HistoricalChristian.Faith

Athanasius of Alexandria on the Old Testament Canon

Dates: c. 367 AD (the thirty-ninth Festal [Easter] Letter, issued for the year 367; Athanasius c. 296/298–373) · Wikipedia · his New Testament canon page
Verdict on the deuterocanon: NO — but Baruch kept inside Jeremiah. The most influential fourth-century Eastern voice ranks the disputed books in a famous three-tier scheme. The Old Testament canon proper is the twenty-two-book Hebrew reckoning; below it stand books "not… included in the Canon, but appointed by the Fathers to be read" — among them Wisdom, Sirach, Judith, Tobit, and (strikingly) Esther; and below those again the "apocryphal" forgeries that are "an invention of heretics." Tobit, Judith, Sirach, and Wisdom fall in the middle, "read" rank, not the canon — yet Athanasius folds Baruch and the Epistle of Jeremiah into Jeremiah within the canon, exactly as Origen and Cyril had done before him.

Claims & evidence

Tier one: the canon — "twenty-two books… the number of the letters among the Hebrews"

Athanasius opens his catalogue with the Old Testament, fixing the count at the Hebrew twenty-two and naming the books in order:¹

There are, then, of the Old Testament, twenty-two books in number; for, as I have heard, it is handed down that this is the number of the letters among the Hebrews… The first is Genesis, then Exodus, next Leviticus, after that Numbers, and then Deuteronomy. Following these there is Joshua, the son of Nun, then Judges, then Ruth. And again, after these four books of Kings, the first and second being reckoned as one book, and so likewise the third and fourth as one book. And again, the first and second of the Chronicles are reckoned as one book. Again Ezra, the first and second are similarly one book. After these there is the book of Psalms, then the Proverbs, next Ecclesiastes, and the Song of Songs. Job follows, then the Prophets, the twelve being reckoned as one book. Then Isaiah, one book, then Jeremiah with Baruch, Lamentations, and the epistle, one book; afterwards, Ezekiel and Daniel, each one book. Thus far constitutes the Old Testament.

Measured against the disputed books, the verdict on the canon proper is plain: no Tobit, no Judith, no Sirach, no Wisdom of Solomon, no Maccabees. This is the same Hebrew-sized count Origen gave and Cyril recited a generation earlier — the standard fourth-century Eastern canon.

The bundled exception — "Jeremiah with Baruch… and the epistle, one book"

As with Cyril, the list is not perfectly clean of the Greek extras. Athanasius's single book of Jeremiah is reckoned "Jeremiah with Baruch, Lamentations, and the epistle, one book" — folding both Baruch and the Epistle of Jeremiah (the "epistle" = Baruch ch. 6 in the Vulgate numbering) into the canonical book, not as separate titles but as appendices, exactly as the Greek manuscripts carried them.¹ So even the bishop who would seal the New Testament against addition or subtraction carried two deuterocanonical works inside his Old-Testament canon — the same gesture by which Origen bundled "the epistle" with Jeremiah and Cyril read Jeremiah "including Baruch… and the Epistle." Baruch's place inside Jeremiah is the recurring Eastern signature, and it is the one thing that keeps Athanasius's "No" from being the bare Protestant thirty-nine.

Tier two: "not… included in the Canon, but appointed by the Fathers to be read"

Between the canon and the heretical apocrypha Athanasius sets a middle class — the books read in the church for instruction but not reckoned Scripture — and it is here that Tobit, Judith, and the rest of the deuterocanon fall:²

But for greater exactness I add this also, writing of necessity; that there are other books besides these not indeed included in the Canon, but appointed by the Fathers to be read by those who newly join us, and who wish for instruction in the word of godliness. The Wisdom of Solomon, and the Wisdom of Sirach, and Esther, and Judith, and Tobit, and that which is called the Teaching of the Apostles, and the Shepherd. But the former, my brethren, are included in the Canon, the latter being [merely] read…

This is the precise datum the later Old-Testament dispute turns on: a generation before the African councils list the deuterocanon among the Scriptures (further down the timeline), the most influential Eastern voice files the very same books one rank below the canon — "appointed by the Fathers to be read," not "accredited as Divine." The distinction is Origen's canonical-versus-read distinction made into an explicit list, and it is the one Rufinus and Jerome will carry forward.

The surprise: Esther demoted, deuterocanon kept inside the canon at one point

Two details cut against the easy "Athanasius = Hebrew canon" reading. First, he places Esther — a protocanonical, Hebrew book — in the second, read-but-not-canonical rank alongside Tobit and Judith, so his twenty-two are not simply the Jewish thirty-nine renumbered; his canon both drops a Hebrew book (Esther) and keeps two Greek ones (Baruch, the Epistle).² Second, the middle tier is a genuinely generous category: the deuterocanonical books are not condemned but commended "for instruction in the word of godliness," sharply distinguished from the third tier — the "apocryphal writings… an invention of heretics," forged and back-dated "that… they may find occasion to lead astray the simple."² So the "No" is a careful, three-graded No, not a flat rejection.

Tier two in his own hand — how Athanasius actually cites the disputed books

The Festal-Letter ranking is one thing; Athanasius's running citations across his apologetic, polemical, and pastoral works are another — and here the gap between his formal list and his working hand is real and wide. Sorting his usage by what it actually does, three classes emerge.

(a) Books his list kept inside the canon — cited as Scripture, no tension. Twice Athanasius cites a disputed book in his own voice as the very speech of God — but each time it is a book his catalogue had folded into the canon. In De Decretis (the Defence of the Nicene Definition) he makes the eternal Word the speaker of Baruch: "when the Word chides Israel, He says, 'You have forsaken the Fountain of wisdom'" (Baruch 3:12), set in a chain with Hebrews, the Psalms, and Jeremiah.³ And in the first Oration against the Arians he ranges the Greek Susanna (the addition to Daniel) exactly alongside Isaiah as parallel testimony to God's eternity: "if Isaiah says, 'The Everlasting God, the Creator of the ends of the earth;' and Susanna said, 'O Everlasting God'" (Daniel 13:42).⁴ Neither cuts against his list: he had reckoned Baruch within canonical Jeremiah and the Greek additions within canonical Daniel — precisely the concession even the modern excluders' spokesmen grant, that the additions travelled inside the canonical books he never knew to separate. So these are Athanasius treating canonical material as Scripture, not the second tier.

(b) Genuine tier-two books cited in his own voice as Scripture — the real and substantial tension. Here list and usage part company. The plainest case is the Wisdom of Solomon, a book his catalogue files expressly below the canon, yet which he quotes again and again as Scripture in his own words. In Against the Heathen (Contra Gentes) he grounds his whole account of the origin of idolatry on a long passage of Wisdom (14:12–21), introducing it as the very voice of Scripture — "Scripture taught us beforehand long ago, when it said, 'The devising of idols was the beginning of fornication…'" — and glossing the same passage as "the wisdom of God testifies beforehand when it says" and, further on, as "what Scripture calls the incommunicable name and honour of God."⁵ In the second Oration against the Arians he marshals Wisdom 9:2 under an express citation formula — "and in the book of Wisdom it is written, 'Having ordained man through Your wisdom, that he should have dominion over the creatures…'" — naming the book and quoting it with "it is written" in a chain of texts "divine Scripture recognises," beside the Psalms, Revelation, 1 Timothy, Colossians, and the words of the Lord.⁶ The same Oration draws on Wisdom 13:5 — "for by the greatness and beauty of the creatures proportionably the Maker of them is seen" — among the texts of which he says in the same breath "hearing the Scriptures we believe."⁷ And in On the Opinion of Dionysius he vouches that Dionysius drew the Son's titles "the word, wisdom, power, breath… and brightness of the Father" "from the Scriptures," the "breath" being Wisdom 7:25.⁸ Sirach fares the same: in the encyclical To the Bishops of Egypt and Libya Athanasius answers the heretic "in the words spoken by the Spirit," chaining Psalm 50:16 with Sirach 15:9 — "Praise is not seemly in the mouth of a sinner" — as Spirit-spoken Scripture.⁹ And Tobit holds too: in the Defence before Constantius he sets "Ask counsel of all that are wise" (Tobit 4:18) under an "it is written" frame.¹⁰ So three of the four deuterocanonical books his Festal Letter ranks below the canon — Wisdom, Sirach, and Tobit — he nonetheless cites in his own voice under Scripture formulae. The strain between his stated canon and his citation practice is not narrow; it is the same wide gap between formal list and working usage that runs through Gregory the Great and John of Damascus further down this timeline.

(c) A moral example, and a contested phrase — no firm claim of Scripture. Two stock proofs are weaker than advertised. In Festal Letter 4 Athanasius holds up "blessed Judith, when having first exercised herself in fastings and prayers, she overcame the enemies, and killed Olophernes" — but as an exemplum of fasting and prayer set beside "blessed Esther," a pattern of conduct to imitate, not a text cited as authoritative Scripture.¹¹ And the line sometimes adduced for Judith as Scripture — "But God is not as man, as Scripture has said" (Oration II.35) — is too uncertain to count: the same words are Numbers 23:19 in the Septuagint, a protocanonical match, and Athanasius elsewhere uses "God is not as man" as a bare maxim with no Scripture tag, so the safer attribution is Numbers, not Judith 8:16.¹²

The honesty correction — and its limit. Two of the lines most often quoted to show Athanasius himself treating the deuterocanon as "holy Scripture" — Wisdom 1:11 ("The mouth that belies slays the soul") and Sirach 30:4 ("though his father die, yet he is as though he were not dead") — are not in fact his own words. Both sit inside documents he compiled into his Defence against the Arians (Apologia contra Arianos), a dossier largely of letters by others: Wisdom 1:11 appears in the Encyclical Letter of the Council of Egypt — the synodal letter the assembled Egyptian bishops wrote on Athanasius's behalf, which speaks of him throughout in the third person, and which in the same breath cites Tobit 12:7 ("It is good to keep close the secret of a king") as "what is written" — and Sirach 30:4 in a letter of Alexander, bishop of Thessalonica.¹³ On those three citations the popular apologetic overreaches: they testify to the council and to Alexander of Thessalonica, not to Athanasius's own pen, and so should be set aside in weighing his practice. But — and this is the correction's limit — setting them aside does not rescue a tidy picture. Athanasius does quote Wisdom and Sirach as Scripture in his own voice; he simply does it elsewhere (Wisdom 14, 9:2, 13:5; Sirach 15:9), not at the two passages the apologists usually cite. So the slogan "he quotes Wisdom and Sirach as holy Scripture" turns out substantially true even though its usual proof-texts are misattributed — and the honest summary is the one the Fathers' practice generally yields: a formal canon that excludes the deuterocanon, and a working hand that reaches for it as Scripture all the same.

Notes

• NO to the deuterocanon's canonicity, but not to its use. Athanasius is rightly cited as a fourth-century witness against the canonical status of Tobit, Judith, Sirach, and Wisdom — he names four of them expressly in the "read but not canon" tier. But the same passage that excludes them also prescribes them for the catechumens, and his canon keeps Baruch and the Epistle of Jeremiah folded into Jeremiah. The honest summary is: not Scripture, but appointed by the Fathers to be read. Whether this makes him a witness for the Protestant canon — as Archibald Alexander, James White, Steve Christie, and Gavin Ortlund variously argue, against Gary Michuta — is weighed on its own argument page (assessment: contested — sound against the books' canonicity, overstated as the Protestant canon).
• The three tiers are the contribution. Earlier Fathers had drawn a two-way line (canon vs. read); Athanasius's letter is the cleanest early statement of the threefold scheme — canonical, ecclesiastically read, and heretical apocrypha — that the whole later dispute presupposes. It lets him exclude the deuterocanon from the canon while defending it against the "apocryphal" forgeries in a single breath. What divides the top two tiers — given that he cites middle-tier books as Scripture, the line cannot be inspiration — is weighed on a separate argument page (assessment: plausible — the read tier reads as catechesis, the canon tier perhaps as the books that carry weight with Jews).
• A witness on the Nicaea question, too. Athanasius sat at Nicaea (325) and still, in 367, ranks Judith below the canon — "appointed by the Fathers to be read," not "included in the Canon." That tells directly against the claim that the Council of Nicaea counted Judith among the Scriptures; the point is weighed on the Nicaea-and-Judith argument page and noted on the Nicaea witness.
• The New-Testament side is its own page. The very same Festal Letter gives the first surviving catalogue identical to the present twenty-seven-book New Testament — the four Gospels, Acts, seven Catholic Epistles, fourteen Paulines, and Revelation, "let no man add to these." That list, and its bearing on the close of the New-Testament canon, is treated on his New Testament canon page; this page weighs only the Old-Testament tiers.

Sources