The Long War Over the Missing Vowels

Ths sntnc hs n vwls, nd y cn rd t nywy.

You can, because you already know English, and your memory of the language fills each gap before you notice it. For most of the alphabet's history, that is simply how reading worked. The Phoenician script, ancestor of the Hebrew, Aramaic, Arabic, Greek and Latin letters, had twenty-two signs, and every one of them was a consonant. The reader supplied the rest.

In July we promised to come back to the moment the Greeks broke that habit. The full story is longer than one clever borrowing. It took well over a thousand years and several scripts to put the vowels on the page, and two of the traditions involved had to do it without changing a single letter.

An alphabet that trusted its readers

Writing only consonants is less reckless than it sounds for a Semitic language. In Hebrew and Arabic, much of a word's meaning lives in a root of three consonants. K-T-B is the idea of writing: Arabic kataba, "he wrote," kitāb, "book," maktab, "office." The vowels carry grammar and nuance, and a fluent reader can usually rebuild them from context.

Scribes still added help where it was cheap. Early Hebrew and Aramaic writers began using a few consonant letters, above all yod and waw, to mark long vowels, and he and aleph at the ends of words. Grammarians later called them matres lectionis, "mothers of reading." They were hints, not a system. A short vowel still went unwritten.

The Greek repurposing

When Greeks adopted the Phoenician letters, conventionally around 800 BCE, the hints became a system. Phoenician had consonants Greek did not use, such as the glottal stop of aleph and the throat sound of ʿayin, and Greek could not be read from consonants alone: too many of its words begin with vowels or turn on them.

So five letters changed jobs. Aleph became alpha. He became epsilon. Yod became iota. ʿAyin became omicron. Waw became upsilon. In dialects that had lost the h sound, heth became eta, a long e, and omega came later. Historians of writing call the result the first alphabet in the narrow sense, with letters for vowels as well as consonants. Latin and Cyrillic inherited the idea, which is why the letters you are reading carry their own vowels.

Even Greek eventually wanted more. Around 200 BCE in Alexandria, the scholar Aristophanes of Byzantium is credited with the accent and breathing marks, at a time when more and more readers of Homer had not grown up hearing his Greek.

Dots around a frozen text

Hebrew faced a harder version of the problem. By late antiquity it was a language of scripture, prayer and study more than of the street, and the consonants of the biblical text were fixed. A scribe could not add a letter, even a helpful one. Yet the traditional way of reading it, passed from teacher to student, had to survive.

The answer came from scholars in Tiberias, on the Sea of Galilee, known as the Masoretes. Between roughly the seventh and tenth centuries they surrounded the letters with small dots and dashes for vowels, signs for chant and stress, and notes in the margins, all placed under, over and inside the letters so that the consonants stayed untouched. Their most famous surviving work, the Aleppo Codex, was written around 920, and Maimonides relied on it. Of its estimated 487 pages, about 294 survive. Most of the Torah was lost after the synagogue in Aleppo that held it was burned in 1947.

Sometimes the points do not describe the written word at all. The four letters of the divine name, יהוה, were not pronounced aloud; readers said Adonai, "my Lord," instead. The Masoretes gave the four letters vowels close to those of אֲדֹנָי, Adonai, as an instruction: read this as Adonai. Christian Hebraists who read the combination literally produced "Jehovah," a form attested in Latin from the thirteenth century and given to English by William Tyndale, who printed Iehouah in 1530. A vowel sign can be an instruction rather than a sound, and a reader outside the tradition who misses that can invent a word no one ever said.

One verse, two endings

Arabic needed two layers of help. Several letters share one shape and are told apart only by dots, called iʿjām, and short vowels were not written. We wrote in March about how the earliest Qur'an manuscripts left readers to carry much of that knowledge themselves.

The tradition remembers why the vowels were added. The grammarian Abu al-Aswad al-Duʾali, who died in Basra around 688, is said to have heard a reciter misread a single ending in the third verse of the ninth sura. In response he devised colored dots, usually described as red: a dot above a letter for a, below it for i, beside it for u. Tradition credits the governor al-Hajjaj with having the consonant dots standardized later, although dated papyri from 643 already use them, sparingly. In the eighth century the grammarian al-Khalil ibn Ahmad replaced the vowel dots with small letter shapes, a little alif for a, waw for u and ya for i. Those marks, the ḥarakāt, are the ones printed Qur'ans and children's books carry today.

Syriac scribes nearby were working on the same problem. By around 600 they used a dot above or below a word to tell apart words spelled alike. The historian Nick Posegay argues that vowel points appeared first in Syriac and soon after in Arabic and Hebrew, and that ideas about them crossed between Christian, Muslim and Jewish scholars.

The language hinge

Here is the verse, Q 9:3. It is worth seeing in Arabic, because the same word appears in it twice.

Near the beginning, the verse speaks of an announcement mina llāhi wa-rasūlihi, "from God and His Messenger." The word is وَرَسُولِهِ, wa-rasūlihi. It ends in -i because it follows "from."

Near the end it says anna llāha barīʾun mina l-mushrikīna wa-rasūluhu, "that God is free of obligation to the polytheists, and so is His Messenger." The word is وَرَسُولُهُ, wa-rasūluhu. It ends in -u because the Messenger is now a second subject, standing beside God.

Without the marks, both words are the same skeleton, ورسوله. Give the second one the ending of the first and the Messenger moves from beside God into the list of those God is free of. That is the misreading the story says Abu al-Aswad heard, and it is the reason the tradition gives for writing vowels down. The whole difference is two small marks: a stroke below the l and the h, or a small curl above them.

A translation simply prints the right meaning. Only the Arabic shows how little ink stands between the two.

What we do not know

The named inventors are traditional. The story of Abu al-Aswad reaches us through later writers, and historians of the script, among them Adam Bursi, treat such origin stories with caution: they explain, after the fact, a process that probably took longer and involved more hands. Datings of the Syriac systems vary from scholar to scholar, and the popular credit given to Jacob of Edessa for the West Syriac vowel signs has been disputed. Radiocarbon dates for early Qur'an manuscripts, such as 568 to 645 for the Birmingham leaves, are ranges with known anomalies, not birth certificates.

And the old habit never died. Hebrew and Arabic newspapers are printed without vowel marks; the marks come back in scripture, poetry, and books for children and learners. Fluent readers today are still doing what Phoenician readers did.

That is the real skill behind reading Biblical Hebrew or Classical Arabic. It is not decoding marks so much as knowing the language well enough to hear what the page leaves out. Both are among PRAVIEL's courses, and that skill is what they are for.

Sources

• Wikipedia, "Greek alphabet"
• Wikipedia, "Mater lectionis"
• Wikipedia, "Masoretes"
• Wikipedia, "Aleppo Codex"
• Wikipedia, "Qere and Ketiv"
• Wikipedia, "Jehovah"
• Wikipedia, "Arabic diacritics"
• Wikipedia, "Abu al-Aswad al-Du'ali"
• Wikipedia, "Al-Khalil ibn Ahmad al-Farahidi"
• Quranic Arabic Corpus, word-by-word analysis of 9:3
• Adam Bursi, "Connecting the Dots: Diacritics, Scribal Culture, and the Qurʾān in the First/Seventh Century," Journal of the International Qur'anic Studies Association 3 (2018)
• Nick Posegay, Points of Contact: The Shared Intellectual History of Vocalisation in Syriac, Arabic, and Hebrew (Open Book Publishers, 2021)
• University of Birmingham, "The Birmingham Qur'an"