Questions and answers
Short, plain answers. Each one links to a longer page if you want the detail.
- What is a Torah code?
- A Torah code is an equidistant letter sequence, or ELS: a word spelled by letters that stand the same number of places apart when the text is written with no spaces. That number is the skip.
- Do Torah codes mean anything?
- We do not say. Torah Codex measures: it finds sequences and shows what chance alone would produce beside each count. Whether a pattern means something is a question the measurement cannot settle, and the published research on it is disputed. Our critique page gives both sides.
- Are Torah codes and Bible codes the same?
- Yes. Torah codes is the Jewish term. Bible code is the name a 1997 book made popular. Both mean equidistant letter sequences in the Hebrew text. See our Bible code guide.
- Is a sequence evidence of anything if it is rare?
- Not by itself. A rare result is expected somewhere if you try enough words, spellings, skips and books. That is why we count your tries and rank results among control texts.
- Can Torah codes tell the future?
- We make no such claim, and nothing here supports one. A search finds what you thought to type. The authors of the 1994 paper rejected the idea that sequences can be read for coming events.
- Which text do you search?
- The Westminster Leningrad Codex, as the Open Scriptures Hebrew Bible publishes it. It has 304,850 letters in the Torah and 1,197,042 in the Tanakh. We search the 24 books in the traditional order.
- Why 304,850 letters and not 304,805?
- 304,805 is the count of the Koren edition, which the 1994 paper used. Editions differ by a few letters in full and defective spellings. Our text has 45 more in the Torah. We do not call either count the correct one.
- Why does the edition matter?
- One letter added or removed inside a sequence's span moves everything after it, so the sequence is gone. A result is a statement about one text, and every page that shows a count says which text was counted.
- What about ketiv and qere?
- We search the ketiv, the letters as written. There are 1,254 places in the Tanakh where the traditional reading differs from the written form. The verse panel shows the reading.
- How are final letters handled?
- A final letter counts as its regular form: ך as כ, ם as מ, ן as נ, ף as פ, ץ as צ. Your term is folded the same way.
- What is a skip?
- The distance between letters of a sequence. At skip 50 you read every 50th letter. Skip 1 is the text as written, so searches begin at skip 2 unless you ask otherwise. Our guide to skips has more.
- What does a negative skip mean?
- The sequence reads against the text. At −50 each next letter stands 50 places earlier.
- How is the chance figure computed?
- For each skip, we multiply the places a term can start by the product of its letters' shares of the scope, and add that up over the skips and directions you searched. The method page works through an example by hand.
- What are shuffled texts?
- We mix the letters of the scope with a numbered seed and run the same search again. Many runs show what the search does on a text with no order. It is one of several controls, and the controls page explains them all.
- What is the trial counter?
- It counts each term, spelling, skip range, scope and matrix width you try in a session. A best-looking result out of 20 tries is not the same as one try, and the counter says so beside every figure.
- What does locking a search do?
- You fix the terms and settings before running them. The laboratory then runs that exact search once, and the link and export carry a badge saying it was locked first. This is the fairest way to test an idea.
- What is the WRR experiment?
- A 1994 paper by Witztum, Rips and Rosenberg in Statistical Science. It tested whether rabbis' names and dates stand unusually close together as sequences in Genesis, and reported a rank of 4 in a million. We explain it here.
- Did anyone answer that paper?
- Yes. A 1999 reply in the same journal argued that the result came from choices of names and date forms. It showed a list with a similar score in War and Peace and found nothing in new replications. Proponents dispute it. A 2004 committee's two tests also came back at chance level, with two members dissenting.
- Can I search the Divine Name?
- Yes. It is searched on its true letters. By default it is displayed in a reverent form in results, exports and links. You can change this in Settings. We never write the Name in English letters.
- Can I search English words?
- You can type English and choose a Hebrew spelling from a list. What is searched is always the Hebrew. A tag on each line says where the spelling comes from, and sound-only spellings are labeled as transliterations.
- Can I search dates?
- Yes. Type a date and the laboratory converts it to the Hebrew calendar. It offers each written form, labeled. Each form counts as a separate try. The dates guide explains the forms.
- How long can a search term be?
- From 2 to 12 letters, with up to 6 terms at a time.
- Where does the search run?
- On your device. The engine works in your browser, and in our apps, over the whole letter stream, so every count is exact and no server does the searching.
- Why does every letter link to a verse?
- Because the text is the point. Tap any letter of a result and its verse opens in the reader, with that letter marked and the JPS 1917 English beside it. Back to your result returns you to the same matrix.
- Why JPS 1917?
- It is a Jewish translation in the public domain. We use only Jewish translations.
- Do you show matrices about current events or living people?
- No. Our gallery holds classic observations and published experiments. You can search what you like, but we do not publish that kind of matrix.
- What is free?
- The Tanakh reader, the gallery, the guides and the term pages are free to everyone. Searching in the laboratory needs a free account, which comes with 10 searches. Any search you have already run stays open at no charge, for good. Every feature is included.
- What does Premium add?
- Unlimited searches, on the web and in the apps, on one account. See pricing.
- I think a number here is wrong. What should I do?
- Send us the share link. A share link carries the whole search, so we can repeat it exactly. If a figure on this site is wrong, we want to fix it.
Sources
- Doron Witztum, Eliyahu Rips and Yoav Rosenberg, "Equidistant Letter Sequences in the Book of Genesis", Statistical Science 9(3), 1994, 429-438.
- Brendan McKay, Dror Bar-Natan, Maya Bar-Hillel and Gil Kalai, "Solving the Bible Code Puzzle", Statistical Science 14(2), 1999, 150-173.
- The text: the Westminster Leningrad Codex, as published by the Open Scriptures Hebrew Bible project.
- The English: The Holy Scriptures According to the Masoretic Text, Jewish Publication Society, 1917.
Checked against its sources on 2026-10-31.