There was an experiment published in a journal called Statistical Science in 1994 that found that certain medieval Rabbis and their birth & yahrtzeit are referenced to in sefer Bereshit by running a search through equally-spaced letters(if I recall correctly).

I have seen many similar codes that produced surprising results that lead me to wonder if they may be considered valid as scientific proof that the letters in the Torah are not placed randomly since the findings are highly improbable according to the rules of probability.

I would like to know if anyone has a way of putting these or any Torah codes to test in a mathematical equation to see if the results are statistically significant(a probability of less than 5% to occur randomly). Does anyone have a way of independently testing such data (codes) mathematically to prove or disprove their statistical significance? Leaving beliefs aside and letting the mathematics speak for themselves.

  • 3
    Which version of the Torah did they use?
    – Double AA
    Sep 16, 2018 at 12:57
  • 1
    Comments are not for extended discussion; this conversation has been moved to chat.
    – Isaac Moses
    Oct 3, 2018 at 5:01
  • The short answer is that the codes were supported by real mathematicians, and were also attacked by real mathematicians. The rest of us have trouble following their discussions but they are available if you want to study them.
    – MichoelR
    Jul 13 at 16:19
  • I myself lost interest in the topic 40 years ago when a Torah Codes presenter mentioned that other religions have found "interesting results" in their own holy books.
    – MichoelR
    Jul 13 at 16:20

5 Answers 5


I think there’s a bigger problem with Torah Codes than the one Al Berko mentioned; namely, even if we were guaranteed accuracy in our Sifrei Torah, we would have this problem.

That problem is that there are several ways to refer to things, any number of letters to start from, and any number of letters to skip. In other words: you can’t prove the Divinity of a text using criteria which can be completely arbitrarily filled in.

Take, as an example, 9/11. The code hinting to 9/11 spots the word התאום, “twin,” if you count from the letter ה in לעיניהם, Bamidbar 20:8, every 36 letters; מגדלי, “towers,” if you count from the letter מ in למות, Bamidbar 20:4, every 71 letters; מטוס, “airplane,” if you count from the letter ם in ובעירם in Bamidbar 20:11, every -33 letters (oh, I forgot to mention that counting backwards is also legal); and מתקפת, “attack,” if you count from the letter ם in פעמים in Bamidbar 20:11, every -144 letters.

So let me get this straight. It says “twin,” “towers,” “airplane,” and “attack,” if you squint hard enough, so that must be a reference to 9/11 and therefore the Torah is from G-d?

Here’s the flaw with this argument. What are the words chosen? התאום and מגדלי - which I assume are meant to be read together as מגדלי התאום, “twin towers.” Why are they referred to as such, and not, say, מגדלים תאומים? Or מגדלי התאומים? Or some other conjugation? Well, it’s because that’s how it’s referred to in Hebrew - Modern Hebrew, which certainly doesn’t have any Divinity.

What about מטוס? Well, once again, that’s the wording used in Modern Hebrew for airplane. But why not use מטוסים?

The final word was מתקפת. Why not use מלחמה, “war”? Or הכאה, “smite”?

Okay, that’s cute, “it’s in Modern Hebrew, therefore it doesn’t count.” But the whole point is that it’s predicting the future - even using the language spoken then! Tell me, then: the attack was in America. Why not refer to it in English? Instead of using מטוס, transliterates it as אירפלאין (or whatever other combination you feel like using), and figure out how that fits in.

The answer, of course, is because that was the only way to make it work.

There’s more to 9/11 than the fact that airplanes crashed into the Twin Towers in an attack. I don’t see anyone conjuring up י״א ספטמבר ב׳א (“September 11, 2001”), which might also be referred to as עשתי עשר instead of י״א, or שני אלפים ואחד instead of ב׳א, or, for that matter, why not use the Julian date, or, better yet, the actual Hebrew date? What about the attack on the Pentagon, or the plane that crashed in Pennsylvania, or Al Qaeda, or Osama bin Laden, or, or...

The reason those are ignored is because they didn’t show up. So we handwave it, point to the few words that we got to work, and shout jubilantly in the streets, “The Torah is Divine!”

I don’t see anyone doing the same experiment on any other sufficiently long text (like, say, the King James Version, or all the works of Shakespeare lined up in chronological order of how he wrote them). You’d eventually be able to do the same thing; if there are enough variables to tweak, you will be able to make it work.

As the old quote says, “If you torture the data, they will confess.”

  • 1
    Comments are not for extended discussion; this conversation has been moved to chat.
    – Isaac Moses
    Oct 3, 2018 at 5:03

No, the Torah Codes have no merit.

There are two core issues:

1) Statistical Improbability - On the surface it does seem to be highly improbable to find hidden "codes" through the use of Equidistant Letter Sequencing" (ELS). The issues is that these same "miraculous finds" have been duplicated using Moby Dick.

2) ELS Coding only works if the text is 100% immaculate. Even one letter off will ruin the entire sequencing. Think of shifting each box in crossword puzzle over one space - each word/sequence will fall apart.

The belief that the Torah's text has not undergone even the slightest alteration is an untenable position.

So why does it seem to work? It's called data dredging.

Data dredging is defined as: "The misuse of data analysis to find patterns in data that can be presented as statistically significant when in fact there is no real underlying effect."

A great article was written on this folly; Bible Codes: Enigma for Dummies.

See also Scientific Refutation of the Bible Codes.

  • I was unaware that these codes could be derived from Moby Dick. Now I know not to take these code “scholars” seriously.
    – Jonathan
    Oct 31, 2019 at 11:55
  • @Jonathan while I have generally concluded like this answer, I wouldn't write it off based on that. Best to look at the research
    – Rabbi Kaii
    Jul 13 at 13:34

No. The Torah codes and any books about them are fruitless as the Torah text has been changed by many translations. True, that these are small variations and only instances where one letter might have been changed. But by default, if just one letter is changed, it messes with the entire process and forfeits any codes which might have otherwise been derived.


Thus, I would caution anyone delving into the scheme of Torah codes. They do not exist because translation is bound to mess up at some point. Of course, these are small, minor variations and do not change the plain meaning of the text but if one letter is off, the entire process is fruitless. If they were codes, we would not be able to derive any benefit from them now.



Our TorahBibleCodes ELS Search Software is available for FREE DOWNLOAD via the web site https://TorahBibleCodes.com and provides Free, Open-Source Python ELS Software Research Tools to make possible for others to confirm and/or extrapolate continued research in this field and academic discipline(s) by providing fully-customizable, expandable, scalable, free software in the Python language. The precise mathematical data that is produced-by means of standardizing both identification and measurement-allows for scientific, academic peer-review through standardization of test variables, test-measurements, units, IDs, and controls across software tools and experiments for scientific, peer-reviewed, academic biblical research.

Offered here is mathematical proof that G-D exists. For anyone who understands (or interested in) statistics, chance, coincidence – and that things just don't happen randomly – the findings visually presented here are both mathematically precise and scientifically confirmable. All the research data and findings presented here are mathematically, statistically, as well as visually-verifiable.

TorahBibleCodes: Free, Open-Source Python Equidistant LetterSequences (ELS) Hebrew Bible Search Software:

ACADEMIA: https://www.academia.edu/104334275

LINK TO RESEARCH DATA: https://github.com/TorahBibleCodes/ResearchData_KorenCodexVsLeningradCodex

  • 2
    Other than a flat "yes," this post doesn't appear to address the question substantively. Could you please edit to reduce the ad copy and increase support for the assertion that addresses the question?
    – Isaac Moses
    Jul 13 at 12:59
  • 2
    Send this to xAI, quick, so it can explain to everyone the proof impartially. As @IsaacMoses said, could you work in the proof to the answer itself, or at least a summary of it and you'll benefit us all
    – Rabbi Kaii
    Jul 13 at 13:30
  • We have pioneered the analysis of the two codices (Leningrad vs. Koren). We have provided the data of the exact differences between the two codices; one concludes how reliable were the Hebrew scribes; the differences are only matres lectionis which are trivial since these are the only differences between the two codices. This answers the question with scientific, mathematical proof: Jul 13 at 17:00
  • Research software, data are provided as well as accompanying academic “white paper” to explain the software and data ID and measuring system. Anyone can search and share data for peer-review and community building. Jul 13 at 17:04
  • Please take a look at our Help page on "How not to be a spammer". Your recent posts include one very on-point answer and a few others, including this one, that appear to be more focused on self-promotion than on answering the question at hand.
    – Isaac Moses
    Jul 13 at 17:53

I think the problem covered by my answer is far more serious refute of the research than that of DonielF.

  1. (My comments contd) The Torah scrolls that we have today (circa 2000 years ago) are far from being the exact replicas of the Torah scrolls at least from the times of the Sofrim (circa 2400 years ago), the fact invalidates the very basis for research. The Sages claim it openly and indisputably in the Gemmorah in Kiddushin 30a brings it:

"בעי רב יוסף וא"ו דגחון מהאי גיסא או מהאי גיסא א"ל ניתי ס"ת ואימנינהו מי לא אמר רבה בר בר חנה לא זזו משם עד שהביאו ספר תורה ומנאום א"ל אינהו בקיאי בחסירות ויתרות אנן לא בקיאינן

... They did not move from there until they brought a Torah scroll and counted the letters? Therefore we can do the same. Rav Yosef said to them: They were experts in the deficient and plane forms of words and therefore could count the letters precisely. We are not experts in this regard, and therefore we would be unable to resolve the question even if we were to count the letters.

As the passage shows, the Torah scroll of the Sages was different from that of the Sofrim at least concerning full and deficient writing, that would changes the codes (if anything) entirely. (I'm not mentioning that "ו דגחון" has moved some 5000 letters!)

Furthermore, Meiri and Rashb"A in place explain that our books went thru many many iterations, even if the changes were not as great as in the days of the Second Temple, and the final version of today was finalized as late as some 1000 years ago.

Read here about Masora iterations and changes. I don't even mention the revolutions of the Torah as in Sanhedrin (21b):

"אמר מר זוטרא, ואיתימא מר עוקבא: בתחלה ניתנה תורה לישראל בכתב עברי ולשון הקודש, חזרה וניתנה להם בימי עזרא בכתב אשורית ולשון ארמי"

My point is that the codes can only be relevant if we have a firm tradition of never-changing scrolls, which we don't. Even the small differences (today it's 9 letters per scroll) make the whole approach invalid.

I would also want to address the true concerns of DonielF (in comments) that this approach might seem problematic regarding the Divinity of our Torah scrolls, but I suggest that the very same way we believe in the Divinity of the Oral Torah thru the Mishnah and the Talmuds we should believe in the Divinity of the Written Torah that the same Sages left us.

Please, don't kill the messenger, but I think we should wholeheartedly admit the above-mentioned point and stop arguing on the authenticity of our Torah scrolls.

NB: My unanswered question anti-altering-measures-in-the-written-torah addressed exactly that point: how come that the Torah did not include any measures for anti-altering? Since it didn't there's no way of telling its authenticity, which supports my point above.

  • 1
    Comments are not for extended discussion; this conversation has been moved to chat.
    – Isaac Moses
    Oct 3, 2018 at 5:04
  • ו דגחון does not mean the middle letter in the Torah, it means the middle big/small letter in the Torah.
    – user25726
    Jun 15, 2021 at 21:30

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .