On Overinterpreting the Bible

In David Carr’s meticulously argued and methodologically sensitive new book The Formation of the Hebrew Bible (see Angela Roskop’s positive and constructive review), he draws attention to an important observation that has significant interpretive implications—particularly for those interested in literary borrowing, whether construed inner-biblically or more broadly within the ancient Near Eastern literary world.

Carr observes that the transmission and growth of the Hebrew Bible is not entirely a literary phenomenon. Nor is it accurate to speak of an oral stage of transmission followed by a literary stage of transmission. Rather, what we have is “a mix of oral and written dynamics” (17) with writing-supported memorization. Carr argues that memorization during the formative stages of the Hebrew Bible was neither exhaustive nor aimed at word-for-word reproduction, that “memory varients” preserved in the text are a clue to the oral dynamics involved in textual expansion and transmission.

What are the interpretive implications of this observation? Carr writes,

I maintain that it would be a mistake to try to explain most or all of the above variations [e.g., Prov6:10-11~24:33-34; 16:2a~21:2a~12:15a; 16:2b~21:2b; 12:11b~28:19b; 19:24b~26:15b; 19:5~19:9] as part of a conscious process. There is a temptation to attempt this, particularly because so many biblical scholars work in religious contexts where the goal is to find meaning in every aspect of the text no matter how seemingly insignificant. Indeed, there is a certain magic attached in scriptureal contexts to finding meaning in every non-understood variation of a line or saying. That is part of what biblical scholars, particularly those working in traditions that view the Bible as Scripture, do: find new meanings in initially obscure ancient texts, often under the (ancient) preconception that every aspect of the canonical text is significant. Set enough such scholars loose on variations such as those discussed above, and it is only a matter of time before many are explained as a result of either error or specific sorts of exegetical/theological modification. (33)

Whether or not Carr’s characterization of religiously interested scholars is fair, his critique is worthy of consideration. In reading the book of Amos, I noticed what I believe are likely “memory variants” in the text

  • Amos 1:4 – So I will send (שׁלח) a fire on the house (בית) of Hazael, and it shall devour the strongholds of Ben-hadad.
  • Amos 1:7 – So I will send (שׁלח) a fire on the wall (חומה) of Gaza, fire that shall devour its strongholds.
  • Amos 1:10 – So I will send (שׁלח) a fire on the wall (חומה) of Tyre, fire that shall devour its strongholds.
  • Amos 1:12 – So I will send (שׁלח) a fire on Teman, and it shall devour the strongholds of Bozrah.
  • Amos 1:14a – So I will kindle (יצת) a fire against the wall (חומה) of Rabbah, fire that shall devour its strongholds.
  • Amos 2:2a – So I will send (שׁלח) a fire on Moab, and it shall devour the strongholds of Kerioth.
  • Amos 2:5 – So I will send (שׁלח) a fire on Judah, and it shall devour the strongholds of Jerusalem.

If Carr is right—and I am inclined to believe that he is—variations like this that occur across a number of verses are best explained as memory variants, different ways in which someone at some point remembered and transmitted these oracles against the nations. This is the same line, despite the differences in terminology (שׁלח verses יצת and בית versus חומה) or variations in sentence construction. Of course, examples like this can be multiplied many times over. Carr is bringing us to a more “methodologically modest” (4) place in which to engage the Hebrew Bible.


13 thoughts on “On Overinterpreting the Bible

  1. Joseph,

    Interesting post. In your examples from Amos, is “wall” or “house” or # (in three places neither term is used) the memory variant? When are we looking at poetic variation and when “memory variant?” I am not entirely opposed to the notion of memory variants as explaining some textual variants (especially between documented texts), but its application to the earliest period of the formation of the Bible seems more like methodological speculation than modesty to me.

    1. Both the verbs and the nouns (or absence thereof) suggest to me memory variation. Carr introduces the notion of memory variants in texts outside the biblical corpus, particularly the epic of Gilgamesh. Amos may be early insofar as the formation of the Hebrew Bible is concerned, but this is not “early” for the phenomenon that Carr is describing. I don’t think Carr is speculating when he can find this same kind of thing occurring in much older texts than Amos. On the issue of poetic variation, do you think the differences between these lines are better explained as poetic variation? I would expect greater variation for that (e.g., Psalm 19:7-9).

      1. How many variants from an “original” can there be? In the case of the nouns, you are now positing three possible “texts.” This is very improbable, for in documented cases there is usually only two and only in relatively few cases are there three. If memory variants are multiplied in the way you are suggesting, then hypothetically there will be no end to the positing of variants caused by forgetfulness. There must be more evidence to make the claims that Carr is making, and there simply is not the evidence at this stage in the Bible’s formation.

        In the case of EoG, are these documented memory variants or are they simply asserted based on “trends” in one text? Or has Carr pushed the methodological problem back a level? In the situation of undocumented cases, my problem with positing memory variants is that there will always be a competing explanation such as poetic variation or the like and thus this is only an interesting proposal and not one which will further our understanding.

      2. What is this “original” of which you speak? Let us posit that Amos does indeed recite these oracles against the nations? Does he do so once? Twice? Recurrently? What then is the original line? The first time he recites these lines? The last? Which line is correct? Is it the one that says “wall” not “house,” “send” not “kindle?” Carr’s point is that these memory variants, far from being errors or forgetfulness, show how we remember and recall things. Has the scribe forgotten that God will “send” the fire when he writes “kindle,” or does this constitute the same “text” for the ancient ear and reader? Carr argues the latter (on which I may post again this week).

        Trust me, my brief post does not do justice to Carr’s argument, and I knew it wouldn’t. I want to raise awareness and interest, but not to be the grounds upon which someone rejects his thesis. And to your question about documented memory variants, yes, Carr is relying on documented memory variants.

  2. Ok, fair enough :). “Original” is the slippery term in the discussion, yet if we are going to talk about variants, there is no way around positing some type of ur-text.I think all of these oracles are in tact textually speaking unless there is some documented evidence which truly SUGGESTS a variant. A proposed pattern and then speculative deviations from said pattern does not suggest anything to me except freedom and creativity on the part of the prophet. But at this point I need to read Carr :). So, Joseph, your post did its job :).

    1. John, now that I have Carr’s book in front of me, I can provide a snippet that may help address some of your questions. “There is always the possibility that certain sayings existed in yet more forms than those reflected in Proverbs. In these cases, the parallel sayings may not be related to each other in a unilinear sort of development. Instead, they may be reproductions of earlier forms of such sayings that were also different. In sum, in so far as the sayings in Proverbs were reproduced—in whole or in part—through memory, the search for and Ur-text and clear lines of dependence and revision often will be fruitless” (33).

      So to expand Carr’s argument to my examples from Amos, what if the Ur-speech used the word עיר two of the seven times the line recurred, with בית occurring twice and חומה occurring three times. The scribe is not manipulating or otherwise “changing” the Ur-text by remembering only the two terms and using one or the other for all seven sayings. This is the same text to an ancient reader (see the quote in today’s post, “On Authorial Intention”).

      I don’t know what evidence you have to suggest that memory supported writing has produced something intact if by intact you mean a word by word dictation of the Ur-speech/text.

  3. If I understand the concept and its application to these Amos passages correctly, you are suggesting that there was one formulaic utterance leveled at one or more foreign nations and that the variation among its various appearances can be explained by people remembering it differently. What I don’t get is what allows you to make the assumption that the formula originally involved one set of terms? You dismiss merely “poetic” variation, but one has to take into consideration the possibility that the variations actually mean something related to the political or geographical context. Is it not relevant that the oracles directed at great cities, and only those, use the term “wall”? Rather than a variation in memory of a single formula, perhaps the use of wall in those oracles has reference to the fact that the entity imagined is a city rather than a larger region. Finally, what does your idea about “memory variants” here imply about the composition of Amos 1-2? Does it imply that the formulae for each oracle was remembered separately and semantically fixed before being brought together? Does the rhetorical and poetic coherence (of most) of this section of Amos make this credible? I don’t think it does.

    1. Hashavyahu, good comments. To clarify, I do not assume that the formula originally involved one set of terms. I find the idea of an “original” problematic (not something to be avoided, but something to treat delicately). For example, these kinds of oracles could have been historical utterances, and if they were the idea of an “original” is particularly problematic. Is the “original” the first utterance (which history could easily forget and certainly neglect to record), is it the stable form it achieved after multiple utterances, is it the first time they were written down (which may or may not be related to the book of Amos), or is it the initial composition of this portion of the book of Amos?

      Historically, I think one of two things have occurred. Scholars have either identified exegetically significant differences between such minor variations (as you propose concerning “wall” and its absence) or they have said there are no exegetically significant differences and explained the differences that do exist as “poetic variations.” I believe Carr has suggested a viable third option, “memory variants.”

      What does my application of Carr’s “memory variants” to this example imply about the composition of Amos 1-2? I’m not drawing any conclusions yet (nor do I see any initial significance in such an observation to the composition history of the book of Amos). Carr’s argument (which I have yet to reach in my reading) is that the early form of the OAN followed a 4+1 patter (Damascus, Philisitia, Moab, Moab, Israel). I’m not sure where he is getting this from, so I will withhold judgment for now.

      To be clear, Carr does not point to Amos 1-2 as exemplifying “memory variants.” This is my attempt to apply his insights.

  4. By suggesting “memory variants” as an explanation for variation in the OAN in Amos 1-2 you are already drawing conclusions about the composition of those chapters. Unless you can defend a theory of composition that involves verbally fixed oral or written traditions about these oracles that were originally separate and then gathered together into Amos 1-2, then invoking memory variation to account for the variations there makes no sense. As valid as this approach may be elsewhere, such as in Proverbs, it does not look like a good fit here because it implies a theory of composition for these chapters for which there is 1) no evidence, and 2) a literary coherence that functions as counter-evidence.

    1. Perhaps I have not articulated myself well, but it would not be necessary for the OAN to be verbally fixed oral or written traditions that were originally separate and later gathered together in order for the variations among them to be regarded as “memory variants.” I believe reading the first 36 pages of Carr would clarify this. I don’t mean to be dismissive, but I don’t feel the force of your argument.

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 )

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s