 (Ehrman does not mention this text directly, but see n29 above).

b. Text that supports Ehrman’s reading (i.e., does not call Jesus θεός): Gal 2.20.

V. W (032)

a. Corrupted texts according to Ehrman (i.e., calls Jesus θεός): John 1.1; John 20.28.

b. Text that supports Ehrman’s reading (i.e., does not call Jesus θεός): John 1.18.

In light of these five examples, which are only a small sampling, I first conclude that much more work needs to be done in the realm transmissional history. More importantly though, just given my examples above, my second, and main, conclusion is that no one would have received a purely truncated view of the deity of Christ if they only received their manuscript. Meaning, each manuscript listed above has at least one “Jesus-θεός” verse that affirms the deity of Christ. It is inconsequential, then, that every potential “Jesus-θεός” passage in every manuscript affirm the same. This evidential conclusion causes another major problem in Ehrman’s overall orthodox corruption thesis.

In the end, I believe that the preponderance of evidence (geographically, genealogically, and internally) points to the true textual reading, “but to the Son [he declares], ‘Your throne, O God, is forever and ever, and a righteous scepter is the scepter of your kingdom.’ ” The probability, then, is high that Heb 1.8 explicitly calls Jesus θεός.

2 Peter 1:1

Second Peter 1.1 is another NT verse potentially calling Jesus θεός. Some MSS (a Ψ 398 442 621 l596 syr^ph vg^mss cop^sa)¹³⁶ read κυρίου instead of θεοῦ in v. 1:

ἐν δικαιοσύνῃ τοῦ θεοῦ ἡμῶν καὶ σωτῆρος Ἰησοῦ Χριστοῦ

ἐν δικαιοσύνῃ τοῦ κυρίου ἡμῶν καὶ σωτῆρος Ἰησοῦ Χριστοῦ

The external support, however, overwhelmingly favors of θεοῦ. In fact, the NA27 and Editio Critica Maior together only list nine witnesses for κυρίου (mentioned above, with only the NA27 listing vg^mss). This means virtually all other witnesses support θεοῦ.

Nevertheless, there are several reasons which appear to support κυρίου. First, θεοῦ could have arisen due to a scribal oversight of the nomen sacrum: K=-=-u vs. q=-u.¹³⁷ Second, the phrase “Lord and Savior” is statistically superior when referring to Christ in 2 Peter. Four times it reads “Lord and Savior” (1.11; 2.20; 3.2, 18) while only once it reads “God and Savior” (if one accepts it in 1.1). Third, a shift to θεοῦ could have been a motivated orthodox corruption to make the text speak unambiguously of Jesus as θεός due to the Christological controversies during the early centuries. Fourth, κυρίου maintains the alleged parallelism between 1.1 and 1.2, distinguishing θεός and Jesus. Fifth, θεός is rarely used of Jesus in the NT.

As for θεοῦ, most of the critiques above can be justifiably reversed while adding a few more arguments. First, although κύριος and θεός are among the earliest nomen sacra,¹³⁸ no other viable variants for κύριος or θεός exist in 2 Peter (1.14; 2.4, 9; 3.9). Second, “Lord and Savior” is the NT (and 2 Peter) norm and a scribe could have easily harmonized it. Third, κυρίου might have been sought to maintain the alleged parallelism between 1.1 and 1.2 (even though the alleged parallelism would be extremely rare in the NT). Fourth, θεοῦ is the harder reading as the opposing critiques reveal. Fifth, the construction is different when an author desires to distinguish two persons (e.g., 2 Pet 1.2; τοῦ θεοῦ καὶ Ἰησοῦ τοῦ κυρίου ἡμῶν).¹³⁹ Sixth, the doxology in 3.18 and the phrase in 1.1 are attesting to Jesus’ exalted status and are both consistent Christologically with the rest of the NT.¹⁴⁰ It should not be argued that the differing words (“God” in 1.1 and “Lord” in 3.18) refute this concept since similar parallels can be shown elsewhere with differing words (e.g., Matt 1.23 and 28.20; Mark 1.1 and 15.39; John 1.1 and 20.28). Seventh, this phrase might be in sync with Hellenistic religious language in order to communicate the gospel meaningfully to Gentile converts.¹⁴¹ Eighth, the external evidence is far better and earlier (not to mention the existing unanimity within all major published NT Greek texts, e.g., NA27, Tischendorf, UBSGNT4, Bover, Merk, von Soden, Westcott and Hort, Vogels, and Weiss). Ninth, the identification of Jesus as θεός here is entirely realistic in light of progressive revelation (2 Peter being one of the last NT books written).¹⁴² Tenth, the Granville Sharp Rule undoubtedly applies to this construction, thereby referring both titles (“God” and “Savior”) to Jesus Christ.¹⁴³ “It is hardly open for anyone to translate 1 Pet 1:3 ‘the God and Father’ and yet here decline to translate ‘the God and Saviour’.”¹⁴⁴ Likewise, Hiebert concludes, “Elsewhere, this epistle never uses the word Savior alone but always coupled with another name under the same article (cf. 1:11; 2:20; 3:2; 18).”¹⁴⁵

At the end of the day, I believe θεοῦ best accounts for all the evidence. If this verdict is correct, it is highly probable that Jesus is explicitly called θεός in 2 Pet 1.1.

CONCLUSION

No one contests that the NT usually reserves the title θεός for God the Father. Yet this usage, though dominant, is not exclusive.¹⁴⁶ The textual proof of the designation θεός as applied to Jesus in the NT merely confirms what other grounds have already established. In fact, the title θεός only makes explicit what is implied in other Christological titles such as κύριος and υἱὸς θεοῦ. Harris adds:

Even if the early Church had never applied the title θεός to Jesus, his deity would still be apparent in his being the object of human and angelic worship and of saving faith; the exerciser of exclusively divine functions such as creatorial agency, the forgiveness of sins, and the final judgment; the addressee in petitionary prayer; the possessor of all divine attributes; the bearer of numerous titles used of Yahweh in the OT; and the co-author of divine blessing. Faith in the deity of Christ does not rest on the evidence or validity of a series of ‘proof-texts’ in which Jesus may receive the title θεός but on the general testimony of the NT corroborated at the bar of personal experience.¹⁴⁷

The question now before us is not whether the NT explicitly ascribes the title θεός to Jesus, but how many times he is thus identified and by whom.¹⁴⁸ Therefore, with at least one text that undoubtedly calls Jesus θεός in every respect (John 20.28), I will conclude by answering my initial question: When did this boldness to call Jesus θεός begin? It began in the first century. It was not a creation of Constantine in the fourth century. It was not a doctrinal innovation to combat Arianism in the third century. Nor was it a sub-apostolic distortion of the apostolic kerygma in the second century. Rather, the church’s confession of Christ as θεός began in the first century with the apostles themselves and/or their closest followers and therefore most likely from Jesus himself.

Table 4. Jesus as Θεός¹⁴⁹

¹ Special thanks are due to Drs. J. K. Elliot, Gordon D. Fee, P. J. Williams, Daniel B. Wallace, Tommy Wasserman, Darrell Bock, and Chrys Caragounis for looking at a preliminary draft of this manuscript and making valuable suggestions.

² For a detailed list of many such views see Daniel B. Wallace, Granville Sharp’s Canon and Its Kin: Semantics and Significance (Bern: Peter Lang, 2008), 27-8.

³ Bart Ehrman, in at least three published books and one published lecture series, even suggests that the Ausgangstext does not necessarily teach the deity of Christ. He bases these allegations on alleged textual problems that he attributes to manipulative scribal activity; most often pointing to textual problems behind such verses. He almost exclusively leans toward the manipulation of early proto-orthodox scribes in the development of a high Christology in his book The Orthodox Corruption of Scripture: The Effect of Early Christological Controversies on the Text of the New Testament (Oxford: OUP, 1993). Cf. also Larry Hurtado, Lord Jesus Christ: Devotion to Jesus in Earliest Christianity (Grand Rapids: Eerdmans, 2003).

⁴ As Raymond Brown hypothesizes, “The slow development of the usage of the title ‘God’ for Jesus requires explanation…The most plausible explanation is that in the earliest stage of Christianity the Old Testament heritage dominated the use of ‘God’; hence, ‘God’ was a title too narrow to be applied to Jesus…” I am unconvinced that that is the “most” plausible explanation given the predominately Jewish context which may have dictated the early evangelistic terminology (e.g., Matthew’s “kingdom of heaven”). Nevertheless, Brown adds, “… we do maintain that in general the NT authors were aware that Jesus was being given a title which in the LXX referred to the God of Israel” (Raymond Brown, “Does the New Testament call Jesus ‘God’?” TS 26 [1965], 545-73).

⁵ In fact, Mark 10.18 records that He differentiates Himself from God (= the Father) [cf. Matt 19.17; Luke 18.19; Mk 15.34; Matt 27.46; John 20.17]. H. W. Montefiore, in his essay “Toward a Christology for Today,” notices this as he postulates that Jesus seems to have explicitly denied that he was God (published in Soundings [1962], 158). In addition, R. H. Fuller, similar to Bultmann, believes that Jesus understood himself as an eschatological prophet (Reginald H. Fuller, The Foundations of New Testament Christology [New York: Charles Scribner’s Sons, 1965], 130). While none of these texts or interpretations portray a complete NT Christology (Jesus does identify himself with God [e.g., John 10.30; 14.9], he never explicitly rejects that he is God, and Jesus understood himself to be more than an eschatological prophet), it is true that Jesus never uses the term θεός for Himself.

⁶ Acts 20.28 is in a speech (and the only one) addressed to a Christian audience. “All the others are either evangelistic sermons . . . legal defenses . . . or the five speeches before the Jewish and Roman authorities . . . ” (John R. W. Stott, The Message of Acts: The Son, the Church, and the World [Downers Grove, IL: Inter-Varsity, 1990], 323). Cf. Richard I. Pervo, Acts: A Commentary (Minneapolis: Fortress Press, 2009), 515-531.

⁷ Raymond Brown, however, insightfully notes that a danger in judging usage from occurrence exists because NT occurrence does not create a usage but testifies to a usage already extant. And none of the passages considered below give any evidence of innovating (Raymond E. Brown, Jesus: God and Man [Milwaukee: Bruce Publishing Co., 1967]).

⁸ With Rom 9.5 probably occurring first, if one could be certain of its punctuation/grammar (see discussion below).

⁹ In a recent popular book, Reinventing Jesus, the authors note that, “there are at least forty-eight (and as many as fifty-nine) Greek New Testament manuscripts that predate the fourth-century.” In an endnote, the authors go on to explain that these are only Greek New Testament MSS and do not include the early versions or the pre-fourth-century patristic writers. Even so, only four “Jesus-θεός” passages (Rom 9.5; John 1.1, 18; 20.28) are included in these MSS (Ed Komoszewski, M. James Sawyer, and Daniel B. Wallace, Reinventing Jesus: What The Da Vinci Code and Other Novel Speculations Don’t Tell You [Grand Rapids: Kregel, 2006], 116).

¹⁰ Ibid., 114, notes, “If a particular verse does not teach the deity of Christ in some of the manuscripts, does this mean that that doctrine is suspect? It would only be suspect if all the verses that affirm Christ’s deity are textually suspect.” Unfortunately, regarding the explicit “Jesus-θεός” passages, that is the case here. At the same time, the authors continue, “And even then the variants would have to be plausible.” This further reveals the importance of this study.

¹¹ Inf. Gos. Thom. 7.4. From the Greek text of Constantin von Tischendorf, Evangelia Apocrypha (Hildesheim: George Olms, 1987; original: Leipzig, 1867). For a more recent text-critical work on it, see T. Chartrand-Burke, “The Greek Manuscript Tradition of the Infancy Gospel of Thomas,” Apocrypha 14 (2003): 129-151.

¹² Pliny, Letters and Panegyricus in Two Volumes: Letters, Books VIII-X and Panegyricus, trans. by Betty Radice, Loeb Classical Library (Cambridge, MA: Harvard University Press, 1969), 2.288-9.

¹³ “Ignatius designates Jesus as ‘God’ on at least eleven occasions,” notes Weinandy, “Thus, Ignatius effortlessly and spontaneously wove within his understanding of the relationship between the Father and the Son the simple and unequivocal proclamation that Jesus Christ is God” (Thomas Weinandy, “The Apostolic Christology of Ignatius of Antioch: The Road to Chalcedon,” in Trajectories Through the New Testament and the Apostolic Fathers [New York: Oxford University Press, 2005], 76). Here are 14 such occurrences in Ignatius: Eph. prol.; 1.1; 7.2; 15.3; 18.2; 19.3; Rom. prol. (2x); 3.3; 6.3; Smyrn. 1.1; 10.1; Trall. 7.1; Pol. 8.3.

¹⁴ θεὸν ἀληθινὸν ἐκ θεοῦ ἀληθινοῦ (Philip Schaff, The Creeds of Christendom: A History and Critical Notes 3 vols. [New York: Harper and Brothers, 1877], 2.57).

¹⁵ I am discussing the origin of the title θεός as applied to Jesus and not the origin of understanding Jesus as divine. That understanding was early and expressed in various ways (see, among others, C. F. D. Moule, The Origin of Christology [Cambridge: Cambridge University Press, 1977]; Hurtado, Lord Jesus Christ: Devotion to Jesus in Earliest Christianity [Grand Rapids: Eerdmans, 2003]). As for the title θεός, “On the one hand, the dominant Greco-Roman ethos assumed that there were many gods and that human beings could be deified. Many emperors refused to be called gods during their lifetimes, yet were named gods after their deaths. The term “god” was also used for living rulers, like Agrippa (Acts 12:21-22; Josephus, Ant. 19.345) and Nero (Tacitus, Annals 14.15). On the other hand, the Jewish tradition centered on faith in one God (Deut 6:4), who was not to be portrayed in human form or to be identified with a human being (Exod 20:4; Deut 5:8; 2 Macc 9:12; cf. John 5:18; 10:33)” (Craig R. Koester, Hebrews: A New Translation with Introduction and Commentary [New York: Doubleday, 2001], 202). Further, one should note that the majority of passages in which Jesus is potentially called θεός appear in writings attributed to Jewish settings, whereas only a few might be Pauline (see, e.g., Richard N. Longenecker, The Christology of Early Jewish Christianity [Naperville, IL: A. R. Allenson, 1970], 139).

¹⁶ I will employ a reasoned eclecticism method, the currently reigning view among textual critics. Several limitations exist, however, on the scope of my research. For example, I did not exhaustively examine each critical apparatus to find other variants that potentially affirm Jesus as θεός. I did not work extensively with foreign literature. I relied heavily on the manuscript collations of others. I created no comprehensive comparative analysis of the manuscript relationships for the Pauline corpus or for any individual book(s). I did not determine the scribal habits of every MS or witness cited. I also depended heavily on those whose academic acumen regarding textual criticism far exceeds mine, and whose scholarly contributions are highly regarded.

¹⁷ A handful of other verses are sometimes used to implicitly equate Jesus with θεός (Luke 8.39; 9.43; 1 Thess 4.9; 1 Tim 1.1; 5.21; 2 Tim 4.1; Titus 1.3; 3.4; Heb 3.4; Jas 1.1), yet I did not think enough academic support existed to merit their inclusion in this work.

¹⁸ Douglas J. Moo, “The Christology of the Early Pauline Letters,” in Contours of Christology in the New Testament (Grand Rapids: Eerdmans, 2005), 190. Similarly, Ehrman concludes, “Nor will I take into account variant modes of punctuation that prove christologically significant, as these cannot be traced back to the period of our concern, when most manuscripts were not punctuated” (Bart D. Ehrman, The Orthodox Corruption of Scripture: The Effect of Early Christological Controversies on the Text of the New Testament [New York: Oxford University Press, 1993], 31). Cf. Robert Jewett, Romans (Minneapolis: Fortress, 2007), 555, 566-69; Bruce M. Metzger, A Textual Commentary on the Greek New Testament, 2nd ed. (Stuttgart: German Bible Society, 1994), 459-62; Murray J. Harris, Jesus as God: The New Testament Use of Theos in Reference to Jesus (Grand Rapids: Baker, 1992), 150-1; and F. W. Danker, reviser and editor, A Greek-English Lexicon of the New Testament and other Early Christian Literature, 3rd ed. (Chicago: University Press, 2000 [from here on, BDAG]), s.v. “θεός.” One must wait, then, to see if any new evidence or manuscript(s) is(are) evinced to reverse this scholarly consensus.

The earliest MS of Romans to-date (P46, ca. 200 [Cf. Kurt Aland, Kurzgefasste Liste der griechischen Handschriften des Neuen Testaments, 2nd ed. (Berlin: Gruyter, 1994), 31-32]) does not contain any punctuation here. Nevertheless, Lattey shows that a fifth-century codex (C/04) contains a small cross between σάρκα and ὁ ὤν designates some form of a stop, which the NA27 and UBSGNT4 texts reflect with a comma (Cuthbert Lattey, “The Codex Ephraemi Rescriptus in Romans ix. 5,” ExpTim 35 [1923-24]: 42-43).

¹⁹ Metzger, Textual Commentary, 167.

²⁰ Listed conveniently in The Text of the New Testament (New York: Oxford University Press, 1992), 236-7.

²¹ The text is overwhelmingly certain here as the author cites Isa 7.14 in relation to the birth of Jesus. Yet, in spite of its textual certainty, we cannot be sure that the evangelist takes “God with us” literally and attempts to call Jesus θεός. See, among others, R. T. France, The Gospel of Matthew (Grand Rapids: Eerdmans, 2007), esp. 49-50 and 56-58.

²² Note the discussion of the grammatical issues relating to this phrase in Harris, Jesus as God, 258-59. The text, nonetheless, should be considered certain.

²³ The textual evidence is solid here. Ehrman accurately explains, “In the text that is almost certainly original (‘the Kingdom of Christ and God’), Christ appears to be given a certain kind of priority over God himself. This problem is resolved by all of the changes, whether attested early or late” (Ehrman, Orthodox Corruption, 269). See Harris, Jesus as God, 261-63, for grammatical issues.

²⁴ The textual issue in this verse does not pertain to the clause in question. Leaving one, therefore, with two possible Greek genitive translations: (1) “according to the grace of our God and Lord, namely Jesus Christ” or (2) “according to the grace of our God and the Lord Jesus Christ.” I favor the latter, which does not attribute the title θεός to Jesus, primarily for the following reason, “Second Thessalonians 1:12 does not have merely ‘Lord’ in the equation, but ‘Lord Jesus Christ.’ Only by detaching κυρίου from Ἰησοῦ Χριστοῦ could one apply [Granville] Sharp’s rule to this construction” (Wallace, Sharp’s Canon, 236).

²⁵ The attestation for the variants is not strong enough to warrant serious consideration. Towner notes, “. . . the change to ὅ (D* and Vg plus some Latin Fathers) was a gender adjustment to accord with τὸ μυστήριον; another late solution was the change to θεός (a² A^c C² D² Ψ 1739 1881 TR vg^mss), which supplies the antecedent thought to be lacking in ὅς” (Philip Towner, The Letters to Timothy and Titus [Grand Rapids: Eerdmans, 2006], 278). Cf. W. M. Zoba, “When Manuscripts Collide,” ChristToday 39, no. 12 (1995): 30-1. Cf. also Robert H. Gundry, “The Form, Meaning and Background of the Hymn Quoted in 1 Timothy 3:16” in Apostolic History and the Gospel: Biblical and Historical Essays Presented to F. F. Bruce (eds. W. Ward Gasque and Ralph P. Martin [Exeter: Paternoster Press, 1970]), 203-222.

²⁶ Though I strongly feel that this verse attributes the title θεός to Jesus, a textual examination is unnecessary since the only viable variant concerns the order of the last two words: Ἰησοῦ Χριστοῦ or Χριστοῦ Ἰησοῦ. The debate, then, will have to continue congregating around syntax. See Gordon Fee, Pauline Christology: An Exegetical-Theological Study (Peabody, MA: Hendrickson, 2007), esp. 442-6. Against Fee’s position, see Wallace, Sharp’s Canon, 256-264; Robert M. Bowman, Jr., “Jesus Christ, God Manifest: Titus 2:13 Revisited,” JETS 51 (2008): 733-52.

It should also be noted that several NT scholars put an asterisk by this book because they consider it deutero-Pauline. Yet even if one assumes that Paul did not write Titus, it still would have been written in the first century and, therefore, impervious to some of the critiques often given for such texts; e.g., orthodox corruption(s) due to the third-century Arian controversy. As a matter of fact, although Ehrman did not mention Titus 2.13 specifically in Orthodox Corruption, by his own argument regarding 2 Pet 1.1, Titus 2.13 would explicitly equate Jesus with θεός, “Because the article is not repeated before Ἰησοῦ (in 2 Pet 1:1), it would be natural to understand both ‘our God’ and ‘Savior’ in reference to Jesus [our ‘God and Savior’]” (Orthodox Corruption, 267). In other words, Ehrman recognizes that one article with two nouns joined by καί refers to the same person; making Titus 2.13 an explicit reference to Jesus as θεός.

²⁷ Of the two notable variants in this verse, neither of them effectually touches our present topic. The crux interpretum is the antecedent of οὗτος, but it is not clear whether it represents a reference to God the Father or Jesus Christ (See Wallace, Sharp’s Canon, for a discussion of the syntax of 1 John 5.20). Even so, Augustine used this verse to support his argument that Jesus was “not only God, but also true God” (The Trinity: Introduction, Translation, and Notes [New York: New City Press, 2000], 71). Likewise, Rudolf Schnackenburg argues strongly from the logic of the context and the flow of the argument that “This is the true God” refers to Jesus Christ (Die Johannesbriefe, in Herders theologischer Kommentar [2^nd ed.; Freiburg: Herder, 1963], 291). “But even if we do not accept the equation (Jesus as God) as explicitly present in this verse, it remains true that there is an association between God and his Son that is articulated here more clearly than anywhere else in 1 John” (Stephen S. Smalley, 1, 2, 3 John [Nashville: Thomas Nelson, 2007], 295).

²⁸ I kept this text in the condensed list primarily because several MSS contain the word θεόν (e.g., K L P S Ψ 049 104 syr^{h, ph}). Landon persuasively argues that the internal evidence supports δεσπότην θεόν rather than simply δεσπότην, and that the expression refers only to God (“The Text of Jude and a Text-Critical Study of the Epistle of Jude,” JSNTSup 135 [Sheffield: Academic Press, 1996], 63-67). What makes his argument strong is that if Ehrman is correct about the direction of corruption away from adoptionist heresies, noting the text of 2 Pet 1:2 in P72, then this reading alone resists orthodox interference (i.e., shortened by scribes who wish to show God and Jesus as the same entity, thereby stressing Christ’s divinity). Yet even with Landon’s well thought out thesis, of which I did not list all his perceptive reasons, I still reject the longer reading for the following reasons: (1) the earliest and best MSS support the shorter reading [e.g., P72 P78 a A B C 0251 33 1739 Lect^pt it^ar vg cop^{sa, bo} geo], (2) it is probable that a scribe sought to clarify the shorter reading and/or stay within the NT’s normal pattern [i.e., Luke 2.29; Acts 4.24; 2 Tim 2.21; Rev 6.10], and (3) it is the more difficult reading. Therefore, my preference is for the shorter reading: δεσπότην (used of God in Luke 2.29; Acts 4.24; Rev 6.10 and of Christ in 2 Pet 2.1 and here). For exhaustive MS evidence see Tommy Wasserman, The Epistle of Jude: Its Text and Transmission (Stockholm: Almqvist & Wiksell, 2006), esp. 251-4. Cf. also C. A. Albin, Judasbrevet: Traditionen, Texten Tolkningen (Stockholm: 1962), 148 and 596.

In addition, the shorter reading in Jude 4 (where Christ is described as the ruling Master, δεσπότην) would comport well with Jude 5 if “Jesus” is indeed the original reading. This would clearly highlight the pre-existence of Christ and thus implicitly argue for his deity. Therefore, both verses taken together make a compelling argument for the pre-existence, as well as the deity, of Jesus Christ (without giving the title θεός to Jesus). For in-depth textual discussion of Jude 5 see, Philipp F. Bartholomä, “Did Jesus Save the People out of Egypt?–A Re-Examination of a Textual Problem in Jude 5” NovT 50 (2008): 143-158. For an opposing view on Jude 5, see James R. Royse, Scribal Habits in Early Greek New Testament Papyri (NTTSD 36, ed. Bart D. Ehrman and Eldon J. Epp; Boston: Brill, 2008), 610-12.

²⁹ I recognize that the anarthrous θεός denotes the pre-existent λόγος and not explicitly Jesus (yet?). I also acknowledge that some scholars have argued well that John 1.1 is a part of the hymn exalting God’s σοφία (the hmkh of Proverbs 8; cf. Sir 1.1-10) and/or have shown that Philo periodically uses the term θεός without the definite article for λόγος (e.g., Somn. 1.230). Nevertheless, without taking the referent for λόγος for granted (even though, for example, σοφία is never designated the title θεός and Philo’s over 1300 uses of λόγος are systematically different from John’s meaning), I still believe the pre-existent λόγος eventually points to Jesus, the λόγος incarnate (i.e., John 1.14, 17; cf. Rev 19.13) and therefore pertains to this paper’s examination. For similar (recent) conclusions about the pre-existent λόγος eventually pointing to Jesus, see, among others, Martin Hengel, “The Prologue of the Gospel of John as the Gateway to Christological Truth,” in The Gospel of John and Christian Theology (ed. Richard Bauckham and Carl Mosser; Grand Rapids: Eerdmans, 2008), 271; Uwe-Karsten Plisch, The Gospel of Thomas: Original Text with Commentary (Stuttgart: Deutsche Bibelgesellschaft, 2008), 76-77; Douglas J. Moo, The Letters to the Colossians and the Philemon (Grand Rapids: Eerdmans, 2008), 118; Andreas J. Köstenberger and Scott R. Swain, Father, Son and Spirit: The Trinity and John’s Gospel (ed. D. A. Carson; Downers Grove, IL: InterVarsity, 2008), 113; Douglas W. Kennard, Messiah Jesus: Christology in His Day and Ours (New York: Peter Lang, 2008), 503.

³⁰ Aland, Kurzgefasste Liste, 29-33. What is more, this statistic was without the benefit of many more John papyrus fragment discoveries to-date (see, e.g., J. K. Elliott, “Five New Papyri of the New Testament,” NovT 41 [1999]: 209-213; idem, “Four New Papyri Containing the Fourth Gospel and their Relevance for the Apparatus Criticus,” Journal of Theological Studies 59 [2008]: 674-678; Peter Head, “P. Bodmer II (P⁶⁶): Three Fragments Identified. A Correction,” NovT 50 [2008]: 78-80).

³¹ This unanimity continues today, for example, in such specialized (i.e., single book) text critical works as the IGNTP edition of the Gospel of John (i.e., The American and British Committees of the International Greek New Testament Project, The New Testament in Greek IV, The Gospel According to St. John Volume One: The Papyri [New York: Brill, 1995], 123; idem, Volume Two: The Majuscules [Boston: Brill, 2007], 189).

³² Reference is made to the publication year (1993) of Bart Ehrman’s Orthodox Corruption.

³³ Ehrman, Orthodox Corruption, 179n187.

³⁴ Ibid.

³⁵ Ibid. On the other hand, John could have used θεῖος (e.g., Acts 17.29; 2 Pet 1.3, 4), or some other word meaning “divine,” had he wished to convey Jesus as simply divine. Keener helpfully points out, “Regarding Jesus as merely ‘divine’ but not deity violates the context; identifying him with the Father does the same. For this reason, John might thus have avoided the article even had grammatical convention not suggested it; as a nineteenth-century exegete argued, an articular θεός would have distorted the sense of the passage, ‘for then there would be an assertion of the entire identity of the Logos and of God, while the writer is in the very act of bringing to view some distinction between them’... Scholars from across the contemporary theological spectrum recognize that, although Father and Son are distinct in this text, they share deity in the same way” (Craig S. Keener, The Gospel of John: A Commentary [Peabody, MA: Hendrickson, 2003], 374).

³⁶ Contra Modalism/Sabellianism (and the Jehovah’s Witnesses rendering of John 1.1c in their New World Translation). Philip Harner, after probing the Fourth Gospel for passages which use predicate nouns, points out that the qualitative force of the predicate is more prominent that its definiteness or indefiniteness in 40 of the 53 cases which use anarthrous predicates preceding the verb. Specifically, “In John 1:1 I think that the qualitative force of the predicate is so prominent that the noun cannot be regarded as definite.” He also suggests “… the English language is not as versatile at this point as Greek, and we can avoid misunderstanding the English phrase only if we are aware of the particular force of the Greek expression that it represents” (“Qualitative Anarthrous Predicate Nouns,” JBL 92 [1973], 75-87). Cf. J. G. Griffiths, “A Note on the Anarthrous Predicate in Hellenistic Greek” ExpTim 62 (1950-1), 314-316; Robert W. Funk, “The Syntax of the Greek Article: Its Importance for Critical Pauline Problems” (Dissertation, Vanderbilt, August, 1953), 148; Robertson, Grammar, 767-68; Wallace, Greek Grammar, 266-69.

On a similar note, over 50 years ago, Bruce Metzger explicitly rejected the rendering “a god” in John 1.1c as reflected in the Jehovah’s Witnesses’ own translation of the New Testament, The New World Translation (Metzger, “On the Translation of John i.1,” ExpTim 63 (1951-52): 125-6; idem, “Jehovah’s Witnesses and Jesus Christ,” ThTo 10 (1953): 65-85). His main argument (in both noted publications) focused on Greek grammar (i.e., Colwell’s Rule). Unfortunately, that argument, though still a popular one today (e.g., Andreas J. Kostenberger and Scott R. Swain, Father, Son and Spirit: The Trinity and John’s Gospel [Downers Grove, IL: InterVarsity, 2008], 49; Douglas W. Kennard, Messiah Jesus: Christology in His Day and Ours [New York: Peter Lang, 2008], 473), is misleading. For clarification, see Wallace, Exegetical Syntax, 257-262.

³⁷ Gordon D. Fee, “P75, P66, and Origen: The Myth of Early Textual Recension in Alexandria” (New Dimensions in New Testament Study [ed. R. Ν. Longenecker and M. C. Tenney, Grand Rapids: Eerdmans, 1974], 44). Cf. also Peter M. Head, “Christology and Textual Transmission: Reverential Alterations in the Synoptic Gospels,” NovT 35 (1993): 105-29, esp. 112-3.

³⁸ Clark, “The Gospel of John in Third-Century Egypt,” NovT 5 (1962): 24. Cf. also S. A. Edwards, “P⁷⁵ under the Magnifying Glass,” NovT 18 (1976): 190-212.

³⁹ Ehrman, Orthodox Corruption, 112.

⁴⁰ Only Merk’s critical NT text contains Codex Freerianus (W[032-S]). Then again, as Dr. Daniel Wallace perceptively pointed out to me, Codex W was not discovered until after Tischendorf wrote his critical work and while von Soden was producing his work (i.e., its publication was shortly before von Soden’s final volume).

⁴¹ “[T]he first quire of John. . . is a later (probably eighth-century) replacement quire that bears no relation to the rest of the manuscript and made up for the (presumably) lost original portion” (James R. Royse, “The Corrections in the Freer Gospels Codex,” in The Freer Biblical Manuscripts: Fresh Studies of an American Treasure Trove, ed. Larry W. Hurtado [Atlanta: Society of Biblical Literature, 2006], 186). Cf. Metzger-Ehrman, Text of the New Testament, 77-81; Edgar J. Goodspeed, “Notes on the Freer Gospels,” AJT 13 (1909): 597-603, esp. 599.

⁴² See Matthew P. Morgan’s chapter, “Egregious Regius: Sabellianism or Scribal Blunder in John 1:1c?”, in this book.

⁴³ Ibid.

⁴⁴ “Today we count about 182 Coptic MSS of the Gospel of John in the Sahidic dialect” (Karlheinz Schüssler, “Some Pecularities of the Coptic (Sahidic) Translations of the Gospel of John,” Journal of Coptic Studies 10 (2008): 41-62). That number, Schüssler continues, includes five complete MSS of John’s Gospel (i.e., sa 505, 506, 508, 561, 600), 38 lectionaries, and three other liturgical MSS. His recent MS calculation helps explain the “1057 Coptic citations of John’s gospel in the 27^th edition of the Nestle-Aland Novum Testamentum Graece” (Christian Askeland, “Has the Coptic Tradition Been Properly Used in New Testament Textual Criticism?” [paper presented at the annual meeting of the SBL, Boston, MA, 22 November 2008], 1).

⁴⁵ Frederik Wisse, “The Coptic Versions of the New Testament,” in The Text of the New Testament in Contemporary Research: Essays on the Status Quaestionis (Grand Rapids: Eerdmans, 1995), 137. Cf. Metzger, Textual Commentary, 15; idem, The Early Versions of the New Testament: Their Origin, Transmission, and Limitations (Oxford: Clarendon, 1977), esp. 132-37; Bruce M. Metzger and Bart D. Ehrman, The Text of the New Testament: Its Transmission, Corruption, and Restoration, 4th ed. (New York: Oxford University Press, 2005), 110-15.

⁴⁶ Bohairic was a different (but new) Coptic translation from Greek. More importantly, though, it is an important witness to the secondary Alexandrian type of text (see, e.g., Metzger, Textual Commentary, 15).

⁴⁷ George W. Horner ed., The Coptic Version of the New Testament in the Southern Dialect, Otherwise Called Sahidic and Thebaic, with Critical Apparatus, Literal English Translation, Register of Fragments and Estimate of the Version, 7 vols. (Oxford: Clarendon, 1911-1924), 3.2. Cf. Hans Quecke, Das Johannesevangelium saïdisch: Text der Handschrift PPalau Rib. Inv.-Nr. 183 mit den Varianten der Handschriften 813 und 814 der Chester Beatty Library und der Handscrift M569 (Barcelona; Rome: Papyrologica Castroctaviana, 1984), 73. For Bohairic, see George W. Horner ed., The Coptic Version of the New Testament in the Northern Dialect, Otherwise Called Memphitic and Bohairic, with Introduction, Critical Apparatus, and Literal English Translation, 4 vols. (Oxford: Clarendon, 1898-1905), 3.2.

⁴⁸ Ibid., Sahidic, 3.3.

⁴⁹ Ibid., 3.376 (italics mine).

⁵⁰ Thomas Lambdin, Introduction to Sahidic Coptic (Macon, GA: Mercer, 1983), 5. Cf. also Bentley Layton, A Coptic Grammar: Sahidic Dialect (Wiesbaden: Harrassowitz, 2000).

⁵¹ Horner, Sahidic, 3.307. A few other examples from the Gospel of John include: 1.16, 26, 33; 3.5, 6; 5.39; 6.10; 16.33.

⁵² The following statistics were produced via the base-text of the NA27 and UBSGNT4 in Bibleworks 7.0. For the Sahidic Coptic version, I examined the George W. Horner ed., The Coptic Version of the New Testament in the Southern Dialect, Otherwise Called Sahidic and Thebaic, with Critical Apparatus, Literal English Translation, Register of Fragments and Estimate of the Version (7 vols. Oxford: Clarendon, 1911-1924); Hans Quecke, Das Johannesevangelium saïdisch: Text der Handschrift PPalau Rib. Inv.-Nr. 183 mit den Varianten der Handschriften 813 und 814 der Chester Beatty Library und der Handscrift M569 (Barcelona, Rome: Papyrologica Castroctaviana, 1984); the Herbert Thompas ed., The Coptic Version of the Acts of the Apostles and The Pauline Epistles in the Sahidic Dialect (Cambridge, 1932) [Chester Beatty MSS.]. Admittedly, a slight distortion of the database might occur due to text-critical issues (e.g., John 8.54). I did not exhaustively examine each critical apparatus, MS, or witness, to find other viable variants that attest to a nominative singular θεός/ⲛⲟⲩⲧⲉ. My purpose here was merely to obtain a highly probable snapshot of occurrences and patterns via several modern Greek NT editions. More exhaustive work, then, is needed (and forthcoming) in this area.

⁵³ θεός occurs twice in this verse (with the second occurrence placed in brackets in both the NA27 and UBSGNT4), but only the second (anarthrous) one is reflected in the Coptic. The other three are John 1.18; 8.54; Rev 21.7.

⁵⁴ Every citation is from 1 John (the only Johannine epistle with a nominative singular θεός).

⁵⁵ Harris, Jesus as God, 29.

⁵⁶ Daniel Rathnakara Sadananda, The Johannine Exegesis of God: An Exploration into the Johannine Understanding of God (New York: Walter de Gruyter, 2004), 177.

⁵⁷ Kurt Aland, Barbara Aland, and Klaus Wachtel, Text und Textwert der griechischen Handschriften des Neuen Testaments: Johannesevangelium (New York: Gruyter, 2005), 3-5.

⁵⁸ Several exegetical and historical details exist that will not be canvassed here.

⁵⁹ John 1:18 is actually the only verse listed under textual issues in both major works on this topic: the standard work by Murray Harris, Jesus as God, lists only three problems as “textual” (Heb 1:8; 2 Pet 1:1; John 1:18) and Raymond Brown, in An Intro to NT Christology, lists three under “textual”: Gal 2:20; Acts 20:28; John 1:18.

⁶⁰ Several major published Greek NT texts are evenly divided here as to the Ausgangstext (e.g., von Soden, Bover and Tischendorf choose ὁ μονογενὴς υἱός, and the UBSGNT4, NA27 and Merk favored μονογενὴς θεός).

⁶¹ For the chief characteristics regarding the copying activity of the scribes of both P66 and P75, consult James R. Royse, Scribal Habits in Early Greek New Testament Papyri (NTTSD 36, ed. Bart D. Ehrman and Eldon J. Epp; Boston: Brill, 2008), esp. 544 and 704.

⁶² Nevertheless, as Clark admonishes, “We are mindful that these papyri cannot claim unquestioned priority on the ground alone of their greater antiquity… [nor can we] blindly follow their textual testimony even when the two are in agreement with one another” (Kenneth W. Clark, “The Gospel of John in Third-Century Egypt,” NovT 5 [1962]: 23). Cf. also Eldon J. Epp, “A Dynamic View of Textual Transmission,” in Studies in the Theory and Method of New Testament Textual Criticism, ed. Eldon J. Epp and Gordon D. Fee (Grand Rapids: Eerdmans, 1993), 274.

⁶³ Ehrman, Orthodox Corruption, 112.

⁶⁴ Kostenberger and Swain, Father, Son and Spirit, 78.

⁶⁵ Ehrman, Orthodox Corruption, 79 (italics added). Later, we shall see that he reverses the same external appraisal he employs here (see Heb 1:8 discussion below).

⁶⁶ Sinaiticus (a) aligns with the “West” in John 1.1—8.38. See Gordon Fee, “Codex Sinaiticus in the Gospel of John: A Contribution to Methodology in Establishing Textual Relationships” in Studies in the Theory and Method of New Testament Textual Criticism, ed. Eldon J. Epp and Gordon D. Fee (Grand Rapids: Eerdmans, 1993), 221-43.

⁶⁷ Kurt Aland and Barbara Aland, The Text of the New Testament: An Introduction to the Critical Editions and to the Theory and Practice of Modern Textual Criticism, 2nd ed. (Grand Rapids: Eerdmans, 1989), 54. “Hardly anyone today refers to this putative Western text without placing the term in quotation marks, i.e., as the ‘Western text’” (ibid.). Likewise, Scrivener concludes, “. . . the text of Codex Bezae, as it stands at present, is in the main identical with one that was current both in the East and West. . .” (Frederick H. Scrivener, Bezae Codex Cantabrigiensis [London: Bell and Daldy, 1864], xlv).

⁶⁸ Aland-Aland, Text of the New Testament, 67. Cf. Roger L. Omanson, A Textual Guide to the Greek New Testament: An Adaptation of Bruce M. Metzger’s Textual Commentary for the Needs of Translators (Peabody, MA: Hendrickson, 2006), 22.

⁶⁹ Bart Ehrman, “Heracleon and the ‘Western’ Textual Tradition,” NTS 40 (1994): 178-79.

⁷⁰ See Bruce Metzger, “The Caesarean Text of the Gospels,” Chapters in the History of the New Testament Textual Criticism (Grand Rapids: Eerdmans, 1963), 42-72; Larry Hurtado, Text-Critical Methodology and the Pre-Caesarean Text: Codex W in the Gospel of Mark (Grand Rapids: Eerdmans, 1981), 24-45; Eldon Epp, “Issues in New Testament Criticism,” in Rethinking New Testament Textual Criticism (ed. David A. Black; Grand Rapids: Baker, 2002), 39.

⁷¹ Ehrman-Metzger, The Text of the New Testament, 118-19. Cf. Robert Blake and Maurice Brière, “The Old Georgian Version of the Gospel of John” PO 26/4 (Paris: Firmin-Didot, 1950).

⁷² For certain cautions when using Syriac, see P. J. Williams, Early Syriac Translation Technique and the Textual Criticism of the Greek Gospels (Piscataway, NJ: Gorgias, 2004).

⁷³ Ehrman-Metzger, Text of the New Testament, 98.

⁷⁴ See, for example, Quecke, Das Johannesevangelium Saïdisch, 75.

⁷⁵ Ehrman, Orthodox Corruption, 79.

⁷⁶ Contra Sadananda, The Johannine Exegesis of God, 210. Cf. Metzger, Textual Commentary, 15.

⁷⁷ Paul McReynolds, “John 1:18 in Textual Variation and Translation,” in New Testament Textual Criticism (Oxford: Clarendon, 1981), 118. Cf. Carroll Osburn, “Methodology in Identifying Patristic Citations in NT Textual Criticism,” NovT 47 (2005): 313-43.

⁷⁸ On the other hand, “Ἦν ποτε ὅτε οὐκ ἦν—“At one time he did not exist”—became the slogan that best expresses the core of his theology, which he shaped in the Thalia (θάλεια = “banquet”), his main work, which is cited in almost all of the sources dealing with Arius” (Hubertus R. Drobner, The Fathers of the Church: A Comprehensive Introduction [Peabody, MA: Hendrickson, 2007], 236). Cf. also Raymond E. Brown, The Gospel According to John (i-xii): Introduction, Translation, and Notes (Garden City, NY: Doubleday and Company, 1966), 17.

⁷⁹ Valentinus (another theologian deemed heretical) also accepted the same reading: Valentinians^{acc. to Irenaeus and Clement}. Furthermore, no Church Father accuses him of changing the text. Hort argued here that μονογενὴς θεός was original because the Gnostics (such as Valentinus) did not invent this phrase; instead, they quoted it (F. J. A. Hort, Two Dissertations [Cambridge: Macmillan, 1876]).

⁸⁰ William Rusch, The Trinitarian Controversy (Philadelphia, PA: Fortress, 1983), 29-30. For Greek text see Urkunden zur Geschichte des arianischen Streites, ed. by H. G. Opitz (Berlin: de Gruyter, 1934). Cf. R. P. C. Hanson, The Search for the Christian Doctrine of God: The Arian Controversy, 318-381 (Grand Rapids: Baker, 2005), 6. Lewis Ayres, Nicaea and its Legacy: An Approach to Fourth-Century Trinitarian Theology (New York: OUP, 2004), 105-126.

⁸¹ See also James R. Royse, Scribal Habits in Early Greek New Testament Papyri (NTTSD 36, ed. Bart D. Ehrman and Eldon J. Epp; Boston: Brill, 2008), esp. 19-27.

⁸² Given my methodology stated above, merely counting MSS is inadequate; one must also weigh them.

⁸³ Harris, Jesus as God, 92.

⁸⁴ For a more extensive grammatical examination, especially whether the adjective μονογενής can be used substantivally when it immediately precedes a noun of the same inflection, see Stratton L. Ladewig, “An Examination of the Orthodoxy of the Variants in Light of Bart Ehrman’s The Orthodox Corruption of Scripture” (Th.M. thesis, Dallas Theological Seminary, 2000), 51-62. Cf. also Ehrman, Orthodox Corruption, 81.

⁸⁵ Certain texts (John 5.44; 17.3; Rom 16.27; 1 Tim 1.17; Jude 25) do not legitimately belong here since they all use μόνος and not μονογενής.

⁸⁶ See BDAG 556-57 and L&N 34.18.

⁸⁷ See, for example, Larry W. Hurtado, “The Origin of the Nomina Sacra: A Proposal” JBL 117/4 (1998): 655-673; Dirk Jongkind, “Scribal Habits of Codex Sinaiticus,” in Text and Studies: Contributions to Biblical and Patristic Literature 3/5 (ed. D. C. Parker and D. G. K. Taylor; Piscataway, NJ: Gorgias, 2007), 62-84.

⁸⁸ Hurtado, “The Origin of the Nomina Sacra,” 655 and 657.

⁸⁹ E.g., it^q Ambrose^1/11 Irenaeus^{lat 1/3} Origen^pt

⁹⁰ I consider ὁ μονογενής so poorly attested externally (vg^ms Ambrose Aphrahat Cyril of Jerusalem Diatessaron Ephraem Jacob of Nisibis Nestorius Ps-Athanasius Ps-Ignatius Ps-Vigilius^1/2 Victorinus-Rome) and too easily explainable transmissionally to necessitate the reverse hypothesis of starting with it.

⁹¹ For similar conclusions see Jack Finegan, Encountering New Testament Manuscripts: A Working Introduction to Textual Criticism (Grand Rapids: Eerdmans, 1974).

⁹² The deity of Christ is not jeopardized even if υἱός is original. Although that shouldn’t be a factor in the discipline of textual criticism, this does allow certain (evangelical?) textual critics to follow the evidence to a relatively objective conclusion. On the other hand, Ehrman does have a tremendous problem if θεός ends up being the best reading because it would contradict his overall thesis and would put a major dent in his a priori assumption that Jesus is not called θεός in the NT. For example, Ehrman specifically states that if μονογενὴς θεός is the original text in John 1.18 then “the complete deity of Christ is affirmed” (Ehrman, Orthodox Corruption, 78). Yet this is a theological belief he does not support at this time.

⁹³ ἀπεκρίθη αὐτῷ Ναθαναήλ· ῥαββί, σὺ εἶ ὁ υἱὸς τοῦ θεοῦ, σὺ βασιλεὺς εἶ τοῦ Ἰσραήλ.

⁹⁴ N. T. Wright, The Resurrection of the Son of God (Minneapolis: Fortress, 2003), 664.

⁹⁵ For example, some have felt that Jesus allowed this statement in order not to “ruin the moment.” Yet Jesus quotes Deut 6:13, “You are to worship the Lord your God and serve only him,” in Matt 4:10 and Luke 4:8. Therefore, his teachings and convictions seem to strongly negate this option. For the most plausible interpretative options of Thomas’s confession, see Sadananda, The Johannine Exegesis of God, 11-44. Cf. Bruce M. Metzger, “Jehovah’s Witnesses and Jesus Christ,” Theology Today 10 (1953): 65-85, esp 71.

⁹⁶ Cf. Frederick H. A. Scrivener, Bezae Codex Cantabrigiensis (London: Bell and Daldy, 1864), 156; IGNTP, Gospel According to St. John Volume Two, 541.

⁹⁷ Harris lists four solid reasons in Jesus as God (109). Though I believe his third reason can be stated much stronger since ὁ κύριος is never used of God the Father in John’s Gospel except in two OT quotations (12.13, 38).

⁹⁸ Aland-Aland, Text of the New Testament, 110.

⁹⁹ “By actual count, there is a parsimonious use of the article in D; in fact, this situation obtains in each book except Luke” (James D. Yoder, “The Language of the Greek Variants of Codex Bezae,” NovT 3 [1959]: 245).

¹⁰⁰ Wallace, Sharp’s Canon, 132. For brief discussion of personal pronouns within these constructions (e.g., μου following θεός), see footnote 144. In addition, regarding ‘my God’, Schnabel points out that “according to Aristotle, ‘It would be an absurdity to profess a friend’s affection for Zeus’ (φιλεῖν τὸν Δία [Mag. mor. 1208 b 30]). The exclamation of Euripides’ choir upon seeing the sculptures at the temple in Delphi, ‘I see Pallas, my own goddess’ (Euripides, Ion 211), is one of the very few references in Greek literature that uses the phrase ‘my God.’ Burkert [“‘Mein Gott?’ Persönliche Frömmigkeit und unverfügbare Götter,” in Geschichte – Tradition – Reflexion: Festschrift für Martin Hengel zum 70 (Tübingen: Mohr Siebeck, 1996), 3-14] points out, however, that this exclamation must be understood as the ‘aesthetic wow-experience’ of a collective. There is very little archaeological evidence for Greek cults of a ‘personal’ god who was interested in or connected with the individual person. In the context of the Anatolian cult of Men, dedications refer to, for example, ‘the Men of Artemidoros’ (CMRDM III 67-70), but such formulations do not imply a particular ‘pact’ with the god. In a Greek polis the gods of the city were important, not the god worshiped by the individual. In everyday life people established contacts with gods only when needed. Burkert concludes that ‘insofar a person fulfills his religious obligations, there remained normally a realm of freedom, of the ὅσιον [‘profane’], in which religious concerns vanished. This would be contradicted by a unique or comprehensive obligation or affiliation. Resort to the gods becomes important in a time of need, however. . . . The pious person was prepared for being saved, but he does not have a revelation on a document and no treaty with ‘his’ god. Gods are not at his disposal” (Eckhard J. Schnabel, Early Christian Mission: Jesus and the Twelve, vol. 1 [Downers Grove, IL: InterVarsity, 2004 ], 615).

¹⁰¹ Ibid.

¹⁰² Moreover, these two OT expressions are frequently juxtaposed when referring to Israel’s one God. Waltke further defines, “The distinct meaning of these two names [the appellative Elohim and Yahweh] is widely recognized: whereas the title Elohim contrasts God with man in their natures, the name Yahweh presents God as entering into a personal relationship with man and revealing Himself to him” (“The Book of Proverbs and Old Testament Theology” BSac 544 [1979]: 305). Mastin also notes, “John’s ‘My Lord and My God’ directed to Jesus reflects the LXX, where it represents <yhla hwhy and similar expressions, but also makes contact with an expression fairly common in pagan religion” (B. A. Mastin, “Theos in the Christology of John: A Neglected Feature of the Christology of the Fourth Gospel,” NTS 22 [1975-76]: 32-51, esp. 37-41). Cf. G. Deissmann, Light from the Ancient East (New York: George H. Doran, 1927) 366-67; Barrett, Gospel according to St. John (Philadelphia, PA: Westminster, 1978), 572-73; H. D. Betz, Lukian von Samosata und das Neue Testament (Leiden: Brill, 1961), 102.

¹⁰³ For a better understanding of the individual witnesses for the Book of Acts, see Joseph A. Fitzmyer, The Acts of the Apostles: A New Translation with Introduction and Commentary (New York: Doubleday, 1998), 66-79; C. K. Barrett, A Critical and Exegetical Commentary on The Acts of the Apostles (New York: T&T Clark, 2004), 2-29; Darrell L. Bock, Acts (Grand Rapids: Baker, 2007), 28-29.

¹⁰⁴ Metzger, Textual Commentary, 425-27. Cf. Pervo, Acts, 523.

¹⁰⁵ Aejmelaeus, Die Rezeption der Paulusbriefe in der Miletrede: Apg 20:18-35 (Helsinki: Suomalainen Tiedeakatemia, 1987), 132-142. Cf. Pervo, Acts, 524.

¹⁰⁶ Adolf Harnack, History of Dogma (London: Constable, [ca. 1900]: reprinted New York: Dover, 1961), 1.187 n.1. For further discussion on the imagery of “the blood” in the history of the church, as related to the work and person of Christ as God, see Jaroslav Pelikan, Acts (Grand Rapids: Brazos Press, 2005), 221-22. Cf. also Charles F. Devine, “The ‘Blood of God’ in Acts 20:28,” CBQ 9 (1947): 381-408.

¹⁰⁷ Metzger, Textual Commentary, 427.

¹⁰⁸ Also, “εκκλησια του θεου occurs eleven times in Paul; εκκλησια κυριου occurs seven times in the LXX but never without v.l. θεου in the New Testament” (J. Keith Elliott, “An Eclectic Textual Study of the Book of Acts,” in The Book of Acts as Church History: Text, Textual Traditions and Ancient Interpretations, ed. Tobias Nicklas and Michael Tilly [New York: Walter de Gruyter, 2003], 29).

¹⁰⁹ “The text [Acts 20.28] caused such puzzlement (God’s own blood?) that some of the scribes responsible for making copies of Luke’s book evidently attempted to improve or clarify it—particularly by reading ‘the church of the Lord, which he obtained through his own blood’ (cf. Heb. 9.12)” (James D. G. Dunn, The Acts of the Apostles [London: Epworth, 1996], 272.

¹¹⁰ The decision ultimately comes down to one’s understanding and interpretation of the phrase διὰ τοῦ αἵματος τοῦ ἰδίου: “with the blood of his own [Son]” or “with his own blood.” For more sources, a better understanding of these phrases, or both, see Harris, Jesus as God, 131-41. Cf. “blood of God” as used in the Apostolic Fathers: Ign. Eph. 1.1; Ign. Rom. 6.3 (cf. Tertullian [sanguine dei; Ad uxor. 2.3.1]).

¹¹¹ Two primary reasons encouraged me to include Gal 2.20 in this study: (1) most standard works on this topic include this passage (e.g., Raymond Brown, “Does the New Testament Call Jesus God?,” Theological Studies 26 (1965): 545-73; Harris, Jesus as God, 259-61; A. W. Wainwright, “The Confession ‘Jesus as God’ in the New Testament,” Scottish Journal of Theology 10 (1957): 274-299) and (2) it is possible (though I do not think highly probable) to translate two of the textual variants as either “God even Christ” (θεοῦ καὶ Χριστοῦ) or “God the Son” (τοῦ θεοῦ τοῦ υἱοῦ) (see, for example, Ehrman, OCS, 86).

¹¹² Meaning, “Although Paul certainly has an elevated Christology, explicit ‘God-talk’ with reference to Jesus does not normally appear until the later books” (New Testament: New English Translation, Novum Testamentum Graece [Dallas: NET Bible, 2004], 860).

¹¹³ Ibid.

¹¹⁴ Cf. Philip W. Comfort and David P. Barrett, The Text of the Earliest New Testament Greek Manuscripts: New and Complete Transcriptions with Photographs (Wheaton, IL: Tyndale House, 2001), esp. 27-9; E. C. Colwell, “Method in Establishing the Nature of Text-Types of New Testament Manuscripts,” in Studies in Methodology in Textual Criticism of the New Testament (vol. 9 of New Testament Tools and Studies, ed. Metzger, Leiden: Brill, 1969), 45-55, esp. 48.

¹¹⁵ Metzger, Textual Commentary, 524. For other scribal possibilities see Ernest C. Colwell, Studies in Methodology in Textual Criticism of the New Testament (Grand Rapids: Eerdmans, 1969), 106-24. Cf. J. R. Royse, “Scribal Tendencies in the Transmission of the Text of the New Testament,” The Text of the New Testament in Contemporary Research, 239-252; J. R. Royse, “The Treatment of Scribal Leaps in Metzger’s Textual Commentary,” NTS 29 (1983) 539-51; Metzger and Ehrman, Text of the New Testament (Oxford: OUP, 2005), 250-71.

¹¹⁶ Douglas Moo, The Epistle to the Romans (Grand Rapids: Eerdmans, 1996), 287. Cf. Moo, Colossians and Philemon, 84 n. 13. The question remains, however, whether Paul, anywhere in his writings, speaks of both God and Christ Jesus together as the object of faith, which is the case in Gal 2.20 if the authentic reading is θεοῦ καὶ Χριστοῦ.

¹¹⁷ Contra Ehrman, “. . . neither of the other expressions (“God even Christ,” “God the Son”) occurs in this way in Paul” (Ehrman, Orthodox Corruption, 86). One should also note that the position of the pronoun does not affect the sense. On the other hand, 1 Tim 5.21 is surely not ascribing the title θεός to the chosen angels as well by adding καὶ τῶν ἐκλεκτῶν ἀγγέλων after θεοῦ καὶ Χριστοῦ Ἰησοῦ.

¹¹⁸ Paul seems to adhere to a bidirectional life for the believer with the two foci being God and Christ.

¹¹⁹ Although Hebrews’ author is anonymous, the author was at least a male (11.32) contemporary of the Apostle Paul’s protégé Timothy (Heb 13:23); placing Hebrews in the first century.

¹²⁰ There are two other variants in this verse that do not need further discussion here (the omission of the conjunction καί and the word order of ἡ ῥάβδος τῆς εὐθύτητος). The second one in no way affects our question of whether Jesus is explicitly called θεός and the first one, according to Metzger and others, would only slightly reduce the difficulty of the last variant if it were to read αὐτοῦ. Still, for clarity’s sake, I feel confident that these two variants together should read καὶ ἡ ῥάβδος τῆς εὐθύτητος (maintaining the καί and subsequent word order).

¹²¹ “Grammatically, no valid objection may be raised against these renderings, but conceptually they are harsh. . . . To render ὁ θρόνος σου ὁ θεὸς by ‘Your throne is God’ is implausible in light of the articular θεὸς . . . No more probable is the translation ‘God is your throne’” (Harris, “Elohim in Psalm 45,” 72). Even more, though, nowhere else is the phrase “God is your throne” ever used. The expression, according to T. K. Cheyne, is not “consistent with the religion of the psalmists” (The Book of Psalms: A New Translation with Commentary [London: Kegan Paul, Trench, Trubner and Co., 1888], 127). Cf. Peter Craigie, Psalms 1-50 (Waco, TX: Word, 1983), 336-37. For opposing view, see K. J. Thomas, “The Old Testament Citations in the Epistle to the Hebrews,” NTS 11 (1965): 303-325, esp 305, and A. Nairne, The Epistle to the Hebrews (Cambridge: Cambridge University Press, 1917), 31-34.

¹²² For example, Mitchell claims, “The predicate nominative is preferred here to the nominative as a vocative, so that God is not directly addressing the son as ‘God’” (Alan C. Mitchell, Hebrews [ed. Daniel J. Harrington; Collegeville, MN: Liturgical Press, 2007], 49). Harris lists two other commentators, Hort and Nairne, who hold to this view (Harris, Jesus as God, 212). On the other hand, though, Wallace points out, “As to which of these two options is better [subject or predicate nominative], we have already argued that with two articular nouns, the first in order is the subject . . . . Hence, ὁ θρόνος σου would be the subject rather than ὁ θεός (contra most NT scholars who opt for either of these views)” (Wallace, Exegetical Syntax, 59).

¹²³ Ehrman, Orthodox Corruption, 265.

¹²⁴ Ibid., 265.

¹²⁵ Harold W. Attridge, Hebrews, 58. Cf. Cheyne, Psalms, 127.

¹²⁶ See, for example, Murray Harris, “The Translation and Significance of Ὁ ΘΕΟΣ in Hebrews 1:8-9,” TynBul 36 (1985): 129-162.

¹²⁷ Murray Harris, “The Translation of Elohim in Psalm 45,” TynBul 35 (1984): 77-78. Cf. Gert J. Steyn, “The Vorlage of Psalm 45:6-7 (44:7-8) in Hebrews 1:8-9,” HTS 60/3 (2004), 1088.

¹²⁸ Chrys C. Caragounis, The Development of Greek and the New Testament: Morphology, Syntax, Phonology, and Textual Transmission (Grand Rapids: Baker, 2006), 141-3.

¹²⁹ Idem., “The Use of the Nominative ὁ Θεός as Vocative in the Septuagint and the New Testament,” in A Festschrift for John Galanis (forthcoming, 2009).

¹³⁰ Wallace, Exegetical Syntax, 59. Similiarly, F. F. Bruce says, “Whatever be said of the force of δέ in v. 6, there is no doubt about its strongly adversative force here, where it harks back to μέν in v. 7 (καὶ πρὸς μὲν τοὺς ἀγγέλους...πρὸς δὲ τὸν υἱόν)” (The Epistle to the Hebrews, Rev. ed. [Grand Rapids: Eerdmans, 1990], 59).

¹³¹ Attridge, Hebrews, 57.

¹³² Ibid. Cf. George H. Guthrie, “Hebrews,” in Commentary on the New Testament Use of the Old Testament, ed. G. K. Beale and D. A. Carson (Grand Rapids: Baker, 2007), 936; Wallace, Exegetical Syntax, 59; et al.

¹³³ Ibid., 59.

¹³⁴ This assessment was kept even after recognizing that the combination of P46a B “has the original reading in eleven other cases of minority readings in Hebrews” (Harris, Jesus as God, 210). For detailed understanding of the MSS for Hebrews, see Attridge, Hebrews, 31-32. Cf. Beare, “The Text of the Epistle to the Hebrews in P⁴⁶,” JBL 63 (1944): 379-96; Ceslas Spicq, L’Épître aux Hébreux, 3rd ed. (Paris: Gabalda, 1952), 1:412-32.

¹³⁵ Ehrman, Orthodox Corruption, 265.

¹³⁶ NA27 and Tischendorf differ on 2 Pet 1.1 regarding a. Nevertheless, after personally checking a high-resolution digital photograph posted online by the Center for the Study of New Testament Manuscripts (online: http://csntm.org/Manuscripts/GA%2001/GA01_122a.jpg; accessed: 25 September 2008), I determined that the NA27 is correct. In other words, a attests to κυρίου. Cf. also Novum Testamentum Graecum: Editio Critica Maior, vol. 4: Catholic Letters, ed. by Barbara Aland, Kurt Aland, Gerd Mink, and Klaus Wachtel; Part 1: Text; Installment 2: The Letters of Peter (Stuttgart: Deutsche Bibelgesellschaft, 2000), 204.

¹³⁷ These forms are reflecting their textual inflection in the verse since that is the way they would have been written (i.e., the last letter changes to reflect the form).

¹³⁸ Hurtado, “The Origin of the Nomina Sacra,” 655, 657.

¹³⁹ The Granville Sharp Rule does not include proper names and thus 2 Pet 1.2 does not fit the rule (“Jesus” and “Lord Jesus Christ” are both proper names). Cf. Wallace, Sharp’s Canon. See also B. Weiss on the use of the article with the name of God, “Der Gebrauch des Artikels bei den Gottesnamen,” TSK 84 (1911): 319-92, 503-38.

¹⁴⁰ See, e.g., Thomas R. Schreiner, 1, 2 Peter, Jude (Nashville: Broadman and Holman, 2003), 287.

¹⁴¹ See Tord Fornberg, “An Early Church in a Pluralistic Society: A Study of 2 Peter” (Ph.D., Uppsala University, 1977), 143. Cf. Michael Amaladoss, Making All Things New (Maryknoll: Orbis, 1990).

¹⁴² In addition, although 1.1 is the only explicit place Jesus is called θεός in 2 Peter, “other things 2 Peter says about Jesus more or less imply this same understanding. One of the clearest instances is 1,3 where the author of 2 Peter speaks of τῆς θείας δυνάμεως αὐτοῦ, and the antecedent of αὐτοῦ is probably Jesus, the last named substantive (in v. 2)” (Terrance Callan, “The Christology of the Second Letter of Peter” [Biblica 82 (2001): 253-263], 253).

¹⁴³ Furthermore, “The construction occurs elsewhere in 2 Peter [cf. 1.11 and 2.20], strongly suggesting that the author’s idiom was the same as the rest of the NT authors’” (New English Translation, 608). Of course, as some scholars note, one can hardly overlook the significance of the personal pronoun ἡμῶν added to θεός in 2 Pet 1.1 (arguably disrupting the Granville Sharp construction). Yet after exhaustively examining 2 Peter, the NT, and non-biblical papyri, Wallace states, “In all such instances the possessive pronoun had no effect on breaking the construction. The fact, then, that a possessive pronoun attached only to the first substantive never nullifies Sharp’s principle—either in 2 Peter or in the NT or in the papyri that I have examined—is strong confirmation of the validity of the rule in 2 Pet 1:1. In this case, as always, presumption must give way to evidence” (Wallace, Sharp’s Canon, 266).

¹⁴⁴ Michael Green, 2 Peter and Jude (Grand Rapids: Eerdmans, 1993), 69.

¹⁴⁵ D. Edmond Hiebert, Second Peter and Jude: An Expositional Commentary (Greenville, SC: Unusual Publications, 1989), 37.

¹⁴⁶ I should also note that an argument based on the NT’s usage or non-usage of the title θεός for Jesus is different from the claim that the NT authors were so entrenched with Jewish monotheism that they could not have thought of Jesus as θεός. Such a claim assumes that they could not reconcile two truths or break away from their prior presuppositions. Even though they may use “contradictory” terminology, they appear to believe in the divinity of Jesus, sometimes even in preexistent categories (e.g., 1 Cor 8.6; Col 1.15-17; Phil 2.6-11).

¹⁴⁷ Murray Harris, “Titus 2:13 and the Deity of Christ” in Pauline Studies: Essays Presented to FF Bruce, ed. Donald A. Hagner and Murray J. Harris (Grand Rapids: Eerdmans, 1980), 271.

¹⁴⁸ A conceptual fallacy exists for any scholar to reject every possible text to show that the original author(s) did not support this concept. Nevertheless, I feel the answer to this question will inevitably boil down to the presuppositions of each scholar (cf. Robert H. Stein, Jesus the Messiah: A Survey of the Life of Christ [Downers Grove, IL: InterVarsity, 1996], 17).

¹⁴⁹ This second chart reveals my level of certainty whether the passage explicitly refers to Jesus as θεός.

Jesus as Θεός (God): A Textual Examination