Copyright © 1998 by
THE EARTH OF GENESIS 1:2
ABIOTIC OR CHAOTIC?
The famous German scholar Hermann Gunkel (1862-1932), well-known
advocate of Formgeschichte, tried to demonstrate that the battle in which
Yahweh defeated the sea monster of the chaos was related to the Hebrew
account of creation in Genesis 1. He assumed that the Babylonian creation
account, with its Chaoskampf or battle between the creator-god and the powers
of the chaos, was the basis for the mythical imagery that appears in the Bible.1
Since the discovery of the Ugaritic myths, the existence of a conflict
between Yahweh and the sea dragons (Leviathan and Rahab in poetical texts
of the OT) has been widely accepted.2 This Canaanite conflict motif has
been related to the biblical creation story as "a missing link" which supports
the apparent Chaoskampf in Gen 1:2. Frequently, the Chaoskampf that appears
in the Babylonian Enuma elish and the Ugaritic Baal myth is considered the
main foundation of any cosmogony in the Ancient Near East (ANE).3 For
instance, J. Day assumed that Gen 1:2 is a demythologization of the original
Chaoskampf myth of ancient Canaan.4 R. J. Clifford and J. J. Collins have
proposed that Genesis 1 begins with a mythical combat between the dragon
1 H. Gunkel,
and erklart, HKAT 3/1 (
Ruprecht, 1901); reprinted with introduction by W. F. Albright in The Legends of Genesis:
The Biblical Saga and History (New York: Schocken, 1974).
2 A. Cooper, "Divine Names and Epithets in the Ugaritic Texts," in Ras Shamra
Parallels, ed. Loren Fisher (Rome: Pontificium Institutum Biblicum, 1981), 3:369-383.
3 See C. Kloos, Yhwh's Combat with the Sea: A Canaanite Tradition in the Religion of
Echoes of a Canaanite
Myth in the Old Testament (
4 Day, 53.
260 SEMINARY STUDIES 35 (AUTUMN 1998)
of chaos and the divine sovereign.5
Gunkel stated that the Hebrew term tehom in Gen 1:2 had a Babylonian
background.6 He suggested that tehom derived directly from Tiamat, the Babylonian
goddess of the primordial ocean in the Enuma elish. Since Gunkel's statement,
many scholars have assumed some kind of direct or indirect connection between
the Babylonian Tiamat and the Hebrew tehom.7 Many have accepted that
the Hebrew tehom in Gen 1:2 has a mythological foundation in Tiamat, the
goddess of the Enuma elish, in which Marduk the storm god fights and defeats
Tiamat the sea dragon, thus establishing the cosmos.8
The expression tohu wabohu, "emptiness and waste," in Gen 1:2 is of-
ten considered a reference to this primordial "chaos," in strict opposition
to "creation." The phrase is taken to refer to the earth in an abiotic or lifeless
state, with no vegetation, animals, or human beings.9
Gunkel also posited the theory, later supported by other scholars, that
the ruah elohim in Gen 1:2c corresponds to the winds that Marduk sends
against Tiamat, thus assuming that it is an expression that describes the pri-
The object of this three-part article is to discover whether in Gen 1:2
there is any evidence for the mythological battle between the creator-god
and the powers of the chaos, Chaoskampf, such as Gunkel and many other
scholars maintain.10 If we found such evidence, we would need to take heed
5 R. J. Clifford and J. J. Collins, eds., Creation in the Biblical Traditions, CBQ
Monograph Series 24 (Washington, DC: Catholic Biblical Association of America, 1992), 32-
33. See also R. J. Clifford, Creation Accounts in the Ancient Near East and in the Bible, CBQ
Monograph Series 26 (Washington, DC: Catholic Biblical Association of America, 1994).
6 H. Gunkel, "Influence of Babylonian Mythology upon the Biblical Creation Stories,"
in Creation in the Old Testament, ed. B. W. Anderson, Issues in Religion and Theology 6
(Philadelphia: Fortress, 1984), 25-52; first published in Schopfung und Chaos in Urzeit und
7 B. S. Childs, Myth and Reality in the Old Testament (London: SCM, 1960), 36; B. W.
Anderson, Creation versus Chaos: The Reinterpretation of Mythical Symbolism in the Bible
(Philadelphia: Fortress, 1987), 15-40; K. Wakeman, "The Biblical Earth Monster in the
Cosmogonic Combat Myth," JBL 88 (1969): 313-320; idem, God's
Study in Biblical Imagery (Leiden: Brill, 1973), 86ff.
8 For a translation and discussion of this text, see A. Heidel, The Babylonian Genesis, 2d
ed. (Chicago: University of Chicago Press, 1951); see also the translation by E. A. Speiser in
Creation Epic," ANET, 60-72. The
most recent translation can be seen in
Myths from Mesopotamia:
Creation, The Flood, Gilgamesh, and Others (
University Press, 1991), 233-274.
9 See D. T. Tsumura, "The Earth in Genesis 1," in I Studied Inscriptions from Before the
Flood, ed. R. S. Hess and D. T. Tsumura (Winona Lake: Eisenbrauns, 1994), 326-328.
10 See for example, B. K. Waltke, Creation and
Conservative Baptist Seminary, 1974). This author points out that there are three main
THE EARTH OF GENESIS 1:2: ABIOTIC OR CHAOTIC? 261
to Gunkel's affirmation: "If it is the case, however, that a fragment of a
cosmogonic myth is preserved in Genesis 1, then it is also no longer allowable
to reject the possibility that the whole chapter might be a myth that has
been transformed into narrative."11 But if, on the contrary, there is no linguistic
or biblical foundation for that assumption, the creation account would no
longer be a myth or compilation of myths similar to those of ANE literature.
The creation story would then be a true, reliable, literal, and objective account
of the origin of life on this planet.
To achieve this goal, these articles about the earth described in Gen 1:2
will analyze the Hebrew terms tohu wabohu, tehom, and ruah elohim in the
OT and their equivalents in the ANE literature.
The Hebrew Text of Gen 1:2
Weaares hayeta tohu wabohu wehosek al--pene tehom
weruah elohim merahepet ‘al--pene hammayim
Now the earth was formless and empty, darkness was
over the surface of the deep, and the Spirit of God was
hovering over the waters (NIV).
Gen 1:2 is formed by three circumstantial clauses:
(1) We ha’ares hayeta tohu wabohu: "Now the earth was formless and empty"
(2) wehosek al---pene tehom: "darkness was over the surface of the deep"
(3) weruah elohim merahepet ‘al- pene hammayim: "and the Spirit of God
was hovering over the waters."
In Semitic languages a circumstantial clause describes a particular con-
dition.12 Verse 2 presents three clauses that describe three circumstances
or conditions that existed at a particular time, which is defined by the verb
interpretations of Gen 1:1-3 within Protestant thinking. These he calls the theory of the
postcreation chaos (or theory of the restitution), in which chaos occurred after the original
creation; the theory of the initial chaos, according to which chaos occurred in connection
with creation; and the theory of the precreation chaos which he himself defends, according
to which chaos occurred before the original creation (18, 19); and other authors such as: A.
P. Ross, Creation and Blessing (Grand Rapids: Baker, 1988), 106-107, 723; V. P. Hamilton,
The Book of Genesis, Chapters 1-11, NICOT (Grand Rapids: Eerdmans, 1990), 117. As can
be seen, the explanation and interpretation of Gen 1:2 are founded on chaos, whether
before, during, or after creation.
11 Gunkel, "Influence of Babylonian Mythology," 26-27.
12 For a discussion of the function of the circumstantial phrase in Hebrew, see W.
Gesenius' Hebrew Grammar, trans. A. E. Cowley (
Press, 1910), 451, 489; Paul Jouon and T. Muraoka, A Grammar of Biblical Hebrew, Subsidia
Biblica 14 (Roma: Editrice Pontificio Istituto Biblico, 1991, 2:581.
262 SEMINARY STUDIES 35 (AUTUMN 1998)
form of the three clauses.13 In this verse the three coordinated clauses begin
with a waw followed by a noun that functions as the subject of the clause.
The theme of the verse 2 is the earth; this is the great central theme,
not only in the rest of Genesis 1, but also of the whole Bible.14 The earth
is the center and object of biblical thought.15
The exegesis of Gen 1:2 has been considered by scholars such as M.
Beauchamp,17 V. P. Hamilton,18 D. Kidner,19
P. Ross,21 N. M. Sarna,22
Westermann,26 and E. J. Young.27
15 "Clauses describing concomitant circumstances are introduced by the conjunction v of
accompaniment.... When the circumstances described are past or future, a finite form
of a verb is employed. For the past a perfect aspect is used, e.g. Uhbv Uht htyh Crxhv ‘the
earth having been a formless void' (Gen 1:2)" (R. J. Williams, Hebrew Syntax: An Outline,
in Qal perfect 3 feminine singular hayeta. As C. F. Keil and F. Delitzsch point out: "The
three statements in our verse are parallel; the substantive and participial construction of the
second and third clauses rests upon the htyhv of the first. All three describe the condition
of the earth immediately after the creation of the universe" (Commentary on the Old
Testament, trans. J. Martin ([
14 For further bibliographical references on Gen 1:1-3 from 1885/86 to 1966, see C.
1-11: A Commentary, trans. J. J. Scullion (
15 So Keil and Delitzsch, 1:48.
16 M. Alexandre, Le Commencement du Livre: Genese I- V (Paris: Beauchesne, 1988), 76-87.
17 P. Beauchamp, Creation et Separation (Paris: Desclee de Brouwer, 1969), 149-174.
19 D. Kidner, Genesis (Leicester: Inter-Varsity, 1967), 44-45.
20 S. Niditch, Chaos to Cosmos (Atlanta: Scholars, 1985), 18.
21 Ross, 106-107.
22 N. M. Sarna, Understanding Genesis (New York: Schoken, 1970), 22, 34 n. 23; idem.,
Genesis, The JPS Torah Commentary (Philadelphia: Jewish Publication Society, 1989), 6-7.
23 L. I. J. Stadelmann,
The Hebrew Conception of the World, Analecta Biblica 39 (
Biblical Institute, 1970), 12-17.
24 G. von Rad, El Libro del Genesis (Salamanca: Sigueme, 1988), 58-60.
25 G. J. Wenham, Genesis 1-15, WBC (Waco: Word Books, 1987), 15-17.
26 Westermann, 102-111.
27 E. J. Young, Studies in Genesis One (
THE EARTH OF GENESIS 1:2: ABIOTIC OR CHAOTIC? 263
The Semichiastic Structure of Gen 1:2
The Hebrew text of Gen 1:2 presents an incomplete antithetical chiastic
structure (i.e., a quasi- or semichiastic antithetical structure, because it
lacks the section A' which is antithetical to A) marked by the following
linguistic and semantic parallelism:
A Weha’ares hayeta tohu wabohu: "Now the earth was formless and empty"
B wehosek ‘al--pene tehom: "darkness was over the surface of the deep"
B' weruah elohim merahepet ‘al--pene hammayim: "and the Spirit of God
was hovering over the waters."
The grammatical, semantic, and syntactic chiastic parallelism is clearly
defined by the microstructures B \\ B'(\\ stands for antithetic parallelism)
in which the expression "over the surface" ‘al - pene is repeated. Grammatically
speaking, this expression is a preposition + plural masculine noun construct
(prep. + p.m.n. cstr.).28
The grammatical and semantic parallel ‘al --pene tehom // ‘al - pene
hammayim represents a second example of paired words, tehom // ham-
mayim that appears in Ezek 26:19 and Ps 104:6; and mayim // tehom that
appear in Ezek 31:4; Hab 3:10; Jonah 2:6; Ps 33:7; 77:17; Job 38:30. Notice
also the parallelism between mayim // tehomot and ruah in Exod 15:8.29 The
antithetic concept is clearly indicated by the opposite or contrasting pair
of words hosek "darkness" \\ ruah elohim "Spirit of God." The noun hosek
is grammatically a masculine singular (m.s.n.), and ruah elohim is a feminine
singular noun construct (f.s.n.cstr.) plus a masculine plural noun (m.p.n.).
However, they present an exact syntactic correspondence and parallelism.
Both have the same syntactic function, that of a subject.30
Another syntactic aspect is important in this antithetic chiasm: the construct
relation in ‘al - pene tehom and ‘al pene hammayim.31 This aspect of the Hebrew
syntax is of great importance to the significance and the semantic and etymological
origin of tehom, as will be seen in the second part of this article.
A particular type of parallelism used in prose is the gender-matched
parallelism. Gen 1:2 is an example of this type of parallelism, since it represent
28 Williams, 10-11.
29 J. S. Kselman, "The Recovery of Poetic Fragments from the Pentateuchal Priestly
Source," JBL 97 (1978): 163.
30 For a study of the biblical grammatical, semantic, and syntactic parallelism, see A.
31 See B. K. Waltke and M. O'Connor, An Introduction to Biblical Hebrew Syntax
(Winona Lake: Eisenbrauns, 1990), 240-241.
264 SEMINARY STUDIES 35 (AUTUMN 1998)
the gender-matched pattern: Feminine + masculine // masculine + feminine
Tohu wabohu in the Old Testament and
the Literature of the Ancient Near East
Before specifically considering this point, we must briefly analyze the
Hebrew terms ha’ares and hayeta in Gen 1:2. The most used Egyptian term
for "earth" is t3. The antithesis for this term is the formula pt-t3, "heaven"
and "earth," by which it makes reference to the whole cosmos. The usual
hieroglyphic symbol t3 represents a flood plain with grains of sand all around.
In Sumerian and Akkadian there is a distinction between "earth" (ki or ersetu)
and "country" (kur, kalam, or matu). In Akkadian ersetu means "earth," in
opposition to "heaven." "Heaven and earth" (samu u ersetu) means the universe.
In Ugaritic ‘rs means "earth, ground, inferior world." The earth is also opposed
to "heaven" and the clouds.33 Ugaritic literature also gives an extraordinary
example of a pair of words, ars // thmt, chiastically related as in Gen 1:2:
tant s'mm ‘m ars // thmt ‘mn kbkbm.34
The pair of words ‘eres // tehom also reveals an example of inclusive
structure in the six days of the creation, where ‘al -- pene tehom before the
first day (Gen 1:2) matches ‘al -pene ha’ares after the sixth (Gen 1:29).35
The Hebrew ‘eres occupies the fourth place among the most frequent
nouns in the OT. The term appears 2,504 times in Hebrew and another 22
32 See W.G.E. Watson, Classical Hebrew Poetry, JSOT Supplement Series 26 (
JSOT, 1986), 53.
33 TDOT, 1:388-392.
34 R. E. Whitaker, A Concordance of the Ugaritic
University Press, 1972), 613.
35 Kselman, 164. For this type of inclusion or construction see D. N. Freedman's
"Prolegomenon" to G. B. Gray, The Forms of Hebrew Poetry (New York: KTAV, 1972),
xxxvi-xxxvii. However, according to D.T. Tsumura the nature of the relationship between
ha’ares "earth" and tehom "abyss, ocean" in Gen 1:2 is a hyponym. According to Tsumura, in
modern linguistics, the relationship of meaning is called hyponym which sometimes is
explained as inclusion. (i.e., what is referred to in the term A includes what is referred to in
the term B). The former is preferred over the latter because a relationship of sense exists
among lexical items rather than a relationship of reference. Thus the hyponym can be used
also in a relationship between terms that have no reference. In Tsumura's own words: "Our
term ‘hyponym' therefore means that the sense [A] of the more general term ‘A’ (e.g. ‘fruit')
completely includes the ‘sense’ [B] of more specific term ‘B’ (e.g. ‘apple'), and hence what
‘A' refers to includes what ‘B’ refers to. In other words, when the referent [B] of the term
‘B’ is a part of/belongs to the referent [A] of the term ‘A’, we can say that ‘B’ is hyponymous
to ‘A,’ ("A 'Hyponymous' Word Pair: 'rs and thm (t) in Hebrew and Ugaritic" [Bib 69
(1988): 258-269, esp. 259-260]). Therefore, in Gen 1:2 there is a hyponym in which tehom
"ocean" is a part of the ha’ares "earth."
THE EARTH OF GENESIS 1:2: ABIOTIC OR CHAOTIC? 265
times in the Aramaic sections. The word tires designates: (1) cosmologically,
the earth (in opposition to heaven) and solid ground (in opposition to water);
(2) physically, the soil on which humans live; (3) geographically, certain regions
and territories; (4) politically, certain sovereign regions and countries. In
the most general sense, ‘eres designates the earth that together with the "heaven,"
samayim, comprises the totality of the universe. "Heaven and Earth" is an
expression designating the whole world (Gen 1:1; 2:1, 4; 14:19, 22; etc.).
In addition to a bipolar view of the world, there is also a tripolar view:
for instance, heaven-earth-sea (Exod 20:11; Gen 1:10, 20 and others); heaven-
earth-water beneath the earth (Exod 20:4; Deut 5:8). But what is important
to the OT is not the earth as part of the cosmos but what lives on it (Deut
33:16; Isa 34:1; Jer 8:16; etc.): its inhabitants (Isa 24:1, 5-6, 17; Jer 25:29-30;
Ps 33:14; etc.), nations (Gen 18:18; 22:18; 26:4; Deut 28:10; etc.), and kingdoms
(Deut 28:25; 2 Kgs 19:15; etc.). Thus the term "earth" may designate at the
same time--as it does in other languages--the earth and its inhabitants (Gen
6:11; etc.). In its physical use, ‘eres designates the ground on which human
beings, things, dust (Exod 8:12), and reptiles (Gen 1:26; 7:14; 8:19; etc.) are.36
The verb haya (to be) that appears in Gen 1:2 as hayeta in Qal perfect
3 f.s. is translated by the majority of the versions as "was" but may also be
translated "became," as it appears in some versions. However, the syntactic
order and the structure of the clause do not allow this translation here. The
syntactic order in Gen 1:2 (first the subject and then the verb) is used to indicate
the addition of circumstantial information and the absence of chronological
or sequential occurrence. For that reason the translators of the LXX translated
hayeta as "was" and not as "became."37 Besides, the Hebrew letter waw that
appears at the beginning of Gen 1:2 is a "circumstantial waw" because it is
joined to the subject "the earth" and not to the verb. Therefore it is better
translated as "now." The translators of the LXX, who were very careful in
the translation of the Pentateuch, translated it in that way.
The initial state of the earth in Gen 1:2 is described as tohu wabohu.
This expression is translated into English as "formless and empty" (NIV).
In the Greek versions it is translated as aoratoj kai akatskeuastoj,
"invisible and unformed" (LXX); kenwma kai ouqen, "empty and nothing"
36 E. Jenni
and C. Westermann, Diccionario Teologico Manual
trans. J. A. Mugica;
Clines, ed., The Dictionary of Classical Hebrew (Sheffield: Sheffield Academic Press, 1993),
esp. 392, which gives specific references to
37 F. Delitzsch comments that the perfect preceded by the subject is the most usual way
of describing the circumstances in which the subsequent account takes place (A New
Commentary on Genesis [
266 SEMINARY STUDIES 35 (AUTUMN 1998)
kai adiakriton, "unproductive and indistinguishable" (Symmachus).38
Etymology and Usage of Tohu in the OT
Tohu is a masculine singular noun (m.s.n.) that means "formlessness,
confusion, unreality, emptiness,... formlessness of primaeval earth in Gen
1:2";39 "wasteland, solitude or emptiness";40 "emptiness, waste, desert, chaos,
confusion";41 "Wuste, Ode, Leere,... Gen 1:2 es ‘bedeutet die ode Wuste,
and ist als Grundbegriff zur Schopfung gebraucht";42 "caos, lo que no tiene
forma ni medida, informe, inmensidad. Lo desmesurado; formulacion clara
y directa de la negacion: nada, la nada, vacio, el vacio, nulidad,... caos informe
en Gen 1:2."43
The term tohu appears 20 times in the OT, 11 of them in Isaiah.44 The
different uses of the term can be classified, according to Westermann, in three
groups that go from the concrete meaning of "desert" to the abstract "emptiness":
(1) "Desert," the terrible and barren desert that leads to de-
struction: Deut 32:10; Job 6:18; 12:24 = Ps 107:40; (2) "Desert or devastation
that threatens": Isa 24: 10; 34:11; 40:23; Jer 4:23; "the state that is opposed
to the creation and precedes it": Gen 1:2; Isa 45:18; Job 26:7. 3; (3) "Nothing":
1 Sam 12:21 (2x); Isa 29:21; 40:17; 41:29; 44:29; 45:19; 49:4; 59:4.45
The first and third groups are simple enough to define and describe. In
the first, tohu is "earth, desert ground" (Deut 32:10), the "untilled land" where
caravans die Gob 6:18), a "barren ground without roads" where people wander
(Job 12:24; Ps 107:40). Therefore, the term refers to the desert as a "barren ground
38 J. W. Wevers, Septuaginta: Genesis (Gottingen: Vandenhoeck & Ruprecht, 1974), 75;
cf. A. Rahlfs, Septuaginta (Stuttgart: Deutsche Bibelgessellschaft, 1979).
39 F. Brown, S. R. Driver, and C. A. Briggs, A Hebrew and English Lexicon of the Old
Testament (BDB) (Oxford: Clarendon Press, 1951), 1062.
40 W. L. Holladay, ed., A Concise Hebrew and Aramaic Lexicon of the Old Testament
(Grand Rapids: Eerdmans, 1989), 386.
41 E. Klein, A Comprehensive Etymological Dictionary of the Hebrew Language for
Readers of English (Jerusalem: University of Haifa, 1987), 692.
42 L. Koehler, W. Baurngartner, and J. J. Stamm, eds., Hebraisches and Aramaisches
Lexikon zum Alten Testament (Leiden: Brill, 1967-1994), 1557.
43 L. A. Schokel, Diccionario Biblico Hebreo-Espanol (Madrid: Trotta, 1994), 792.
Translation: "Chaos; what has no shape or measure: shapeless, immensity, the excessive; a
clear and direct formulation of the negation: nothing, the nothingness, empty, the
emptyness, nullity, . . . shapeless chaos in Gen 1:2."
44 See A. Even-Shoshan, A New Concordance of
the Old Testament (
Sefer, 1990), 1219. The 20 texts are: Gen 1:2; Deut 32:10; 1 Sam 12:21 (2x); Job 6:18; 12:24;
26:7; Ps 107:40; Isa 24:10; 29:21; 34:11; 40:17, 23; 41:29; 44:9; 45:18-19; 49:4; 59:4; Jer 4:23.
45 Westermann, 102-10:3.
THE EARTH OF GENESIS 1:2: ABIOTIC OR CHAOTIC? 267
or land." In the third group tohu refers to a situation in which something that
ought to be there is lacking. It is used in an abstract sense in which it
appears in parallel with other nouns such as ‘epes, "nothing"(Isa 41:29), riq,
"empty" (Isa 49:4), and "empty arguments" (Isa 59:4, NIV).46 In these passages
tohu is better understood as "lack or emptiness" rather than "nothing."
Of special interest to this study are the uses of tohu in Westermann's
second group, where the word describes the situation or condition of places
such as the planet earth, land (region), or city. In Isa 24:10 we have qiryat-
tohu, referring to the "desolate or deserted" state of a city, almost equivalent
to the term samma in v. 12, which refers to the desolation of a city: "The
ruined city lies desolate; the entrance to every house is barred" (NIV).
In job 26:7, Westermann thinks 'al -- tohu is directly opposed to the
creation, though he does not translate it as chaos.47 But the expression
al -- tohu is parallel to the expression ‘al - beli -- ma "a place where there
is nothing." Therefore, in this context a possible translation of tohu would
be "a desert-like or empty place."48
Westermann points out that in Isa 45:18 lo- tohu is in direct opposition
to the creation.49 However, here tohu is in parallelism with lasebet, Qal infinitive
construct (Qal inf. cstr.), "to be inhabited" (NIV), from the verb yasab "to dwell.50
The text does not indicate anything about a chaotic state in the earth: "he did
not create it to be empty, but formed it to be inhabited" (NIV). Instead, tohu
in this text also means "a desert, an uninhabited place." Thus this verse may
be better translated as "[earth] not to be a desert or uninhabited place he created
it, to be inhabited he formed it."51 In other words, this verse explains that God
46 E. J. Young translates tohu in Isa 44:9 as "unreality" and explains that the word
"suggests an absence of all life and power" (The Book of Isaiah, NICOT [
Eerdmans, 1972], 3:172).
47 Westermann, 103.
48 Job 26:7a: noteh sapon a1-tohu //Job 26:7b: toleh ‘eres al-beli-ma.
49 Westermann, 103.
50 BDB, 442;
51 Isa 45:18f: to -tohu be ra’ah // Isa 45:18g: lasebet yesarah. We can verify that it is a
structure in parallel panels which is marked by the following microstructure:
A lo--tohu [Earth] not to be a desert or uninhabited place
B bera’ah he created it
A' lasebet to be inhabited
B' yesarah he formed it
We observe a clear antithetical parallelism between A \\ A', lo'- tohu "[Earth] not to
be a desert or uninhabited place" //lasebet "[Earth] to be inhabited." As Watson points
out when referring to the parallel types of words: "antonymic word pairs are made up of
words opposite in meaning and are normally used in antithetic parallelism" (131). At the
same time, there is a synonymous parallelism between B // B', bera’ah "he created it" //
268 SEMINARY STUDIES 35 (AUTUMN 1998)
did not create the earth to be uninhabited or desert but to be inhabited. Gen
1:2 can be understood in the same sense, that God created the earth to be
inhabited, but "it was still desert or uninhabited" during the initial stage of the
creation though it was in no sense in a chaotic state.
In Isa 45:19 the term tohu has been interpreted in two ways: concrete
(locative) and abstract. The syntax is always understood in the same way:
tohu as an adverb that modifies the verbal clause bagqesuni, as part of the
direct speech.52 The Tg. Isa. analyzes tohu in the same way: "!Buscad en vano
(lryqnw) mi temor!"53 However, its meaning and grammatical function must
be analyzed by considering the parallel structure of the complete verse.54
Therefore, from the literary structure in parallel panels, B' tohu is parallel
with B bimeqom ‘eres hosek "in a land of darkness" (NIV). In Tsumura's words:
"Tohu without a preposition directly corresponds either to ‘eres hosek or
to hosek.... In this case, the term tohu, corresponding directly to hosek ‘darkness,'
probably means ‘desolation.’”57 To conclude, we must point out that in the
Targums, the Talmudic and the Midrashic literature tohu is interpreted as
"waste, desolation; vanity, idleness."57
*Thw in Ugaritic Literature
Once we have analyzed the etymology and the usage of tohu in the OT,
we consider its etymology and usage in the Ugaritic literature. Until recently,
yesarah "he formed it." In Watson's words: "synonymous word pairs comprise a large class
with a broad spectrum.... Its components are synonyms or near-synonyms and therefore
almost interchangeable in character" (ibid.).
52 D. T. Tsumura, tohu in Isaiah XLV 19," VT 38 (1988): 361-364, esp. 361.
53 J. Ribera Florit, El Targum de Isaias (Valencia: Institucion San Jeronimo, 1988), 192.
54 Isa 45:19a: lo'basseter dibbarti // Isa 45:19c: lo' amareti lezera 'ya aqob. Isa 45:19b:
bimeqom 'eres hosek // Isa 45:19d: tohu baqqesuni. We can observe that it is a structure in
parallel panels that is marked by the following microstructures:
A lo'basseter dibbarti I have not spoken in secret
B bimeqom 'eres hosek from somewhere in a land of darkness
A' lo' amareti lezera 'ya’aqob I have not said to Jacob's descendants
B' tohu baqqesuni’ Seek me in vain' (NIV)
The syntactical and morphological parallelism is evident between A \\ A' in the nega-
tive sentence, and the tense and the person of the verb, lo' dibbarti negative+Pi'el perfect
1 common singular // lo' amareti negative+Qal perfect 1 common singular. Meanwhile,
there is a semantical parallelism between B // B', ‘eres hosek // tohu, with the same
nouns as in Gen 1:2 (for a linguistic study of the different types of biblical parallelisms, see
57 Tsumura, 362-363.
57 M. Jastrow, A Dictionary of the Targumim, the Talmud Babli and Yerushalmi, and the
Midrashic Literature (New York: Title, 1943), 1651.
THE EARTH OF GENESIS 1:2: ABIOTIC OR CHAOTIC? 269
recently, the etymology of tohu was explained in the light of the Arabic
tih, waterless desert, trackless wilderness.58 However, as Tsumura points
out, the Arabic term, with a second weak consonant h, does not explain
the final long u of the Hebrew tohu.59
The Ugaritic term equivalent to the Hebrew tohu is the thw nominal
form that appears only once in the Ugaritic literature,60 in the cycle of Baal
and Mot as follows:
pnp.s.nps.lbim  thw
"'But my appetite is an appetite of lions (in) the waste,
"’just as the longing of dolphin(s) is in the sea.61
Del Olmo Lete presents the following translation of the same text: "Tengo,
si, el apetito
In the context of the two lines of Ugaritic text, lbim.thw "of a lion in the
steppe [desert]" corresponds to anhr.bym, "of a shark in the sea," since nps
and brlt are a well known idiomatic pair.63 Del Olmo Lete maintains that
the Ugaritic term thw is a cognate of the Heb tohu.64
Considering the evidence presented, we can affirm that the Ugaritic
term thw is a cognate of the Heb tohu and both have a common meaning:
"desert." They are probably nouns with a common Semitic root, *thw. In
relation to this, Huehnergard points out that the text or alphabetical form
thw is probably /tuhwu/ "wasteland."65
58 Klein, 692.
59 D. T. Tsumura, The Earth and the Waters in Genesis 1 and 2: A Linguistic Investigation,
JSOT Supplement Series 83 (Sheffield: JSOT Press, 1989), 17.
60 See C. H. Gordon, Ugaritic Textbook, Analecta Orientalia 38 (Roma: Pontificium
Institutum Biblicum, 1965), 178. It is the transliteration of the text 67.1.15:
thw.ham; brlt.anhr; also M. Dietrich, O. Loretz and J. Sanmartin, Die keilalphabetischen Texte
1.5 115: thw.hm.brlt.anhr.
text 5 115, in J.C.L. Gibson, Canaanite Myths and Legends (
& T. Clark, 2d ed., 1978), 68.
Translation: "I have, yes I do, the appetite of a lion on the steppe, the longing of a shark
(who lives) in the sea."
63 On p. 635 Del Olmo Lete says: "thw: n.m., ‘estepa, desierto' (cf. heb. tohu; cf. Gibson,
64 Dietrich, Loretz and Sanmartin, 1.18 IV 25, 36-37, 55, 58. Del Olmo Lete notes that
thw "steppe, desert" is antonymous to ym, "sea."
65 J. Huehnergard, Ugaritic Vocabulary in Syllabic Transcription, Harvard Semitic Series
32 (Atlanta: Scholars, 1987), 84, 287.
270 SEMINARY STUDIES 35 (AUTUMN 1998)
Etymology of *bhw
Bohu is similar to tohu because it is a m.s.n. which means "’emptiness’
of primeval earth";66 "emptiness (// formlessness, + earth) ... formlessness
and emptiness";67 "Heb. bohu ‘vacuite, vide'; Arab. ‘bahw- ‘espace degage,
trouee, etc.', bahiya ‘etre vide, desert', bahi ‘vide, desert'";68 "void, waste";69
"emptiness, chaos";70 "Leere, Ode";71 "vacio, caos, caos informe."72
The term bohu appears only 3 times in the OT, always with tohu: Gen
1:2; Isa 34:11; Jer 4:23. Its meaning will be considered in the section on the
usage of phrase tohu wabohu. In the Targums, as well as the Talmudic and
the Midrashic literature, Jastrow finds that bohu is interpreted as "chaotic
condition; always with vht."73
*Bhw in the Ancient Near Eastern Literature
The etymology of bohu has been explained through the Arabic bahiya,
"to be hollow, empty."74 This Arabic term is used to describe the "empty"
state of a store or house that has little or nothing in it.75 Therefore, its meaning
is more concrete than abstract, "nothing, empty."
Albright suggested that the Akkadian term bubutu, "emptiness, hunger,"
comes from *buhbuhtu and is possibly a cognate of the Heb bohu.76 However,
the Chicago Assyrian Dictionary does not list "emptiness" as a meaning of
bubutuA. It translates the term as: "famine, starvation, want, hunger, sustenance"77
66 BDB, 96.
67 D.J.A. Clines, ed., The Dictionary of Classical Hebrew (Sheffield:
1995), 2:97; in the
68 D. Cohen, Dictionnaire des
70 Klein, 65.
71 Koehler and Baumgartner,107.
72 Schockel, 102. Translation: "empty, chaos, shapeless chaos."
73 Jastrow, 142.
74 According to Klein, bohu comes from the root of hhb , Arabic bahw, "hollow,
76 W.F. Albright, "Contributions to Biblical Archaeology and Philogy," JBL 43 (1924):
77 CAD, B:301-302.
THE EARTH OF GENESIS 1:2: ABIOTIC OR CHAOTIC? 271
and Von Soden suggests "hunger" as a possible meaning of bubutu. Neither
of these Akkadian terms is a cognate of Heb bohu.78
It has been also suggested that the term bohu is related to Phoenician
divine name baau, the goddess of "night."79 Tsumura. indicates that it is
phonologically possible to propose an original "Canaanite" form /bahwu/ for
both Heb bohu and Phoenician /bah(a)wu/, which was apparently re-
presented in Greek script as ba-a-u.80 But he adds that there is no evidence
that the Hebrew term had any connection with the Phoenician divine name,
except for its possible origin. in a common root, *bhw.81 Likewise, Cassuto,
after indicating that the word is found in the earlier Canaanite poems, adds:
"but there is no connection apparently with the Mesopotamian goddess Ba-u.”82
Recently Gorg suggested that tohu and bohu must be explained by the
Egyptian terms th3 and bh3.83 This proposal is highly speculative since no
hendiadys of these terms in is known."
In conclusion, taking into account available evidence, although there
is no final etymological explanation, the Heb bohu seems to be a Semitic
term based on the root *bhw and is probably a cognate of Arabic bahiya,
"to be empty."
*Thw and *bhw in the OT
Albright's affirmation that the clause tohu wabohu means "chaos" and
W. von Soden, Akkadisches Handworterbach (
79 Albright, 366, n. 7.
80 Tsumura, The Earth and the Waters, 22. This author proposes the following evolution
of the original form for the Heb bohu: */bdhwu/ > /buhwu/ > /buhuu/ > /buhu/ >
/bohu/. But he immediately adds the possible origin of bohu in an original form */bihwu/
from a Ugaritic example written syllabically (ibid., n. 26).
82 U. Cassuto,
A Commentary on the Book of Genesis: From
Adam to Noah (
Magnes, 1961; reprinted 1989), 22.
83 M. Gorg, "Tohu wabohu: ein Deutungsvorschlag," ZAW 92 (1980): 431-434; see also
"Zur Struktur von Gen 1.2" Biblische Notizen 62 (1992): 11-15.
84 Hendiadys is defined as: "The use of two substantives, joined by a conjunction, to
express a single but complex idea. The two words may be collocated, be joined by a copula
or be in apposition. Hendiadys is used very often in Hebrew.... The important aspect of
hendiadys is that its components are no longer considered separately but as a single unit in
combination" (Watson, 324-325). Such is the case of tohu wabohu in Gen 1:2. E. A. Speiser
explains: "The Heb. pair tohu wa--bohu is an excellent example of hendiadys, that is, two
terms connected by ‘and’ and forming a unit in which one member is used to qualify the
other" (Genesis, AB [New York: Doubleday, 1962], 5, n. 2a).
272 SEMINARY STUDIES 35 (AUTUMN 1998)
that tohu refers to a watery chaos is shared by many modern scholars, includ-
ing Cassuto.85 According to most modern scholars, the expression tohu
wabohu in Gen 1:2 is understood as the primeval "chaos, confusion,
disorganization" and is, therefore, in direct opposition to creation.86 On the other
hand, Burner--Klein points out that tohu wabohu describes the state of the earth
immediately after God had created the world. From the LXX and the ancient
versions, as well as the
refers to a created, yet shapeless earth.87
To complete the study we must consider Isa 34:11 and Jer 4:23, where
tohu and bohu appear. In Isa 34:11 tohu and bohu appear in parallel expressions 88:
qaw - tohu "the measuring line of thw" (NIV) II 'abne --- bohu "the plumb
line of bhw" (NM." This passage clearly refers to an uninhabited place. Basic
85 Cassuto, 23. See also B. K. Waltke, "The Creation Account in Genesis 1:1-3, Part 3,
The Initial Chaos Theory and the Precreation Chaos Theory," Bibliotheca Sacra 132 (1975):
225-228. Waltke interprets tohu wabohu as the chaotic state before creation. For a recent
answer to Waltke's arguments, see M. F. Rooker, "Genesis 1:1-3: Creation o Re-Creation?
Part 1," Bibliotheca Sacra 149 (1992): 316-323; and "Genesis 1:1-3: Creation or Re-Creation?
Part 2," Bibliotheca Sacra 149 (1992): 411-427. Wenham speaks of "total chaos" (15-16).
86 See Alexandre, 77; Beauchamp, 162-163; Hamilton, 108; Kidner, 44; Niditch, 18; Ross,
106; Sarna, 6; Stadelmann, 12; Wenham, 15; Westermann, 103; Young, 33-34.
87 D. Burner-Klein, "Tohu u and bohu: Zur Auslegungsgeschichte von Gen 1,2a," Henoch
(1993): 3-41. Burner-Klein analyzes the LXX, Origen,
Theodotion, which use a variety of images to translate the clause: "the earth was invisible,"
"uncultivated," "a desert," "an empty
space," "nothing." His study of
renders the following interpretations: "a desolate country," "vanity" and "empty." Rabbinic
literature interprets the clause as a negative principle, primeval matter that God already
found at creation, i.e., a substratum of the creatio ex nihilo, created matter but shapeless yet.
In a Karaite commentary on Genesis he found the idea of an empty earth, without buildings.
His study included Christian Bible commentaries that develop similar concepts in
opposition to Aristotle's doctrine of the eternity of the world.
88 See W. G. E. Watson, Traditional Techniques in Classical Hebrew Verse, JSOT
Supplement Series 170 (Sheffield: Sheffield Academic Press, 1994), 148, 153, 161, 165.
89 Isa 34:11a: wiresuha qaat weqippod //Isa 34:11b: weyansop we ‘oreb yiskenu-bah; Isa
34:11c: wenata aleyha qaw-tohu // Isa 34:11d: we’abne--bohu. The structure in parallel
panels is marked by the following microstructures:
A wiresuha qaat weqippod The desert owl and screech owl will possess it
A' weyansop we ‘oreb yiskenu --- bah the great owl and the raven will nest there
B wenatd a1eyha qaw-tohu ... the measuring line of chaos
B' we ‘abne - bohu and the plumb line of desolation (NIV)
There is a semantic and syntactic synonymous parallelism between A // A', wiresuha
qaat weqippod "The desert owl and screech owl will possess it" // weyansop we ‘oreb
yiskenu - bah "the great owl and the raven will nest there." In both cases, at a semantic level,
the lines refer to birds. On the syntactic level, there is also a subject+verb (+suffix) //
subject+verb (+suffix) parallelism, but with the components of the clauses inverted.
Likewise, there is semantic and syntactic synonymous parallelism between B // B', wenata
THE EARTH OF GENESIS 1:2: ABIOTIC OR CHAOTIC? 273
to the understanding of Isa 34:11 as a land uninhabited by human beings
is the grammatical and semantic parallelism of the verbs wry, "take possession
of,"90 Qal perfect 3 common plural wire-suha "will possess it"; and Nkw "live
in, settle,"91 Qal imperfect 3 masculine plural yiskenu, "will dwell," in Isa
34:11a and Isa 34:11b. Besides, an exegesis of the immediately preceding verse,
Isa 34:10cd, clearly shows the meaning of Isa 34:11: an un-
inhabited land." In Young's words: "the land will become a desolation and
waste so that it can no more receive inhabitants."93 Therefore, in Isa 34:11
we do not find linguistic or exegetic evidence for any chaotic situation.
Jer 4:23 contains the following parallel structure:94
A raiti et –ha’ares I looked at the earth,
B wehinneh---tohu wabohu and it was formless and empty;
A' we ‘el -hassamayim and at the heavens,
B' we’ en ‘oram and their light was gone (NIV).
It has often been stated that Jer 4:23-26 describes a return to the primitive
chaos.95 But this point of view is highly influenced by the traditional exegesis
of the expression tohu wabohu as "chaos" in Gen 1:2 and not on the analysis
of the context of Jer 4:23. In vv. 23-26, each of the verses begins with raiti,
‘aleyha qaw- tohu: "the measuring line of chaos"// we‘abne- bohu "and the plumb line of
desolation." In both lines we find the same nouns that appear in Gen 1:2, tobu and bohu.
Finally, both nouns are in a construct relation (on grammatical, semantic, and syntactic
90 BDB, 439;
92 Isa 34:10cd: middor lador teherab lenesah nesahim eyn ‘ ober bah "From generation
to generation it will lie desolate; no one will ever pass through it again" (NIV). Thus Isa
34:10d interprets Isa 34:10c and 34:11 in a definite semantic parallelism to: middor laddor
teherab, "From generation to generation it will lie desolate."
93 Young indicates that the prophet Isaiah uses the language of Gen 1:2 (Book of 1saiah,
94 There is an antithetical semantic parallelism between A // A', raiti ‘et- ha’ares "I
looked at the earth" // weel-hassamayim "and at the heavens." These are the basic
components of the Hebrew conception of the bipartite structure of the universe, earth and
heavens. There is also a grammatical and semantic parallelism between B // B', wehinneh-
tohu wabohu "and it was formless and empty" // we ‘en ‘oram "and their light was gone."
This parallelism can be observed at a grammatical level between the nouns tobu and bohu
in 4:23b, and or in 4:23d, both are m.s.n.; at a semantic level, both concepts imply the lack
of something, both on the earth ("formless and empty") and the heavens ("light").
the world again become the chaos before creation began" (W. L. Holladay, Jeremiah
Commentary on Jeremiah [
274 SEMINARY STUDIES 35 (AUTUMN 1998)
"I saw," and the word wehinneh, "and behold," is repeated in each verse.
The exegesis of verse 23 is completed and confirmed by the interpretation
of verses 25-26, which are translated: "I looked, and there were no people;
every bird in the sky had flown away. I looked, and the fruitful land was
a desert; all its towns lay in ruins before the Lord" (NIV).
There is a precise positive-negative syntactic parallelism96 between the
vv. 23 and 25-26, "I looked at the earth" (4:23 a) // "I looked and there were
no people (4:25a); "I looked, and the fruitful land was a desert" (4:26a) and
"and at the heavens" (4:23c) // "every bird in the sky had flown away" (4:25b).
Therefore, v. 23a, "I looked at the earth," is interpreted in vv. 25a-26a, "I
looked, and there were no people"; "I looked, and the fruitful land was a
desert." Likewise, v. 23c, "and at the heavens" is also interpreted by v. 25b,
"every bird in the sky had flown away." Therefore, the
4:23 was uninhabited, with no human beings on it; "there were no people."
It was also arid and unproductive: "the fruitful land was a desert." On the
other hand, the heavens of Jer 4:23 are empty, without light ("their light
was gone") and without birds ("every bird in the sky had flown away").97
The interpretation of tohu wabohu in the Targums also helps solve
the difficulties inherent in the interpretation of Gen 1:2. On Gen 1:2 the
Tg. Neof reads as follows, according to two translators: Diez Macho and
Y la tierra estaba tehi’ y behi' deshabitada de hombres y bestias y vacia
de todo cultivo de plantas y arboles.98
Now the earth was tehi' and behi' [meaning it was] desolate (sdy) with
respect to people and animals and empty (rygn’)in respect to all manner of
agricultural work and trees."
On his translation of Tg. Neof.
This text first reproduces the Aramaic equivalent of the Hebrew pair tohu
wabohu and then interprets them. The first term, tohu, is interpreted
to mean an absence of faunal life; the second term, bohu, the absence of
97 Jer 4:23a: raiti 'et---ha’ares //Jer 4:25a-26a: raiti wehinneh 'en ha’adam ... raiti
wehinneh hakkarmel hammidbar; Jer 4:23c: we 'el-hassamayim // Jer 4:25b: of kol- op
hassamayim nadadu. The following microstructures are evident.
A raiti et -haares I looked at the earth
B we ‘el--hassamayim and at the heavens
A'ra itI wehinneh en ha’adam ... raiti wehinneh hakkarmel hammidbar I looked, and
there were no people ... I looked, and the fruitful land was a desert
B'wekol- op hassamayim nadadu every bird in the sky had flown away (NIV).
98 A. Diez Macho, Neophyti: Targum Palestiniense (Madrid: CSIC, 1968), 1:2.
THE EARTH OF GENESIS 1:2: ABIOTIC OR CHAOTIC? 275
floral life. No longer do tohu wabohu connote a primeval substrate "chaos."
Rather they simply describe the earth in an unfinished state. The earth
was not created as a state of chaos; rather it is simply devoid of the living
matter which will be created in days 3, 5 and 6. Exegesis has brought order
to the unordered. All other targums follow this general exegetical
In brief, the expression tohu wabohu refers to a "desert-uninhabited"
(Isa 34:11; Jer 4:23) and "arid or unproductive" (Jer 4:23) state.101 Neither
text gives any linguistic or exegetical evidence to support the existence of
a situation of mythic chaos in the earth.
*Thw and *bhw in the Ugaritic Literature
Several studies have pointed to the similarity between the Heb tohu
wabohu and the Ugaritic tu-a-bi[u(?)].102 Tsumura proposes a possible explanation
of the morphological correspondence between the Hebrew expression
tohu wabohu and the Ugaritic tu-a-bi[u(?)].103 It is, therefore, possible that
the Ugaritic tu-a-bi [u(?)] and the Hebrew tohu wabohu are two versions of
the same idiomatic expression in the Northwestern Semitic.104
However, scholars such as J. Huehnergard have proposed a different
morphological relation, considering the Hebrew expression tohu wabohu
as an equivalent of the Ugaritic tu-a pi [ku(?)],105 since the verb form *hpk,
"to upset or overthrow," is identified in the Ugaritic alphabetical texts.106
In this way, both interpretations to-a-bi (u(?)land to-a pi [ku(?)] are possible
from a phonological and morphological point of view.
To conclude, considering OT and ANE literature, the expression tohu
101 See also Tsumura, The Earth and the Waters, 41.
"'See, for example, J. C. de Moor, "El, the Creator," in The Bible World: Essays in
Honor of Cyrus H. Gordon, ed. G. Rendsburg et al. (New York: KTAV, 1980), 183, and n.
58; Tsumura, Earth and the Waters, 24.
102 According to Tsumura, the first half of the syllabic orthography, tu-a, probably
represents /tuha/ since in the Ugaritic syllabic ortography the grapheme < a > can be used
as a syllable /ha/. In the second half of the syllabic orthography, bi [u], if the second sign is
correctly restored, it can represent /bihu/ since the grapheme < u > of the syllabic
orthography is used in syllables /hu/ (ibid.)
105 UVST, 84, 121, 315, 322.
106 Ibid; Gordon, 392a n° 788; Dietrich et al., 1.103:52. Sumerian: BAL = Akkadian: na-
bal-ku-tu, = Hurrian: tap-su-hu-um-me = Ugaritic- tu-a pi [ku(?)].
276 SEMINARY STUDIES 35 (AUTUMN 1998)
wabohu in Gen 1:2 must be interpreted as the description of a "desert, uninhabited,
arid and unproductive" place. 107 The earth of Gen 1:2, which "was" hayeta
tohu wabohu, refers to the earth in an "empty" state with no vegetation,
animals, or people. Hence the title of this series of articles: "The Earth of
Genesis 1:2: Abiotic or Chaotic." The concept that appears in Gen 1:2 is
an abiotic concept of the earth; i.e., Gen 1:2 describes an earth in which
there is no life; it presents the absence of life-vegetable, animal, and human.
That life appears in the following verses of Genesis 1 by the fiat of God.
The Hebrew idiomatic expression tohu wabohu refers to an earth that is
"uninhabited and unproductive," owing to the absence of life, of fauna, and
of flora at this stage of the creation. At a later stage the earth will be "inhabited
and productive." In no case does the phrase describe a chaotic state of the
earth as the result of mythical combats between the gods of the myths and
The main reason why the author describes the earth as tohu wabohu
is to inform the audience that the earth "is not yet" the earth such as they
know it. Westermann puts it this way: "Creation and the world are to
be understood always from the viewpoint of or in the context of human
existence."108 In other words, it is necessary to use literary language and
figures common to the audience to communicate to human beings the theme
of creation. Therefore, the author uses in this verse language originating
in his life experience (desert, empty, uninhabited, unproductive places) to
explain the initial situation or condition of the earth.
The words of Westermann summarize well the findings on Gen 1:2:
There is no sign of either personification or mythological allusion in
the biblical use of Uht.... The course of the debate about the mythical
explanation of vhbv vht indicates clearly that the arguments for a mythical
background are becoming weaker and weaker. The discussion can now be
107 See also N. H. Tur-Sinai, The Book
of Job: A New Commentary (
Sepher, 1967), 381: "in Gen 1:2 ... [tohu] describes the barrenness of the earth before
anything grew on it."
108 Westermann, 104.
109 Westermann, 103.
This material is cited with gracious permission from:
Please report any errors to Ted Hildebrandt