The Garden and the Command

The differences between Genesis 1:1 to 2:3 and Genesis 2:4 to 25 are not an invention of hostile critics, and a teacher who denies that they exist loses the class the first time somebody reads the chapters side by side. The first account calls God Elohim, God, thirty-five times and never anything else. From 2:4 the name changes to YHWH Elohim, the LORD God, and stays that way through chapter 3. Chapter 1 is cosmic, patterned and liturgical, with refrains and counted sevens. Chapter 2 is a narrative set on a patch of ground, with dust, a potter's verb, a rib or a side, and a man who talks. In chapter 1 vegetation comes on day three and humanity on day six, male and female together, after the animals. In chapter 2 there is no vegetation at the start because there is no rain and no one to work the ground, the man is formed first, the garden is planted for him, the animals are brought to him, and the woman comes last. The verbs differ too: bara' and asah in chapter 1, yatsar, to shape as a potter shapes clay, in 2:7, and banah, to build, of the woman in 2:22.

The documentary explanation of those differences has a long and respectable history, and you should be able to state it without caricature. In 1753 Jean Astruc, a French physician, published his observation that the alternation of divine names in Genesis looked like the trace of two sources. The analysis was developed through the nineteenth century and given its classical form by Julius Wellhausen in the Prolegomena to the History of Israel, where Genesis 1:1 to 2:3 is assigned to a Priestly writer, marked by order, genealogy, exact numbers, the name Elohim and a strong interest in the sabbath, and Genesis 2:4b to 3:24 to the Yahwist, marked by narrative, anthropomorphism and the name YHWH. On this reading the two accounts are two compositions, preserved side by side by a later editor who did not feel entitled to choose between them. And the argument was never built on Genesis 1 and 2 alone, which is why answering it from these two chapters only does not touch it. Genesis has further pairs that behave the same way: two interwoven strands in the flood narrative, which the next lesson in this module takes up; two accounts of Hagar's flight, in chapter 16 and chapter 21; two naming stories for Beersheba, in 21:31 and 26:33. A reader who explains one such pair as a literary device still has the pattern to account for. Two things make the case stronger than the divine-name observation alone, and a fair summary has to include them. First, the material assigned to the Priestly writer coheres across the whole Pentateuch and not just here: the same interest in genealogies, exact measurements, dated chronology, covenant formulae and the sabbath turns up in Genesis 5, Genesis 17, the tabernacle chapters and Leviticus, which is a pattern rather than a coincidence of vocabulary. Second, there is an internal datum that source analysis explains neatly and that any unified reading has to work at: Exodus 6:3 says that God appeared to the patriarchs as God Almighty but was not known to them by the name YHWH, although Genesis has the patriarchs using that name repeatedly, beginning in Genesis 4:26.

What has to be said next is that this hypothesis is no longer in the shape it had in 1900, and the revision came from inside critical scholarship rather than from its opponents. From the 1970s onward Rolf Rendtorff and John Van Seters attacked the coherence of the sources, R. N. Whybray's The Making of the Pentateuch argued that the criteria were being applied inconsistently, and much continental scholarship abandoned the Yahwist as an identifiable document altogether. A group sometimes called the neo-documentarians, including Joel Baden and Jeffrey Stackert, defends a revised four-source model with considerable rigour. Two further difficulties belong on the record and they are not answered by the revision. The first is evidential: no manuscript of J, E, D or P has ever been found, and no ancient writer mentions one, so the documents are reconstructions from internal features of the text we have rather than objects anyone has seen. That does not make the reconstruction illegitimate, since historians infer unseen sources all the time, but it does mean the hypothesis cannot be tested the way a textual variant can. The second is that the divine-name criterion, which started the whole enterprise, does not hold cleanly: there are passages assigned to the Yahwist that use Elohim and passages assigned to the Priestly writer that use YHWH, and once the criterion admits exceptions the question of what is controlling the division becomes sharper. So the state of play is that the textual observations which generated the hypothesis are solid data that nobody disputes, and the explanation of those observations is contested among the very scholars most committed to source analysis. Report both halves of that.

The literary explanation starts from Genesis 2:4 itself. 'This is the account of the heavens and the earth when they were created' is, in the BSB's wording, the first of roughly ten occurrences of the same formula in Genesis (the Hebrew noun is toledot, which older versions such as the KJV render 'generations'), at 5:1, 6:9, 10:1, 11:10, 11:27, 25:12, 25:19, 36 and 37:2, and every one of them narrows the field: from the cosmos to Adam's line, from Adam's line to Noah, from Noah to the nations, from the nations to Shem, from Shem to Terah, and so to Abram. On this reading 2:4 is the book's own structural device rather than a seam, and chapter 2 is not a second creation account at all. Look at what it is actually about: it never mentions the sea, the stars, the sky or the animals of the water. It is an account of a garden, a man, a command and a marriage, told at ground level because those are the things chapter 3 will need. Umberto Cassuto argued this case in detail in lectures published as The Documentary Hypothesis, and in different forms it is carried by Kenneth Kitchen, John Sailhamer and Jacques Doukhan. Two further arguments belong to it. The divine names are not random across the two units but appropriate to them: Elohim, the generic word for God, suits an account of the whole cosmos made by the God of everything, and YHWH Elohim, the covenant name joined to the generic one, suits an account of a particular God dealing with a particular man in a particular garden, which is how the same pairing works later in Exodus. And the compound form YHWH Elohim is itself awkward for the source division, since it is the very combination you would expect a writer to use who was deliberately identifying the God of chapter 2 with the God of chapter 1. Gordon Wenham's commentary is a useful model of a third option: he grants source analysis real work to do and still reads the final text as a deliberate literary whole.

The literary reading has its own internal difficulties and they should be stated with the same frankness. The clearest is that 2:4 does not behave like the other nine toledot headings. Everywhere else the formula introduces the descendants of a named person, and the heavens and the earth do not beget, so the first occurrence is already an analogical use of the formula it is being asked to anchor. A related problem is direction: several scholars have argued that the formula functions as a colophon closing the section before it rather than a heading opening the section after it, which, if right, makes 2:4 the end of the creation account rather than the start of a second one, and the toledot argument then supports something different from what it is usually enlisted for. And the reading is a minority position in the academic field, which is not an argument against it but is a fact a student should hear from you rather than discover later. Two further difficulties are methodological. The literary reading has no explanation of Exodus 6:3 to offer, and a reading that leaves the sharpest internal datum in the file untouched is not yet a complete account. And the device it relies on, a deliberate narrowing of focus, is capable of absorbing any discrepancy at all: once any difference between two passages can be explained as the author choosing a different lens, there is no observation the reading could fail, and a reading that cannot fail is not doing much work. What the literary reading still has, and what no amount of source analysis takes away from it, is the observation that chapter 2 never mentions the sea, the sky, the stars or the water creatures, and so is not a competing account of the cosmos whatever its origin.

Two further points belong to an honest treatment. First, a general statement followed by a detailed expansion of one part of it is a recognised compositional pattern in ancient Near Eastern narrative, which weakens the inference from repetition to multiple authorship. Second, the sequence difficulty in 2:19 can be handled grammatically: Hebrew narrative verbs do not always advance the timeline, so some versions, the ESV among them, render the verb in 2:19 as an English pluperfect, a past-before-the-past form of the verb, which removes the clash with chapter 1. The BSB does not: it reads 'And out of the ground the LORD God formed every beast of the field and every bird of the air' as a simple past, so on its page the apparent clash stays in view, and you can compare the renderings on Bible Hub or Bible Gateway, both free. Critics call the pluperfect reading harmonising; defenders call it ordinary Hebrew syntax, and point out that the verse is not making a claim about when the animals were made but about why they were brought; critics reply that a narrative verb in sequence normally advances the action, so the simple past is the default reading. Both sides are entitled to their objection. What you must not do is tell a class that one of these accounts of Genesis 2:4 is obvious and that the people on the other side have not read the text.

Genesis 2:7 is where the chapter plants its anthropology. The LORD God forms man from the dust of the ground and breathes the breath of life into his nostrils, and the man becomes a living being. Two things follow. The wordplay is unmistakable even in English transliteration: adam is made from the adamah, the ground, which is why 3:19 can say he returns to it. And the phrase the BSB translates 'living being' in 2:7, nephesh chayyah, is the same Hebrew phrase used of the animals, which the BSB renders 'living creatures' in 1:20 and 1:24 and 'every living thing' in 1:21. It does not mean that a human has been given an immortal soul; it means the man is now a living being, as the fish are. Whatever the Bible teaches about human constitution, it is not taught by this phrase, and students who have been told otherwise have been told something that will not hold.

The garden itself is in Eden, in the east, and the Hebrew gan means an enclosed plot, an orchard. The Septuagint translated it paradeisos, from a Persian word for a walled royal park, and the choice is apt, because walled parks stocked with trees and water were exactly what Near Eastern kings built beside their palaces and temples. Genesis 2:10 to 14 gives four rivers, Pishon, Gihon, Tigris and Euphrates. Two are recognisable and two have never been identified with any confidence, and the attempts to map Eden from this list have all failed. Notice what the geography does supply: gold, mentioned twice and the second time called good, and bdellium and onyx, which is to say precious metal and precious stone, the materials of a sanctuary rather than a farm.

That hint is part of a larger cluster, and it was set out most influentially by Gordon Wenham in a 1986 essay on sanctuary symbolism in the Eden story, and developed by G. K. Beale in The Temple and the Church's Mission. Genesis 2:15 says the LORD God placed the man in the garden 'to cultivate and keep it', and the pair of verbs, abad and shamar, is the pair used in Numbers 3:7 to 8 and 18:5 to 6 of the Levites' duty at the tabernacle. Genesis 3:8 has the LORD God walking in the garden, using a verb form applied in Leviticus 26:12 and Deuteronomy 23:14 to God's presence moving in the camp. Genesis 3:24 posts cherubim at the east of the garden, and cherubim are the figures over the ark and woven into the veil, while the tabernacle and the temple are both entered from the east. The lampstand of Exodus 25 is a stylised tree. Ezekiel 28:13 to 14 puts Eden on the holy mountain of God and sets precious stones there. Ezekiel 47 and Revelation 22 bring back the river and the tree of life, and put them in a temple and a city. Be honest that the individual links are of uneven strength, and that which text is influencing which is disputed, but the cumulative case that Eden is being presented as the first sanctuary and the man as its first priest is strong.

Which leads directly to work. Genesis 2:5 explains the absence of plants by two lacks, no rain and no man to cultivate the ground, so the world of chapter 2 is introduced as a place waiting for a worker. In 2:15 the man is given a job before anything has gone wrong, and the job is cultivation and guarding. There is a small but real grammatical nicety here: the verb the BSB translates 'placed' in 2:15 is not the same Hebrew verb as the one it also translates 'placed' in 2:8, and the one in 2:15 is a form built on the root for rest or settle, so that the man is settled in the garden. Whatever weight you give that, the plain point stands. Work is not a punishment, it is not a consequence of the fall, and it is not a regrettable necessity. It is the first assignment given to a creature in a world God has just called good, and it is priestly in its vocabulary.

That reading is what makes the curse in 3:17 to 19 legible. The curse does not invent labour; it frustrates it. Thorns and thistles come up, the ground resists, bread is eaten in sweat. A student who thinks work is the curse will misread both chapters and will also misread his own Monday morning. The biblical complaint about work after Genesis 3 is about futility, which is the subject of Ecclesiastes, not about effort. And the trajectory runs the other way at the end of the book: Revelation 22:3 has the curse gone and God's servants serving him (the Greek verb is the one for service; the BSB renders it 'worship'), with the same river and the same tree.

Genesis 2:18 is the one verse in the first two chapters that says something is not good: 'It is not good for the man to be alone. I will make for him a suitable helper.' Two Hebrew words carry the solution, and both are mistranslated by the ordinary English reader even when the translation on the page is defensible. The first is ezer. It occurs twenty-one times in the Hebrew Bible, and in the large majority of those occurrences the one doing the helping is God. Exodus 18:4 names a son for the God who was Moses's help. Deuteronomy 33:7, 33:26 and 33:29 all use it of the LORD. Psalm 33:20 says he is our help and our shield, Psalm 121:2 locates help in the maker of heaven and earth, and Psalm 146:5 blesses the one whose help is the God of Jacob. In Isaiah 30:5 it describes a military ally Judah hoped to lean on. There is no occurrence in which the word means a junior assistant or a domestic servant, and that is not a tendentious reading of the data; it is the data. English 'helper' has drifted since 1611 and now carries exactly the sense the Hebrew lacks.

The second word is kenegdo, which appears in this form only in 2:18 and 2:20. It is built on neged, meaning in front of, over against, corresponding to, so the phrase says something like a helper matching him, or his counterpart. And the narrative immediately shows you what the phrase is for. The animals are formed and brought to the man in 2:19 to 20, he names them all, and 2:20 ends by saying, 'But for Adam no suitable helper was found.' The whole parade exists to establish a negative result: nothing in the animal world corresponds to him. So the function of ezer kenegdo in its own paragraph is to assert likeness of kind, not to assign a role. You should also know, and then handle carefully, R. David Freedman's 1983 proposal in Biblical Archaeology Review that the phrase should be rendered 'a power equal to him', on the basis of a second root behind ezer meaning to be strong. It is widely quoted in popular writing and it is not widely accepted by Hebraists. Mention it as a proposal; do not build an argument on it.

Then the making of her, which has its own vocabulary. Genesis 2:21 has God take one of the man's tsela', which the BSB translates 'ribs' ('one of the man's ribs'), and 2:22 has him build the rib into a woman. Tsela' elsewhere in the Hebrew Bible means a side: the sides of the tabernacle frames in Exodus 26:20 and 26:26 to 27, the side chambers of Solomon's temple in 1 Kings 6:5 to 8. Genesis 2:21 is the only place a rib is the obvious sense, and the Septuagint rendered it 'side'. If side is right, the picture is not a spare part but a division: she is taken from his flank, which is why his first words in 2:23 are 'This is now bone of my bones and flesh of my flesh.' (The Hebrew for 'now' is happa'am, literally 'this time'.) That is not a compliment on her appearance. The formula 'bone and flesh' is a declaration of kinship elsewhere in the Hebrew Bible, though the BSB renders it idiomatically as 'flesh and blood' in those places, so the Hebrew is literally 'my bone and my flesh': Laban says it to Jacob in Genesis 29:14, Abimelech says it to his mother's relatives at Shechem in Judges 9:2, and the tribes say it to David in 2 Samuel 5:1. The man is identifying her as family, which is the result the animals could not supply.

Now the dispute this passage carries, which divides serious interpreters and which you must not settle for your students. The complementarian reading holds that Genesis 2 sets out a created order in which the man has a particular responsibility that the woman does not: he is formed first, the command of 2:16 to 17 is given to him, the woman is made as his ezer, he assigns the word Woman in 2:23, and Paul reads the chapter this way in 1 Corinthians 11:8 to 9 and 1 Timothy 2:13, grounding his argument in the sequence of creation rather than in the fall. The position insists in its own literature, for instance in Raymond Ortlund's essay in Recovering Biblical Manhood and Womanhood, that this is a difference of role within an unqualified equality of worth, both image-bearers and both commissioned in 1:28.

The egalitarian reading holds that the hierarchy is not in chapter 2 at all. Its case: ezer is a word for a rescuer, not a subordinate; kenegdo says corresponding, which asserts parity; the commission of 1:26 to 28 is given to both without distinction; 2:23 is not a naming, because the Hebrew naming formula, literally 'he called her name', which Genesis uses when naming happens, does not appear until 3:20, and Phyllis Trible argued in God and the Rhetoric of Sexuality that 2:23 is recognition rather than an exercise of authority; and the asymmetry of rule enters the narrative in 3:16, in a list of consequences rather than a list of institutions. Richard Hess, Philip Payne and the essays in Discovering Biblical Equality develop these lines in detail. Both readings are being argued from the Hebrew by people who have read it, both have a serious answer to the other's best point, and 2:24 belongs to neither party exclusively: one flesh is the conclusion the chapter draws, and it is the verse Jesus quotes in Matthew 19:4 to 6 alongside 1:27 when he is asked about divorce.

Read 2:16 and 2:17 in the order they are given, because the order is the argument. The permission comes first and it is vast: every tree of the garden, with the Hebrew doubling the verb for emphasis, which is why the BSB has 'You may eat freely' (literally, 'eating you may eat'). The prohibition comes second and it is single: one tree. A great deal of bad preaching on Genesis 3 proceeds as though the garden were a minefield with one safe path. It is the reverse. The restriction is one item on an otherwise unrestricted list, and the serpent's first move in 3:1 is to misquote the proportions.

What is the tree of the knowledge of good and evil? Four readings are in circulation and three are live. The first takes the knowledge as moral knowledge, the capacity to tell right from wrong, which the man and woman lack until they eat. Its difficulty is that a creature who cannot tell right from wrong cannot meaningfully be given a command. The second takes 'good and evil' as a merism, a figure in which two poles stand for everything between them, so that the tree offers comprehensive knowledge or mature discernment. This has solid support: in 2 Samuel 14:17 and 14:20 the king's wisdom is described as discerning good and evil, Solomon asks for the same discernment in 1 Kings 3:9, and Deuteronomy 1:39 describes children as those who 'did not know good from evil', which is plainly about maturity. The third, and the one that fits the narrative most closely, takes the knowledge to be not information but prerogative: the right to determine for oneself what is good and what is evil, which is the autonomy the serpent offers in 3:5 when he says they will be like God, knowing good and evil, and which 3:22 says they have obtained. The fourth, an older reading that took the knowledge to be sexual, is in trouble from the start, since 1:28 blesses fruitfulness and 2:24 commends union before anybody eats anything.

Then the sentence, and here you must not tidy. 'For in the day that you eat of it, you will surely die', and in Genesis 5:5 Adam dies at nine hundred and thirty. Four answers are given and more than one of them may be right at once. The first notes that the Hebrew construction, the infinitive with the finite verb, is the standard formula of the law courts: in Exodus 21:12 and 21:15 to 17 the same shape means he must surely be put to death. On that reading 2:17 pronounces a capital sentence, and the question the narrative then raises is not whether God meant it but why it is commuted. The second takes the death as the alienation the next chapter describes, hiding in 3:8, expulsion in 3:23 to 24, and a relationship broken on the day itself. The third takes it as the onset of mortality through loss of access to the tree of life, which is explicitly what 3:22 to 24 secures. The fourth reads the phrase as doomed to die, a sentence passed that day and executed later. Notice that the serpent's flat contradiction in 3:4, 'You will not surely die', turns out to be true at the shallowest possible level and false at every other, which is how the best lies work.

Now the question the brief of this lesson has been building toward. What does a prohibition imply about the creature who receives it? At minimum, that the creature can do otherwise. Commands are not issued to rocks. Genesis 2 presents a human being who is addressed, who is capable of obeying or not, and whose obedience is therefore something rather than nothing. And notice where the tree stands. Genesis 2:9 puts the tree of life and the tree of the knowledge of good and evil in the middle of the garden, and 3:3 has the woman place the forbidden tree there too. The limit is not at the perimeter, where it could be forgotten; it is at the centre, where it is passed every day. Dietrich Bonhoeffer, lecturing on these chapters in Berlin in 1932 and 1933 in what became Creation and Fall, made exactly this point: that the boundary stands in the middle of human existence, and that a creature who knows its limit and knows it is loved within it is in a position no unlimited creature could occupy.

There is a counter-case, and students will raise it, so raise it first. No reason is given for the prohibition. The tree is not said to be poisonous, dangerous or unclean; the only thing wrong with eating it is that it is forbidden. Some readers find that arbitrary and say so. The reply available from the text is that an unexplained command is the only kind that can test trust rather than calculation, since a command with a published reason is obeyed on the strength of the reason. Whether that reply satisfies you is a question this lesson leaves open, and the larger problem it points at, why a good God permits the possibility he has just created, belongs to the Systematic Theology track, where it is given a lesson of its own and no tidy answer. What Genesis 2 gives you is a generous permission, one limit, a creature capable of keeping it, and a God who states the stakes in advance. Everything in chapter 3 happens inside that frame.