“When God Began to Create…”: Nouns Bound to Verbs

In order to post on this, I have to break an informal but firm rule I have kept for myself: not to blog on Genesis 1–3. I have no wish to attract any debate or controversy about the truth/historicity/inerrancy of the story/myth/history/polemic/whatever of Gen 1. That said, my Hebrew students are at a point now where they can make sense of the Bible’s very first phrase, so let me have this, please.

As for my Hebrew students: sit up and pay attention. You won’t know the Akkadian, but you can still follow the argument: just give the Hebrew special attention. I am throwing in the Akkadian for my enjoyment, and because I am up to my neck in some texts at the moment.

The text in view is Genesis 1:1:

בראשית ברא אלהים

Traditionally, “In the beginning, God created…” But, more recently, “In the beginning, when God created…” Or even, “When God began to create…” Students naturally want to see a definite article in the first form: BA-rēšīt (“in THE beginning”). They are told (if they are told anything) that the form is in construct, and therefore unable to take a definite article. The obvious question is, “To what is it in construct? It precedes, not a noun, but a perfect verb.”

In Hebrew relative clauses, one permissible construction is a construct noun followed by a finite verb: קרית חנה דוד “The-district-of David encamped,” that is “The district where David encamped” (Isa 29:1). Similarly, for Akkadian relative clauses, one permissible construction is a bound (“construct”) form followed by a finite verb (with -u of subordination): bīt ēpušu “the house-of I-built,” that is “the house (which) I built.” (Note that, in both Hebrew and Akkadian, we will more often see the relative particle in such a clause: Hebrew אשר, Akkadian ša.)

This construction is often used with verbs of time. So in Hebrew: ביום הציל יהוה “In-the-day-of YHWH delivered,” or “In the day when YHWH delivered” (2 Sam 22:1). That the noun of time is in construct is more clearly shown in examples with distinct bound forms: בליל שדד ער “in-the-night-of Ar was devastated,” or “in the night when Ar was devastated” (Isa 15:1). In Akkadian: UD-um É.GAL KUG.BABBAR i-r-i-[šu], or Ūm ekallum kaspam irrišu: “The-day-of the-palace receives silver,” or “The day when the palace receives silver.” (CT 8 36a, cited in Huehnergard exercise 19.G.2.)

An excellent parallel to our text is found in Hosea 1:2: תחלת דבר יהוה “The-beginning-of YHWH spoke,” that is “When YHWH first spoke” or even “When YHWH began to speak.”

By now, it is clear that these constructions are parallel to that of Genesis 1:1. Looking at it again, we can translate:

“In-the-beginning-of God created,” that is, “In the beginning, when God created,” or even “When God began to create.” In other words, the Bible begins with a subordinate clause, preceding the main clause. Where, then, would you say that the first main clause in the Bible begins?

7 Responses

  1. Well, if you translate רשית as “beginning” then it seems like it needs a definite article, but not if you translate it as “first”. It works fine as “At first, God created…”

    Personally, I don’t really have a preference between the traditional reading and the construct.


    • Hi, Pete, thanks for reading. When I talked about the absence of a definite article, I was just trying to articulate the difficulty that beginning Hebrew students often have with the form: they hear “the” in English, and look for it in the Hebrew.

      My reservation about “at first” is that, sure, it relieves the trouble in English, but the grammatical warrants for it from the Hebrew aren’t self-evident: we’re translating a Hebrew noun (ראשית) with an English adjective (“first”) without that English adjective even being clearly substantive (well, now or is it? Basically the phrase is elliptical for “at the first”…I dunno). It’s a possibility, but we’ve got this very well-attested grammatical feature—construct forms bound to finite verbs to form subordinate clauses—with many examples parallel to Gen 1:1. I think much might depend on whether one *really wanted* the remainder of verse 1 to be a main clause rather than a subordinate clause.

      Pete, I never thanked you for your posts on definiteness. I’ll thank you here, if you’ll promise not to tell everyone how interesting I think definiteness in Hebrew is.

      • i’d like to suggest that ברא is mispointed: that it shouldn’t read בָּרָא but rather בְּרֹא as in genesis 5:1. this would make it an infinitive construct, and make it grammatically similar to both genesis 2:4b and genesis 5:1, which are both (temporal?) infinitive constructs, preceded by complex prepositions. this makes the most sense, especially considering that genesis 5:1,2 was written as a direct compliment to genesis 1. and frankly, i think that makes a whole lot more sense than having a finite verb.

        this would render the verse, “when god began creating…” which is similar to the nJPS translation you quoted above. for the record, i happen to think that the third verse is the independent clause, and the waw-consecutive can safely be ignored as it can in genesis 6:2.

  2. In light of your comments, I’d be curious what you think of my argument about the syntax of this verse: VT 58(1):56-67. (You can download it from my site.)

    Ellen van Wolde (JSOT 34(1):3-23) likes my analysis but takes me to task for a poor final translation (it was openly bumpy due to source and target language differences, which I discuss in a final footnote, but I clearly failed in fully justifying that rhetorical move).

  3. Mark S. Smith in his new book, THE PRIESTLY VISION OF GENESIS sums up the case nicely, and agrees with your translation as do Jewish translators and some recent Christian Bible translators. Indeed, according to Mark, your translation of Genesis 1 has become the majority opinion.

  4. The noun ראשׁית may indeed be a construct for בָּרָא the finite verb which follows. If we pay attention to the first clause only, it might be read as subordinate.
    There is no need to vocalize ברא as an infinitive construct, even if it is possible. However, if one would emend the Masoretic vocalization, why not emend the vowel of the initial word בָרֵאשִׁת bAre’shit ?
    My understanding is that ראשׁית does not require definite article, as well as other nouns like תהום , תֵּבֵל , שׁאוֹל . נ . In Is 46:10, this noun is also used without the definite article, though it is definite by itself: מַגִּיד מֵרֵאשִׁית אַחֲרִית. The drop of definite article is common in poetry (archaic language). It seems that the noun ראשׁית receives definite article only when it means “first-fruts” (לָרֵאשִׁית Ne 12:44). There is no other instance where ראשׁית receives definite article. Its antonym אַחֲרִית is also used without definite article, as well as its synonym ראשׁ (beginning) in Pr 8:23; Is 40:21; 41:4,26; 48:16; (מראשׁ); Qoh 3:11 (מראשׁ ועד סוף). The noun ראשׁ receives article when it means “head”, “chief” etc., but not with temporal meaning.
    And there is another problem that appears if we choose to see ראשׁית as a construct in Gen 1:1. What about the syntax of verse 2 ? If reading ראשׁית as a construct is OK, then in verse 2 the first clause must have been ותהי הארץ , not וְהָאָרֶץ הָיְתָה … Thus my conclusion is that the traditional translation (cf. Old Greek, Vulgate, Targum) is the best solution, up to this time.

    • > If reading ראשׁית as a construct is OK, then in verse 2 the first clause must have been ותהי הארץ , not וְהָאָרֶץ הָיְתָה …

      i don’t understand your logic. wouldn’t deviation from the standard syntax be a good hint that verse 2 is NOT the independent clause, and that it is part of a larger subordinate structure?

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: