HistoricalChristian.Faith

Martin Luther on the Old Testament Canon

Dates: c. 1534 (the complete Luther Bible, Wittenberg; the prefaces to Judith and Tobit date with the 1534 edition, the Table Talk remarks from notes taken 1531–1546) — but the view it crowns developed across 1518–1519, hinging on the Leipzig Disputation · Wikipedia
Verdict on the deuterocanon: NO — not equal to Holy Scripture, but "useful and good to read." In his 1534 German Bible Luther gathered the disputed books out of the Old Testament proper and printed them in a separate section between the Testaments under the heading Apocrypha — "books which are not held equal to the Holy Scriptures, and yet are profitable and good to read." It is the Hebrew-sized canon of Jerome made a feature of the printed page: the seven deuterocanonical books segregated, demoted from canon, yet kept and commended for edification. The vocabulary — canonical Scripture on one side, profitable-but-not-doctrinal reading on the other — is the formula the Thirty-Nine Articles and the Westminster Confession will borrow. But that settled 1534 position was reached only gradually: as late as 1518 Luther was citing the deuterocanon as ordinary Scripture, and the turn came at the 1519 Leipzig debate over purgatory (below).

Claims & evidence

The development of the view, 1518–1519: the Leipzig turn

The 1534 layout was the settled end of a position Luther reached only gradually. In his earliest Reformation writings he cites the disputed books much as the medieval schoolmen did — as serviceable Scripture, without flagging them as a lower tier. In the Explanations of the Ninety-Five Theses (the Resolutiones, 1518) he quotes Sirach in the ordinary run of his proof-texts — "Be not without fear concerning sin forgiven" (Sir. 5:5) and "Do not delay to turn to the Lord" (Sir. 5:8) — and in his 1518 Response to Silvester Prierias he appeals to Tobit; the Heidelberg Disputation of the same year likewise draws on Sirach.¹ At this stage nothing on the page marks these books as sub-canonical.

The hinge was the second Leipzig Disputation (July 1519). Debating purgatory against Johann Eck, Luther was pressed with 2 Maccabees 12:43–45 — the offerings for the dead — as a scriptural proof. Rather than re-read the passage, Luther denied the book's standing, and in doing so made what scholars take to be the first explicit Reformation challenge to a deuterocanonical book's canonicity:

There is no proof of purgatory in any portion of Sacred Scripture, which can enter into the argument and serve as a proof; for the book of Maccabees, not being in the canon, is of weight with the faithful, but avails nothing with the obstinate.

— Luther at Leipzig, July 1519. The familiar English faithfully renders the printed Latin acta of the disputation, where Luther first grants the proof-text — 2 Maccabees 12:46, "Sancta est et salubris cogitatio pro defunctis exorare" ("It is a holy and wholesome thought to pray for the dead") — only to deny the book any probative force: "nam et liber Machabeorum, cum non sit in canone, pro fidelibus potens est, contra pertinaces nihil facit" ("for the book of Maccabees, since it is not in the canon, has force for the faithful but avails nothing against the obstinate").² Eck rejoined that Maccabees had always belonged to the Christian canon — rejected by the Jews but received by the Church — and Luther fell back on Jerome and the Hebrew canon (the canon apud Hebreos against the wider canon apud ecclesiam), the same Hieronymian ground he would hold for the rest of his life. He repeated the demotion in his own printed treatise two years later: in the 1521 Defense and Explanation of All the Articles he writes of 2 Maccabees that "that book is not among the books of Holy Scripture, and, as St. Jerome says, it is not found in a Hebrew version… this book deserves little authority."²

What this development means is genuinely disputed, and a witness page is better served by marking the dispute than by pretending to settle it. The Catholic apologist Gary Michuta reads Leipzig as a sudden, motivated reversal — Luther treated the deuterocanon as Scripture in 1518, then, cornered on purgatory, demoted 2 Maccabees to escape the proof-text, his "theological convictions" deciding the canon. Protestant critics answer that Luther cited Sirach and Tobit by the identical method before and after 1519 — and went on doing so to the end of his life, in the Lectures on Genesis — so the Leipzig statement reads better as his first explicit articulation of the long-standing Jeromist two-tier view than as a volte-face; they add that Luther still believed in purgatory in 1519 and for some years after, which sits awkwardly with the "rejected 2 Maccabees in order to defeat purgatory" motive. A recent academic treatment splits the difference: from 1516 to 1519 Luther used deuterocanonical and protocanonical texts interchangeably, and Leipzig marks a real first explicit denial — but a selective and gradual one (2 Maccabees named, other deuterocanonical texts still used), not a wholesale rejection.³ The bare "Luther removed the books to defeat purgatory" version of the story is weighed in the Arguments on the summary timeline — the 2 Maccabees proof-text and the "did Luther remove the deuterocanon?" pages; the cross-tie is to purgatory, where Leipzig 1519 is the proof-event.

The 1534 Bible: the Apocrypha gathered, segregated, and labelled

Luther's translation did to the canon physically what the Hebrew-minded Fathers had done verbally. Where the Latin Vulgate had interleaved Tobit, Judith, Wisdom, Sirach, Baruch, and 1–2 Maccabees among the Old Testament books, Luther lifted them out and set them together in their own section between the Old and New Testaments — all but 1–2 Esdras, which he omitted entirely — under a single heading:

Apocrypha — that is, books which are not held equal to the Holy Scriptures, and yet are profitable and good to read.

— Luther Bible (1534), heading to the Apocrypha section.⁴ The placement is itself the verdict: the disputed books are no longer numbered among the canonical Old Testament, but they are not discarded either. The arrangement became the model for the vernacular Protestant Bibles that followed — the Swedish (1541), Danish (1550), and others — all carrying the Apocrypha as a distinct, sub-canonical appendix with Luther's heading and prefaces.⁴

The prefaces: "a fine, good, holy, useful book"

To each apocryphal book Luther wrote a preface pointing out the devotional good a Christian might draw from it — precisely the use Jerome had allowed and the canonical rank he had refused. For Judith he weighs the historical difficulties, finds the book will not square with the chronology of Jeremiah and Ezra, and concludes that it is best read as a sacred poem or drama rather than a history — yet commends it warmly:

If one could prove from established and reliable histories that the events in Judith really happened, it would be a noble and fine book, and should properly be in the Bible.… Therefore this is a fine, good, holy, useful book, well worth reading by us Christians.

— Luther, Preface to the Book of Judith (1534).⁵ For Tobit he reads the book the same way — not as history but as edifying fiction about married life, teaching patience under hardship:

Tobit shows how things may go badly with a pious peasant or townsman, and there may be suffering in married life, yet God always graciously helps and finally crowns the outcome with joy, in order that married folk should learn to have patience and, in genuine fear of God and firm faith, put up gladly with all sorts of hardships.

— Luther, Preface to the Book of Tobit (1534).⁵ The praise is real and the demotion is real at once: these are "holy, useful" books to be read, but not the canonical Scripture by which doctrine is established.

The Table Talk: Judith a "tragedy," Tobit "a comedy concerning women"

The same reading turns vivid in his recorded dinner-table conversation, where Luther classifies the two books by the literary genres of the ancient poets:

The book of Judith is no history, nor does it accord with geography… I believe it is a poem… 'Tis a figurative work, like that of Homer about Troy, and that of Virgil about Aeneas, wherein is shown how a great prince ought to be adorned with surpassing valor… It is a tragedy, setting forth what the end of tyrants is. I take the book of Tobit to be a comedy concerning women, an example for house-government.

— Luther, Table Talk XXIV.⁶ The same passage is sharper still on two protocanonical-margin books: "I am so great an enemy to the second book of the Maccabees, and to Esther, that I wish they had not come to us at all, for they have too many heathen unnaturalities."⁶ The remark shows that Luther's exclusion of the deuterocanon was bound up with a wider readiness to rank books by their edifying and evangelical value — a criterion he applied even inside the protocanon (his low estimate of Esther, and famously of James in the New Testament).

Notes

• NO — but a graded, Jerome-style "no." Luther does not anathematize the deuterocanon or call it heretical. His verdict is the old patristic middle position of Jerome, Rufinus, and Athanasius — read in the Church for edification, not used to establish doctrine — now realized as the physical layout of a printed Bible: the books gathered, set apart, and labelled. He differs from those Fathers chiefly in printing the line where readers could see it.
• How settled was the change? Two things are clear and not in dispute: that by 1521 Luther was in print denying 2 Maccabees canonical authority on Jerome's Hebrew-canon ground, and that by 1534 he had segregated the whole deuterocanon as "Apocrypha." What is disputed is the story of how he got there — a sudden, purgatory-driven reversal at Leipzig (Michuta) or the gradual surfacing of a Jeromist distinction he is alleged to have held all along (Swan and others). The witness above keeps that an open question; the motive thesis is adjudicated in the Arguments, not asserted here.
• Why the demotion mattered doctrinally. The Reformers' return to the Hebrew-sized canon was not a neutral antiquarian choice: the deuterocanon carried proof-texts for contested practices — 2 Maccabees 12 for prayer and offerings for the dead, Tobit 12 for almsgiving that "saves from death" — which is why the canon question and the justification controversy ran together. Those proof-texts are weighed in the Arguments on the summary timeline.
• The vocabulary handed forward. Luther's "not equal to Holy Scripture, yet useful and good to read" is the seed of Article VI of the Thirty-Nine Articles (the Apocrypha read "for example of life and instruction of manners," but not to "establish any doctrine") and of the Westminster Confession's flat exclusion (1.3). The English Reformation borrows the heading; the Reformed confessions harden it. Luther himself kept the books in the Bible, and they stood in the Protestant Bibles — the 1611 King James included, which printed the Apocrypha in its own section between the Testaments — for nearly three centuries. Their actual disappearance from most Protestant Bibles was not Luther's act but a nineteenth-century one: amid the "Apocrypha controversy" of the 1820s the British and Foreign Bible Society resolved in 1826 to apply its funds "to the printing and circulation of the Canonical Books of Scripture, to the exclusion of those Books and parts of Books usually termed Apocryphal," and the American Bible Society followed in 1827. That Bible-society decision — driven by Reformed-Puritan conviction and printing economy alike — is the mechanism by which the deuterocanon left the common Protestant Bible, some three centuries after Wittenberg; the "Luther removed the books" trope that telescopes it back onto Luther is weighed in the Arguments.⁷
• Provenance. The Apocrypha heading is given as standardly translated and documented by Bruce M. Metzger; the prefaces are quoted from Luther's Works, vol. 35 (the American Edition, trans. Charles M. Jacobs); the Table Talk from the William Hazlitt translation (paragraph XXIV).

Sources