World Religions

Scripture & Interpretation

Day 23ComparativeFrom oral transmission to canon

Reading scripture sounds like the simplest religious act there is: open the book, look at the words. But every scripture had already travelled a long production line before anyone could open it — centuries of deliberately unwritten transmission, then a fixed text, then fights over which books count, then translation bans and translation drives, and only at the very end the question of how to read it. One counterintuitive pattern: the more sacred a tradition holds its scripture to be, the later it tends to allow writing it down — and the less likely it is to recognize a translation.

"Turn it and turn it again, for everything is in it."— Pirkei Avot 5:22, the rabbinic tradition on the Torah

History & Origins

A production line with five stages

Stage one: oral — and deliberately so. The Vedas are called shruti, "that which was heard," traditionally held to have no author at all; what matters is sound, not script. The recitation systems reorder the same passage word by word (padapatha) and in interlocking permutations (krama, jata, ghana) — a checksum built into memory, designed so that a single dropped syllable shows up immediately. UNESCO inscribed the tradition on its intangible heritage list in 2008. Likewise the Pali Canon, by traditional account, was not written down until 29 BCE, at the Aluvihare monastery in Sri Lanka; and "Qur'an" itself means recitation. In these traditions writing was not originally the vehicle of authority — it was a backup copy of memory.

Stage two: the fixed text. Islam's standardization was carried out under Uthman (r. 644–656), with copies sent to the major garrison cities and divergent codices ordered destroyed. Surviving early material is broadly consistent with this: the Birmingham folios radiocarbon-date to 568–645 CE (95.4% probability), and the lower text of the Sana'a palimpsest to 578–669 CE — the latter being the only known manuscript carrying substantial non-standard readings. On the Hebrew side, the Dead Sea Scrolls found from 1947 onward comprise some 15,000 fragments representing 800–900 manuscripts, spanning the 3rd century BCE to the 1st century CE. The Great Isaiah Scroll, the most complete of them, agrees with the later Masoretic Text roughly 95% of the time; of its 2,600-plus variants, the overwhelming majority are spelling and word choice.

Stage three: canon. Which books count was a separate argument, and it ran for centuries. Athanasius's 39th festal letter of 367 CE is the earliest known list matching today's 27-book New Testament exactly; the North African councils of Hippo (393) and Carthage (397) confirmed a list; the Catholic Church settled its Old Testament, deuterocanonical books included, at the Council of Trent on 8 April 1546. Three canons still differ today: 39 Old Testament books for Protestants, 46 for Catholics, somewhat more across the autocephalous Orthodox churches, and the broadest of all in the Ethiopian Orthodox Tewahedo tradition (often cited as 81, counted differently in different sources).

Oral Fixed text Canon Translation Reading Vedas · Pali Canon memory as medium Masoretic Text Uthmanic codex 27 books, 367 CE 3 canons differ 776 full Bibles or none allowed literal · symbolic who may interpret Traditions brake at different stages: the Vedas stay oral, a translated Qur'an is not the Qur'an, and three canons still differ.
The stages do not replace one another — recitation continues after the text is written, and argument continues after the canon is closed. Schematic, not to scale.
Where It Stands Today

The most translated book, and the book that refuses translation

The Bible is the most widely translated text in human history, and still expanding. Per the Wycliffe Global Alliance count of August 2025: of roughly 7,396 living languages worldwide, 776 have the full Bible, 1,798 have the New Testament, and 1,433 have some portion — over 4,000 languages with Scripture in some form. Just 544 languages have no translation work under way at all, down from more than 5,000 in 1999.

The Qur'an runs the opposite way. The mainstream position holds that "the Qur'an" means the Arabic text; renderings into other languages are usually titled interpretations of the meaning rather than the Qur'an itself, and ritual recitation is in Arabic. Its internal variety lies not in translations but in recitations (qira'at): ten readings accepted as canonical, each with its own chain of transmitters, differing mostly in vowels, occasional word choice, and pause points. Hafs 'an 'Asim dominates in practice today — estimates of its reach range from well over 70% to above 90% — while Warsh 'an Nafi' is common in North and West Africa.

More manuscripts means more variants; this is arithmetic, not scandal. Over 5,800 Greek New Testament manuscripts have been catalogued, and the variants among them total roughly 400,000 — genuinely more than there are words in the New Testament. Yet scholars on opposite sides of the argument broadly agree that the vast majority are spelling, word order, and obvious slips, and that variants which are both meaningful and viable amount to under 1%. Compare the Sikh Guru Granth Sahib: fixed at 1,430 pages, so that the same page number is the same line in every copy on earth — a stability only the printing press made possible.

How many people actually read? Only scattered national data exist, nearly all of it from the United States. In the American Bible Society's State of the Bible 2025, 41% of adults counted as "Bible Users" (reading Scripture outside of worship at least three times a year). In Gallup's 2022 survey, 20% said the Bible is the literal word of God — a record low for that question — while 49% called it inspired but not to be taken literally throughout, and 29% called it fables, legend, history and moral precepts recorded by man, a record high. Such figures are highly sensitive to question wording and should be read for order of magnitude only.

Core Beliefs

What a book can be

Four different kinds of sacred

When traditions say "scripture," they are not pointing at the same kind of object:

These four positions settle everything downstream: whether it may be translated, whether it may be revised, who gets to interpret it, and how it is physically kept.

Layered reading: three strikingly similar structures

Nearly every tradition with a large scripture developed a surface-to-depth reading scheme — and the number of layers is often exactly four. The Jewish PaRDeS (the word means "orchard"; the scheme settled from the 12th century on) runs peshat (plain sense), remez (hint), derash (homiletic), sod (hidden). The medieval Latin Quadriga runs literal, allegorical, moral, anagogical. In Shia and Sufi tradition, a fourfold scheme attributed to Ja'far al-Sadiq (702–765) runs 'ibara, ishara, lata'if, haqa'iq. The three do not map neatly onto one another, but they share two rules: move from outside in, and never skip the literal layer — abolishing the plain sense means abolishing the law, a standard charge inside all three.

Judaism · PaRDeS Latin West · Quadriga Shia & Sufi · four layers Sod — the hidden anagogia haqa'iq — realities Derash — homiletic tropologia lata'if — subtleties Remez — hint allegoria ishara — allusion Peshat — plain sense littera 'ibara — the wording the mystical layer where it all points for the prophets sermon and legal reasoning what to do for the saints what the wording hints at it points to Christ for the elite grammar and context what happened for everyone surface depth The three schemes do not map one-to-one, but all run outside-in — and all forbid skipping the bottom layer.
Layered reading is the shared technology for handling an old text in new circumstances: the words stay put, the depth of reading changes.

Literalism is younger than it looks

Treating "read it literally" as the oldest and most faithful approach is a common chronological illusion. Origen (c. 185–253) held that Scripture has body, soul and spirit; Augustine, in The Literal Meaning of Genesis, wrote that where a literal reading collides with what is known of the natural world, the reading should give way. Strict inerrancy in its modern form is largely a reaction to 19th-century historical criticism and evolutionary science: Wellhausen set out the documentary hypothesis of the Pentateuch in 1878, and the evangelical side put its position on paper in 1978, when nearly 300 scholars signed the Chicago Statement on Biblical Inerrancy at a conference there. Both lines are modern.

Practice

Scripture is used, not merely believed

Common Misreadings

Misreading: literal reading is the oldest and most orthodox approach.
Layered and allegorical reading is just as old — Origen and Augustine both argued for it. Modern strict literalism is largely a 19th- and 20th-century response to historical-critical scholarship.

Misreading: "the Bible says" points at something settled.
Ask first: which canon (39 books, 46, or the longer Orthodox and Ethiopian lists), which text tradition (Masoretic, Septuagint), and which translation. All three variables change the sentence.

Misreading: oral transmission is unreliable; writing makes it safe.
It cuts both ways. Vedic interlocking recitation builds redundancy checking into transmission itself, while copying steadily introduces errors. Each medium has its own failure mode.

Misreading: hundreds of thousands of variants mean the text is unrecognizable.
There are many variants mainly because there are many manuscripts, and the great majority carry no difference in meaning. Nor should one overcorrect to "no effect at all" — the longer ending of Mark and the woman taken in adultery are absent from the earliest manuscripts, and that is scholarly consensus.

Cross-Tradition Reading · Today

Every change of technology reshuffles interpretive authority. In a manuscript world very few could read Scripture at all, so authority sat naturally with a trained class; print plus vernacular translation put the text in ordinary hands, and half the energy of the Reformation is in that fact (Day 5). Search, parallel translations and machine-generated commentary push this further, and the trade-off is symmetrical: going around the scholars has never been easier, and neither has circulating a verse stripped of its context.

Traditions handle modern critical scholarship very differently. Catholicism encouraged original-language and historical study after the 1943 encyclical Divino Afflante Spiritu, and the Pontifical Biblical Commission called the historical-critical method indispensable in 1993. Conservative and Reform Judaism broadly accept it; most Orthodox Jewish authorities do not. Protestantism spans a continuous range from mainline to fundamentalist. Critical study of the Qur'an's textual history remains highly sensitive across much of the Muslim world. These differences are different answers to the question of where authority comes from, not a ranking of who is more advanced.

It connects back to the last two issues. Day 21 covered conceptions of the divine and Day 22 mystical experience; in both cases the disagreements end up being fought over texts, because in most traditions the final say about what counts sits with scripture. And scripture has to be read by someone — which is why "who may interpret" usually determines a tradition's direction more than "what the text says."

Going Deeper

Why do the traditions that revere scripture most tend to write it down last?
At least three reasons. Control of distribution: oral transmission requires a teacher, so who learns and how deeply stays governable, whereas a written copy can be picked up by anyone. A technical judgment about fidelity: before printing, copying accumulated errors, while a well-designed recitation system exposed them on the spot. And suspicion of writing is itself a theological position — in the Indian tradition sound is the reality of scripture and script is only its shadow. Worth noticing: these arguments return in new form whenever text becomes trivially copyable. What is scarce then is not the text but the reading.
Is there really such a thing as reading "only literally"?
Strictly, it is hard to sustain. Any reading starts by judging genre — is this poetry or chronicle, statute or parable — and that judgment is already interpretation. Self-described literalists generally accept that "his eyes were like a flame of fire" is figurative; they simply draw the line elsewhere. So the real dispute is rarely whether to interpret, but where the line falls and who has standing to draw it. Which is why internal arguments about the text so often turn into arguments about authority.
Do variants and multiple translations weaken a tradition, or strengthen it?
Both, depending on where the tradition puts them. Stake inerrancy on "not one word has changed" and every variant becomes a threat; the rabbinic tradition, which preserves minority opinions alongside majority ones in the Talmud, can treat divergent readings as a resource instead. Translation works the same way: refusing it keeps linguistic unity at the cost of most believers meeting the text through a screen, while translating widely buys reach at the cost of "the Bible says" always needing a follow-up question. There is no free option, only different structures of cost — and which cost a tradition chose often tells you more about how it works than its doctrines do.

This piece introduces how scriptures were formed and are interpreted across traditions from a neutral, comparative standpoint, aiming to present each tradition's internal logic and its disagreements with others, without praise, blame, or ranking, and without taking a position on any scripture's truth or authority. Dates and figures are recent scholarly and survey estimates and vary by method and definition.