Did God Create Heaven and Earth—or Just Separate Them? An Analysis of Ellen van Wolde’s “Hypothesis,” Part 1 (of 2)

Late last year, the popular press published a news release announcing a retranslation of the biblical Hebrew by Ellen van Wolde, a professor at Radboud University in the Netherlands. They reported her conclusion that God did not “create” the universe, but rather “spatially separated” preexisting materials. The news release ends with a stunning quotation from van Wolde: “The traditional view of God the Creator is untenable now.”

We sought out and reviewed van Wolde’s article: a pamphlet in Dutch, Terug Naar Het Begin. Our conclusion is that it does not live up to the English-language news release—nor to her claim that the traditional translation is “untenable.” In fact, we find the article deeply flawed and inadequate in justifying her proposed retranslation.

Old Testament scholar Arie Leder commented that van Wolde’s “translation and reading is not without problems… caveat lector [‘Let the reader beware’],”¹ and we heartily agree with that conclusion. However, the purpose of this article is not to discuss van Wolde’s Hebrew linguistics, but to critique her methodology. Even if we agree with van Wolde’s translation and reading of the Bible—which most of the time we don’t—she fails to prove her point because her analytical process is so flawed.

Van Wolde approaches her analysis of biblical Hebrew as if using the scientific method: she proposes a “test” of her “hypothesis.”² Hence it seems fair to critique her article on that basis.

Many non-scientists, such as van Wolde, believe the essence of the scientific method is testing a hypothesis, but that’s only one aspect of it. The key to the scientific method is falsifiability, a concept well articulated by Karl Popper. A hypothesis is proven true only if it passes tests designed to prove it false; thus, a valid test must include at least one outcome that will prove the hypothesis false. As an example, the proper way to prove the hypothesis “all swans are white” is to search for a non-white swan; counting only white swans proves nothing. Even scientists sometimes fall into this trap by trying to devise tests that will confirm pet theories, but Popper says:

So how should this principle be applied to biblical Hebrew? One approach includes, (1) finding a portion of text where the word in question seems to demand a meaning to the exclusion of the accepted definition, (2) hypothesizing a new meaning for the word in that text, and (3) testing the revised meaning against all other texts to see if it is plausible. If the new meaning doesn’t work in all other texts, it is falsified, unless a rational explanation for discontinuities exists.

However, van Wolde does not follow this pattern or anything like it in her analysis. She proposes a hypothesis and then goes to great lengths to construct a confirmation of it—even an obscure one. Then she fails to make the most rudimentary effort to falsify it. Hebrew linguist Joel Hoffman says, “Van Wolde’s ‘evidence’… is essentially her conclusion.” It’s as if she wishes to prove that “all swans are black” and after finding one black swan, triumphantly proclaims: “I proved it.”

Another shortcoming in van Wolde’s work is the tendency to interpret Genesis 1 as a science textbook. One must walk a tightrope here. A document understood by ancient peoples from Moses to Ezra to Malachi could not incorporate modern scientific theories, yet if the Bible is supernaturally inspired, one would nevertheless expect its broad outlines to be consistent with modern science when correctly interpreted. Van Wolde falls off this tightrope.

The essence of Van Wolde’s argument is that the Hebrew word bara—traditionally translated “create”—should be retranslated as “separate.” She observes that Genesis begins with bereshith (“in a/the beginning”) rather than with bareshith (“in the beginning”).³ Hoffman says this is an example of trying to force rules of modern European grammar upon ancient Hebrew. One feels certain the Jewish sages who compiled the Septuagint—and other scholars for more than 2,000 years since—understood the relative implications of these two words when they translated Genesis 1:1 as “In the beginning God created.” Moreover, bareshith never appears in the Bible, and reshith never has a definite article⁴ when it refers to temporal matters.⁵ Nevertheless, van Wolde claims the use of bereshith mandates that Genesis 1:1 does not speak of an absolute beginning of time, but of the beginning of a specific act of God.⁶

This is a statement about science. The twenty-first century standard theory of cosmology holds that time and space (“space-time”) began in an instant about 13.7 billion years ago in a big bang event. Generations of Jewish scholars at least as far back as the Septuagint (LXX) similarly believed the universe had an absolute beginning, initiated by God in a creation event. But van Wolde does not accept the traditional view of God—nor do her earlier writings suggest she understands the big bang theory.⁷

Rather, Van Wolde seems to equate the God of the Bible with the deities of Mesopotamian mythology. Using three examples from ancient Near Eastern (ANE) texts, she points out that the first act of a Mesopotamian deity is to separate heaven and earth.⁸ In other words, because of a parallel between Genesis 1 and these ANE texts, she leaps to the conclusion that the parallel is determinative: Genesis 1 must follow the pattern of these texts.⁹ She retranslates Genesis 1:1 accordingly, but in so doing, ignores the observations of numerous scholars that parts of Genesis were written specifically to contradict Mesopotamian mythology. The Bible is not just another myth from a common ANE oral tradition; it was often formed as a polemic to counter them. Additionally, van Wolde does not seem to recognize that the concept of discontinuity may be at work. As an example, in Mesopotamian mythology the sea monster preexists the world, but in the Mosaic polemic, sea monsters (tanninim) are a later creation of God.¹⁰ Yet van Wolde makes no effort to falsify her hypothesis by addressing objections raised by legitimate scholars.

Interpreting Genesis based on words or literary parallels in ANE texts is risky business; not only do the words come from a different time and place but also from a different culture. Usage of a word must dominate as one seeks to discern meaning; the origin of a word is not nearly as relevant as the way in which the word is used. As one seeks to interpret a word found in Genesis, the most important criterion is its use in Genesis and in other Mosaic literature.¹¹ The next most important criterion is use of the word elsewhere in the Old Testament, but this is secondary because non-Mosaic texts were written by different authors and in different contexts (if not at a much later time). Individual authors do not always use words in the same way and language usage changes substantially with time. For example, words like “web” and “tweet” mean different things in a nature magazine than from Internet jargon; and consider how much the use of the word “green” has changed in recent years!

In addition to the above, van Wolde attempts some arguments based on Hebrew linguistics to justify retranslating bara as “separate.” However, as with van Wolde’s use of the scientific method’s principles, her analytical process in making these linguistic arguments is faulty. This will be discussed next week in part 2 of this series.

Dr. Hugh Henry, Ph.D.

Dr. Hugh Henry received his Ph.D. in Physics from the University of Virginia in 1971, retired after 26 years at Varian Medical Systems, and currently serves as Lecturer in physics at Northern Kentucky University in Highland Heights, KY.

Daniel J. Dyke, M.Div., M.Th.

Mr. Daniel J. Dyke received his Master of Theology from Princeton Theological Seminary 1981 and currently serves as professor of Old Testament at Cincinnati Christian University in Cincinnati, OH.