The Bible as a Science Text

Book Reviews by Matt Young

Department of Physics

Colorado School of Mines

Golden, Colorado 80401

For a book and more articles, return to

Matt Young's home page

Posted November 20, 2001.

The Genesis Question:

Scientific Advances and the Accuracy of Genesis

(a book by Hugh Ross; Navpress, Colorado Springs, Colorado; 1998)

reviewed by Matt Young

Matt Young is Adjunct Professor in the Department of Physics at the Colorado School of Mines and a former Physicist with the National Institute of Standards and Technology. He is the author of a skeptical book on science and religion.

[Originally published in Rocky Mountain Skeptic, July, 2000, pp. 2-4. reprinted in shortened form as "Specious Arguments: Twisting Scientific Theory and the Bible," Skeptical Inquirer, March-April, 2001, pp. 51-52.]

Ross earned a PhD in astronomy from the University of Toronto and practiced for a short time as a physicist. But soon he became more interested in the Bible and founded Reasons to Believe, a proselytizing religious organization. Ross's thesis, like that of Gerald Schroeder,1 is that the Bible is literally true in all details. Ross and Schroeder are more sophisticated than most Biblical literalists: Instead of perpetrating the lies that evolution, geology, and the big bang theory are wrong, they accept these theories, with their own variations, and instead try to force agreement between the Bible and scientific theory.

Ross begins by knocking down a straw man: That most people have heard of the Bible but have never read it. They have heard of contradictions but cannot actually cite one. I do not know whom Ross has been talking to, but contradictions in the Bible are invisible only to those who have eyes but see not. Just one: Noah took two of each kind into the Ark (Genesis 6:19, 20); no, Noah took seven of each clean animal into the Ark (7:2, 3); no, Noah took two of every kind (7: 14, 16).

Ross's critics think, for example, that the Bible teaches the flat earth, geocentrism, male superiority. Well, doesn't it? As I noted in the Schroeder review, I have found in the Hebrew Bible about a dozen references to the heaven above and the earth below, the ends of the earth, the corners of the earth, and the length of the earth. Consider for example Daniel 4:20, "The tree that thou sawest, which grew, and was strong, whose height reached unto the heaven, and the sight thereof [reached] to all the earth." Can you build a tower so tall that it can be seen from anywhere on earth? Not unless the earth is flat. Or maybe concave.

Similarly, "And the Lord God said, It is not good that the man should be alone; I will make him an help meet for him" (Genesis 2:18) strongly suggests subservience of women. Ross is at pains to argue that the Hebrew word 'ezer means an ally, but the root meaning is to help or assist.

Ross's book is full of hackneyed old probability arguments, similar to Schroeder's. For example, he claims without evidence that life must have arisen at least 50 times between 3.86 and 3.5 billion years ago. Therefore, it must be easy to create life. Why then can we not manufacture even a single strand of DNA? Ross thinks that the earth is so finely tuned that life could not have arisen here accidentally. The argument ignores that there might be billions of other planets that are not quite so finely tuned and on which life did not arise. The odds that life will arise on any one planet are indeed small, but the odds that life will arise somewhere may not be so small, given the vast number of somewheres. Ross displays all the naiveté of the journalist who quotes the odds that a woman from New Jersey will win the lottery twice, when the relevant probability is the odds that someone, somewhere will win twice. Given the vast numbers of lottery tickets sold, those odds are actually fairly good.

Ross accepts uncritically the Bible's claim that lifetimes were typically 900 years around the time of Adam. Indeed, people were forbidden to eat meat because toxins contained in the meat, but not in plants, would have poisoned them well before their allotted 900 years were up. Evidently God was unaware at the time of the carcinogens in mushrooms.

What changed things, so that our lifetimes are now a mere 120 years? The Vela supernova. Ross dates the Vela supernova at 20,000 to 30,000 years ago, a bit early for Adam or Noah, but who's counting?

According to Ross, the Vela supernova bathes the earth in cosmic rays to this day and is the major source of cosmic radiation. Cosmic rays are one cause of cancer, so God adjusted the rate of programmed cell death, or apoptosis, to protect us from cancer. The adjustment incidentally shortens our lives, but God did that deliberately, because people had become wicked and short lives limit the damage that can be done by a single wicked person. Thus does Ross think he has demonstrated the plausibility of the 900 year lifetimes in the Bible. The argument would be more convincing if cosmic rays were the major source of radiation to the gonads; they are not. At sea level, cosmic rays account for roughly 25 to 40 % of the radiation exposure to the gonads.2

Ross cites the discovery of circumstellar disks around ten or more stars as evidence that the primordial earth had an opaque atmosphere. Indeed, he notes that planets have been discovered around several stars and wrongly makes it appear that the planets have been observed directly and their atmospheres studied ("[t]he theory and observations [my italics] both confirm that all planets start with opaque atmospheres"). Thus, he considers the earth's transparent and rarified atmosphere a mystery: In general, he claims, as a rule of thumb (but with no reference), the farther from the sun and the heavier the planet, the thicker the atmosphere. Since Venus's atmosphere is very thick, Earth's should be thicker. The mystery is solved by the Moon miracle: The Moon crashed into the Earth about 4.25 billion years ago, cleared the atmosphere of dust and debris, and let the light from the sun through. It is at that time that God said, "Let there be light."

The rule, the farther from the sun, the thicker the atmosphere, however, is not true in the solar system. Venus has a very dense atmosphere owing to a runaway greenhouse effect that is fairly well understood. Presumably, Earth's atmosphere should be denser, since the two planets are roughly equally massive. Mars is substantially smaller than the Earth, but farther from the sun; why is its atmospheric pressure less than 1 percent of Earth's anomalously low pressure? Simply put, because the histories of the three planets are very different. There is no mystery as to why the Earth's atmosphere is less dense than that of Venus, and we need not invoke a Moon miracle to explain a nonexistent mystery.

Ross relies occasionally on exact translations from the Hebrew, yet his understanding of Hebrew seems limited to isolated words. For example, he translates sheretz nephesh chayyah ve'oph, a swarm of living (or wild) creatures and birds (or fowl), as if it were punctuated sheretz, nephesh chayyah, ve'oph, swarming things, living creatures, and birds that fly. That is, he seems unaware of the construct case in Hebrew, in which two consecutive nouns act as a prepositional phrase in English: sheretz nephesh chayyah = a swarm of living things, not swarming things and living things. The point is important because, later, he uses nephesh to mean "soulish creature," or higher animal that can show emotion, whereas the text makes clear that the waters also swarm with creatures with nephesh. Thus, nephesh need not mean soulish creature but rather any creature with the spark of life.

On the sixth day, God created chayyat ha-aretz, behema, and remes ha-adamah, animals of the earth, livestock, and animals that creep on the earth. Ross argues, incorrectly, that chayyah necessarily means long-legged land mammal and therefore that remes must necessarily refer to short-legged land mammals such as rodents, hares, and armadillos. Such a translation is a stretch that is required by Ross's insistence that God is leading up to the creation of humans, and is typical of literalists' thinking: force the translation to agree with your preconceived notion, no matter what.

The myth that God created Eve from Adam's rib is too much even for Ross. Ross claims that it is based on a mistranslation. In fact, says Ross, God took a biopsy from Adam and used genetic engineering to create Eve. He'd need more than "just a few million modifications here and there," however; he'd have to add a whole X chromosome to a genome that contained only one X chromosome and one Y chromosome with relatively little genetic material. If God did it Ross's way, he should have made Eve first, then taken a biopsy from Eve and removed a few million base pairs to make Adam.

In any case, that's not how God did it, and Ross is plainly wrong. The text states in so many words that God removed one of Adam's ribs and filled the void with flesh. You don't need to fill a small biopsy with flesh.

Speaking of DNA, Ross uses "mitochondrial Eve" to verify the date of Biblical Eve. Mitochondrial Eve is a hypothetical female who is the putative ancestor of all living women. Her existence is inferred from studies of mitochondrial DNA, a form of DNA that is passed through the mother only.

Less well known than mitochondrial Eve is what I will call "chromosomal Adam," the putative ancestor of all living men. Chromosomal Adam has been inferred from studies of the Y chromosome and is analogous to mitochondrial Eve.

Ross dates chromosomal Adam to 35,000 to 47,000 years before the present; mitochondrial Eve, a few thousand to a few tens of thousands of years earlier. He explains this discrepancy by noting that all the men on the Ark were related, whereas the women were not. Thus, the most recent common ancestor of living men was Noah, whereas the female lineage goes through the wives of Noah and his sons to Eve.

The existence of mitochondrial Eve is still controversial but even so does not imply a time when there was only a single woman. At most, we can say that there was once an evolutionary bottleneck, a time when there were only a relatively few women, and that since then all lineages but mitochondrial Eve's have died out. Mitochondrial Eve may or may not be the ancestor of all living women, but she was certainly not the ancestor of all women who ever lived. Nor was chromosomal Adam the ancestor of all men who ever lived.

Ross knows or ought to know all this, but his gullible readers do not, and he exploits them with many such specious arguments.

What is truly appalling about this book, however, is the cavalier way in which Ross justifies mass murder and, indeed, genocide. Instead of a primitive people callously using God as justification for slaughtering their enemies and furthering their conquest, Ross sees God using a holy people to make surgical strikes against "reprobation." Ross assures us that reprobation abounded. When reprobation is bad enough, it infects everything around it, even the soulish animals. Everything has to be destroyed, much as a surgeon removes cancerous tissue. The cancerous tissue, however, consists of humans beings: usually Israel's enemies but in the Noachian flood, everyone on earth except a handful of people. Most are guilty, if at all, by association. I could not help wondering, what will happen if Ross and his coreligionists decide that Jews or skeptics or liberal Protestants or critical book reviewers are reprobate and need to be excised?

Many thanks to Mike Grant of the University of Colorado for checking my biology, Shirley and Gideon Weisz for checking my Hebrew, and Lowell Yemin for supplying Reference 2.

Copyright © 2000 by Matt Young. All rights reserved.


1. Young, Matt, 1998, "The Bible as a Science Text," Rocky Mountain Skeptic, November/December, pp. 2-4; review of Schroeder, Gerald L., 1997, The Science of God, Free Press.

2. K. Z. Morgan and J. E. Turner, Principles of Radiation Protection, Krieger, Huntington, N. Y., 1973.

The Bible as a Science Text

(a book review)

-Matt Young

[Originally published in Rocky Mountain Skeptic, November, 1998, pp. 2-4.]

The following is a review of the book The Science of God, by Gerald Schroeder, Free Press, New York, 1997. Matt Young is a physicist who lives and works in Boulder, belongs to two synagogues, and has completed a manuscript on science and religion.

In the opening passages of an earlier book, Genesis and the Big Bang, Gerald Schroeder makes crystal clear that his intention is to reconcile the creation myth in Genesis with modern science, not to examine the truth of either. For example, Schroeder, a mathematical physicist, describes relativistic time dilation in layman's terms and then states flatly that the six days of creation are six days when viewed from the proper frame of reference but are billions of years when viewed from earth. He provides no argument to support this claim but rather seems to think that stating it is enough.

Readers of Rocky Mountain Skeptic may recall Gerald Schroeder from my article "Did God Write the Bible-or Was It a Computer?" in the March-April issue. Schroeder is the author of one of the numerological arguments that tries to find hidden meaning in samples of letters from the Hebrew Bible.

In spite of my inauspicious introduction to his works, I decided to read Schroeder's newest book, The Science of God. His agenda is the same: to find harmony between the Bible and other ancient Jewish texts, and modern science.

Schroeder has a PhD from MIT and teaches at the Weizmann Institute of Science in Israel. The Weizmann Institute may not be familiar to American readers; it is a graduate research institution similar to the Rockefeller University in New York.

Schroeder states at the outset that he will examine modern scientific texts and ancient religious texts and try to bring them into agreement. If that sounds sensible to you, try turning it around: What if he was going to limit his study to modern theological texts and ancient scientific texts and try to bring them into agreement? You would very possibly recommend that he extend his study to include modern scientific texts. But Schroeder evinces no interest in religious studies later than those of the Jewish sage Nachmanides (1194-1270).

If you go into an investigation with what I will call a thesis, you will very probably prove that thesis to your satisfaction, whether or not the evidence would be convincing to a more or less neutral observer. Schroeder seems a lot like a man with a thesis: that the account of the creation in the Bible is literally true. In support of his thesis, he latches very firmly onto Big Bang theory but rejects evolutionary biology. To my mind, Big Bang theory, although it is probably right in outline, is less well supported by evidence. It is in the last analysis an extrapolation to densities and temperatures at which our physics has simply never been tested. Extrapolating general relativity too far back in time and therefore density may be like extending a gas law to temperatures or pressures at which the substance has liquefied. The law breaks down and becomes completely inapplicable. Evolutionary biology, by contrast, is established fact, even if there is yet some controversy about the details. Ultimately, as we will see, not even Schroeder can escape the fact of evolution.

Schroeder believes that the first chapter of Genesis accurately describes the creation as viewed from the proper relativistic frame of reference. He calls time as seen from this frame of reference cosmic time. Cosmic time is not entirely arbitrary but is based on a philosophical argument developed by Jean-Marc Lévy-Leblond. According to Schroeder, Chapter 1 of Genesis was written from the point of view of a being living in cosmic time. Only at the end of Chapter 1 is the clock turned over to earthly beings. Specifically, Schroeder argues that the first day of Genesis corresponds to the first 8 billion years of earth time, the second day to 4 billion years, and so on.

The correlation between cosmological and paleontological fact and the six days of creation is fairly good, but there are problems. In the Book of Genesis, there is water above the sky; dry land and plants are created before the sun; and fowl before reptiles. Schroeder notes that flowering plants would not have appeared on the third day but on the fifth, but solves his dilemma with an ad hoc argument taken from Nachmanides: the plants developed during the next days. What did they develop from? Single-celled plants. Sounds a good bit like evolution.

Similarly, there is no geological evidence for a Noachian flood, so Schroeder applies a textual argument to suggest that the flood was only local. In brief, he notes a change in terminology from eretz to adamah. These words mean earth or land and, in context, are synonyms, but to Schroeder they signify that God changed his mind about a worldwide flood. Finally, at the beginning of the sixth day, the fossil record shows a major mass extinction, but it is not mentioned in Genesis, and Schroeder all but ignores it.

Though he admits to microevolution (changes within a species), Schroeder does not believe in macroevolution (speciation). To "prove" that speciation is not possible, Schroeder carries out a back-of-the-envelope calculation to suggest that the required number of mutations could not take place in a short enough time. The calculation uses a simple probabilistic argument and assumes that a fortuitous mutation takes place every so many generations. It is flawed by the assumption of independent probabilities, but it has other problems as well.

Schroeder is correct in that some biologists argue that evolution takes place as the result of fortuitous mutations. Others, however, argue that at least some speciation is the result of what Schroeder calls microevolution operating on isolated or stressed populations. Computer simulations that are far more sophisticated than any back-of-the-envelope calculation suggest, for example, that an eye can easily form from a light-sensitive spot within a finite time. Schroeder is nobody's fool, and he surely understands evolution, so I find it hard to believe that he is wholly ingenuous describing evolution as simplistically he does.

Schroeder seems positively triumphant when he discovers that eyes may not have evolved independently several times. The discovery allows him to argue that eyes did not evolve in the sense that evolutionists mean "evolve" but rather that the deck was stacked from the beginning. It would be harder to argue in this way if we thought that eyes had evolved several times, rather than once, because then there would have had to be several stacked decks.

Specifically, researchers have discovered similar genes in the visual systems of vertebrates and fruit flies. When the discovery was made, however, long-lived gene clusters called homeoboxes had been known for approximately a decade, so the discovery of the related genes was not exactly a bolt from the blue. In addition, many phyla have light sensors if not actual eyes, and most or all of these rely on similar light-sensitive molecules. There is no reason not to believe, therefore, that the light-sensing chemicals in most or all eyes have a common origin.

The differences among the structures and the developmental pathways among eyes in different phyla, however, suggest that the eyeballs themselves do not share a common origin and that eyes indeed evolved several times. For example, in some phyla, the light-sensing layer lies above the nerves and blood vessels; in others, including ours, it lies below the layer that contains the nerves and blood vessels. In particular, in the human eye, the light has to travel through a layer of blood vessels and nerve cells before it hits the retina - not exactly my idea of a thoughtful design, by the way.

Schroeder is well aware of homeoboxes and uses them to suggest that evolution has been directed in the sense that the homeoboxes were prepared for precisely the purposes to which nature later put them: a typical God-of-the-gaps argument (an argument that points to gaps in our knowledge and attempts to fill those gaps by appealing to a creator). Unfortunately, he does not develop the argument further, but rather is content attacking sophisticated computer simulations with back-of-the-envelope calculations. Not even Schroeder can escape the fossil record, however, and he clearly indulges in a form of evolution, even while pretending that he is not. Indeed, Schroeder is performing a service in that he may be helping to drag other Biblical literalists into the twentieth century; it would be churlish of me to complain that it is only the first half of the century.

Schroeder fixes on the absence of transitional forms in the fossil record and suggests that such gaps disprove evolution. One transitional form that excites him, however, is Archeopteryx, a birdlike creature with teeth. He associates Archeopteryx with tinshemet, which is categorized both as a bird (Leviticus 11:18) and as a reptile (11:30). The Jewish Publication Society renders the two instances as horned owl and chameleon but notes that many of the words in this section have been lost and their meanings are uncertain. Eliding quickly from speculation to fact, Schroeder informs us that an understanding of Archeopteryx/tinshemet may provide an insight equivalent to the Rosetta stone. There are, however, many intermediate forms, as long as you do not demand an intermediate between distantly related species, like horse and human. Indeed, the progression from eohippus to modern horse, which is depicted in many textbooks, shows a series of intermediate forms. If you examine the DNA of different species, you find practically a continuum of intermediate forms.

The Hebrew word for soul is neshamah. (It may have the same root as tinshemet; what are we to make of that?) Schroeder believes that Adam and Eve were the first humans in the sense that they were the first hominids to be endowed with a neshamah. He does not deny the existence of prehuman and therefore pre-Adamic hominids whose fossils are indistinguishable from those of humans. Only after hominids were given souls did they become humans and develop writing and civilization. Part of the evidence is from the language of the Bible. For example, Genesis 2:7, "And the Lord God formed man of the dust of the ground," spells "formed" VYYZR (va-yiytzer; Schroeder incorrectly transcribes it as ya-tsar), whereas Genesis 2:19, "And out of the ground the Lord God formed every beast of the field," spells it VYZR (va-yitzer). The Hebrew letter Y (yud) is the first letter in YHVH, the name of God, and is often used as an abbreviation for God. According to Schroeder, it is put into the verb in 2:7 to show that humans have an extra spiritual input. If you had asked me, I might have noted that the Bible was written over a long time and then copied over and over by hand. Post-Biblical Hebrew was developing a system of using Y and V as vowels, and possibly one scribe or another wrote the first instance of VYYZR with the second Y to signify the vowel. Contrary to what you commonly hear, the Hebrew Bible was not stabilized until after the invention of the printing press. But Schroeder has no apparent interest in modern Torah scholarship. Instead, he scrutinizes the text with the dedication of a Stalinist apparatchik combing the Daily Worker for clues about the present party line and gives deep significance to what are probably no more than scribal errors or inconsistencies.

The remainder of the book concerns itself with free will and theodicy. Schroeder uses the fairly standard kabbalistic argument that God withdrew from the world in order to give us free will. Schroeder's answer to the question, "How can we have free will if God knows the future?" is clever but unsupported in any way: God lives in Eternity; Eternity is timeless, so everything happens at once from that vantage point; and the photon, for which time does not exist, is the link between time and Eternity. Why the photon? No answer; again, he seems to think that stating it is enough.

Using the Bible to "prove" or "verify" modern science is a dangerous game, and two can play it. For example, 1 Kings 7:23 says, "And he made a molten sea, ten cubits from the one brim to the other: it was round all about, ... and a line of thirty cubits did compass it round about," with almost identical wording in 2 Chronicles 4:2. Was pi equal to 3 in the days of Solomon?

Similarly, I have found in the Hebrew Bible about a dozen references concerning the heaven above and the earth below, the ends of the earth, the corners of the earth, and the length of the earth. The most telling is perhaps Daniel 4:20, "The tree that thou sawest, which grew, and was strong, whose height reached unto the heaven, and the sight thereof to all the earth." Was the earth flat and rectangular in those days?

I am not so foolish as to believe that such inaccuracies "prove" that the Bible is wholly wrong or even untrustworthy. A great deal of modern scholarship and archeology have established the rough accuracy of, let us say, those parts of the Hebrew Bible that postdate David or perhaps Deborah. Many of the earlier portions may be accurate in outline as well. Other parts, such as the Books of Esther, Jonah, or Job, are fictions designed to make a point. Yet other parts, such as the Song of Songs, are poetry. And other parts, such as the first chapter of Genesis, are allegory. None of these observations diminishes the Bible as a historical or religious document. What diminishes the Bible is the insistence, against all odds, that every word in it is literally true and that God did not have the imagination to write allegorically.

Acknowledgement: Thanks to Gideon Weisz and Michael Grant for reading the manuscript and correcting some inaccuracies. Carla Selby edited the final version and improved the clarity of several paragraphs.

Copyright © 1998 by Matt Young.

The Bible as a Science Text

(Correction and Addendum)

- Matt Young
[Originally published in Rocky Mountain Skeptic, September, 1999, p. 2.]

In my review, "The Bible as a Science Text," in the November issue of Rocky Mountain Skeptic,1 I was too quick to dismiss Gerald Schroeder's claim that many vertebrate and invertebrate eyes share a common origin. In fact, the development of eyes in animals as distinct as vertebrates, cephalopods, and insects may be governed by the same gene, a gene known as Pax-6. If that is so, then eyes may not have evolved independently in different phyla, as has been commonly thought.2, 3 A biologist I consulted tells me, however, that the existence of a common early gene (Pax-6 in this case) does not necessarily demonstrate common ancestry and that you can still make a sensible argument that eyes developed differently in different phyla, even though they all use the Pax-6 gene. (I immediately thought of different cultures using mud to make bricks, even though they had no contact with each other.) The matter is not settled, and my consultant thinks it will take years before a consensus is reached.

In addition, during my talk to the May meeting of the Rocky Mountain Skeptics, someone asked how often the word va-yitzer or va-yiytzer appears in the Hebrew Bible.1 If the word appeared with both spellings in different contexts, we might be able to shed light on Schroeder's argument that the difference in spelling is purposeful and not a scribal inconsistency. (Schroeder claims that the spelling va-yiytzer, with the second Y, is significant when the verb refers to the creation of Adam. While researching the question, incidentally, I noticed that the very next time that the root YZR is used (Genesis 2:8), it is in connection with Adam, and it is written yatzar, with a single Y.)

Although the root YZR can be found elsewhere in the Bible, va-yitzer or va-yiytzer cannot. In Biblical Hebrew, YZR in the common (qal) conjugation means to make or create, whereas, in the (pi'el) conjugation of va-yiytzer, it means to form, as a potter does with clay. There are many words that appear only once or twice in the Bible. Thus, if Schroeder wants to claim that va-yiytzer has special significance, he needs to demonstrate two things: that other unique words have special significance and that what I call scribal inconsistencies are, to the contrary, purposeful. He has addressed neither question.

1. Matt Young, "The Bible as a Science Text," Rocky Mountain Skeptic, November 1998, pp. 2-4.

2. Stephen Jay Gould, "Common Pathways of Illumination," Natural History, December 1994, pp. 10-20.

3. Charles S. Zuker, "On the Evolution of Eyes: Would You Like It Simple or Compound?" Science, 5 August 1994, pp. 742-743.

The Bible Code

Did God Write the Bible?

Or Was It a Computer?

- Matt Young
[Originally published in Rocky Mountain Skeptic, March-April, 1998, pp.1, 4-6.]

Take the Hebrew Bible. Go to the first instance of the letter tav. Count 50 letters (a skip of 49 spaces) and write each fiftieth letter. You get the Hebrew word TVRH, or Torah (the five books of Moses). Do the same thing with the Book of Exodus, the second book of the Hebrew Bible, and you get the same result. Now take Deuteronomy and Numbers, the fourth and fifth books, and you will get TVRH spelled backward. [Schroeder, 1992]

In Leviticus, the central book, begin at the first yod and count eight letters, and you get YHVH, the name of God in Hebrew. Thus, the four words TVRH in the first and last two books point toward the name of God in the central book.

The procedure is not as cut and dried as Schroeder claims. In Genesis and Exodus, the rule works. In Numbers, however, you do not begin from the first heh but rather from the third (remember that TVRH is spelled backward). In Deuteronomy, you have to begin counting in the fifth verse, and you count only 48 spaces, not 49. It looks suspiciously as if you get to make up the rules as you go along. Schroeder is almost certainly dredging for correlations and then ascribing significance to correlations that occur by chance alone.

Very probably, so does Michael Drosnin in his best seller The Bible Codes. [Horovitz, 1997; Hendel, 1997; Odenheimer, 1997] Drosnin, a journalist, looks for pairs of words at different skips. He claims that he can predict the future by evaluating such pairs; most spectacularly, he claims to have predicted the assassination of Yitzhak Rabin by correlating Rabin's name with the words "assassin that will assassinate."

Rabin's name appears in Genesis (in Hebrew) with a skip of 4772. [Sternberg, 1997] If you arrange the letters with the right line length (see companion article by David Thomas), you see the words Yitzhak Rabin running vertically and crossing Deuteronomy 4:42,

41 Then Moses set aside three cities ..., 42 to which a manslayer could escape, one who unwittingly slew a fellow man without having been hostile to him in the past.... [literally, without knowledge and he did not hate him]
What the Jewish Publication Society renders "manslayer who slew," Drosnin translates as "assassin that will assassinate," which is perhaps a plausible translation if the words are taken wholly out of context. Sternberg's translation, "a slayer who happens to have killed," however, is more in keeping with the sentiment that the killing was unintentional and without malice.

But the verse as a whole refers to a man who unknowingly kills someone he has nothing against. Even if you believe that Rabin's assassin, Yigal Amir, intended only to wound Rabin, it would take a wild leap to argue that Amir held nothing against Rabin and that the killing was wholly unintentional. Even if the correlation were significant, it would be hard to see how these verses refer to Rabin's assassination. In addition, a simple correlation between "Yitzhak Rabin" and "assassin who will assassinate" tells us nothing quantifiable. Will an assassin assassinate Rabin? Or is Rabin himself the assassin?

Ancient Hebrew did not have tenses in the way we think of tenses in modern English. [Chomsky, 1957] Instead, ancient Hebrew verbs signified either completed action or continuing action. Consider, for example, the story of Hannah, the favored but childless wife (1 Samuel). Elkanah's wife, Penninah, bore children, but Hannah had none, even though Elkanah favored Hannah.

6 And her rival [Penninah] also provoked her ... because the Lord had shut up her womb. 7 And as he [the Lord] did so year by year, when she went up to the house of the Lord, so she provoked her; therefore she [Hannah] wept, and did not eat. 8Then said Elkanah her husband to her, Hannah, why weepest thou? and why eatest thou not?
The Hebrew words for weepest and eatest are tivki and tochli. Out of context, they would be translated as "you will weep" and "you will eat." But here these translations make no sense, inasmuch as Hannah is already weeping. Instead, the meaning is, "Why do you keep on crying? Why do you keep on not eating?"

The words "assassin who will assassinate" similarly imply continuing action and would be better translated as "assassin who keeps on assassinating." Amir killed only once, whereas Rabin, in many eyes, was responsible for many deaths. Thus, the assassin could be Rabin himself. I have no doubt that a Muslim journalist dredging for Bible codes would embrace that interpretation, not Drosnin's.

Jewish fundamentalists do not accept Drosnin's claim that he can use the codes to make predictions. They argue that the codes are planted only to refer to events past; since there is no syntax, you cannot interpret the codes as anything other than signs. Their argument is based on the Biblical injunction (Deuteronomy 18:10-12)

10 There shall not be found among you any one that ... useth divination, or an observer of times, or an enchanter, or a witch, 11 Or a charmer, or a consulter with familiar spirits, or a wizard, or a necromancer. 12 For all that do these things are an abomination unto the Lord.
Drosnin [1997] replies that no one has proven that the codes are not real. Indeed, he not only found the correlation before the assassination but also warned Rabin a year ahead of the event.

Drosnin's argument, "You can't prove it is wrong," is worthy of discussion. You can never disprove certain kinds of arguments, usually negative statements like Drosnin's. I can never prove that an infusion of some herb did not cure your cold; all I can do is cite evidence that the same herb worked no better than a placebo in careful studies. Similarly, I can never prove that Drosnin's theory is wrong; all I can do is show that his correlations could have been caused by chance and that his interpretation is not the only possibility.

The embedded codes are cryptic at best - a few words here, a few words there. If God wanted to convince us unequivocally, then why did he not encode a substantive message, instead of playing parlor games and encoding messages of no more than a few letters? At the risk of anthropomorphizing, may I suggest that, if God wanted to be truly convincing, he would have gone straight to the first aleph in Genesis and encoded a significant statement such as "I am the Lord your God who brought you forth from Egypt." To be more convincing, he would have made the skip significant as well: 49 spaces, for example.

To show that the embedded words have real meaning, we would have to show that they appear in the text with greater frequency than would be expected by chance. That is precisely what Eliyahu Rips of the Hebrew University of Jerusalem and his colleagues Doron Witztum and Yoav Rosenberg have attempted to do. [Satinover,1995; Witztum et al., 1994]

Witztum and his colleagues [1994] searched for the names of historical figures and their birth dates. The historical figures were eminent rabbis. Their names and the dates were taken from a standard Hebrew reference work and chosen according to the length of the citation; only the figures with the longest citations qualified. All the historical figures lived long after the completion of the Book of Genesis, so they could not have been known to its authors. Whenever Witztum and colleagues found the name of one of the historical figures at a given skip, they looked for the birth date, nearby and with the same skip. They found that date with a higher probability than was predicted by chance. This result seemed so bizarre to the referees of the paper that they insisted that the authors perform other tests, such as rearranging the letters randomly and looking again for correlations. The authors complied but found no correlations in the randomized texts.

According to an exchange of mail on the Internet and an article in a Hebrew-language science magazine, [Bar-Hillel et al., 1997] Witztum and colleagues have dredged for correlations by adjusting the input parameters to create the result he wanted.

How? Maya Bar-Hillel, of the Center for the Study of Rationality at the Hebrew University, and her colleagues, mathematicians Dror Bar-Natan and Brendan McKay, point out that people are often called by more than one descriptor. For example, Robert F. Kennedy was called Bobby Kennedy, Robert Kennedy, Senator Kennedy, Sen. Kennedy, RFK, or Robert Francis Kennedy. Similarly, beloved rabbis usually have nicknames. Often these nicknames are acronyms. Thus, Rav Shlomo ben Yitzchak (Rabbi Solomon son of Isaac) is most commonly known as Rashi. If you wanted to dredge for correlations, you could search for Shlomo, Rav Shlomo, Rav Shlomo ben Yitzchak, Shlomo ben Yitzchak, or Rashi and retain only the one that works best.

In addition, Hebrew spelling was not standardized until very recently, so we know many of the eminent rabbis by several variant spellings, much as we know Shakespeare as Shakespear and Shakspere, as well as by other variants. The name Jonathan, for example, appears in the Bible as YVNTN and as YHVNTN; you could use both and ignore the spelling that did not work.

Witztum and his colleagues used three ways to express the dates: In Hebrew, letters substitute for numerals, so you can find dates embedded in a text as easily as you can find words or names. If a rabbi was born on the fourth of the month Tevet, you may express that as 4 Tevet, on 4 Tevet, or 4 in Tevet, where 4 is indicated by the fourth letter dalet. Compare these with 4 July, on the fourth of July, and the fourth of July. Witztum and his colleagues used all three descriptors, thereby increasing their odds of finding a match. They did not use the fourth possibility, on 4 in Tevet.

Bar-Hillel and her colleagues performed the same calculations as Witztum, except that they carefully varied the spellings of the names and the form of the dates. They got the best correlations when they used the names and dates in the forms chosen supposedly a priori by Witztum. That is, as Bar-Hillel expresses it, "... wonder of wonders, it turns out that almost always, if not always, the allegedly blind choices paid off: just about anything that could have been done differently than it was actually done would have been detrimental to the [result]."

Perhaps we cannot dismiss Witztum's work quite as easily as Schroeder's or Drosnin's, but it certainly seems probable that, consciously or otherwise, they have finagled in order to get the "right" result. In addition, there is a great deal of evidence that the Bible was assembled, or edited, rather than written as a single work. To believe Witztum's interpretation uncritically, we would have to believe that God guided the hands of countless scribes over many generations in order to get precisely the right text to prove his point to us, over 1000 years later, and ignored intervening generations entirely.

Excerpted from the manuscript No Sense of Obligation: What Science Says to Religion, by Matt Young. Copyright © 1998 by Matt Young. All rights reserved. I thank Brendan McKay and David Thomas for their comments and corrections.


Bar-Hillel, Maya, Dror Bar-Natan, and Brendan McKay, 1997, "There Are Codes in War and Peace Too," Galileo, November-December (in Hebrew). I thank Maya Bar-Hillel of the Hebrew University for supplying me with an English translation of this article.

Chomsky, William, 1957, Hebrew: The Eternal Language, Jewish Publication Society of America.

Drosnin, Michael, 1997, "Battling over the Codes" (letter to the editor), Jerusalem Report, 2 October, p. 2.

Hendel, Ronald S., 1997, "The Secret Code Hoax," Bible Review, August, pp. 23-25. See also Anonymous, 1997, "The Bible Code: Cracked and Crumbling," Bible Review, August, p. 22.

Horovitz, David, 1997, "Busting the Bible Code Breakers," Jerusalem Report, 4 September, pp. 14-21.

Odenheimer, Micha, 1997, "False Positive, There's a Price to Be Paid for Wanting to Be Proven Right," Jerusalem Report, 16 October, p. 59.

Schroeder, Gerald L., 1992, Genesis and the Big Bang, Bantam.

Sternberg, Shlomo, 1997, "Snake Oil for Sale," Bible Review, August, p. 24.

Witztum, Doron, Eliyahu Rips, and Yoav Rosenberg, 1994, "Equidistant Letter Sequences in the Book of Genesis," Statistical Science, Vol. 9, No. 3, pp. 429-438. See also editor's remark on p. 306.