The Epistle to the Greeks

They dwell in their own countries, but as sojourners; they partake of all things as citizens, and endure all things as strangers; every foreign land is their country, and every country a foreign land. (Greeks 5:5)

Known in scholarship as Πρὸς Διόγνητον (Pros Diognēton, "To Diognetus").

🔗The Title

Scholarship has always called this text "The Epistle to Diognetus," after the named recipient. The UPDV calls it "Greeks" — after the audience.

The reason for the scholarly title is not recorded. The sole known manuscript witness, the Codex Argentoratensis (Strasbourg), was a collection of works attributed to Justin Martyr and already contained two separate works titled Πρὸς Ἕλληνας ("To the Greeks"): Justin's Discourse to the Greeks and his Exhortation to the Greeks; the anonymous epistle was copied among them as the fifth item, under Justin's name, with the heading Πρὸς Διόγνητον — which matches the letter's own opening address to "most excellent Diognetus." Any further motive (a scribe avoiding a title collision) is conjecture.¹

The UPDV includes none of Justin's works. The collision does not exist in this context. The natural title is what the letter actually is: an epistle to the Greeks. This follows the same naming convention as Hebrews — an anonymous author whose work is identified by its audience rather than its recipient. The author himself confirms this reading: the UPDV's plural-you markers (⁺) reveal that the idol critique in chapter 2 addresses Greek pagans as a class (second person plural verbs throughout), not Diognetus alone.

🔗Overview

The Epistle to the Greeks is an early Christian apology, a defense of the faith addressed to a pagan inquirer, known in scholarship as the Epistle to Diognetus. The author, who is anonymous, responds to questions from a high-ranking Greek named Diognetus: What God do Christians worship? Why do they reject both Greek idolatry and Jewish ritual? Why did this new movement appear when it did, and not before?

The epistle is remarkable both for what it says and what it does not. The author never uses the names "Jesus" or "Christ." The word "Christians" occurs fourteen times (Lienhard's count), but the person at the center of the faith is never named — only described: "his Child," "his Son," "a Savior," "the Lord," "king," "God." Lienhard calls this deliberate restraint; Meecham attributes it to the author's "desire not to irritate his pagan interrogator by protruding peculiarities of the Christian faith which might prove uncongenial or incredible."²³ The Holy Spirit is also entirely absent from chapters 1–10 — not mentioned, not hinted at, not implied. The author presents a complete theology (preexistence, creation, incarnation, atonement, future judgment) structured entirely around the Father and the Son.

The literary quality is extraordinary. Chapters 5 and 6 contain one of the earliest and most striking portraits of the Christian community in the Roman world. The extended analogy, that Christians are to the world what the soul is to the body, has no parallel in the New Testament. Chapter 9 contains one of the earliest articulations of substitutionary atonement outside the apostolic writings: "O sweet exchange! O unsearchable workmanship! O unexpected benefits! That the iniquity of many should be hidden in one righteous; that the righteousness of one should justify many lawless!" (Greeks 9:5).

For a deeper exploration of these ideas (the unnamed Christ, the soul-of-the-world analogy, the sweet exchange, and more) see Reading the Epistle to the Greeks: A Guide to the Text.

🔗Why It Is Included

The Epistle to the Greeks is included in the UPDV as scripture: a faithful early text preserving the beliefs and life of the Christian community in the period immediately following the apostles. It makes no claims of apostolic authorship, engages in no legendary embellishment, and contains no theological innovations incompatible with the New Testament writings. Its theology is grounded entirely in the apostolic tradition — specifically the Pauline epistles and Johannine theology.

The epistle was evaluated using a three-pass review process applied to all candidate texts from the post-apostolic period: a straight reading for coherence and theological content, a systematic search for red flags (pseudepigraphy, legendary material, anti-Judaism, Gnosticism), and an assessment of whether the text contributes something not already present in the canon. The epistle passed all three tests. Its portrait of everyday Christian life in the pagan world and its articulation of the faith to an outsider are without parallel in the existing UPDV texts.

Other early Christian texts were evaluated by the same process and did not advance. First Clement is church administrative correspondence about a specific leadership dispute in Corinth — historically valuable but not theologically generative. The Epistle of Barnabas relies on pseudepigraphic attribution. The Shepherd of Hermas is visionary fiction composed significantly later than the apostolic period. The Didache, initially the strongest candidate on historical pedigree, was rejected on close reading for mandatory fasting schedules, baptismal technique requirements, and a triadic baptismal formula reflecting a later era of institutional liturgical standardization rather than the earliest apostolic practice. Greeks is the only text from this period that passed all three tests: it is theologically faithful, free of red flags, and fills a genuine gap in the canon.

🔗Date and Authorship

The date of composition is debated, but the epistle is most plausibly placed in the first half of the second century. Meecham reports a date around 150 CE, following Lightfoot, Bardenhewer, and Krüger, based on the epistle's theological vocabulary and its relationship to other second-century apologetic works.² Lienhard treats the author's lack of a stand on even primitive theological questions as "an argument for an early date," while explicitly declining to press for a particular year or decade; he separately considers the Christology less developed than later patristic formulations.³ Yet the epistle's theology is substantive. The author teaches that the Son existed before creation, made all things, was sent by the Father, died as a ransom for sinners, and will return as judge. He declares that God can be seen "through faith, through which alone it is granted to see God" (8:6). These themes are all found in the New Testament. The sharpest linguistic dating tool is the author's use of two distinct Christological titles in deliberate alternation: παῖς (pais, "Child") in 8:9–9:1, where the Son shares in the Father's eternal counsel, and υἱός (huios, "Son") from 9:2 onward, where he is actively sent as ransom. The Theological Dictionary of the New Testament documents that παῖς θεοῦ for Christ survived after the New Testament only as a fixed liturgical formula — almost always the stereotyped phrase "through Jesus your Child" in prayers.²⁰ By the time of Ignatius (~110 CE), υἱός had become the dominant title; by Justin (~150 CE), παῖς had virtually disappeared from free theological prose, surviving primarily in liturgy and direct Old Testament exegesis. Yet the author of Greeks deploys both terms in free theological argument, not liturgy — placing him in the narrow window where the transition from παῖς to υἱός was still live. No other second-century text combines both titles in a single argument with distinct functions. This pattern, first identified by Meecham (1949) and formalized by Lienhard (1970), anchors the epistle firmly within the ~130–150 CE range.²²

The epistle's vocabulary confirms this window. The compound παντοκτίστης (pantoktistes, "creator of all") in 7:2 is a hapax within the surveyed second-century evidence: BDAG and Lampe's Patristic Greek Lexicon cite this verse alone, the word has no entry in Liddell-Scott-Jones, and searches of the Perseus Digital Library locate no other second-century occurrence. Located reuse begins only after a gap of half a millennium: Andrew of Crete (~660–740 CE) applies the word to God in Marian homilies on the nativity and the dormition, and Peter of Sicily (ninth century) follows in the same Marian register. A search of the Patrologia Graeca returns no further witness: the epistle's own text, printed at PG 2, cols. 1175–1176, is the word's only occurrence in that volume. No reuse earlier than Andrew was located in the searched corpora.²³ Three major lexicons either omit the word entirely or cite only Greeks — an apparent coinage that no located writer adopts for five hundred years.

The letter is a formal protreptic, an exhortation to adopt a way of life, a philosophical genre going back to Aristotle's lost Protrepticus. The five-part structure (address, refutation of rival positions, positive exposition, theological grounding, exhortation to conversion) is followed with precision. The author's prose style is Asianic, not Atticist: sustained antithetical parallelism, balanced clauses with matching verbal endings, exclamatory sequences, and short cola strung in chains — long periodic structures, Blomqvist finds, are virtually absent — the hallmarks of the rhetorical tradition associated with the Roman province of Asia (western Asia Minor).¹¹ The paradox catalogue in chapter 5 and the soul-body analogy in chapter 6 maintain formal structural discipline across eight consecutive verses, a density of rhetorical figuration that exceeds anything in contemporary Christian literature. The closest stylistic parallel is Melito of Sardis (On the Passover, ~170 CE), who shares the Asianic profile.¹² Notably absent are all markers of the Atticist revival that dominated the contemporary Second Sophistic movement: the epistle contains zero oblique optatives, zero wish optatives, zero purpose optatives, and zero dual forms — every feature an Atticist would deploy to signal classical education. The author's ten optatives are all Koine-survival forms (potential with ἄν, polite softeners), never the revivalist constructions an Atticist would choose. His vocabulary is broadly that of Hellenistic philosophical Greek; yet several expressions have close Septuagint comparanda — the creator-of-all wording at 7:2 (2 Maccabees 1:24; Sirach 24:8), the maker of heaven and earth at 3:4 (Exodus 20:11; Psalm 145:6 LXX), 7:6 (Malachi 3:2), and 10:2 (Genesis 1:26). The profile points to a secular rhetorical education; it does not prove one, since the absence of explicit quotation cannot by itself measure the author's scriptural knowledge.

However, the theology and the language may not originate from the same moment. The complete absence of the Holy Spirit, the lack of any Trinitarian formula, the absence of church structure (no bishops, elders, deacons, sacraments, or creed), and the raw Pauline atonement theology of chapter 9 (unprocessed by the Irenaean "Recapitulation" framework already emerging by ~150 CE) all point to an earlier theological tradition, perhaps as early as ~70–100 CE. The author knows Paul's letters (2 Corinthians 6 is echoed in chapter 5; Philippians citizenship language appears in chapters 5 and 10) and Johannine theology (Father-Son sending language throughout chapters 7–9). The epistle nowhere recounts Pentecost, Paul's conversion, or the missionary journeys; but these silences do not establish ignorance of Acts — Meecham proposes Paul's Lystra speech in Acts 14:15 as the immediate source of 3:4.² The παῖς/υἱός alternation reinforces this profile. While standard reference works like the Theological Dictionary of the New Testament catalog later liturgical uses of παῖς, they omit Greeks entirely.²⁰ The author's non-liturgical, functional deployment of the term may preserve a Christology prior to its fossilization into pure liturgy, a transitional usage sitting in a gap that the standard reference works have not yet mapped.²¹

A systematic vocabulary mapping of chapters 7–9 confirms this profile quantitatively. Every theological term in the epistle's densest section was classified by chronological distribution — whether it belongs to pre-Christian Greek, first-century Christian usage, transitional vocabulary, or the characteristic lexicon of the mid-second-century Apologists. The result: 32% of the theological vocabulary in the soteriological core (8:5–9:6) aligns with early Christian usage absent or exceptionally rare among the surveyed Apologists. Terms like λύτρον (lutron, "ransom"), μακρόθυμος (makrothymos, "long-suffering"), and φανερόω (phaneroō, "reveal") peak in the New Testament and Apostolic Fathers and are absent or exceptionally rare in the extant apologetic corpus; their later citation histories differ by term. Not a single term belongs to the apologetic vocabulary characteristic of Justin, Theophilus, or Athenagoras. The theological lexicon of the epistle is not natively aligned with the extant mid-second-century apologetic idiom; it preserves the soteriological vocabulary of an earlier era. No baseline has been computed for what percentage a comparable second-century text would score under the same classification; until one is, the 32% figure should be read as descriptive of this passage rather than as distinguishing it from its contemporaries. For the full study, see Theological Vocabulary of the Epistle to the Greeks.²⁴

This suggests the author may be a trained second-century Greek rhetor faithfully transmitting a theological tradition one or two generations older than his prose style — old theology in new clothes. This is a tradition-historical hypothesis — Traditionsgeschichte, the modern scholarly method for tracing the oral and written traditions behind a composition, applied here to the epistle's own genesis — with direct parallels in the Didache (second-century Greek redaction wrapping first-century Jewish-Christian tradition) and the Odes of Solomon (later manuscripts preserving deeply Johannine, pre-institutional theology from an early Syrian milieu).

The author is anonymous. The text was traditionally grouped with the Apostolic Fathers (a designation used for early Christian writers thought to have had direct contact with the apostles) though this classification is now recognized as imprecise. The epistle was not attributed to any known author in antiquity. The Latin preface to the epistle in Migne's Patrologia Graeca vol. 2 argues for Apollos as the most likely candidate (an Alexandrian, eloquent, and a contemporary of Paul); the epistle's lack of any explicit mention or direct quotation of the Old Testament sits awkwardly with the early tradition's portrait of Apollos as scripturally learned — though silence about explicit citation is thin evidence either way.¹ Quarry (1896) proposed Hippolytus of Rome on stylistic and theological parallels — a probability, as he stressed, not a demonstrated identification — and Connolly (1935) later made a separate, more tightly delimited case for the same hypothesis; neither has gained wide acceptance.⁴

The identity of Diognetus himself is unknown. The name Διόγνητος (Diognētos, "God-born") was a common upper-class Greek theophoric name, attested in Aeschines, Demosthenes, and Plutarch across the fifth through second centuries BCE. One prominent later bearer was Marcus Aurelius's painting teacher (mentioned in the Meditations 1.6), but there is no evidence connecting him to this epistle.

🔗The Manuscript

The Epistle to the Greeks survives from a single manuscript, the Codex Argentoratensis (Strasbourg), a thirteenth- or fourteenth-century codex written on charta gossypina (cotton paper, as the early cataloguers described it), with 260 leaves by a later numbering. The codex gathered five works under Justin Martyr's name — among them this anonymous epistle, the fifth item in the list — but not Justin's complete corpus: the two Apologies and the Dialogue are absent. Athenagoras's Plea and On the Resurrection follow in the same early hand; the sibylline verses and oracle collections were added by a later hand.¹ It was destroyed by fire on August 24, 1870, during the Prussian bombardment of Strasbourg in the Franco-Prussian War — an irreplaceable loss for early Christian studies.

The Excerpta ex præfatione editoris Ienensis in Patrologia Graeca vol. 6 (col. 222, continuing at col. 224) preserves the manuscript's chain of custody, from observations communicated by Eduard Cunitz of Strasbourg. The codex was owned by Johann Reuchlin (1455–1522), who inscribed the wooden cover in his own hand: Liber Graecus Johannis Reuchlin Phorcen. ("Greek book of Johann Reuchlin of Pforzheim"). By 1560 the manuscript was at Maursmünster Abbey in Upper Alsace, and it later reached the Strasbourg public library, as the Jena editor reports, through the violent political upheaval of the French revolutionary era. Mice had gnawed the volume, making their nest in it, a muribus nidum sibi in eo parantibus admodum corrosus est, though only one part suffered damage.¹

The text is preserved through two channels. First, Henri Estienne (Stephanus) published the editio princeps in 1592, working directly from the Strasbourg manuscript. Second, several scholars made independent collations of the manuscript before its destruction. These collations, compared against Estienne's edition, allow modern editors to reconstruct the text with reasonable confidence.⁵

Where the manuscript is corrupt or has gaps, scholarly conjectures are adopted. These are noted in the footnotes. The most significant conjectural restoration occurs at 9:1, where the manuscript text is corrupt and most modern editors follow Lachmann's reconstruction of the opening clause. A lacuna (gap in the text) also exists after 7:6, where an unknown amount of text has been lost.⁶

🔗The Question of Authenticity

The epistle possesses no known patristic citations. It is not quoted by Eusebius, Photius, or any ancient writer, and it survived in only one medieval manuscript. Meecham states plainly that the epistle is "known neither to Eusebius nor to Photius nor indeed to any ancient or medieval writer."² Eusebius's own silence carries little independent weight: in Historia Ecclesiastica 4.18 he catalogs the works of Justin Martyr while admitting to "many other works current among the brethren" that he does not list. A catalog its author calls incomplete cannot date an attribution; its silence about the epistle simply provides no evidence that Eusebius knew the work or the Justin heading under which the Strasbourg manuscript later transmitted it. The text survived in obscurity — preserved by copyists but never discussed by theologians.

In biblical textual criticism, this narrow transmission history naturally invites scrutiny regarding the text's authenticity. The forgery hypothesis has been formally proposed — by Donaldson, who suggested Estienne himself fabricated it but later gave up that "strange hypothesis," and by Overbeck, who published a monograph arguing it was a post-Constantinian fiction (Ueber den pseudo-justinischen Brief an Diognet, 1872). Overbeck's theory was, in Schaff's summary, "refuted by Hilgenfeld, Keim, Lipsius, and Draseke" — the last writing against Overbeck and Donaldson together.⁷

The cumulative internal evidence makes fabrication extraordinarily implausible. The epistle's theology centers entirely on the Father and the Son, with no mention of the Holy Spirit as a person. It contains no reference to baptism, the eucharist, bishops, priests, deacons, church councils, or creedal formulas. These omissions are natural for a second-century text addressed to a pagan outsider; they are psychologically inexplicable for a medieval or Renaissance writer, for whom Trinitarian theology, sacramental practice, and institutional hierarchy were the defining structures of Christian life. The author would have had to deliberately suppress over a thousand years of doctrinal development — and do so with such consistency that the resulting text would fool generations of scholars trained in patristic Greek.

A second line of evidence is chronological. The epistle shares structural parallels with the Apology of Aristides (~147 CE) — both clear the field by critiquing pagan idolatry and Jewish ritual before presenting Christianity as a "third way."⁸ An earlier generation of scholars assumed direct dependency, but both texts may independently draw from a common apologetic catechesis attested in the Kerygma Petrou (Preaching of Peter), a lost work preserved in fragments by Clement of Alexandria, which contains the same tripartite structure and the same checklist of Jewish practices critiqued.^8a The Apology of Aristides was itself completely lost until 1878 (Armenian fragments) and 1889 (full Syriac text discovered by J. Rendel Harris at St. Catherine's monastery). Since Diognetus was published in 1592 and the Strasbourg manuscript was destroyed in 1870, both before Aristides was rediscovered, a forger could not have imitated a text that was not yet available to the modern world. The brief Kerygma fragments preserved by Clement supply a general catechetical outline, but the specific literary convergences between Aristides and Greeks, the detailed chapter-level execution of the tripartite structure, go far beyond what those fragments contain. Whether through direct dependency or a shared early tradition, the deep structural match between the two texts was only identified after both became accessible to modern scholars. This convergence is an accidental proof of antiquity.

Finally, the manuscript itself bears the marks of genuine transmission. The text contains a lacuna after 7:6, corrupt readings at 9:1 requiring conjectural reconstruction, and dozens of passages where scholarly emendation is necessary. A forger does not damage his own work. The physical break between chapters 10 and 11 in the Strasbourg manuscript, with the appended chapters written in a markedly different style, is characteristic of real manuscript traditions in which texts accumulate layers over centuries — not of deliberate fabrication.

For the UPDV, authenticity is determined by internal theological integrity and textual evidence, not by manuscript volume. The UPDV excluded the Book of Acts, John 19:36–21:25, and Luke 1–2, all texts attested by thousands of manuscripts, on internal textual and theological grounds. The same standard applied in reverse permits the inclusion of a thinly attested text when the internal evidence of genuine antiquity and theological fidelity to the apostolic writings is overwhelming.

🔗Exclusion of Chapters 11–12

Only chapters 1 through 10 are included. Chapters 11 and 12, which appear in the manuscript after the epistle proper, differ from it in ways that point to a separate origin. Lightfoot and Meecham treated them as a distinct work by a different author; other editors have held that the same author wrote both sections as separate compositions (Barnard) or that chapters 11–12 belong to the same work (Marrou, Thierry). Blomqvist and Lienhard, who examined the Greek most closely, both declined to draw an authorship conclusion from the stylistic evidence. Computational analysis of particle distribution and function-word density is consistent with separate authorship, though the distance metric does not discriminate at this text length. The case rests on the particle and density evidence, not on Cosine Delta.

Physical break in the manuscript. The Strasbourg manuscript itself contained a physical break between chapters 10 and 11, suggesting that the scribe recognized these as distinct works. Lightfoot, who examined the manuscript tradition closely, treats chapters 11–12 as a separate document appended to the epistle.⁹

Style and vocabulary. The language of chapters 11–12 differs markedly from that of chapters 1–10. Meecham's conclusion: "Chapters 11 and 12 form an appendix by a later hand," and among his grounds is vocabulary — ninety-three words occurring in chapters 11–12 are absent from chapters 1–10, and "in the two appended chapters the proportion of post-classical words is higher (2 percent)."² The author of 1–10 writes concise, antithetical prose; chapters 11–12 shift to an ornate, repetitive, almost homiletic style. The Christological vocabulary shifts with them: in Holmes's Greek of chapters 11–12, παῖς (pais) never occurs, where chapters 8–10 carefully alternate it with υἱός (huios); the Son-title itself survives once (11:5), while λόγος appears eight times. This is the article's own comparison, not Lienhard's: he expressly excluded chapters 11–12 from his study, so the παῖς/υἱός shift cannot be attributed to him.³

Particle analysis. The stylistic difference is measurable at the level of individual Greek particles. Four key discourse particles that appear frequently in chapters 1–10 (γε (ge), μέν (men), οὖν (oun), and ἄν (an)) are completely absent from chapters 11–12. These are not rare words: μέν alone occurs 27 times in chapters 1–10 (12.2 per thousand words), and ἄν occurs 12 times (5.4 per thousand). Together, these four particles appear 50 times in 2,205 words. In a 408-word sample drawn from the same author, one would expect approximately 9 occurrences; the observed count is zero. A Poisson probability test yields p = 9.6 × 10⁻⁵ — the probability of this absence occurring by chance is less than one in ten thousand.¹⁰

This figure assumes the four particles occur independently. They do not: μέν anticipates δέ, and οὖν tracks argumentative structure, so a writer shifting to causal chaining suppresses the class together rather than each word separately. The true probability is higher than the Poisson estimate. The observation (an entire class of discourse particles at zero where roughly nine occurrences were expected) does not depend on the exact figure.

The remaining particles tell a complementary story. While δέ (de) is present in both sections, its rate drops by half (17.7‰ in chapters 1–10 versus 9.8‰ in chapters 11–12). Conversely, γάρ (gar) doubles in frequency (7.3‰ to 14.7‰). This inversion (the near-total loss of contrastive and concessive particles (μέν, δέ, ἄν) alongside the doubling of causal γάρ) suggests a different mode of argument: the tight antithetical rhetoric of chapters 1–10 gives way to the flowing causal reasoning of a homily. Blomqvist noted this particle absence parenthetically while analyzing hiatus avoidance strategies in the epistle: γε and μέν, he observes, "do not appear at all in chs. 11-12." His footnote 21 separately calls the divergent hiatus rate in chapters 11–12 "another mark of the divergent character of these chapters." Blomqvist declined to draw an authorship conclusion from these observations; see note 11.

Function word density. A deeper computational analysis reveals an even more significant divergence. Chapters 1–10 use function words (articles, prepositions, conjunctions, particles, and copular verbs) at a rate of 44.5% of all words. Chapters 11–12 use them at 31.6%, a relative drop of nearly 30%. A two-proportion z-test yields z = 4.83, p = 1.35 × 10⁻⁶ — the probability of this difference occurring by chance is approximately one in a million.¹⁰

The natural objection (that a shift in genre from apologetic argument to homily might explain the drop) can now be tested directly. The Greek text of Melito of Sardis's Peri Pascha (On the Passover, mid-second century), an Asianic rhetorical homily roughly contemporary with the epistle, provides an exact genre control. Melito's function word density is 44.0% — squarely within the range of the apologetic texts (42–45%) and virtually identical to chapters 1–10. A second homiletic control, 2 Clement (the oldest surviving Christian homily (~140–160 CE), a paraenetic rather than rhetorical work) measures 45.9%. Two independent homilies from the same era, representing two different homiletic sub-genres, both maintain normal function word density. The 31.6% rate in chapters 11–12 is not a feature of homiletic Greek.¹²

Particle profile under genre control. The Blomqvist particle test gains additional force from the homiletic controls. If the genre shift from apology to homily were responsible for the disappearance of contrastive particles, one would expect Melito's homily to show a similar suppression. It does not. Melito uses μέν at 4.5 per thousand words, οὖν at 2.3‰, and δέ at 21.3‰ — the last figure actually higher than chapters 1–10 (17.7‰). Similarly, 2 Clement uses οὖν at 12.3‰ and μέν at 1.3‰. Both homilies retain contrastive particles at measurable rates. The combined rate for the four absent particles (γε + μέν + οὖν + ἄν) is 7.7‰ in Melito, 13.6‰ in 2 Clement, and 0.0‰ in chapters 11–12. The complete suppression of an entire class of discourse particles is not a genre phenomenon.¹²

The causal particle γάρ tells the complementary story. Chapters 11–12 use γάρ at 14.7‰ — double the rate of both chapters 1–10 (7.3‰) and Melito (7.9‰). If this were merely a genre shift, Melito's rate should be elevated as well. Instead, the doubling of γάρ alongside the complete absence of μέν/δέ antithesis reveals a different cognitive syntax: the author of chapters 1–10 structures thought through contrast and concession, while the author of chapters 11–12 structures thought through relentless causal chaining, a shift in argumentative habit that the homiletic controls do not reproduce.

Nearest-neighbor analysis. The forensic analysis applied Cosine Delta (z-scored relative frequencies with cosine distance) to a corpus of 53 text chunks drawn from eight early Christian authors: Holmes's critical Greek text of Diognetus, Melito of Sardis (Peri Pascha, Hall critical text), Justin Martyr (1 Apology and Dialogue with Trypho, providing a cross-genre same-author control), Athenagoras, 1 Clement, 2 Clement, Barnabas, and five letters of Ignatius. Feature set: 132 Greek function words. Bootstrap consensus over 1,000 iterations with 70% random feature sampling found that the nearest neighbor to chapters 11–12 is the final section of chapters 1–10 (31.0% of iterations), followed by Melito (23.4%). The remaining 45.6% is scattered across Athenagoras, Ignatius, and Justin. Chapters 11–12 are a single 408-word chunk, below the size where Delta stabilizes, and the scatter reflects this limitation rather than a positive signal of authorial homelessness. For comparison, the five Ignatius letters self-cluster at 100% (12/12 chunks), and the four Diognetus 1–10 chunks not adjacent to the break point cluster with each other at 100%. The nearest-neighbor result is consistent with a short text whose closest match in the corpus is the longer work it was transmitted with, in a metric that does not resolve at this length.¹⁰

A same-author cross-genre control tests the methodology. Justin Martyr's two stylistically different works (forensic apology versus exegetical dialogue) give a Cosine Delta distance of 1.1768 (sd=0.0814, n=25 chunk pairs of 400 words each). The Diognetus 1–10 vs. 11–12 distance is 1.043 — but chapter 11–12 is a single 408-word chunk, below the size where Delta stabilizes, and short texts inflate Delta distances. The observed split is approximately 1.6 standard deviations below the same-author baseline: chapters 11–12 are well within the range a single author produces across genres. Cosine Delta does not discriminate between the authorship hypotheses in this case. The particle evidence (the complete absence of γε, μέν, οὖν, and ἄν from chapters 11–12, against an expected count of approximately nine) and the function-word density gap carry the stylistic case independently of the Delta result. The forensic method follows Eder (2015) on short-text attribution. It is the same general approach used in the UPDV's Luke-Acts analysis, where a Josephus register control found that a single author shifting from contemporary historiography to Septuagintalizing narrative changed his τε rate by only a factor of 1.3. That result concerns a single function-word rate rather than the Cosine Delta distances used here, but it is consistent with the premise that function-word frequencies resist deliberate alteration.¹⁰

The controls above test whether the genre shift from apology to homily can account for the stylistic divergence; they show that it cannot. They do not test the hypothesis advanced by Barnard — that the same author wrote both sections as separate compositions, possibly years apart and in different modes. A within-author, across-works baseline would be needed to rule that out, and the Justin control, which measured that, placed the Diognetus split inside single-author range. The particle evidence is the strongest remaining argument for separate authorship, and it is not genre-dependent; whether it rules out Barnard's hypothesis is a question the present analysis cannot resolve from the available controls.

Date. The authorship conjectures attached to chapters 11–12 cluster in the late second and third centuries: Holmes reports that Hippolytus of Rome (ca. 170–236) is frequently suggested and that Lightfoot hazards Pantaenus (d. ca. 190).⁶ Holmes's own summary of the compilation history is that the two chapters "almost certainly belong to some other work" that was attached to the epistle in the manuscript tradition — a claim about transmission, not about whether the same author wrote both sections.⁶ Meecham, who likewise treats 11–12 as a later appendix, describes chapter 12 as "a little homily."²

🔗Consistency with the Apostolic Witness

The Epistle to the Greeks was never included in any ancient canon list. The Lexham Bible Dictionary notes that "while never included in a canon list, this letter was accepted as authentic in the early church."¹³ On the transmission evidence — no ancient or medieval writer is known to have mentioned the epistle at all — the second half of that sentence has no attested basis and is retained here only to mark how the work is grouped in some modern references.² Green, writing on 2 Peter rather than on this epistle, asks "If early date and orthodox content [were the criteria], it is hard to see why Hermas, the Epistle of Diognetus, the Epistle of Barnabas and the First Epistle of Clement were not included. Yet the very Councils that accepted 2 Peter into the Canon decisively rejected Clement, Barnabas, Diognetus and Hermas."¹⁴ That is Green's rhetorical grouping of texts he treats together as non-canonical; since no ancient or medieval writer is known to mention this epistle, there is no evidence any council actually considered it. The UPDV includes the text on its own standard, which — unlike the canon lists' operative criterion of apostolic authorship — does not require it.

Marshall, in New Testament Theology: Many Witnesses, One Gospel, acknowledges that "there could be a case that some of the second-century writings (e.g., the Epistle of Diognetus) stand very close to [the New Testament writings] in spirit and quality."¹⁵

The epistle is thoroughly familiar with the apostolic writings, and its theology contains no contradictions with them. Ellicott's commentary on 1 John states that "six of [Diognetus's] chapters contain indisputable reminiscences of the First Epistle" of John.¹⁶ Specific parallels include Greeks 10:3 and 1 John 4:9-11 (God's prior love as the ground of human love), and the cosmological theology of chapter 7 echoes both John 1 and Colossians 1:15-17. The Cambridge Greek Testament notes that Greeks 5:9 — ἐπὶ γῆς διατρίβουσιν, ἀλλʼ ἐν οὐρανῷ πολιτεύονται (epi gēs diatribousin, all' en ouranō politeuontai, "they dwell on earth, but have citizenship in heaven") — is "probably" written with Philippians 3:20 in mind.¹⁷ The atonement language of chapter 9 closely parallels 2 Corinthians 5:21 and Romans 5:19.

These parallels are evidence of corroboration, not the basis of its standing; the UPDV includes the text on its own standard (see "The Question of Authenticity" above). The author of Diognetus wrote in the generation after the apostles, and his theology is entirely consistent with the apostolic writings. What the epistle adds is perspective: the view from outside, looking at the Christian community through the eyes of a curious and intelligent pagan. This perspective is absent from every other book in the UPDV.

🔗Placement

The Epistle to the Greeks is placed between the Gospels and the Epistles. This is the position formerly occupied by the Book of Acts, which is not included in the UPDV (see The Book of Acts).

Greeks occupies this position not as a replacement for Acts (it is an apologetic epistle, not a historical narrative) but because this is the natural point in the canon where an introduction to the Christian movement belongs. The Gospels tell the story of Christ. The Epistles address the internal theology and life of the churches. Greeks stands between the two, answering the questions an outsider would ask after reading the Gospels: What is this new movement? Why do its followers reject both Greek religion and Jewish ritual? Why did it appear when it did?

The opening verse of the epistle frames precisely this transition. The author addresses Diognetus as one "exceedingly eager to learn the godliness of the Christians," inquiring "In what God do they trust, and in what way do they worship him, that they all scorn the world and despise death?" (Greeks 1:1). This is the question that naturally arises between the narrative of the Gospels and the theological arguments of the Epistles.

🔗Theology

🔗The Critique of Idolatry and Jewish Ritual (Chapters 2–4)

The epistle opens with a sharp critique of Greek idol worship. The author's approach is distinctive: rather than theological argument, he uses pointed material observation. Are not the idols made of stone, bronze, wood, silver, iron, and clay — the same materials used for common vessels? Could not the same craftsmen reshape them into ordinary household objects? The idols stand condemned in a drumbeat of questions: "Are they not all deaf? All blind? All without life? All without touch? All motionless? All rotting? All decaying?" (Greeks 2:4).

The critique of Jewish ritual follows a different logic. The author does not reject monotheism ("The Jews then, if they abstain from this previously described service, rightly choose to worship the one God over all" (3:2)) but rejects the sacrificial system as a misunderstanding of God's nature. God "who made the heaven and the earth and all things in them, and supplies us all with whatever we need, himself needs none of those things" (3:4). The author similarly dismisses dietary laws, Sabbath observance, circumcision, and calendrical observance as "laughable, and worthy of no account" (4:1).

🔗Christians in the World (Chapters 5–6)

Chapters 5 and 6 are the literary and theological heart of the epistle, and the primary reason for its inclusion in the UPDV. No other early Christian text provides a comparable portrait of the community's self-understanding in relation to the surrounding world.

The description is built entirely on paradox. Christians are "distinguished from the rest of men neither by country, nor by language, nor by customs" (5:1). They "dwell in their own countries, but as sojourners" (5:5). They "are in the flesh, but do not live after the flesh" (5:8). They "obey the public laws, and in their lives go even further than the laws [require]" (5:10). The antithetical structure builds to a comprehensive portrait: poor yet enriching others, persecuted yet multiplying, unknown yet condemned.

One verse in this catalog carries particular weight. "They marry, as do all. They do not throw away what is born, but acknowledge the children" (5:6). In the Roman world, exposure of unwanted newborns was a familiar reality — a private letter from Roman Egypt (P.Oxy. IV 744, 1 BCE) instructs a husband with total casualness that a boy child is to be kept and a girl exposed, a document Meecham cites in his note on this very verse.² The author's Greek embeds this cultural divide in the grammar itself: the verb for bearing children, τεκνογονοῦσιν (teknogonousin), has the word for "child" (τέκνον, teknon) built into it, while the word for what is discarded, τὰ γεννώμενα (ta gennōmena, "what is born"), strips it out, reducing the infant to the bare biological event of birth. And for the act of disposal, the author reaches past the polite Greek word for infant exposure (ἐκτιθέναι, ektithenai) and grabs the most violent word available: ῥίπτω (rhiptō) — throw, hurl. Meecham notes that Otto called it a verbum magis odiosum, a more hateful word, chosen deliberately.²

Chapter 6 extends this into the soul-body analogy. "But to speak simply: what the soul is in the body, Christians are in the world" (6:1). The analogy is sustained through ten parallel statements: the soul is sown through the body, Christians through the cities; the soul dwells in the body but is not of it, Christians dwell in the world but are not of it; the flesh hates the soul, the world hates Christians; the soul holds the body together, Christians hold the world together. Freeman (2022) analyzes the epistle's antitheses and paradoxes as protreptic technique: opposed phrases seize the reader's attention, and the paradoxes draw the reader onward into the substance of Christianity.¹⁸

🔗The Cosmic Christ (Chapters 7–8)

Chapter 7 contains a remarkable theological statement. The author describes God sending his Son into the world — not through any intermediary ("He did not, as one might suppose, send to men some attendant, or angel, or ruler") but through "the very craftsman and builder of all things" (7:2). What follows is a cosmic catalogue: the one "by whom he created the heavens, by whom he shut the sea within its own bounds, whose mysteries all the elements faithfully keep, from whom the sun has received the measures of its daily course to keep, whom the moon obeys" (7:2).

This high Christology is tempered by restraint. God sent his Son "in gentleness and meekness. As a king sending his son, a king, he sent him; sent him as God; sent him as to men; sent him as one saving, as one persuading, not forcing; for violence is not with God" (7:4). As Lienhard observes, the Christology of Diognetus is less developed than later patristic formulations — the author makes no explicit statement of the incarnation, and the relationship between the Son and the cosmic creator is left suggestive rather than dogmatic.³

🔗The Sweet Exchange (Chapter 9)

Chapter 9 contains the epistle's most theologically dense passage. The author describes God's salvific plan in terms of progressive revelation: during the "former time," God allowed humanity to follow its own desires, "Not that he took pleasure in our sins — he bore with them" (9:1). When "our unrighteousness was made complete," God acted: "having mercy, he himself took upon himself our sins, he himself sold his own Son as a ransom for us, the holy for the lawless, the harmless for the evil, the righteous for the unrighteous, the incorruptible for the corruptible, the immortal for the mortal" (9:2).

The passage climaxes in the "sweet exchange" (9:5) — one of the earliest patristic articulations of what later theology would call substitutionary atonement. The language closely parallels Paul's formulation in 2 Corinthians 5:21 ("him who knew no sin he made [to be] sin on our behalf") and Romans 5:19 ("through the obedience of the one, the many will be constituted righteous").

🔗Key Translation Decisions

🔗Child and Son (παῖς and υἱός)

The author of Greeks deliberately alternates between two Greek words for the Son: παῖς (pais, "Child") in 8:9, 8:11, and 9:1, and υἱός (huios, "Son") in 9:2, 9:4, and 10:2. This is not random variation. The distribution is theologically motivated — Meecham's observation, in his words: "Christ as παῖς shares and reveals the Father's plan of salvation (8:9, 11; 9:1); as υἱός he is 'sent' and effects it (9:2, 4; 10:2)"; Lienhard reports the distinction as Meecham's and builds on it.²²³

The UPDV preserves this distinction throughout, rendering παῖς as "Child" and υἱός as "Son." This follows the UPDV's general principle of never rendering παῖς as "Son", a translation that would obscure the author's intentional word choice. The distinction is invisible in most English translations of the epistle, which render both words as "Son."

🔗Word, Not Speech (λόγος in 7:2)

In 7:2, the Greek λόγος (logos) is translated "word" (lowercase), not "Speech." The UPDV reserves the capitalized title "Speech" for the Johannine corpus (John 1:1, 1 John 1:1, Revelation 19:13), where λόγος functions as a recognized Christological title connected to the Targumic Memra tradition (see The Speech in John 1:1).

In Greeks, λόγος more likely refers to the message or revelation sent through the Son, not a title of the Son himself. Lienhard concludes that the Christological use of λόγος in Greeks 7:2 "is doubtful" and that "the author is more probably speaking of the revelation of God's plan than applying a title to Christ."³ The UPDV follows this reading.

🔗Governs (πολιτεύομαι in 10:7)

The Greek πολιτεύομαι (politeuomai) carries three distinct senses in the UPDV, all attested in BDAG: (1) citizenship, the state of being a citizen (Greeks 5:9); (2) governs, political administration when the subject is God (Greeks 10:7); and (3) live as citizens — conduct of life with a civic connotation (Philippians 1:27).

In 10:7, the subject is God: "you will see that God governs (πολιτεύεται) in heaven." Bitner, in his review of Jefford's edition of Diognetus, singles out exactly this verse as a puzzling translation choice: Jefford renders it "reigns" rather than "orders the constitution" or "governs," softening the word's ordinary sense.¹⁹ The UPDV renders this as "governs" to distinguish it from the citizenship sense used earlier in 5:9.

🔗The Manuscript Reading at 10:1

The manuscript reads: Ταύτην καί σύ τήν πίστιν ἐάν ποθήσῃς, καί λάβῃς πρῶτον μέν ἐπίγνωσιν πατρός (tautēn kai su tēn pistin ean pothēsēs, kai labēs prōton men epignōsin patros). Most modern editors, from Gebhardt (1875) to Holmes (2007), have diagnosed this sentence as broken: two subjunctive clauses joined by καί (kai) with no main clause. Holmes prints κατάλαβε (katalabe, "grasp"), an aorist imperative of καταλαμβάνω — replacing λάβῃς (labēs, "receive"), from λαμβάνω, with a completely different verb. Other editors proposed different repairs: Otto changed the mood to a future indicative; Lachmann and Bunsen rewrote both verbs as optatives; Meecham kept the manuscript text but diagnosed aposiopesis, a lost ending.²

The UPDV follows the manuscript reading without emendation. The sentence is not broken. The second καί is not a conjunction joining two protases; it is the apodosis marker, a well-attested construction in biblical Greek influenced by Hebrew conditional syntax. Blass-Debrunner-Funk §442.7 documents this pattern explicitly: καί introducing a result clause after ἐάν is a calque of the Hebrew construction where וְ (waw) marks the apodosis after אִם ('im, "if"). The same pattern appears in Revelation 3:20, where NA28 brackets the apodosis καί — indicating that even ancient scribes were uncomfortable with the construction. Luke 2:21 shows the same phenomenon: Codex D omits the second καί, and the Homilies of Clement 2.43-44 contain twenty-four consecutive examples of the pattern.

The theological logic closely parallels John 7:17: ἐάν τις θέλῃ ... γνώσεται (ean tis thelē ... gnōsetai, "if anyone desires ... he will know") — the same pattern of human desire triggering knowledge of God. The author of Greeks was already steeped in Johannine language: λόγος in 7:2, μονογενής (monogenēs, "only begotten") in 10:2, and the distinctive verb προαγαπήσαντα (proagapēsanta, "loved first") in 10:3 — which echoes 1 John 4:19. The Semitic syntax came with the Johannine theology. The Syriac Peshitta of John 7:17 supports an Aramaic substrate beneath John's Greek: the same root ܨܒܐ (ṣbā) serves for both "desires" and "will," a natural Aramaic idiom paralleling the Greek θέλῃ / θέλημα pattern.

The rendering maps to the Greek structure carefully. "This faith" (τήν πίστιν, accusative) echoes "put faith" (πιστεύειν) in 9:6 — the same πιστ- root, deliberately chained by the author across the verse break. "Once" captures both the conditional ἐάν and the decisive, single-act nature of the aorist verb ποθήσῃς — not an ongoing state of desire, but a single snap of wanting. "Then" renders both the apodosis καί (result marker) and πρῶτον (sequence marker) — not as the literal word "first," but as a time marker indicating the next step in the chain of knowledge → joy → love → imitation that unfolds through 10:3-6. "Will be received by you" renders λάβῃς as a functional passive: while the Greek verb is morphologically active, it occupies the apodosis — the result clause, marked as such by the Semitic καί. The desiring (ποθήσῃς) is the reader's one genuine act of agency; the receiving (λάβῃς) is God's response to that act. Making the English explicitly passive preserves the theology the author built into the sentence structure: you do one thing, want it, and God does the rest.

🔗Brackets: [objects]

Square brackets are used in two ways throughout the translation. First, neuter substantival adjectives in the Greek (such as τὰ κωφά ("the deaf [things]") and τὰ ἀναίσθητα ("the senseless [things]")) are supplemented with "[objects]" to prevent the English reader from understanding these as references to people rather than to idols. The author deliberately refuses to dignify idols with religious vocabulary; the brackets preserve this rhetorical strategy. Second, other brackets mark words added for clarity that are not present in the Greek, following standard UPDV practice.

🔗Source Text

This translation was not derived from the ASV, which was the starting point for most of the UPDV (except Wisdom of Sirach and First Maccabees). J. C. Moffat's English translation was used as the starting point for The Epistle to the Greeks:

Moffat, J. C. "The Epistle to Diognetus." Biblical Repertory and Princeton Review 25 (1853).

🔗Notes

---

¹ Migne, J.-P., ed. Patrologia Graeca. Vol. 2, cols. 1167–1186 (Greek text and Latin translation), cols. 1159–1166 (editorial preface, Dissertatio proemialis: authorship discussion, Apollos candidacy, recipient identity). Vol. 6, col. 222, continuing at col. 224 (Excerpta ex præfatione editoris Ienensis: Codex Argentoratensis description — charta gossypina/bombycinus, 260 leaves by later numbering, the Reuchlin inscription, Maursmünster 1560, mouse damage to one part — and chain of custody, after observations communicated by Eduard Cunitz of Strasbourg). The same prolegomena's codex-title list runs the epistle under Justin's name as Πρὸς Διόγνητον.

² Meecham, H. G. The Epistle to Diognetus. Manchester University Press, 1949. Meecham provides an edition with a comprehensive introduction covering date, authorship, manuscript history, and notes on selected verses.

³ Lienhard, Joseph T. "The Christology of the Epistle to Diognetus." Vigiliae Christianae 24 (1970): 280-289. Key findings: the παῖς/υἱός distinction is deliberate — an observation Lienhard expressly reports as Meecham's (p. 288); λόγος in 7:2 is doubtful as a Christological title; the Christology is less developed than that of later patristic writers. Lienhard excludes chapters 11–12 from his study because their authenticity is disputed, but explicitly states the authorship question is "avoided, not solved" by his treatment.

⁴ Quarry, J. "The Epistle to Diognetus and Its Possible Authorship, with a Sequel on Novatian's Treatise De Trinitate." Hermathena 9 (1896): 318-357 — Hippolytus proposed as a probability, not a demonstration ("certainty is not to be expected," p. 357). Connolly, R. H. "The Date and Authorship of the Epistle to Diognetus." Journal of Theological Studies 36 (1935): 347-353 — a separate, more tightly delimited case for Hippolytus ("the most probable" of three explanations, p. 352); Quarry is absent from Connolly's article, though non-citation does not by itself establish independent development.

⁵ The manuscript history is summarized in Holmes, The Apostolic Fathers, 3rd ed., introduction to Diognetus; and in Meecham, introduction.

⁶ Holmes, Michael W., ed. The Apostolic Fathers: Greek Texts and English Translations. 3rd ed. Grand Rapids: Baker Academic, 2007.

⁷ The forgery hypothesis and its refutation are summarized in Schaff, Philip. History of the Christian Church. Vol. 2, §170; and in "Diognetus, Epistle To," The New Schaff-Herzog Encyclopedia of Religious Knowledge. Per Schaff: Donaldson "was inclined to assume" that Stephanus fabricated the epistle "but gave up the strange hypothesis," and Overbeck's post-Constantinian theory "has been refuted by Hilgenfeld, Keim, Lipsius, and Draseke" (Schaff's printing) — Dräseke's book being expressly "Against Overbeck and Donaldson." The four-name list is Schaff's assessment; the New Schaff-Herzog entry itself does not carry it.

⁸ "Diognetus, Epistle To," in The New Schaff-Herzog Encyclopedia of Religious Knowledge, ed. Samuel Macauley Jackson. 13 vols. New York: Funk and Wagnalls, 1908–1914. The encyclopedia describes Diognetus as standing in "a secondary or derived relation" to Aristides — an assumption of chronological convenience, since Aristides had a hard date and Diognetus did not. Neither Lienhard (1970), Meecham (1949), nor Jefford (2013) independently substantiates direct dependency.

^8a Clement of Alexandria, Stromata 6.5.39–41, quoting the Kerygma Petrou: "the ways of the Greeks and Jews are old, but we are they that worship him in a new way in a third race, even Christians." The Jewish critique in the same fragment lists angels, new moons, sabbaths, unleavened bread, and feasts — the same items critiqued in both Aristides and Greeks. Dunn, James D. G. Neither Jew nor Greek: A Contested Identity. Grand Rapids: Eerdmans, 2015, notes the link between the Kerygma and Aristides but observes that "which way the influence ran is unclear," dating the Kerygma to "the first three decades of the second century" (§40.2c).

⁹ Lightfoot, J. B., and J. R. Harmer, eds. The Apostolic Fathers. London: Macmillan, 1891.

¹⁰ Forensic stylometry applied to Holmes's critical Greek text using Cosine Delta on 132 Greek function words, following the short-text attribution methodology of Eder (2015). All Greek text was normalized for accent variation (grave → acute) before analysis. Control corpus: Melito of Sardis (Peri Pascha, Hall critical text via Stewart-Sykes 2020, 4,407 Greek words), Justin Martyr (cross-genre), Athenagoras, 1 Clement, 2 Clement, Barnabas, Ignatius (5 letters), balanced to max 5 chunks of 400 words per text (53 chunks total). Bootstrap consensus: 1,000 iterations, 70% feature sampling. Statistical tests: two-proportion z-test (function word density), Poisson (particle absence). Full methodology: Eder, Maciej. "Does Size Matter? Authorship Attribution, Small Samples, Big Problem." Digital Scholarship in the Humanities 30.2 (2015): 167–182.

¹¹ Blomqvist, Jerker. "Apologetics and Rhetoric in the Ad Diognetum." In Apologists and Athens: Early Christianity Meets Ancient Greek Thinking, ed. Gunnar af Hällström, 31–47. Papers and Monographs of the Finnish Institute at Athens XXV. Helsinki: Finnish Institute at Athens, 2020, and "The Ad Diognetum and Contemporary Rhetorical Practice." In Paradeigmata: Studies in Honour of Øivind Andersen, ed. Eyjólfur K. Emilsson, Anastasia Maravela, and Mathilde Skoie, Papers and Monographs from the Norwegian Institute at Athens, Series 4, vol. 2, 203–218. Athens: Norwegian Institute at Athens, 2014 (the latter: periodic structures "virtually absent"). Blomqvist observed the particle absence and the divergent hiatus rate but declined to draw an authorship conclusion from them: "the divergences in language and style are not great enough to warrant the deduction that chs. 11-12 were not written by the same person as chs. 1-10." The present analysis reaches a different conclusion from substantially the same observations. The disagreement is over what inference the data supports, not over the data itself, and Blomqvist's caution is reinforced by the Cosine Delta result reported below, which places the 11–12 split inside the range a single author produces across genres.

¹² Melito of Sardis, Peri Pascha. Greek critical text: Hall, Stuart George, ed. Melito of Sardis: On Pascha and Fragments. Oxford Early Christian Texts. Oxford: Clarendon Press, 1979. English translation and introduction: Stewart-Sykes, Alistair. On Pascha: With the Fragments of Melito and Other Material Related to the Quartodecimans. 2nd ed. Crestwood, NY: St Vladimir's Seminary Press, 2020. The Greek text comprises 105 sections (4,407 words). 2 Clement: Holmes, Apostolic Fathers, 3rd ed. Function word density and particle rates computed from z-scored relative frequencies of 132 Greek function words across the full control corpus. Two-proportion z-test for Melito vs. Diognetus 11–12 density: z = 4.82, p = 1.46 × 10⁻⁶.

¹³ "Diognetus, Letter To," in The Lexham Bible Dictionary, ed. John D. Barry et al. (Bellingham: Lexham Press, 2016).

¹⁴ Green, Michael. 2 Peter and Jude: An Introduction and Commentary. Tyndale New Testament Commentaries. Leicester: Inter-Varsity Press, 1987. Introduction, §1.5.

¹⁵ Marshall, I. Howard. New Testament Theology: Many Witnesses, One Gospel. Downers Grove: InterVarsity Press, 2004. Chapter 1.2, "The Problem of the Canon."

¹⁶ Ellicott, Charles John, ed. A Bible Commentary for English Readers. 8 vols. London: Cassell, 1905. Introduction to 1 John.

¹⁷ Moule, H. C. G. The Epistle of Paul the Apostle to the Philippians. Cambridge Greek Testament for Schools and Colleges. Cambridge: University Press, 1897. The note on 3:20 says the writer of the Epistle to Diognetus wrote "probably with this passage in his mind." Lightfoot, J. B. Saint Paul's Epistle to the Philippians. London: Macmillan, 1868, lists the Diognetus parallel among comparanda on 3:20 without that qualifier.

¹⁸ Freeman, Michelle. "Antithesis and Paradox in the Epistle to Diognetus." Vigiliae Christianae 76 (2022): 37-59.

¹⁹ Bitner, Bradley J. Review of C. N. Jefford, The Epistle to Diognetus (with the Fragment of Quadratus): Introduction, Text, and Commentary. The Classical Review 64 (2014).

²⁰ W. Zimmerli and J. Jeremias, "παῖς θεοῦ," in Theological Dictionary of the New Testament, ed. G. Kittel and G. Friedrich, trans. G. W. Bromiley, vol. 5 (Grand Rapids: Eerdmans, 1967), 654–717. Key findings: παῖς θεοῦ for Christ survived after the NT "only as a fixed liturgical formula anchored in the eucharistic prayer, the doxology and the confession" (p. 700); the stereotyped formula is always διά Ἰησοῦ τοῦ παιδός σου (p. 700); παῖς was "already replaced by υἱός θεοῦ in the Hellenistic Jewish Christian Church" (p. 703); "From the 5th century παῖς disappears completely as a term for Christ" (p. 703). Diognetus is not discussed in this article.

²¹ W. Schneemelcher, "υἱός," in TDNT, vol. 8 (1972), 334–397, section E ("υἱός θεοῦ in Primitive Christian Literature outside the New Testament"). Surveys 1 Clement, Didache, Ignatius, Barnabas, Hermas, Justin, Irenaeus — but not Diognetus. Diognetus is discussed only in the Χριστός article (vol. 9, s.v. "Χριστός," section E.7), where TDNT notes υἱός at 9:2, 9:4, 10:2 but does not mention the παῖς usage at 8:9, 8:11, 9:1.

²² H. G. Meecham, The Epistle to Diognetus (Manchester: Manchester University Press, 1949), notes on 8:9, p. 137: "If we can at all refine here, Christ as παῖς shares and reveals the Father's plan of salvation (8:9, 11; 9:1); as υἱός he is 'sent' and effects it (9:2, 4; 10:2)." Meecham also notes the parallel interchange in Wisdom 2:13, 18 and John 4:46–53. Lienhard, Joseph T. "The Christology of the Epistle to Diognetus." Vigiliae Christianae 24 (1970): 280–289, noted that the alternation "can be taken as an argument for an early date" (p. 289) but did not develop the point; the vocabulary mapping in the supplementary article pursues this lead.

²³ Andrew of Crete (~660–740 CE), Homilia I in Nativitatem B. Virginis Mariae, PG 97, cols. 809–810 (Σήμερον γὰρ τοῦ Παντοκτίστου τὸ κτιστὸν ᾠκοδόμηται τέμενος); Andrew of Crete, In Dormitionem III, PG 97, cols. 1091–1092 (ἡ παναγεστάτη τοῦ παντοκτίστου σκηνή); Peter of Sicily (ninth century), Sermo II adversus Manichaeos, PG 104, cols. 1339–1340 (τοῦ παντοκτίστου καὶ παμβασιλέως τὸ πανιερώτατον κατοικητήριον). These are the earliest later occurrences of παντοκτίστης located in the searched corpora (Patrologia Graeca, with BDAG, Lampe's Patristic Greek Lexicon, and Perseus as supplements); no reuse earlier than Andrew of Crete was located. The epistle's own text is also printed in Migne's second volume (PG 2, cols. 1175–1176; see ¹), which is the only occurrence of the word in that volume.

²⁴ Vocabulary mapping methodology: every theologically weighted term in chapters 7–9 was extracted from Holmes's 3rd edition critical text and classified using BDAG, Lampe's Patristic Greek Lexicon, TDNT, and LSJ. The Patrologia Graeca corpus and Lampe's citation chronology provide the comparison data for each Bucket B assignment against Justin, Theophilus, and Athenagoras. The supplementary article contains the full classification tables, comparison results, and robustness analysis.

🔗Bibliography

🔗Primary Editions

• Holmes, Michael W., ed. The Apostolic Fathers: Greek Texts and English Translations. 3rd ed. Grand Rapids: Baker Academic, 2007.
• Meecham, H. G. The Epistle to Diognetus. Manchester University Press, 1949.
• Marrou, Henri I., ed. A Diognète. 2nd ed. Sources Chrétiennes 33. Paris: Cerf, 1965.
• Migne, J.-P., ed. Patrologia Graeca. 161 vols. Paris, 1857–1866. Epistle to Diognetus in vol. 2, cols. 1167–1186; editorial preface cols. 1159–1166; Codex Argentoratensis description in vol. 6, cols. 222–224.
• Lake, Kirsopp, ed. The Apostolic Fathers. 2 vols. Loeb Classical Library. Cambridge: Harvard University Press, 1912.
• Lightfoot, J. B., and J. R. Harmer, eds. The Apostolic Fathers. London: Macmillan, 1891.
• Brannan, Rick. The Apostolic Fathers Greek-English Interlinear. Bellingham: Lexham Press, 2011.
• Stewart-Sykes, Alistair. On Pascha: With the Fragments of Melito and Other Material Related to the Quartodecimans. 2nd ed. Crestwood, NY: St Vladimir's Seminary Press, 2020.
• Hall, Stuart George, ed. Melito of Sardis: On Pascha and Fragments. Oxford Early Christian Texts. Oxford: Clarendon Press, 1979.

🔗Journal Articles

• Blomqvist, Jerker. "Apologetics and Rhetoric in the Ad Diognetum." In Apologists and Athens: Early Christianity Meets Ancient Greek Thinking, ed. Gunnar af Hällström, 31–47. Papers and Monographs of the Finnish Institute at Athens XXV. Helsinki: Finnish Institute at Athens, 2020.
• Blomqvist, Jerker. "The Ad Diognetum and Contemporary Rhetorical Practice." In Paradeigmata: Studies in Honour of Øivind Andersen, 203–218. Papers and Monographs from the Norwegian Institute at Athens, Series 4, vol. 2. Athens: Norwegian Institute at Athens, 2014.
• Eder, Maciej. "Does Size Matter? Authorship Attribution, Small Samples, Big Problem." Digital Scholarship in the Humanities 30.2 (2015): 167–182.
• Lienhard, Joseph T. "The Christology of the Epistle to Diognetus." Vigiliae Christianae 24 (1970): 280-289.
• Freeman, Michelle. "Antithesis and Paradox in the Epistle to Diognetus." Vigiliae Christianae 76 (2022): 37-59.
• Bitner, Bradley J. Review of C. N. Jefford, The Epistle to Diognetus. The Classical Review 64 (2014).
• Quarry, J. "The Epistle to Diognetus and Its Possible Authorship, with a Sequel on Novatian's Treatise De Trinitate." Hermathena 9 (1896): 318-357.
• Connolly, R. H. "The Date and Authorship of the Epistle to Diognetus." Journal of Theological Studies 36 (1935): 347-353.
• Barnard, L. W. "The Epistle ad Diognetum: Two Units from One Author?" Zeitschrift für die neutestamentliche Wissenschaft 56 (1965): 130-137.
• Thierry, J. J. "The Logos as Teacher in Ad Diognetum XI, 1." Vigiliae Christianae 20 (1966): 146-149.
• Stander, H. F. "A Stylistic Analysis of Chapter 4 of the Epistle to Diognetus." Acta Classica 27 (1984): 129-132.
• Housman, A. E. "A. E. Housman on the Letter to Diognetus VII 2." Harvard Theological Review 45.1 (1952): 1-2.

🔗Secondary Sources

• Barry, John D., et al., eds. The Lexham Bible Dictionary. Bellingham: Lexham Press, 2016.
• Ellicott, Charles John, ed. A Bible Commentary for English Readers. 8 vols. London: Cassell, 1905.
• Green, Michael. 2 Peter and Jude: An Introduction and Commentary. Tyndale New Testament Commentaries. Leicester: Inter-Varsity Press, 1987.
• Marshall, I. Howard. New Testament Theology: Many Witnesses, One Gospel. Downers Grove: InterVarsity Press, 2004.
• Lightfoot, J. B. Saint Paul's Epistle to the Philippians. London: Macmillan, 1868.
• Moule, H. C. G. The Epistle of Paul the Apostle to the Philippians. Cambridge Greek Testament for Schools and Colleges. Cambridge: University Press, 1897.
• Schaff, Philip. History of the Christian Church. 8 vols. New York: Scribner, 1882–1910.
• Jackson, Samuel Macauley, ed. The New Schaff-Herzog Encyclopedia of Religious Knowledge. 13 vols. New York: Funk and Wagnalls, 1908–1914.

🔗Base Text

• Moffat, J. C. "The Epistle to Diognetus." Biblical Repertory and Princeton Review 25 (1853).