HistoricalChristian.Faith

Rufinus of Aquileia on the Old Testament Canon

Dates: c. 404 AD (the Commentary on the Apostles' Creed, written at Aquileia on his return from the East; Rufinus c. 345–411) · Wikipedia
Verdict on the deuterocanon: NO — but the books are "Ecclesiastical," read in church though not used to prove doctrine. Rufinus gives the cleanest, most explicit version of the middle position the Eastern Fathers had been refining. He sorts the books into three named ranks: the Canonical Scriptures (a Hebrew-sized Old Testament, the disputed books absent); the Ecclesiastical books — Wisdom, Sirach (Ecclesiasticus), Tobit, Judith, and the Maccabees — "read in the Churches, but not appealed to for the confirmation of doctrine"; and the Apocrypha, which "they would not have read in the Churches" at all. It is the locus classicus for the canonical-versus-ecclesiastical distinction that the later Reformers and the Thirty-Nine Articles would borrow.

Claims & evidence

The Canonical Old Testament: a Hebrew-sized list, the disputed books absent

Rufinus introduces his catalogue as a list "of the New and of the Old Testament, which, according to the tradition of our forefathers, are believed to have been inspired by the Holy Ghost, and have been handed down to the Churches of Christ."¹ The Old Testament he then runs out is the Hebrew-sized canon:

Of the Old Testament, therefore, first of all there have been handed down five books of Moses, Genesis, Exodus, Leviticus, Numbers, Deuteronomy; Then Jesus Nave, (Joshua the son of Nun), The Book of Judges together with Ruth; then four books of Kings (Reigns), which the Hebrews reckon two; the Book of Omissions, which is entitled the Book of Days (Chronicles), and two books of Ezra (Ezra and Nehemiah), which the Hebrews reckon one, and Esther; of the Prophets, Isaiah, Jeremiah, Ezekiel, and Daniel; moreover of the twelve (minor) Prophets, one book; Job also and the Psalms of David, each one book. Solomon gave three books to the Churches, Proverbs, Ecclesiastes, Canticles. These comprise the books of the Old Testament.

Measured against the disputed books the verdict on the canon proper is plain: no Tobit, no Judith, no Wisdom, no Sirach, no Maccabees.¹ This is the same Hebrew reckoning Origen, Cyril, Athanasius, and Jerome gave. One small difference: where the other Eastern lists fold Baruch and the Epistle of Jeremiah into the canonical Jeremiah, Rufinus names "Jeremiah" bare, with no separate place for Baruch — closer to Jerome's harder line on that one book.

The "Ecclesiastical" books: Wisdom, Sirach, Tobit, Judith, Maccabees — read, but not for proving doctrine

Rufinus does not leave the disputed books unmentioned, nor lump them with heretical forgeries. He gives them a distinct, named middle rank — the heart of his witness:²

But it should be known that there are also other books which our fathers call not "Canonical" but "Ecclesiastical:" that is to say, Wisdom, called the Wisdom of Solomon, and another Wisdom, called the Wisdom of the Son of Syrach, which last-mentioned the Latins called by the general title Ecclesiasticus, designating not the author of the book, but the character of the writing. To the same class belong the Book of Tobit, and the Book of Judith, and the Books of the Maccabees. In the New Testament the little book which is called the Book of the Pastor of Hermas, [and that] which is called The Two Ways, or the Judgment of Peter; all of which they would have read in the Churches, but not appealed to for the confirmation of doctrine. The other writings they have named "Apocrypha." These they would not have read in the Churches.

So Rufinus draws a three-fold line that the earlier Fathers had only sketched: Canonical (inspired, doctrine-proving) — Ecclesiastical (read aloud in worship, but not used to settle dogma) — Apocrypha (not read at all). The deuterocanon sits squarely in the middle tier. The operative formula — "read in the Churches, but not appealed to for the confirmation of doctrine" — is almost word-for-word Jerome's "read… for the strengthening of the people, but not for confirming the authority of ecclesiastical dogmas," the rank Athanasius had called "appointed by the Fathers to be read."

The New Testament canon, and the symmetry of the scheme

For completeness Rufinus's list also gives the New Testament — the four Gospels, Acts, fourteen Pauline Epistles, the Catholic Epistles, and Revelation, "the books which the Fathers have comprised within the Canon, and from which they would have us deduce the proofs of our faith"³ — and then files New-Testament books in the same Ecclesiastical tier (the Shepherd of Hermas and the Two Ways / Judgment of Peter). The structure is symmetric across both Testaments: a doctrine-proving canon, a worship-reading middle rank, and the apocrypha shut out. That makes Rufinus's the most systematic statement of the graded canon on this timeline — the Western counterpart, in tidy Latin form, to the Eastern lists of Origen, Cyril, Athanasius, and Epiphanius.

Definition versus usage: he cites Sirach as "the Scripture says"

The tidy three-fold scheme is Rufinus the cataloguer. Rufinus the exegete is harder to keep inside his own lines. A few years before the Creed commentary, expounding Jacob's blessings in his Commentary on the Blessings of the Patriarchs (c. 398–400), he comes to the blessing of Gad and argues that no one reaches perfection without first being proved in temptation — and clinches the point by quoting Sirach with the plainest possible Scripture-formula:⁴

Nec ullo tempore omnino quis perveniet ad perfectionem, nisi prius in tentationibus suis documenta praestiterit: ita enim Scriptura dicit, Qui non est tentatus, non est probabilis.

— "No one will ever attain perfection at all, unless he has first given proof of himself in his temptations; for thus the Scripture says, 'He that has not been tempted is not approved'" (Sirach 34, the chapter Rufinus cites as "Ecclesiasticus 34"). The book he had filed under "Ecclesiastical… not appealed to for the confirmation of doctrine" he here introduces as "Scriptura" and leans on to settle a point of moral doctrine — exactly the kind of appeal his own definition forbids. This is the sharpest definition-versus-usage strain on Rufinus, and it is the same gap that runs through the other Hebrew-minded Fathers (Cyril, Athanasius, Epiphanius): the books ranked below the canon in the list keep doing the work of Scripture in the argument. The point is weighed on the "did the early church reject the deuterocanon?" argument page.

The other side of the strain: in the Apology he defends the Greek additions as apostolic Scripture

The same tension runs the other way in Rufinus's Apology against Jerome (Book II, c. 401), written in the heat of their quarrel. Defending the old Greek Old Testament against Jerome's new Hebrew-based version, Rufinus treats the Septuagint — including the very Greek-Daniel additions his own catalogue passes over (Susanna and the Song, or Hymn, of the Three Children) — as the apostolic deposit, and charges Jerome with doing "havoc" to it. Along the way he answers, in its own century, the later thesis that the early church embraced the longer Greek books only out of ignorance of Hebrew:⁵

There has been from the first in the churches of God, and especially in that of Jerusalem, a plentiful supply of men who being born Jews have become Christians; and their perfect acquaintance with both languages and their sufficient knowledge of the law is shewn by their administration of the pontifical office. In all this abundance of learned men, has there been one who has dared to make havoc of the divine record handed down to the Churches by the Apostles and the deposit of the Holy Spirit? … the whole of the history of Susanna, which gave a lesson of chastity to the churches of God, has by him been cut out, thrown aside and dismissed. The hymn of the three children, which is regularly sung on festivals in the Church of God, he has wholly erased from the place where it stood.

The catalogue and the Apology do not sit easily together. In the Creed commentary Rufinus runs the Hebrew-sized canon and names "Daniel" bare; here he treats the Greek-Daniel additions as Scripture so settled that to touch them is to "make havoc of the divine record handed down … by the Apostles." It is the same definition-versus-usage gap as the Sirach citation above, sharpened by a generation of polemic: the books ranked below the canon in the list keep doing the work of Scripture once the argument is joined. Two of the props Rufinus leans on do not bear his weight — the legend that the seventy translators, sealed in separate cells, produced word-for-word identical versions under direct inspiration, and the claim that Peter himself delivered the Septuagint books to the Roman church as its sacred deposit. But the sociological core of his rebuttal — that the early church, "especially" at Jerusalem, was never uniformly ignorant of Hebrew, and that its bilingual Jewish-born members never moved to excise the longer readings — stands independent of those legends. It is weighed on the "did the early church accept the Apocrypha out of ignorance of Hebrew?" argument page.

Notes

• NO to canonicity, but a precise middle, not a flat rejection. Rufinus is a clear witness against the canonical standing of Tobit, Judith, Wisdom, Sirach, and the Maccabees — his canon is the Hebrew list. But he gives the disputed books an honored place: "Ecclesiastical," read in church for edification though not cited to prove doctrine. This is the same posture as Jerome and Athanasius, stated with unusual clarity.
• He coined the working vocabulary. Rufinus's three-fold Canonical / Ecclesiastical / Apocrypha is the cleanest articulation the early church produced of the distinction, and the word "Ecclesiastical" for the read-but-not-doctrinal books is largely his. The same logic — useful for life and manners, not for establishing doctrine — resurfaces in the Reformation, when the Thirty-Nine Articles and the Westminster Confession quote Jerome to the same end. What that middle category means — a rank below the canon (John Meade) or a catechetical subset within it (Gary Michuta) — is weighed on the "what did the 'ecclesiastical' category mean?" argument page (assessment: sound — a graded middle rank, read and quotable yet ranked below the doctrine-proving canon).
• Agreeing with Jerome — for once. Rufinus and Jerome were bitter antagonists in the Origenist controversy of these very years, yet on the canon they land in almost exactly the same place: a Hebrew-sized canon, the deuterocanon read but not doctrinal, and even Baruch given no separate canonical slot. The contemporary fault line on the canon was not Rufinus-versus-Jerome but both of them against Augustine and the African councils, who leveled the disputed books up into the canon.
• Translation provenance. The English is the Nicene and Post-Nicene Fathers rendering by William Henry Fremantle (NPNF 2.3); the sections are §37 (the canonical list) and §38 (the Ecclesiastical books and the Apocrypha).

Sources