Skip to main content
← All posts

Research

Did God hide a code in the Bible?

Count every 50th letter from the start of Genesis and you spell "Torah". We tested the Bible codes the way the Bereans tested Paul, against the text itself. Here is what we found, and why faith does not need them.

Carestead Research7 min read

What we found

  • The famous example is real: in Genesis and in Exodus, every 50th letter from the first "t" spells Torah. By chance that happens about 1 time in 10,000.
  • But it never happens again. With 22 other words, across every pair of books in the Hebrew Bible, the device does not recur once.
  • The number puzzles behave the same way, with one lovely exception: John's 153 fish point to Ezekiel's river of life.
  • God's word does not depend on hidden codes. Its power is in what it plainly says, above all about Jesus Christ.

Open a Hebrew Bible to the first verse of Genesis. Find the first letter tav, the Hebrew "t". Now count fifty letters, and fifty more, and fifty more. The four letters you land on spell תורה: Torah, the name of the very books you are reading.

Turn to Exodus and do it again. It works there too.

This is the most famous of the so-called Bible codes, credited to Rabbi Michael Dov Weissmandl in the last century. Books and documentaries have since claimed that the Bible hides thousands of words at equal spacing, so many that only God could have put them there.1

Why test it at all?#

Because we love the Bible, and claims made in its name deserve care. When Paul preached in Berea, Luke commends the listeners not for believing at once, but for checking what he said against Scripture.

These were more noble than those in Thessalonica, in that they received the word with all readiness of mind, and searched the scriptures daily, whether those things were so.
Acts 17:11 (KJV)

There is a practical reason too. When an argument offered on God's behalf turns out to be weak, the damage rarely stops at the argument. People conclude that the Bible itself has failed. So we asked three questions, in order. Is the example real? Is it rare? And, the one that decides it, does it ever happen again?

Yes, it is real#

In Genesis the tav is the sixth letter of the Bible, and the next three letters fall in verses 2, 4 and 5. In Exodus they fall in verses 2, 5 and 6. Both work whether the text is read as written or as traditionally read aloud. The claim is often extended to the other three books of Moses, and there it gets shaky.

The claim, book by book, in the Leningrad Codex
BookWhat is claimedDoes it hold?
GenesisTorah, every 50th letter from the first tavYes
ExodusTorah, every 50th letter from the first tavYes
LeviticusGod's name, YHWH, every 7th or 8th letterEvery 8th, yes; every 7th, no
NumbersTorah backwards, every 50th letterNo
DeuteronomyTorah backwards, every 49th letterOnly if you start counting in verse 5

Yes, it is rare#

We shuffled the letters of Genesis and Exodus 10,000 times, like shuffling a deck of cards, and looked for the same link at any spacing from 2 to 100. Only one shuffle produced it.2

But there is a catch. When a pattern is found by looking, the odds that matter are not the odds of that exact pattern. They are the odds of finding something, anything, that looks meaningful. Someone browsing the five books for a handful of obvious words, forwards and backwards, has roughly a hundred chances. That brings the odds nearer 1 in 300. Unusual, but no longer astonishing.

The real test: does it happen again?#

A trick shot filmed once proves little. The shooter who can do it again, on a court you choose, proves a great deal. If God had built the text with hidden spacing as a sign, we would expect the device to appear more than once, in places nobody picked in advance.

So before looking, we fixed a list of 24 important words, among them God, Israel, covenant, Jerusalem, holy and peace. Then we checked every one of the 741 pairs of books in the Hebrew Bible for the same kind of link.

It never happened. Not with any of the 22 other words in the Torah, and not between any other two books anywhere. Genesis and Exodus, linked by "Torah", is the only case, in the one word it was first noticed with.

We also simply counted. Across every book, we tallied how often the 24 words appear at spacings from 10 to 1,000 letters, and compared that with how often they should appear given the mix of letters in each book. The Torah carries exactly as many as chance predicts, no more than Chronicles or the Psalms.3

Hidden words found, against chanceHidden words found in each book, divided by the number chance predicts. 1.0 means exactly as many as chance. The five books of Moses sit with everything else.
00.511.5GenesisExodusLeviticusNumbersDeuteronomy1 Chronicles2 ChroniclesPsalmsFound / expected
Hidden words found, against chance
CategoryFound / expected
Genesis1.002
Exodus0.981
Leviticus0.989
Numbers1.004
Deuteronomy1.007
1 Chronicles1.004
2 Chronicles1.012
Psalms1.004

Number puzzles, and a river of life#

Hebrew and Greek letters double as numbers, so every word, and every verse, has a value. Readers have long found striking numbers this way.

Genesis 1:1, "In the beginning God created the heaven and the earth", adds up to 2701, which is 37 times 73: a number times its own reverse. John 1:1, the whole verse, "In the beginning was the Word, and the Word was with God, and the Word was God", adds up to 3627, which is 39 times 93. Two openings, two mirrored numbers.

It is a lovely pair. But across 1,187 chapter openings in the Hebrew Bible the same device turns up 4 times, slightly fewer than the 5 that chance predicts. One of them, 1 Samuel 3:1, has exactly the same value as John 1:1. And John's 3627 depends on a small letter that medieval scribes wrote under one word. The oldest complete copy of John, Codex Sinaiticus, does not have it.4

One number did hold up, narrowly, and it leads somewhere beautiful. After the resurrection, the disciples haul in 153 fish on the shore of Galilee (John 21:11). In 1958 J. A. Emerton pointed to Ezekiel 47, the prophet's vision of a river flowing from God's temple into the Dead Sea and bringing it to life. In verse 10 fishermen spread their nets "from En-gedi to En-eglaim", and that verse holds a word worth 17 and a word worth 153, which is 1 + 2 + ... + 17. Of the 33 verses in the Hebrew Bible about fish or fishing, it is the only one that does.5

And it shall come to pass, that every thing that liveth, which moveth, whithersoever the rivers shall come, shall live: and there shall be a very great multitude of fish, because these waters shall come thither: for they shall be healed; and every thing shall live whither the river cometh.
Ezekiel 47:9 (KJV)

Whatever we make of the arithmetic, the connection itself is no puzzle. Jesus spoke of rivers of living water flowing from those who believe in him (John 7:38), and Revelation closes with the river of the water of life (Revelation 22:1). A net full of fish from the hand of the risen Lord is a picture Ezekiel's readers would have known.

What this means for faith#

Every code here has the same shape: one striking coincidence, found by someone looking for it, and no repeat where nobody looked. We cannot say the pattern at the opening of Genesis and Exodus was not intended; God is free to delight in detail. But a device used only once cannot be told from coincidence, so it cannot carry the weight that the codes books put on it.

We do not think that is a loss. God has not hidden his word in a puzzle only a computer can solve. What he has revealed, he has revealed plainly, in words a child can hear.

The secret things belong unto the LORD our God: but those things which are revealed belong unto us and to our children for ever, that we may do all the words of this law.
Deuteronomy 29:29 (KJV)

The Bible tells us plainly why it was written, and it is not to be decoded. It is to be believed.

But these are written, that ye might believe that Jesus is the Christ, the Son of God; and that believing ye might have life through his name.
John 20:31 (KJV)

Every test here was written down, with its pass mark, before it was run. The full reports, with their methods and every figure in this post, are to be published with Carestead Research's open data.

Notes

  1. 1H. M. D. Weissmandl, Torat Hemed (1958). A statistical study by Witztum, Rips and Rosenberg (Statistical Science 9, 1994) and popular books such as J. Satinover, Cracking the Bible Code (1997), took the idea to a wide audience. McKay, Bar-Natan, Bar-Hillel and Kalai (Statistical Science 14, 1999) answered the 1994 study. ↩
  2. 2p = 0.0002. Letter frequencies predict about 3 in 100,000. ↩
  3. 3Torah books: 0.981 to 1.007 of the expected count. The other 34 books: 0.93 to 1.08. p = 0.51. On shuffled text every ratio fell between 0.99 and 1.011, which checks the prediction itself. ↩
  4. 4Opening device: 4 of 1,187 openings against 5.07 expected, p 0.75 (track 90). John 1:1 counts 3627 only with the iota subscript; without it, 3617 (tracks 73 and 73r). ↩
  5. 5Track 94: p = 0.022, 0.045 after correcting for two tests, against a bar of 0.05. Emerton went looking for a meaning of 153, so the true odds are longer than they appear. The claim that John 21 is built on multiples of 153 does not hold (29th of 260 New Testament chapters, p 0.11). ↩

New posts by email

Product notes and Carestead Research, when there is something worth reading.