Genesis 1:1 Is The First Event, Not A Summary

By Vern S. Poythress

[Vern S. Poythress is Professor of New Testament Interpretation at Westminster Theological Seminary.]

Commentaries regularly discuss three main interpretations of Gen 1:1 in relation to the subsequent verses. (1) According to the first, traditional interpretation, Gen 1:1 describes the initial event among God’s acts of creation. Verse 2 then gives circumstantial information about the state of the earth at an early point. (2) According to the second interpretation, Gen 1:1 functions as a temporal subordinate clause: “In the beginning, when God created the heaven and the earth, the earth was without form....” (3) According to the third interpretation, Gen 1:1 is a summary of the entire sequence of divine acts described in vv. 2–31. It does not describe the very first event that led to the creation of the earth and its unformed state in v. 2. Rather, the first act of making things starts with v. 3, and Gen 1 offers no comment on how the unformed earth of v. 2 came into being.[1]

The second interpretation has had a good many advocates, but it seems to be fading, and it has received a number of convincing refutations.[2] For the sake of brevity, we confine ourselves to the debate between the first and the third interpretation. The first interpretation says that Gen 1:1 is the initial event, and accordingly may be designated the initiation view. The third interpretation says that Gen 1:1 is a summary and accordingly may be designated the summary view.

The initiation view was common among earlier Jewish and Christian interpreters,[3] but it is no longer in such favor. In his 1987 commentary, Gordon Wenham indicates that “the majority” of modern commentators favor the summary view.[4]

I. Major Arguments For The Initiation View

The initiation view still has its defenders. The commentaries by Collins, Wenham, and others advocate it.[5] But because of space limitations, these commentaries interact only briefly with the summary view. I propose to take the space to engage more thoroughly with the summary view, focusing especially on its fullest articulation in a key article by Bruce K. Waltke.[6]

In our analysis we will treat Genesis as a literary unity, as Waltke does. By contrast, the historical-critical tradition breaks Genesis apart, and usually finds layers of meanings at times earlier than the extant form of Gen 1. We will not deal with this line of speculation.[7] For simplicity, we will mostly quote from the English Standard Version, but it is to be understood that the arguments are ultimately framed in terms of the underlying Hebrew text.

Let us begin by briefly noting the three main arguments for the initiation view.

1. Cohesion Between Verses 1 And 2: The Initial State Of The Earth As Without Form

The first argument appeals to the close connection between Gen 1:1 and v. 2. The term the earth (הָאָרֶצ) occurs as the last term in v. 1 and the first main term in v. 2. The syntactic linkage between the two verses consists in a waw-conjunctive, which, when followed by a noun and then the main verb of the clause, customarily introduces circumstantial information.[8] (By contrast, the waw-consecutive plus imperfect is the usual way of introducing new main events in a narrative sequence.) Verse 2 is providing circumstantial information.

The significant point here is what kind of circumstantial information is introduced in v. 2. It is information about the state of the earth. Since the earth has just been introduced in the preceding verse, the information specifies the state of the earth that was already mentioned in v. 1. It follows that the act of creation mentioned in v. 1 results in an earth that is “without form and void.” “The earth” is not the formed and filled earth at which the narrative arrives by v. 31, and which is summarized in 2:1: “Thus the heavens and the earth were finished, and all the host [the furnishings, implying that the earth was no longer empty or ‘void’] of them.” The early unformed state of the earth is described by 1:2 with reference to the earth of v. 1. So 1:1 cannot be a summary. That is to say, the expression “the heavens and the earth” in 1:1 does not refer to the heaven and the earth in their completed form (2:1), as a summary might do. Rather, it refers to the heavens and the earth in an immature state.

2. Theological Purpose: The Assertion Of Absolute Divine Sovereignty

A second argument focuses on the theological purpose of Gen 1. It is clear that Gen 1:1–2:3[9] as a whole strongly asserts the full and effective sovereignty of God. God is the one true God who controls and rules over everything that he has made. In its majestic monotheism, the passage contrasts strongly with the polytheism of the cultures of the ancient Near East. It also contrasts with ancient Near Eastern cosmogonic narratives that involve the birth of gods and conflicts between gods. In Gen 1 there is no plurality of gods. There are no birth events. There is no mention of conflict. God personally rules and brings about his will.

It is therefore fitting that the narrative of creation should assert God’s sovereignty not only over some of the things in the world, but over all. God’s sovereignty must include not only ruling over the development of things that already exist, but controlling the very being and constitution of whatever exists. This comprehensive sovereignty must include the original earth, which is without form, and the deep. Otherwise, the earth is left as a potential independent entity. If God did not make it, if it is just eternally there, its original constitution escapes God’s sovereignty, and God just has to do the best he can with material that he did not originally specify. Moreover, the earth may be just as eternal as God himself. Anything coeternal with God, even an impersonal coeternal, is really a rival to complete sovereignty. So it is fitting that the narrative in Gen 1:1 closes this door to rivalry, by indicating that the initial act of creation includes the creation of the earth, and, by implication, the deep that covers its surface. By contrast, the summary view postulates that the earth and the deep are already there, without any explanation, before God begins to create in v. 3.[10] This postulate is in tension with the overall theological purpose of Gen 1.

3. Narrative Structure: The Use Of The Perfect Verb For An Antecedent Event

A third argument focuses on narrative structure in Gen 1:1–2. C. John Collins argues that the use of the Hebrew perfect tense at the commencement of a narrative normally refers to an antecedent event.[11] His case can be strengthened by observing two cases where such a structure occurs at the beginning of a whole book.

בִּשְׁנַת שָׁלוֹשׁ לְמַלְכוּת יְהוֹיָקִים מֶלֶך־יְהוּדָה בָּא נְבוּכַדְנֶאצַּר מֶלֶך־בָּבֶל יְרוּשָׁלַםִ וַיָּצַר עָלֶיהָ

In the third year of the reign of Jehoiakim king of Judah, Nebuchadnezzar king of Babylon came to Jerusalem and besieged it. (Dan 1:1)

The grammatical structure in Hebrew is parallel to Gen 1:1:

בְּרֵאשִׁית בָּרָא אֱלֹהִים אֵת הַשָּׁמַיִם וְאֵת הָאָרֶץ

In the beginning God created the heavens and the earth.

In Dan 1:1 we first have a temporal marker (“In the third year of the reign of Jehoiakim king of Judah”), parallel to the temporal marker “in the beginning” in Gen 1:1. Then, in the Hebrew word order, comes a perfect verb (“came,” בָּא), parallel to the perfect verb “created” (בָּרָא) in Gen 1:1. Then comes the subject, “Nebuchadnezzar king of Babylon,” parallel to the subject “God” in Gen 1:1.

A second, similar example occurs in Ezra 1:1:

In the first year (וּבִשְׁנַת אַחַת) of Cyrus king of Persia, that the word of the Lord by the mouth of Jeremiah might be fulfilled, the Lord stirred up (הֵעִיר יְהוָה) the spirit of Cyrus king of Persia, so that he made a proclamation throughout all his kingdom and also put it in writing.

Unlike Dan 1:1, the verse begins with a waw-conjunctive. But then comes the temporal marker, “in the first year of Cyrus king of Persia,” parallel to the expression “in the beginning” in Gen 1:1. Then comes an infinitive clause of purpose, “that the word of the Lord by the mouth of Jeremiah might be fulfilled,” which is an extra element in comparison with Gen 1:1. Then comes a verb in the perfect, “stirred up” (הֵעִיר), and then the subject, “the Lord.”

In both Dan 1:1 and Ezra 1:1, the opening describes the first event, rather than giving a summary of the subsequent narrative. The grammatical structure in both verses is parallel to Gen 1:1. So, reasoning by analogy, we conclude that Gen 1:1 describes the first event, in relation to the narrative in vv. 2–31.

II. The Summary View

Now we turn to the summary view of Gen 1:1. The summary view has many advocates. For the sake of simplicity, and for the sake of allowing a fuller discussion, we focus on Bruce K. Waltke as the best representative of that view.[12] Waltke opposes each of the three arguments above with a corresponding counterargument. We shall consider each of them in turn.

1. The Heavens And The Earth As Already Ordered

The first counterargument is that the expression “the heavens and the earth” in Gen 1:1 designates “the organized universe, the cosmos.”[13] It is not the unorganized state described in v. 2. If the heavens and earth are organized in v. 1, it follows that the endpoint of God’s activity of creating, as described in the verse, must be the same endpoint at which the narrative arrives in v. 31. This endpoint is then summarized in 2:1, “Thus the heavens and the earth were finished.” So Gen 1:1 gives in summary form the same sequence of activities that is expounded in detail in the rest of the chapter, vv. 2–31.

In favor of this interpretation of the expression “the heavens and the earth,” Waltke has three subpoints: (A) “the heaven(s) and the earth” is a merism, that is, a designation of the whole using two opposite polarities, so the expression must be considered as a whole; (B) the Hebrew expression always designates the ordered or organized cosmos; and (C) consequently, to postulate a distinct meaning in Gen 1:1 would violate standard philology.

On the surface, this line of argument may sound reasonable. But with respect to each of the three points, there is some vagueness in the claims, and some slippery points in the arguments.

a) A Merism. Waltke’s first subpoint is that the expression “the heavens and the earth” (אֵת הַשָּׁמַיִם וְאֵת הָאָרֶץ in the underlying Hebrew) is a merism.[14] Waltke elsewhere defines a merism as “a figure of speech involving opposites to indicate totality.”[15] In discussing Gen 1:1, Waltke illustrates with various expressions: “they came, great and small”; and “the blessed man meditates in God’s law ‘day and night,’ i.e., ‘all the time.’”[16]

The appeal to merism is significant, because Waltke thinks that it lays the groundwork for his point (B) about the reference to the organized universe. I agree that the key expression is a merism, but, as we shall see, it does not help Waltke’s case.

Many merisms are relatively “transparent” in meaning. The meaning of the whole can easily be inferred from the meaning of the two opposites. For example, “day and night” pretty much does cover all the time, if we allow that the expression could cover by implication times of transition between full daytime and full nighttime, that is, the times of twilight. Similarly, the classic merisms in the marriage vows, “for richer, for poorer” and “in sickness and in health,” cover all the human conditions, if we allow for situations that are intermediate (for example, recovering health after being sick). The expression “the heavens and the earth” in Hebrew is similar, because “the heavens” usually refers to what is above, and “the earth” refers to what is below us or at least lower down, below the heavens. Together, the two make up everything that we see. So the meaning of the compound expression “the heavens and the earth” is transparently composed from the meanings of the two main constituents, “the heavens” and “the earth.”

Why is the transparency of meaning significant? Waltke says that the expression “the heavens and the earth” “is a compound phrase that must be studied as a unity.”[17] But what does it mean for a compound to “be studied as a unity”?

It could mean merely that the full import of the compound should not be deduced merely from taking the two main words “in isolation from one another.”[18] But it could also mean that, once we see the unity of the compound, it must be studied in isolation from the two main words that compose it. In other words, we ignore all the occurrences of the words “heavens” and “earth” outside of the compound expression.

This is not a trivial issue. If, in Gen 1:1, we replace the expression “the heavens and the earth” with some other expression, like “all things” (John 1:3) or “visible and invisible” (Col 1:16), we lose the key connection between “the earth” in Gen 1:1 and “the earth” in Gen 1:2, which is important for determining the state of the earth in v. 1. Similarly, if we assume that the compound expression in v. 1 must be isolated from the expression “the earth” in v. 2 (because the compound is somehow a seamless whole), we arrive at a similar result, wherein v. 2 is disconnected from v. 1. We fail to do justice to the significance of the occurrence of “the earth” in both verses. Waltke’s argument never discusses this problem.[19]

Waltke quotes approvingly from Cassuto, who uses the English word broadcast to illustrate the principle that the meaning of a compound (“broadcast”) cannot be deduced from the meaning of its parts (“broad” and “cast”).[20] But the example does not prove what it is supposed to prove. The question is whether the compound “the heavens and the earth” works in the same way as the compound “broadcast.” It does not. As we indicated, many merisms are by their nature transparent in meaning, as the examples “day and night” and “in sickness and in health” illustrate. In determining the meaning of a merism, it is only necessary to adjust to the fact that the two polar opposites, by being adjoined, are meant by implication to encompass any intermediates. So the meaning of a merism is deducible from the meaning of its constituents. This transparency of meaning allows us to multiply merisms indefinitely. For example, we can have any number of merisms to describe humanity: rich and poor, slave and free, big and small, young and old, strong and weak, short and tall, educated and uneducated, employed and unemployed. Given an appropriate context, discerning the meaning of the compound is easy. Our ability to discern the meaning is not significantly affected by whether the compound is already a common, well-known, fixed expression.

Note also that, while all the merisms for humanity have the same referent, namely humanity, none is strictly synonymous with any other. “Rich and poor” draws attention to a financial polarity, “slave and free” draws attention to the polarity between freedom and nonfreedom, and so on. The merism “educated and uneducated” might occur suitably in a context where naïvely it might be thought that education would affect people’s situation, but where in fact it does not—all humanity, both educated and uneducated, belong together.

The same holds for the expression “the heavens and the earth.” We can see that it refers to the whole world precisely because its two major inner constituents have polar meanings that are used to refer to the two major spatial regions of the world. Given that it is a merism, the meaning of the whole is transparently derivable from the meanings of the two parts.

So let us look more carefully at how Waltke treats the compound expression “the heavens and the earth.” Citing Cyrus Gordon with approval, Waltke uses the illustration that “in English, the expression ‘they came, great and small’ means that ‘everybody came.’”[21] There is some vagueness here with the word “means.” The sample statement using the compound expression “great and small” implies that “everybody came.” It “means” that, in a loose sense of the word “means.” But if we substitute “everybody” for “great and small,” we change the sense subtly, precisely by eliminating the fact that the compound expression is transparent to its two inner components, “great” and “small,” and operates by inviting us to conceive of humanity as composed of these two polar parts.

The “meaning,” in a more nuanced sense, includes a focus of attention on the two extremes and then everyone in between. Likewise, the distinct meanings of “heavens” and “earth” do not totally disappear in the compound. They are still “visible”; they are “transparent” in their contribution to the full meaning of the compound expression.

b) The claim of an organized universe. A second element in Waltke’s argumentation is the claim that the compound expression designates “the organized universe, the cosmos.” But here we must be careful to distinguish the sense of an expression from its referent. For the sake of clarity, let us illustrate the difference between sense and referent. The expression “the father of Isaac” refers to Abraham, but the sense or meaning of the phrase “the father of Isaac” is roughly “the first-generation male parent of the person designated Isaac.” The sense does not contain everything we know about the referent (Abraham). Or take another example. I may refer to Philadelphia as “the largest city in Pennsylvania” or “the city where the Delaware River and the Schuylkill River flow together” or “the city where the Liberty Bell is” or “the place where the Declaration of Independence was signed” or “the first U.S. capital.” These expressions all have the same referent, but they differ in what sort of information they provide about that referent. They differ in sense.

Once we make this distinction, we can see that there is a potential problem with how we go about analyzing the expression “the heavens and the earth.” We must distinguish the sense of the expression from what it refers to in any particular case of its use. The great majority of occurrences of this expression in the OT will refer to the world in an organized state. Why? Because, subsequent to the completion of God’s work of creation, the world remains pretty much in an organized state (we may make a partial exception for the flood of Noah). In addition, subsequent to the initial chapters in Gen 1–3, nearly every use of the expression will refer to the world in a state of historical development, brought about by human activities and human births and deaths. Therefore, if we are talking about the referent, we may say that in most cases the compound expression “the heavens and the earth” designates a world already having undergone human historical development. We may also say that it designates a world that shows effects of the fall. But it would be erroneous to take all this information about the referent, and read it back into the sense of the expression, as if the sense included the idea of historical development and the idea of effects of the fall. Such reading back would be just as mistaken as if I were to claim that the sense of the expression “the largest city in Pennsylvania” included all the voluminous information about the referent, namely the actual city with all its material, structural, economic, and social dimensions. Likewise, we must ask with some care, “Does the sense of the expression ‘the heavens and the earth’ include the idea of organization, which idea certainly belongs to the referent, at least in the great majority of cases?” It will not do just to go through the various occurrences, noting that the referent is organized.

Is this an artificial problem? Waltke’s own use of terminology is not reassuring, because it seems sometimes to focus on the referent and sometimes on the sense, without clearly distinguishing the two. Consider the following two paragraphs in Waltke’s argument:

So here, “the heavens and the earth” are antonyms to designate “everything,” and more specifically “the organized universe, the cosmos.” In fact, Wisdom of Solomon [11:17] uses the Greek words ὁ κόσμος to refer to Genesis 1:1.

This is undoubtedly the sense of the compound in the summary statement concluding the creation account: “Thus the heavens and the earth were completed, and all their hosts” (Gen. 2:1). The compound occurs again in this sense in the summary statement introducing the stories about man [Gen 2:4].... This compound never has the meaning of disorderly chaos but always of an orderly world.[22]

The shift from terms about reference (“designate,” “refer”) to terms about meaning (“sense,” “meaning”) does not show awareness of a distinction. And the sentence discussing the Wisdom of Solomon is inexact. More precisely, it should say that the words ὁ κόσμος refer to the universe and allude to Gen 1:1.

We can further illustrate the problem by dipping into the specific evidence concerning the usages of the expression “the heavens and the earth” and other textual expressions that have the pair “heaven” and “earth.” Waltke first mentions Gen 2:1 in the quotation provided above. He says, “This is undoubtedly the sense of the compound.” But once we make the distinction between sense and reference, the evidence fails to have any force. In Gen 2:1, the compound expression refers to the completed heavens and earth. But the sense may still turn out to be little more than “what is above” and “what is below,” taken together. The sense does not automatically accumulate everything that we know about the state of the referent at the time to which the sentence refers when it uses the compound expression. The word “undoubtedly” shows that Waltke is confused. He does not see that the compound could have a fairly minimal sense and still refer to a universe that happened to be organized at the time to which the reference was made.

Even when we focus on the referent of Gen 2:1, and not the sense, we find a subtle difficulty. Genesis 2:1 mentions not only “the heavens and the earth,” but also “all the host of them.” The expression “all the host of them” refers to things in the sphere of the heavens, like the heavenly bodies and the birds, and things in the sphere of the earth, like the plants and the animals. These hosts are distinguished from the heavens and the earth themselves. So the expression “the heavens and the earth” focuses in this context primarily on the spatial regions, in distinction from their “host” or inhabitants (compare Jer 51:48). Consequently, “the heavens and the earth” is not simply a synonym for “everything.” It would be odd to say, “Everything was completed, and all its host.” “Everything” cannot have inhabitants distinct from “everything,” because the inhabitants are already included in the referent of “everything.”

In this respect, the compound expression has the same flexibility in its use as the constituent expression “the earth.” “The earth” sometimes serves to designate the lower region, particularly the solid ground (Gen 1:11, 28, 29; 7:3; 8:17, 19; 11:8; etc.), and sometimes to designate inclusively the region together with everything on it (Gen 2:4; 6:11; 9:11, 13; Exod 19:5; etc.). The flexibility in use confirms that the compound expression is not a rigid, technical term, but transparently reflects the flexibility in its constituent terms.

Another difficulty arises because of a certain vagueness in the idea of being “organized” or “ordered.” How much organization does it take before we consider something organized? Well, it depends on the circumstances and the purposes and interests of those who evaluate a particular object or region. In a minimal sense, the earth in Gen 1:2 already shows organization. There is the deep, which has some kind of surface, but also involves a larger body of material of which the surface is the upper, exposed part. So the deep is “organized” into two parts, the surface and the depth beneath the surface. Second, the deep together with its surface is “organized” in relation to the space above it, in which the Spirit of God[23] is hovering. The space in question is distinct from the deep and is in place over it.

In addition, there is organized movement within this space. The Spirit is hovering. This description of hovering suggests that the space is like our normal space with three distinct dimensions, one of which is the up-and-down dimension. That in itself is a kind of “organization.”

Moreover, as we learn from v. 9, a solid entity is already present underneath the liquid-like “deep.” Verse 9 does not say that God made or created the dry land alongside the waters. Neither does it say that God caused the dry land to be congealed out of the waters. Rather, the waters were “gathered together,” so that the dry land might “appear.” This description implies that the solid material already existed underneath. If so, presumably it already existed in v. 2.

So, in v. 2, the early stage of “the earth” has a vertical arrangement involving at least six distinct elements: the space over the Spirit, the Spirit himself, the space under the Spirit, the surface of the deep, the deep under the surface, and the solid ground underneath the deep. If we combine the space over and under the Spirit, and consider them as a single distinct element in the midst of which the Spirit is hovering, we still have five elements, and they are structurally organized in specific ways in relation to each other.

So the earth in v. 2 already has some degree of organization. To be sure, the earth in v. 2 is “without form and void.” But, as C. John Collins points out, that is not equivalent to saying that it is total chaos. He says, “‘[W]ithout form and void’ (Gen. 1:2) is not a term for ‘disorderly chaos’ but pictures the earth as ‘an unproductive and uninhabited place.’”[24]

The scholars who insist on the meaning “organized universe” for the compound “the heavens and the earth” do not tell us how organized the universe must be before it can appropriately be designated by the compound. If we push in one direction, the situation in Gen 1:2 is already sufficiently organized, and Waltke’s argument loses all force. So let us try to push in the other direction. Let us specify that the universe has to be organized as completely as it is now in order to count as “organized.” Then we create tension with Gen 2:1 itself. Genesis 2:1 says that “the heavens and the earth were finished,” implying that they underwent a process to get to their finished state. And if so, the implication would seem to be that they are appropriately called “the heavens and the earth” while they are still in the process. (Otherwise, the wording would presumably have been something like “the region above was completed, and the region below was completed, and so God made the heavens and the earth”—that is, the organized spaces are called “the heavens and the earth” only at the endpoint.) In reply to this point, the defender of the idea of complete organization could argue that the earlier stages are only indirectly or proleptically being treated as worthy of the appellation “the heavens and the earth.” But even that is a partial concession.[25]

So the actual wording in Gen 2:1 exhibits some tension with the idea that everything has to be “finished” in order for the whole to be called “the heavens and the earth.” This tension further illustrates that Gen 2:1 does not actually provide positive evidence (as opposed to neutral or negative evidence) for the thesis that the meaning of the merism includes as an essential feature the idea of organization.

We find a similar difficulty with Waltke’s interpretation of Gen 2:4. He cites Gen 2:4 as a second indication that “the heavens and the earth” means the organized universe. But, as usual, the meaning needs to be distinguished from the referent. The meaning is not to be equated with everything that we know about the referent. Even if we suppose that the referent in Gen 2:4 is as fully organized as we might want, yet that does not imply by itself that the meaning of the merism includes as an essential feature the idea of organization.

As for the referent for Gen 2:4, it is once again not so clear how organized it has to be. We may see the difficulty by focusing on the expression “the day” in 2:4, that is, “the day that the Lord God made the earth and the heavens.” This period when God makes things may include the entire span of 1:1–31. In that case, the expression “the earth and the heavens” may be designating the universe as it exists all the way through the process, including its beginning as well as its end. Again, therefore, 2:4 does not support the theory that the compound expression includes the idea of organization in its meaning.

Waltke’s interpretation has still another difficulty, which is even more serious. As we observed, the expression “the heavens and the earth” is transparently built out of its two main constituents, “the heavens” and “the earth.” Both of these regions are referred to not only at the end, when they are fully organized, but several times in the course of the narrative in 1:2–31 (the heavens: 1:9, 14, 15, 17, 20, 26, 28, 30; the earth: 1:2, 11 [twice], 12, 15, 17, 20, 22, 24, 25, 26 [twice], 28 [twice], 29, 30 [twice]). And 1:2 provides clear information. The earth is appropriately designated “the earth” even while it is “without form and void.” Since the meaning of the compound “the heavens and the earth” is built out of the constituent meanings of the parts, 1:2 provides evidence that the compound as well as its constituent parts does not innately contain as an integral and essential element of its sense the idea of a thorough organization. The idea that organization is included in the sense is an illusion created by a confusion between sense and reference.

Waltke summarizes his argument by saying:

If this understanding [that the meaning is “organized universe”], based on its extensive and unambiguous usage in the creation account itself and elsewhere is allowed, then Genesis 1:2 cannot be construed as a circumstantial clause.[26]

This summary is not reassuring. Waltke talks about “extensive and unambiguous usage.” He intends thereby to include in principle many other occurrences of the compound expression (he says, “and elsewhere”). But he has cited only two verses, 2:1 and 2:4. And neither verse is unambiguous in its evidence. In a careful analysis that distinguishes sense and referent, neither verse shows any positive evidence that the idea of organization is an essential element of the sense, in distinction from the referent. Waltke has also completely ignored the evidence offered by 1:2. As for the many other occurrences of the compound expression within the OT, we would have to look at them one by one to see just what evidence they offer. The alleged evidence that has been offered through Gen 2:1 and 2:4 is not convincing. In fact, in subtle ways Gen 2:1 and 2:4 both present problems for Waltke’s thesis.

How then do we actually assess whether the idea of organization is an essential component of the sense of the expression “the heavens and the earth”? We have already done so, by our previous observation that the meaning of the compound is transparently composed out of the meaning of the parts. The idea of organization is not an essential component of the meaning of “the earth” in Gen 1:2. So neither is it an essential component of the compound.[27]

Moreover, the compound through its composition out of two polar opposites functions like other merisms. It enables a reference to a larger whole precisely through the use of the opposites. The particular kind of opposition is still visible in the meaning of the whole. For example, the merism “educated and uneducated” refers to humanity in terms of two parts distinguished by means of education. It thereby draws attention, not to the age distinctions within humanity (young and old), nor to the beauty or ugliness of humanity, nor to the “organization” of humanity, but to one criterion only, namely education. Likewise, the merism “the heavens and the earth” refers to the world in terms of two parts distinguished as higher and lower regions within the whole. The focus, if any, is on the world as composed of regions, and the regions are located in two distinct vertical directions with reference to the observer. This focus on the regions and their locations actually counts against the idea that organization rather than space is essential to the meaning.

We could also go through all the occurrences of the expression “the heavens and the earth” and related expressions in the OT. Many of these would have the organized universe as their referent. But because of the distinction between sense and referent, it would be a delicate task, not a trivial one, to show that this information about referent has any implications at all for incorporation of the idea of organization into the sense.

It is nevertheless informative to search through joint occurrences of “heavens” and “earth” to see the range of usage. We easily find many cases where the two terms occur in a paired way, but not in the exact expression “the heavens and the earth.”

May God give you of the dew of heaven
and of the fatness of the earth
and plenty of grain and wine. (Gen 27:28)

And I will break the pride of your power, and I will make your heavens like iron and your earth like bronze. (Lev 26:19; cf. Deut 28:23)

Out of heaven he let you hear his voice, that he might discipline you. And on earth he let you see his great fire, and you heard his words out of the midst of the fire. (Deut 4:36)

Behold, to the Lord your God belong heaven and the heaven of heavens, the earth with all that is in it. (Deut 10:14)

that your days and the days of your children may be multiplied in the land that the Lord swore to your fathers to give them, as long as the heavens are above the earth. (Deut 11:21)

for the Lord your God, he is God in the heavens above and on the earth beneath. (Josh 2:11)

On that day the Lord will punish
the host of heaven, in heaven,
and the kings of the earth, on the earth. (Isa 24:21)

Thus says the Lord:
“Heaven is my throne,
and the earth is my footstool. (Isa 66:1)

There is actually considerable variety. And in a number of instances it is clear that the polarity between spatial locations, above and below, is very much operative.

We can also find passages with the expression “the heavens and the earth” that are quite similar to passages with a looser pairing between the two terms “heavens” and “earth.” Compare, for example, Jer 32:17 with 51:15:

Ah, Lord God! It is you who has made the heavens and the earth by your great power and by your outstretched arm! (Jer 32:17)

It is he who made the earth by his power,
who established the world by his wisdom,
and by his understanding stretched out the heavens. (Jer 51:15)

The similarities suggest that the expression “God made the heavens and the earth” is similar in meaning to “God made the heavens and God made the earth.” The constituent expressions “the heavens” and “the earth” have their normal meanings in both contexts.

Consider also Deut 31:28 in relation to 32:1, which comes only a few verses later:

Assemble to me all the elders of your tribes and your officers, that I may speak these words in their ears and call heaven and earth (אֶת־הַשָּׁמַיִם וְאֶת־הָאָרֶץ) to witness against them. (Deut 31:28)

Give ear, O heavens (השַָּׁמיַםִ), and I will speak, and let the earth (הָאָרֶץ) hear the words of my mouth. (Deut 32:1)

Deuteronomy 31:28 contains the compound expression in its typical form. Deuteronomy 32:1 separates the two expressions “heaven” and “earth.” But the two verses are talking about the same thing. In Deut 31:28 God says that he will call “heaven and earth” to witness. After the assembly is gathered (31:30), in Deut 32:1 God does just what he said he would do, by commanding the heavens and the earth to listen to the succeeding words. In both 31:28 and 32:1, heaven and earth are personified. But, given this figure of speech, the words “heaven” and “earth” have the same function inside and outside of the compound expression used in 31:28. So a comparison of these two verses in Deuteronomy supports the idea that the compound expression is transparent to the meanings of its two main constituents.

We can also find a few passages where the heavens and the earth occur together with the sea, or with the sea and the dry land:

For in six days the Lord made heaven and earth, the sea, and all that is in them, and rested the seventh day. (Exod 20:11)

who made heaven and earth, the sea, and all that is in them. (Ps 146:6)

I will shake the heavens and the earth and the sea and the dry land. (Hag 2:6)

These passages suggest that, even when they are paired, “heaven” and “earth” retain their normal function, according to which each designates a region.

Finally, we can find passages where the organization of the heavens and the earth is threatened:

Therefore I will make the heavens tremble,
and the earth will be shaken out of its place,
at the wrath of the Lord of hosts
in the day of his fierce anger. (Isa 13:13)

Lift up your eyes to the heavens,
and look at the earth beneath;
for the heavens vanish like smoke,
the earth will wear out like a garment,
and they who dwell in it will die in like manner;
but my salvation will be forever,
and my righteousness will never be dismayed. (Isa 51:6)

I looked on the earth, and behold, it was without form and void;
and to the heavens, and they had no light. (Jer 4:23)

The Lord roars from Zion,
and utters his voice from Jerusalem,
and the heavens and the earth quake. (Joel 3:16)

I will shake the heavens and the earth. (Hag 2:21)

The last two verses, Joel 3:16 and Hag 2:21, are particularly telling, because they have the compound expression (Joel 3:16 without definite articles in Hebrew) (cf. also Hag 2:6). In all these verses, the threat to organization is compatible with the regions still being called “heavens” and “earth.”

c) The appeal to philology. In his wrap-up to his first argument, Waltke appeals to philology: “it is impossible to do so [take v. 2 as a further description of the result of v. 1] on philological grounds.”[28] In some ways this conclusion is only a summary of the subpoints we have already discussed. But it is still worthwhile to make two observations about the flexibility of language and the flexibility of meanings within a language.

First, the events described in Gen 1 are unique in the whole history of the world. The world was only created once. It continues under God’s providential rule for ages afterward. The events described in Gen 1 include events involving origination, such as the first creation of light, the first creation of plants, and the first creation of dry land. In their uniqueness, events of origination are necessarily unlike later events under the providential control of God. Moreover, the human beings who are addressed by the narrative in Gen 1 have not themselves been eye-witnesses to the events of origination. So the only way of intelligibly describing such unique events is by way of analogy with events in providence.[29] And analogy is not identity. Therefore, we must not expect that the descriptive usages in Gen 1 will exactly match the later usages with respect to providential events. In particular, the fact that later references to heaven and earth refer to them in an organized state does not force an identical form of organization onto Gen 1.

Second, word meanings in ordinary language include flexibility.[30] They do not function like technical terms, whose boundaries of meaning are precisely fixed. We can use old words in new contexts, and readers adjust.

Let us consider an example, with the English expression “the world.” Among modern English words for referring to the universe, the word world is not often linked with discussions of the origin of the universe. Such discussions take place in the domain of technical science (cosmology) and popularized science. In the context of science, expressions like “the universe” and “the cosmos” are customary. By contrast, the expression “the world” occurs in more commonplace contexts.[31] The result is that almost all the occurrences of the expression “the world” refer to the organized world familiar to us today, or some subdivision of it: “the world in which we live,” “the world of finance,” “the world of music,” and so on. Or the expression can refer to former ages, but still with the organization of human culture: “the world of ancient Greece” or “the world of the Renaissance.”

But it takes only a moment to produce a discourse that stretches out beyond these more customary uses, and refers to a less organized world:

In the beginning God created the world. The world was without form and void, and darkness was all around. Then God said, “Let there be light.”

The use of “the world” in the second sentence seems superficially to “contradict” the normal pattern in which the expression refers to the modern, organized world. But does the average reader see a contradiction? Or does he quickly adjust, by seeing that the author of the discourse has chosen to use the expression in a loose or more extended way?

The same reasoning applies by analogy to the Hebrew expression underlying “the heavens and the earth.” The use of the expression “the heavens and the earth” in Gen 1:1 can be taken in stride by a reader who is accustomed to frequent references to the world of providence, subsequent to the completion of the days of creation. Thus, the philological problem that Waltke finds with the initiation view does not really exist.

The irony is that, rightly assessed, philology weighs heavily against the summary view that Waltke champions “on philological grounds.” A proper understanding of philology notes the transparency of meaning in most merisms, the necessity of analogical use in describing unique events, and the flexibility of meaning. Almost by itself, the first of these principles destroys the claim that the idea of organization is included in the meaning of the compound expression “the heavens and the earth.” And that is at the heart of the argument against the initiation view.

2. The Theological Issue Of God Creating A Formless Earth

In addition to the argument for the meaning “the organized universe,” Waltke has two other supporting arguments in favor of the summary view of Gen 1:1. One of these is an argument from theology. Waltke argues that it is theologically inappropriate to say that God would create a formless entity.

a) Isaiah 45:18. Waltke begins by discussing Isa 45:18:

For thus says the Lord,
who created the heavens
(he is God!),
who formed the earth and made it
(he established it;
he did not create it empty [לֹא־ תֹהוּ בְרָאָהּ],
he formed it to be inhabited!):
I am the Lord, and there is no other.

Waltke claims that this verse is incompatible with the idea that God created the earth initially empty (תהוּ) in Gen 1:2.[32]

Waltke’s interpretation of Isa 45:18 is awkward. It is as if we were to take a single line out of the verse and treat it as a technical discussion. That is, we treat “he did not create it empty” as if it precisely targets the issue of an early unformed state. According to such an interpretation, this poetic line precisely denies that the earth in Gen 1:2 was the result of a creative act.

But Isa 45:18 is not a technical discussion. It is poetry. Within the key line, “create” is not to be construed narrowly as focusing solely on the initial act of bringing something into existence out of nothing. The parallels in the surrounding lines show that “create” is construed more broadly, as parallel to “formed” (twice), “made,” and “established.” In context, the key line is referring in broad fashion to the sequence of events that include everything that God did over the course of the six days in order to prepare the earth to be a suitable environment for man. This broad scope is made particularly evident in the next-to-last line, “he formed it to be inhabited.” By using the words for “created” and “empty” together in the key line, the verse is making an allusion to Gen 1:1–2. But that allusion functions as part of a verse that is commenting on “creation” in a broad sense, including in principle the entire sequence in 1:1–31.

Elsewhere in his articles, Waltke himself seems to agree that Isa 45:18 is speaking broadly about the entire process taking place in the six days, and that the endpoint of the process is the completed work of Gen 1:31: “He [God] did not end up with chaos, as Isaiah noted (Isa. 45:18).”[33] That is, Isa 45:18 is saying that chaos is not the endpoint. But that is consistent with saying that, at an earlier point in time, God might have brought into existence an earth that lacked much of the later organization. So there is no contradiction between Isa 45:18 and any part of Gen 1, whether we hold the initiation view or the summary view of Gen 1:1. Isaiah 45:18 is irrelevant to deciding between the two views.

b) Formless and void. Next, Waltke appeals to the meaning of “formless and void”:

Then too it has been demonstrated from Jeremiah 4:23 and Isaiah 34:11 that תהוּ ובהוּ denotes the antithesis of creation.[34]

But in this statement there is vagueness about the word “creation,” as well as difficulties with the related idea of being organized or unorganized. In the quoted statement from Waltke, does the word creation focus on the initial act of creation? Or does the work of creation include every step in the overall transition, including beginning the work, passing through less unorganized states, and arriving at a well-organized state?

And what kind of “antithesis” do we have in view? To undo the present order of things and to return the world to a less unorganized state is in some respects the antithesis of building up multiple kinds of organization over the course of the six days of creation. It is the “antithesis” of creation as an overall transition. But how could it be the antithesis of the initial act of creation from nothing? Has Waltke’s argument unconsciously slipped-in the assumption that “creation” never means creation out of nothing into a temporary situation that is relatively unorganized?

If God had left the created world in the situation described in Gen 1:2, it would have been unsuitable for human habitation. Given that his purposes included human habitation, the situation in Gen 1:2, as a static situation, apart from further development, is at “odds” with the endpoint that God has purposed. But of course it is a mistake to isolate Gen 1:2 from the overall purposes of God involved in the entire narrative in Gen 1. Once we have the entire narrative, we can see that the initial production of the earth in an uninhabitable state is quite in accord with his purposes. We must just be careful to take into account the theme of development, and to see that the state of Gen 1:2 was intended by God, and created by God, but never intended just to stay that way.

c) God’s order. Next, Waltke appeals to the fact “that elsewhere in Scripture it is said that God created everything by His Word”[35] (Ps 33:6, 9; Heb 11:3). But “no mention is made anywhere in Scripture that God called the unformed, dark, and water state of verse 3 [sic; v. 2] into existence.”[36] This is an argument from silence, and a weak one at that. Waltke’s own words contradict him, because he says that “God created everything by His Word.” “Everything” means everything, as the verses that he cites show. The initial watery state of Gen 1:2 is included by implication in “everything.”

Next, Waltke appeals to the absence of sea and darkness in the new heaven and new earth. “This revelation about the new cosmos [Rev 21:1, 25] suggests that the deep and darkness in verse 2 [of Gen 1] are less than desirable and were not called into existence by the God of order and goodness.”[37] Collins has discerningly replied that in the visionary context of Revelation the sea and darkness are used “as symbols for what fallen man fears rather than as comments on the moral status of sea and night in themselves.”[38] Moreover, Waltke’s aspersions with regard to the deep and darkness are in danger of ignoring the importance of history and development. The deep and darkness are indeed “less than desirable” if they are regarded as endpoints in the development of creation. The earth in Gen 1:2 does not yet present a suitable habitation for man. But what is undesirable as an endpoint may be fully in accord with the will and plan of God for an early stage.[39]

Waltke calls God “the God of order and goodness.” Yes. But there is a goodness and a wise temporal order to be found in Gen 1:2, if we see it in relation to the subsequent narrative in Gen 1. Genesis 1:2 is one phase in the total process, and its “order” and “goodness” cannot rightly be evaluated if we isolate it from the larger narrative in which it is embedded.

Waltke’s language unfortunately opens the door to an unbiblical idea of God, according to which God is only sovereign creator with respect to some pieces of the total picture. Are we supposed to say that he brings about order but not disorder? I should hope not! Moreover, to say that God did not create the deep and the darkness directly contradicts a number of NT texts (Col 1:16; Heb 11:3; Rev 4:11)[40] and distorts our conception of the sovereignty of God.

This particular argument by Waltke is not his finest hour. Fortunately, in his 2001 commentary Waltke shows a change in his position: he affirms that God “made everything.”[41] This change removes the basis for his whole argument in 1975 concerning the alleged theological inappropriateness of God creating a formless earth.

d) Parallels in ancient myths. In his key arguments, Waltke does not appeal directly to the theme of primeval chaos found in some ancient Near Eastern myths.[42] But other interpreters do appeal to this factor in order to argue that the ancient Near East had no concept of creation from nothing, and so such a concept cannot be found in Gen 1:1. Genesis 1:2 therefore describes the original condition that is the starting point for creative activity. Interpreters may also appeal to later poetic biblical texts that use the imagery of God triumphing over the sea and over the sea monster, “Leviathan.”

A brief reply may include four points. First, God, the true God, may say and do something different from and even in contrast to the ancient Near East. Second, there is a partial parallel in some Egyptian texts that have Ptah producing everything, including the primeval waters.[43] Third, in dealing with the ancient Near Eastern myths, one must ask whether the sea or water god(dess) is genuinely primeval. In contexts where gods give birth to other gods, the sea god is not necessarily first. So the waters are not just “there,” but come from something more ultimate. Fourth, the biblical texts that poetically invoke a picture of God defeating a sea monster must be used with sensitivity to their poetry: they are examples of imagery rather than theories about an initial chaos. Moreover, the specific terminology used in these texts often has connections with terms in later verses in Gen 1—the “seas” in v. 10 and “great sea creatures” in v. 21, both of which are clearly created by God and over which he is thoroughly sovereign.[44] Therefore, specific vocabulary choices in poetic biblical language actually count against the idea that Gen 1 is playing with a theory of original chaos outside the creative activity of God.

3. Structural Evidence For Genesis 1:1 As A Summary

Waltke’s final, third argument focuses on parallel structures.

a) A parallel with Genesis 2:4–7. The most impressive parallel in Waltke’s exposition is the one between Gen 1:1–3 and Gen 2:4–7. According to Waltke, each text is composed of three pieces:

(1) “Introductory summary statement” (1:1; 2:4);

(2) “Circumstantial clause of the pattern waw + noun + verb (היה) describing a negative state before creation” (1:2; 2:5–6);

(3) “Main clause of the pattern waw consecutive + prefixed conjugation form describing the creation” (1:3; 2:7).[45]

This parallelism may look impressive. But we must recognize that (2) and (3) represent common ways of producing circumstantial clauses and main clauses, respectively. And of course Waltke’s label (1), “introductory summary statement,” is only fitting if the summary view is correct.

There are also some differences between Gen 1:1–3 and 2:4–7.

(1) The material in 2:4 is not really a heading solely for events in which God creates new things. It is an introductory title for the entire section 2:5–4:26. As the outline and discussion in Waltke’s commentary recognizes,[46] 2:4 is the first of several headings in the form “these are the generations of Noah” (6:9), “these are the generations of Terah” (11:27), “these are the generations of Isaac” (25:19), and so on. (5:1 is only slightly different in wording, “this is the book of the generations of Adam.”) Each section introduced by a heading mainly contains, not the account of the origin of the named person, that is, Noah or Terah or Isaac, but the account of the subsequent history (the “generations”) involving the named person and his descendants. The word “these” points forward to the entire section. It signals that the sentence is a heading for the section. This key use of “these” is unlike 1:1, which does not contain the key word. Nothing in 1:1 clearly marks it out as a heading.

Note also that the section 2:5–4:26 gives no attention to the creation of the heavens or heavenly lights. It is a more focused account. It is about the “generations” of the heavens and the earth, the history and the products that flow from them. So it does not run fully parallel to the creation account in 1:1–2:3. Thus also, vv. 5–6 are not “describing a negative state before creation” of the world as a whole, but an undeveloped state before the creation of Adam and the garden of Eden.

(2) 1:1, unlike 2:4, has a main verb in the perfect. A clause with this structure can naturally be construed as describing the first event in a series unfolding in the subsequent verses. 2:4 does not have this feature.[47]

(3) Typical cases in which Hebrew discourse supplies a heading have clear signals that it is a heading. The use of the word “these” in the headings to the sections of Genesis is such a signal. Similarly we have “These are the names of the sons of Israel” (Exod 1:1); “These are the words that Moses spoke to all Israel beyond the Jordan” (Deut 1:1); “These are the words of the letter that Jeremiah the prophet sent from Jerusalem” (Jer 29:1). Or more simply, a heading may use a phrase instead of a clause: “The proverbs of Solomon, son of David, king of Israel” (Prov 1:1); “The words of the Preacher, the son of David, king in Jerusalem” (Eccl 1:1); “The vision of Isaiah the son of Amoz” (Isa 1:1). All these expressions show by their special form that they are headings. In the absence of such special signals, Gen 1:1 is to be construed as describing an initial event.

(4) As we have observed, the linkage between 1:1 and 1:2 through the key term “the earth” is naturally interpreted as an indication that 1:2 begins with a circumstantial clause linked backward to 1:1. The linkage between 2:4 and 2:5 is not as tight. It is true that the Hebrew for “earth” (ארץ) occurs twice in 2:4, twice in 2:5, and once in 2:6. (ESV translates it “the land” in 2:5–6 and includes an explanatory footnote.) But in all three occurrences in vv. 5–6, the key word has a subordinate, inconspicuous role in the narrative. Verses 5–6 have in focus the lack of bushes, plants, rain, and man, and the presence of mist. They do not start off with the land itself as the subject. In addition, because 2:4 in its structure clearly identifies itself as a heading for the entire subsequent narrative, 2:5–6 can only be construed as circumstantial clauses linked forward to the main clause in 2:7. The same is not true concerning 1:1–3.

(5) Genesis 2:4–7 comes too late to affect the ordinary Israelite’s interpretation of the basic meaning and syntax of Gen 1:1–3. The reader has already found out what it means, long before coming to 2:4–7. To read quite a different significance into the sequence of Gen 1:1–3 on the basis of 2:4–7 is therefore suspect.

b) A parallel with Genesis 3:1. Next, Waltke appeals to parallels between Gen 1:1–3 and 3:1.[48] According to Waltke, in 3:1 the heading is supplied by 2:4. The circumstantial clause is 3:1a, and the main clause begins in 3:1b. But this is a weak analogy, because Waltke finds the heading for 3:1 all the way back in 2:4. Moreover, 2:4 is not the heading for 3:1–7 or 3:1–24 as such, but for the entire section, 2:5–4:26. So observations about 3:1 cannot have much relevance for determining whether 1:1 is the heading for what comes immediately after it.

c) A parallel with the beginning of Enuma Elish. Waltke also appeals to the Enuma Elish, which begins with a circumstantial clause and then a main clause.[49] But there is no heading in the Enuma Elish. So this alleged parallel again does not help us to determine whether 1:1 is a heading.

In various cases describing development, it is natural to start with a description of a relatively undeveloped state. So the transitions from undeveloped to developed state are natural, quite apart from whether some preceding material functions as a heading, a summary, or an earlier event preceding the process of development.

In sum, the parallels that Waltke finds are too loose to serve as persuasive evidence.

III. Conclusion

In conclusion, all three of the main arguments for the summary view have superficial plausibility, but none has weight. In addition, as of 2001, Waltke himself no longer holds to the second argument contained in his earlier work (in 1974 and 1975). The summary view is much weaker than many have taken it to be. By contrast, the initiation view makes good sense of the phrase meanings, theology, and syntax of Gen 1:1–2 in relation to Gen 1:1–2:3 as a whole, and beyond (the rest of Genesis and the rest of the Bible). It is the correct view.

Notes

1. For discussion of these interpretations, see Gordon J. Wenham, Genesis 1-15 (Waco: Word, 1987), 11-13; C. John Collins, Genesis 1-4: A Linguistic, Literary, and Theological Commentary (Phillipsburg, NJ: Presbyterian & Reformed, 2006), 50-55; Edward J. Young, “The Relation of the First Verse of Genesis One to Verses Two and Three,” WTJ 21 (1959): 133-46, reprinted in Edward J. Young, Studies in Genesis One (Philadelphia: Presbyterian & Reformed, 1964), 1-14; Bruce K. Waltke, “The Creation Account in Genesis 1:1-3: Part III: The Initial Chaos Theory and the Precreation Chaos Theory,” BSac 132 (1975): 216-28. This article by Waltke is the third in a total of five articles in which he addresses aspects of Gen 1:1-3. The others are “The Creation Account in Genesis 1:1-3: Part I: Introduction to Biblical Cosmogony,” BSac 132 (1975): 25-36; “The Creation Account in Genesis 1:1-3: Part II: The Restitution Theory,” BSac 132 (1975): 136-44; “The Creation Account in Genesis 1:1-3: Part IV: The Theology of Genesis 1, ” BSac 132 (1975): 327-42; and “The Creation Account in Genesis 1:1-3: Part V: The Theology of Genesis 1—Continued,” BSac 133 (1976): 28-41. An editorial note accompanies each of these articles, noting that they are adapted from the Bueermann-Champion Foundation Lectures at Western Conservative Baptist Seminary, Portland, Oregon, Oct. 1-4, 1974, and published as Creation and Chaos (Portland, OR: Western Conservative Baptist Seminary, 1974). We will focus on the 1975 articles rather than the 1974 book, because the 1975 articles are more widely accessible. (The 1974 and 1975 pieces seem to be almost, but not quite, identical in wording.) Part III in the series in BSac is especially relevant for the purposes of this article, and we will regularly cite it simply as “Waltke, ‘Part III.’” Waltke’s 2001 commentary (Bruce K. Waltke, with Cathi J. Fredricks, Genesis: A Commentary [Grand Rapids: Zondervan, 2001], 58-59), maintains the same basic interpretation of Gen 1:1, but contains only a short version of the first of his three main arguments set forth in 1975. A later work in 2007 also contains a shorter version of the first argument and a piece of the second (Bruce K. Waltke, with Charles Yu, An Old Testament Theology: An Exegetical, Canonical, and Thematic Approach [Grand Rapids: Zondervan, 2007], 179-81). A fourth interpretation, sometimes called “the gap theory,” now receives little attention. But it used to be advocated, and was popularized by the Scofield Bible note on Gen 1:2 (The Scofield Reference Bible, ed. C. I. Scofield, new and improved ed. [New York: Oxford University Press, 1917], 3n3). The theory says that there is a time gap between vv. 1 and 2. Gen 1:1 briefly describes God’s creation of an initial good creation, while Gen 1:2 describes a subsequent ruination (“the earth became without form and void”) of that creation, as an act of judgment. Gen 1:3-31 describe a re-creation after the ruination. In support of this idea, Scofield’s note (ibid.) cites Jer 4:23-26; Isa 24:1; 45:18. But the gap theory is now largely abandoned, because it is does not conform to the natural reading of the Hebrew in 1:2. The word order of v. 2 indicates that the verse introduces an accompanying circumstance rather than an advance in the main events in the narrative. For a critique of the gap theory, see Waltke, “The Creation Account in Genesis 1:1-3: Part II.”
2. Waltke, “Part III,” 221-25; Collins, Genesis 1-4, 50-52; Victor P. Hamilton, The Book of Genesis: Chapters 1-17 (Grand Rapids: Eerdmans, 1990), 104-8; Young, “Relation,” 133-39; Young, Studies, 1-7; Nicolai Winther-Nielsen, “‘In the Beginning’ of Biblical Hebrew Discourse,” in Language in Context: Essays for Robert E. Longacre, ed. Shin Ja J. Hwang and William R. Merrifield (Dallas: The Summer Institute of Linguistics and the University of Texas at Arlington, 1992), 67-80, http://www.sil.org/system/files/reapdata/12/98/61/129861883369277823521029452481206904550/31844.pdf, accessed Nov. 2, 2016.
3. Waltke, “Part III,” 217.
4. Wenham, Genesis, 12.
5. Collins, Genesis 1-4, 51-55; Wenham, Genesis, 12-13. Wenham also cites others (p. 13).
6. Waltke, “Part III.” Collins considers Waltke’s article to be “the strongest case” for the summary view (Collins, Genesis 1-4, 54).
7. The commentaries on Genesis have voluminous discussion of source theories. Source theories can make a difference, because often they treat Gen 1:2 as stemming from a primitive tradition that starts with chaos and thereby repudiates any idea of creation out of nothing. If someone accepts this assumption, and treats Gen 1:2 as still meaning what it meant at the earlier stage, he has already confined himself to only two options: either to say that Gen 1:1 does not describe creation out of nothing or to say that it contradicts Gen 1:2 because two disparate sources have not been satisfactorily united. My approach is to interpret the text as it stands, and to presuppose that, even if there are sources behind it, the meaning of the text can differ from its sources.
8. Waltke agrees that v. 2 is circumstantial, and that “on syntactical grounds” it could be attached backward to v. 1 (“Part III,” 221); but he thinks that it provides circumstantial information connecting it forward to v. 3 (pp. 226-27). On circumstantial clauses, see Joüon, §155nc.
9. On the literary division occurring between 2:3 and 2:4, see Collins, Genesis 1-4, 40-42.
10. E. J. Young advocates a form of the summary view, but also thinks that Gen 1:1, though not directly focusing on the initial act of creation out of nothing, indirectly implies it (Young, “Relation,” 141; Studies, 9).
11. Collins, Genesis 1-4, 51-52.
12. Waltke, “Part III.” Waltke, Genesis, 58-59, and Waltke, Old Testament Theology, 179-81, also contain shortened versions of some of the same arguments.
13. Waltke, “Part III,” 218 (italics mine); Waltke, Old Testament Theology, 179; so also Young, “Relation,” 142n17; Studies, 10n17.
14. Waltke, “Part III,” 218; Waltke, Old Testament Theology, 179.
15. Bruce K. Waltke, A Commentary on Micah (Grand Rapids: Eerdmans, 2007), 456.
16. Waltke, “Part III,” 218. The first of the two illustrative merisms is quoted by Waltke from Cyrus H. Gordon, The World of the Old Testament (Garden City, NY: Doubleday, 1958), 35. See also Waltke, Old Testament Theology, 179.
17. Waltke, “Part III,” 218.
18. Ibid.
19. E. J. Young, who advocates a form of the summary view, does notice the problem. In explanation, he says, “Verse two does obviously [!] connect with verse one and employs the word הָאָרֶץ in a sense different from that which it had in the first verse. In verse two הָאָרֶץ serves as a practical equivalent of our designation ‘the earth.’ It is the earth as we now know it.... Hence, the thought may be paraphrased as follows: ‘And the earth (i.e., the earth we now know) at that time was desolation and waste’” (Young, “Relation,” 142n17; Studies, 10n17). This explanation does not make sense to me. The earth we now know is fully structured and filled with inhabitants; it is not “desolation and waste.” The earth at a stage of “desolation and waste” is simply not the earth we now know. Why then does Young choose to describe the earth at an earlier stage (in v. 2) as the earth we now know (“It is the earth as we now know it”)? And then, with even more confusion, he describes the earlier stage as “the earth (i.e., the earth we now know) at that time”? So which time is it, “now” or “at that time”? If it is the same earth at both times, then Young appears to deny the very difference that he is concerned to describe. If, on the other hand, it is not the same earth, why does he describe it as “(i.e., the earth we now know) at that time”? Of course the early “earth” may be viewed as having some minimal continuity with the earth now. But if that is the way Young is looking at things, and he wants to stress that the earth at the two stages is in some sense the same because of the elements of continuity, why does he say that there are two different senses in v. 1 and v. 2? Rather than provide a supporting argument, Young’s explanation seems to me only to illustrate the difficulty. On the one hand, he says that there is an obvious connection between v. 1 and v. 2. And we can see that the heart of the connection consists in the repetition of the term “the earth.” On the other hand, he thinks there is a radical difference: “the earth” in v. 1 is the organized earth, while “the earth” in v. 2 is yet to be formed and organized (Young, “Relation,” 141-43; Studies, 9-11). He lapses into incoherence when he tries to maintain that “the earth” in v. 2 is and is not the same as “the earth” in v. 1. Other analysts, such as Gunkel, escape the problem by postulating that 1:1 and 1:2 go back to distinct sources (Hermann Gunkel, Genesis, trans. Mark E. Biddle [Macon, GA: Mercer University Press, 1997], 104 [p. 103 in the original German]). The difficulty is generated only if an interpreter—whether Waltke or Young or Gunkel or another—decides that v. 1 is referring to the organized, completed heaven and earth. This move is a common mistake. How it came to be so common is revealed in our subsequent discussion.
20. Waltke, “Part III,” 218. Waltke, Old Testament Theology, 179, offers the example “Butterfly is quite different from butter and fly.”
21. Waltke, “Part III,” 218.
22. Ibid. (italics mine).
23. Hamilton makes the case for the sense “Spirit of God” (Genesis, 111-14).
24. Collins, Genesis 1-4, 54.
25. In another context in his article, Waltke rejects an explanation that appeals to proleptic use of the compound expression “the heavens and the earth” (Waltke, “Part III,” 219). His rejection of proleptic use only increases the difficulty that his view has with 2:1.
26. Ibid. For proper punctuation, the quoted material should have an extra comma after “elsewhere.” I have let it stand as it is printed in the article. Waltke, Creation and Chaos, 26, includes the extra comma.
27. In favor of the idea of organization, Waltke, “Part III,” 218-19, includes a supporting quote from John Skinner, A Critical and Exegetical Commentary on Genesis (Edinburgh: T&T Clark, 1910), 14. But Skinner provides no evidence for his claim.
28. Waltke, “Part III,” 221.
29. Vern S. Poythress, “Correlations with Providence in Genesis 1, ” WTJ 77 (2015): 71-99.
30. See the discussion of variation in Vern S. Poythress, In the Beginning Was the Word: Language—A God-Centered Approach (Wheaton, IL: Crossway, 2009), 154-55; Kenneth L. Pike, Linguistic Concepts: An Introduction to Tagmemics (Lincoln: University of Nebraska Press, 1982), 52-59.
31. The online Merriam-Webster Dictionary provides a simple definition of “world”: “the earth and all the people and things on it : a part of the world and the people and things that exist there : human society” (http://www.merriam-webster.com/dictionary/world, accessed April 28, 2016). This definition clearly focuses on the organized world with people and things already in it. The same online page then provides a “Full Definition” with 14 distinct senses. We consider only the most pertinent. Sense 2 is “the earth with its inhabitants and all things upon it.” The mention of “inhabitants” and “all things” clearly has in mind a structured “world,” more or less like the present. Sense 6 is “the system of created things : universe.” At first glance, this sense might seem to be synonymous with the word universe. But the word system gives the sense of an organized whole, and the word created—in the past tense—indicates that we are thinking of the world in something like the present state. None of the 14 senses naturally brings to mind the kind of physical situation that existed long in the past, before the solar system was formed.
32. Waltke, “Part III,” 220.
33. Waltke, “The Creation Account in Genesis 1:1-3: Part IV,” 342; also: “Isaiah 45:18 has reference to the completed creation at the end of six days” (Waltke, “The Creation Account in Genesis 1:1-3: Part II,” 144).
34. Waltke, “Part III,” 220. Waltke provides a footnote referring readers back to Waltke, “The Creation Account in Genesis 1:1-3: Part II,” 136-44.
35. Waltke, “Part III,” 220.
36. Ibid., 221.
37. Ibid. Waltke, however, wants to assert the complete sovereignty of God over the deep and the darkness (Waltke, “The Creation Account in Genesis 1:1-3: Part IV,” 338-39). God exerts control over them, but still Waltke does not think that God originated them.
38. Collins, Genesis 1-4, 54n55.
39. Waltke is not alone in mistakenly assessing the state described in Gen 1:2. He quotes approvingly from Brevard S. Childs, who says, “It is rather generally acknowledged that the suggestion of God’s first creating a chaos is a logical contradiction and must be rejected” (Childs, Myth and Reality in the Old Testament, 2nd ed. [London: SCM, 1962], 30, quoted in Waltke, Old Testament Theology, 179). If such a view is “rather generally acknowledged,” it is not a good sign for the state of OT scholarship. Such a view appears to ignore the distinction between different kinds and degrees of organization, between complete chaos and being “without form and void,” and between a permanent condition and a starting point for development. In addition, Childs’s choice to use the language of “logical contradiction” is inappropriate.
40. Collins, Genesis 1-4, 53.
41. Waltke, Genesis, 68. In its key section, the 2001 commentary on Genesis does not refer to Waltke’s earlier articles in BSac. My judgment about his change of position is therefore an inference. Waltke, Old Testament Theology, 180, published in 2007, seems to me to be more ambivalent, and on pp. 180-81 revives pieces of the second argument found in Waltke, “Part III” (from 1975), the argument about theological inappropriateness. The distance between Waltke’s three pieces is not too great. In both 2001 and 2007 Waltke calls the state of Gen 1:2 “surd” (Genesis, 68; Old Testament Theology, 180).
42. However, Waltke does mention parallels in the fourth article of his series (“The Creation Account in Genesis 1:1-3: Part IV,” 329), after he has finished the main arguments in “Part III.” And he reviews ancient Near Eastern cosmogonies in the first article in the series (“The Creation Account in Genesis 1:1-3: Part I”). Moreover, Waltke, Old Testament Theology, 181-83, discusses the myths.
43. Wenham, Genesis, 13; Viktor Notter, Biblischer Schöpfungsbericht und ägyptische Schöpfungsmythen (Stuttgart: KBW, 1974), 23-26; Vincent Arieh Tobin, “Creation Myths,” in “Myths,” The Oxford Encyclopedia of Ancient Egypt, ed. Donald B. Redford (Oxford: Oxford University Press, 2001), 2:471. Note, however, that the productions by Ptah appear to be similar to emanations, so there is no clear Creator-creature distinction. Nothing in ancient Near Eastern polytheism is truly parallel to the monotheism of Gen 1.
44. Waltke, Old Testament Theology, discusses the theme of chaos and the primordial waters in the ancient Near East (p. 181) directly on the heels of an argument claiming that Gen 1 does not reveal the origin of “the primordial water” (p. 180). Both discussions fall under Waltke’s section heading, “Negative State of the Earth before Creation (1:2)” (italics mine). So Waltke may be using the ancient Near East as an extra support for his summary view. As a biblical parallel to the ancient Near Eastern theme of chaos, Waltke quotes Ps 74:12-17, and interprets the reference to “the sea” in v. 13 by inserting brackets: “[Yamm]” (pp. 181-82). He intends to indicate that v. 13 has a parallel in the role of the god “Yamm” in Canaanite myth. Yes, there may be allusion to such myths. But such allusion still does not imply that the Bible endorses a theory of original chaos. The Bible must be allowed to speak with its own voice. Unfortunately for the theory of original chaos, the Hebrew word in v. 13 corresponds to “seas” in Gen 1:10. And the parallel line in the second half of the same verse has “the sea monsters,” the same word as in Gen 1:21. The details in Ps 74:13 actually fight against Waltke’s suggested alignment of Ps 74:13 with initial chaos. Waltke also cites Ps 77:17 (English v. 16), where the word for “deep” occurs. But here the context provides a poetic recital concerning the exodus from Egypt, and the waters in question are the waters of the Red Sea, which Gen 1:10 affirms to be part of the created order. The argument for seeing Gen 1:2 as pre-creation chaos has three doubtful steps. First, read into the ancient Near East an affirmation of initial chaos. But in a polytheistic context, the sea god is not necessarily first. Second, transfer the entire theory of initial chaos, rather than looser poetic imagery of triumph, from the myths into OT poetry. Third, project the OT poetry back onto Gen 1:2 rather than onto later events to which it may be more directly related—the exodus, the flood, the creation of seas (Gen 1:10), and the creation of the great sea creatures (v. 21).
45. Waltke, “Part III,” 226. The indented block (1)–(3) as a whole is not a direct quote, but a summary of Waltke’s presentation.
46. Waltke, Genesis, 18.
47. At another point, Waltke carefully notes the differences (Waltke, “Part III,” 225).
48. Ibid., 227.
49. Ibid.