The Law of Conservation of Biblical Characters

Now that that’s out of the way…

After this week’s Torah portion tells the story of Abraham sending his servant to find a wife for his son Isaac, it tells how, after Sarah’s death, Abraham remarried (Gen. 25:1, Revised JPS translation):

And Abraham took another wife, whose name was Keturah.

This woman is not mentioned earlier in the Torah. Rabbinic tradition identifies her with Hagar, Sarah’s maidservant whom she gave to Abraham as a concubine, whom Abraham later banished along with Ishmael, his son by her. This identification is so strong that later authorities simply present it as fact. One of the most prominent Aramaic translations, Targum Pseudo-Jonathan renders this verse as follows (my translation):

And Abraham again took a wife, and her name was Keturah. She is Hagar, who was bound to him from the beginning.

By saying Keturah/Hagar was “bound to him from the beginning”, the Targum highlights the upgrade in her status, from concubine to wife. Rashi simply says, “This is Hagar,” and then summarizes how the Midrash makes that identification (Bereishit Rabbah 61:4): again based on a pun-based reading of her name, which perhaps relates to “perfume” or “incense” (the verbal root is q.ṭ.r.), because she was “perfumed with mitzvot and good deeds”. The Targum explains her name via a pun as well, but in Aramaic: the word for “bound” is ketira. And though the Midrash and Rashi don’t make the connection explicit, Keturah is referred to in 1 Chron. 1:32 as “Abraham’s concubine”, whereas in Genesis she is referred to as another “wife”.

However, the identification of Keturah with Hagar is by no means the consensus of the medieval rabbis: Ibn Ezra, for example, explicitly rejects it, as do Rashbam (Rashi’s grandson), Nachmanides, and others. Still, as far as the “orthodox” (small-“o”) interpretation of Scripture went, Rashi’s full-throated endorsement of the Midrash won the day. The ArtScroll Chumash, a standard text found in synagogues, homes, and schools throughout the Orthodox (big-“o”) world and beyond, quotes Rashi’s view in its commentary like there’s no debate, and even performs some legerdemain to make it look like Nachmanides (Ramban) endorses this view when he actually rejects it:

Keturah. Abraham remarried Hagar, who was given this name because her deeds were as beautiful as incense (ketores), and because she remained chaste (keturah is Aramaic for restrained) from the time she was separated from Abraham (Midrash; Rashi).
Ramban explains that at this point in his life, Abraham felt no need to find a wife from among his family in Haran. That was required of Isaac — as it would be again later of Jacob — because the covenant of Israel would be fulfilled through him.

See what they did? The commentary cites the Midrash and Rashi as the sources for its explanation of the word Keturah, and then leads off the next line with “Ramban explains”—the effect is to harmonize a dissenting perspective. And it is this perspective that I remember from my middle school days: when I learned this story, I was taught, explicitly, that Keturah is Hagar, full stop. I was not informed that this was a subject of disagreement; I was taught that this was the consensus of the rabbis and therefore the correct meaning of the text. It was not until I could direct my own learning by reading the commentaries on my own that I discovered that this was a subject of controversy.

The midrashic identification of Keturah with Hagar is not a one-off thing either. Indeed, this gets to something I like to call “the law of conservation of Biblical characters”. (It’s not really a “law”, it’s a principle, but you get the picture.) There are numerous examples throughout Judaic literature of identifying a minor character with another, usually more major, character. Oftentimes these midrashim are justified based on (rather specious) linguistic clues, revolving around puns and folk etymology. Keturah = Hagar is merely one of the most prominent examples. Others include:

• Pinchas, Aaron’s grandson who kills a couple in flagrante delicto = Elijah the Prophet
• The king of Nineveh in the Book of Jonah = Pharaoh, who did not die at the Red Sea
• Yiscah, daughter of Haran and therefore Abraham’s niece = Sarah, Abraham’s wife
• Dinah, daughter of Jacob = an unnamed wife of Shimon, and/or = mother of Asenath, Joseph’s wife
• Shifra and Puah, the midwives who saved the Hebrew boys in Egypt = Miriam and Yocheved, sister and mother of Moses
• Laban, Jacob’s uncle = Bilaam, the prophet, owner of a talking donkey, who tries to curse Israel
• Hatach, a chamberlain in the royal court of Shushan in the Book of Esther = Daniel, the prophet and official in the Persian court
• Malachi, the final prophet = Ezra the scribe, or = Mordechai from the Book of Esther

And many others. It’s not just limited to people, either: objects get this treatment too: the midrash identifies Esau’s garment as being the one that God created for Adam (Bereishit Rabbah 63:13), and Moses’s staff, with which he performed signs and miracles, is identified with Jacob’s and then Judah’s staff, which very staff was carried through the future by every future king of Israel and will be restored to the rightful ruler in Messianic times (Yalkut Shimoni 763:12).

Why do the rabbis do this? What is the reason that underlies “the law of conservation of Biblical characters”?

One reason is probably a consequence of the traditional view of the Torah as a sacred text in which no detail is trivial. This leads to the investment of seemingly minor characters, objects, places, dates, and so forth with particular significance. One way to do this is to associate these minor details with characters or situations the tradition already cares about. I think this motivation underlies, to some degree, all of the examples of the “law of conservation of Biblical characters”, and is even the simple primary motivation in many of them. For example, the Talmud (Megillah 12b) identifies Memucan, the courtier of Ahasuerus who persuades the king to issue an edict at the end of the first chapter of Esther, with the wicked villain Haman. There seems to be no reason to need this identification except to imbue a minor character with more significance than the plain reading of the text gives him. (Tosafot note that there is also a midrash, also related by the Targum Sheini, that Memucan is Daniel, who is, of course, a hero rather than a villain—so take your choice.)

In many cases, this kind of midrash has the goal of establishing thematic connections and reflecting parallels between aspects of narrative and character. For example, the identification of Laban with Bilaam, also recorded in Targum Pseudo-Jonathan, reflects a parallel between both men’s role in the history of the Israelite nation: both try to undermine the Israelites—Bilaam by cursing them during their wanderings in the wilderness, and Laban by dealing deceitfully with Jacob himself.

Sometimes this kind of interpretation exists as an apologetic for the apparent misdeeds of a major character. This explains why Shimon, whom the Biblical text implies married a Canaanite, is reinterpreted as having married Dinah, whose parentage is not suspect. Of course, this introduces a new problem—specifically, that if this identification is correct, Shimon married his sister, or at least had a child by her!^[1] It’s fascinating that this objection is silently passed over; to the originator of this midrash, evidently, Canaanite ancestry was more problematic than incestuous ancestry.

The converse is that sometimes the identification of a minor character with a major one can be used to redeem someone, either that character or another. And this brings us back to the arch-example of the “law of conservation of Biblical characters”: Hagar/Keturah. Sarah demands Abraham banish Hagar and Ishmael into the wilderness with nothing but a skin of water, and God expressly gives permission for Abraham to do this. By identifying Hagar with Keturah (conveniently after Sarah’s death), Hagar is given a second chance with Abraham, and literally, a good name. This also quietly rehabilitates Abraham and “fixes” anything that may have been problematic about throwing his concubine out in the first place.

But at what cost? In making Keturah Hagar, the tradition erases Keturah. Who was she? We have no idea. The modern commentator Richard Elliott Friedman calls Keturah “the most ignored significant person in the Torah”—and in this traditional reading, she is even more ignored, because she isn’t a person at all, but only a new name for someone we’d already met. Her six sons, the ancestors of the Midianites and others, become, like Ishmael, Hagar’s sons. Keturah’s own story vanishes. The “law of conservation of Biblical characters”, taken seriously, works by deleting people.

And there’s a surprising parallel in the way a flattened reading of text erases the diversity of the commentators. We saw this with how the ArtScroll Chumash smoothly elides Nachmanides’ dissent from this identification and by some deft editorial harmonizing makes it appear that this midrash represents a consensus that does not, in fact, exist. The midrash merges two women, and this modern commentary merges the diversity of rabbinic opinion. Both smooth over a seam—but a seam is where the text invites you in.

Tradition’s tools cut both ways. Midrash can redeem Hagar and erase Keturah all at once. Learning to read tradition well means seeing both, and not treating either as a reason to throw the whole thing out. It also means noticing where disagreement has been smoothed over. And it means making room for the possibility that Abraham and Sarah were imperfect human beings who had to live with the consequences of their actions, without a midrash to tidy them up. Don’t let either woman be sacrificed in the name of harmonization. Hold both Hagar and Keturah.

---

1. In fandom parlance, this is a “voodoo shark”: an explanation offered to patch a plot hole that is itself more implausible, or more troubling, than the hole it was meant to fix. The astute reader will notice that there are quite a few of these that come up as a result of the “law of conservation of Biblical characters”. ↩︎