In Exodus 21:22-25, the consequences for hitting a pregnant woman are discussed. But the Masoretic Text and the Septuagint seem to differ. The MT refers to harm to the child, while the LXX refers to the "formation" of the child:

22 When men strive together and hit a pregnant woman, so that her children come out, but there is no harm, the one who hit her shall surely be fined, as the woman’s husband shall impose on him, and he shall pay as the judges determine. 23 But if there is harm (אָסוֹן), then you shall pay life for life, 24 eye for eye, tooth for tooth, hand for hand, foot for foot, 25 burn for burn, wound for wound, stripe for stripe. (ESV)

22 And if two men strive and smite a woman with child, and her child be born imperfectly formed, he shall be forced to pay a penalty: as the woman's husband may lay upon him, he shall pay with a valuation. 23 But if it be perfectly formed (ἐξεικονισμένον), he shall give life for life, 24 eye for eye, tooth for tooth, hand for hand, foot for foot, 25 burning for burning, wound for wound, stripe for stripe. (Brenton's LXX)

Apparently, Augustine and Theodoret believed that early (unformed or "imperfectly formed") fetal termination was not homicide on the basis of this verse.

What's the deal with the difference between the Greek and Hebrew? Is the Greek a (mis?)translation of the Hebrew? Or is it based on a different manuscript tradition? Is there a way to know?


3 Answers 3


The Idea in Brief

The LXX speaks of two outcomes: one regarding the developed fetus and one undeveloped. The Masoretic Text however simply presents the fetuses in the plural, and in this respect there is no amplification or clarification between developed and undeveloped in the Masoretic Text.

In summary, if the Masoretic Text carries the original literary form, then the LXX would be the wider explanation for the plurality of the fetuses through the two scenarios: one developed fetus and one undeveloped. However, if the plural form refers to an indefinite idea (according to recognized Hebrew scholars), then the answers appear to lead to something else: that is, there is distinct nuance between murder and manslaughter, which Jewish oral tradition makes clear.


The Samaritan Pentateuch and the LXX are contemporaneous witnesses, since both appeared several centuries before the current or common era. The Samaritan Pentateuch, like its LXX counterpart, addresses one fetus.

Exodus 21:22-23 (Samaritan Pentateuch)
22 If men strive, and hurt a woman with child, so that her fruit depart [from her], and yet no mischief follow: he shall be surely punished, according as the woman's husband will lay upon him; and he shall pay as the judges [determine]. 23 And if [any] mischief follow, then thou shalt give life for life.

According to Page 93 of the Vetus Testamentum Hebraicum cum Variis Lectionibus by Benjamin Kennicott, the verses above from the Samaritan Pentateuch are identical to the Masoretic text, except that the fetus here is in the singular and not plural (which is the case of the Masoretic Text).

To recap, the Samaritan Pentateuch provides its translation in reference to one single fetus. Its contemporary, the LXX, also provides for one single fetus, however, also provides for two scenarios regarding the one fetus. In other words, the original proto-Hebrew texts may have referred to a plurality of fetuses and therefore appeared ambiguous.

For example, another witness is the Targum Onqelos, which may have appeared as early as the third century of the current or common era. This Targum provides the plurality of fetuses, but without any amplification or explanation. In this regard, the Targum Onqelos is consistent with the Masoretic Text, which appeared many centuries later (and which also refers to the plurality of fetuses). The proposed translation of this passage from this Targum is as follows:

Exodus 21:22-23 (Targum Onqelos)
22 On account that men fight and strike the woman who is pregnant, and the offspring (lit., unborn children) come forth, and no death results, a claim will be demanded just as what is appropriate to him, the woman's husband, and they will pay according to the words of the judges. 23 And if there may be death, then they shall give life for life.

In summary, one hypothesis for these disparities between these various texts is that the proto-Hebrew texts were ambiguous, and the LXX was one attempt to amplify and clarify the plurality of fetuses in the passage. (The Samaritan Pentateuch eliminated the ambiguity and referred to only one fetus.) The Targum Onqelos, and later the Masoretic Text, preserved the original proto-Hebrew form with its ambiguity.

Why then the plurality of fetuses in the womb? The plural form was to express an indefinite idea in Biblical Hebrew (and therefore would be inclusive of all scenarios). For example, Keil & Delitzsch write the following. Please click on the image to enlarge.

enter image description here

In this regard, Gesenius also notes the following:
enter image description here

In Biblical Hebrew, the indefinite plural refers to the wider idea, and therefore would capture all nuances and scenarios. And so we are back at "square one." What then is the meaning of this passage, if the student relies on the plain and normal reading of Scripture? In this regard, the rabbis who preserved the oral traditions of the Jews provide the most sensible explanation: the idea in this passage concerns the difference between manslaughter, which is intentional, and murder, which is intentional.

In this regard, then, the Babylonian Talmud in Sanhedrin, Folios 78B, 79A, and 79B provides the most comprehensive and authoritative view concerning the passage in question from the perspective of Jewish oral tradition. (The hyperlink to the Talmud reference includes commentary in the last paragraph by the translator, Jacob Neusner.) That is, the rabbis understood from the Biblical passages that there were nuances between manslaughter and murder. In this regard, the intent to kill is what discriminated manslaughter from murder. In other words, the life of the mother and the life of the fetus have equal par in this passage, because in the event of either the death of the mother or of the death of the fetus, the result would have been manslaughter (since the men who were fighting had no intent of killing either one or the both of them). The passage therefore does not equate the life of the fetus to the level of common chattel, but to the same level of manslaughter as if the life of the mother too were killed (since both mother and fetus appear together in this passage as innocent bystanders). The point here is: is there deliberate intent to kill? The following graph provides the visualization of the rabbinic logic as found in Sanhedrin concerning Exod. 21:22-25. Please click on the image to enlarge.

[enter image description here


If the proto-Hebrew texts were ambiguous, then both the LXX and the Samaritan Pentateuch appear to deal with this literal ambiguity of the plurality of fetuses in the passage. In the LXX, the translation provides for two scenarios, which discriminate between the developed and the undeveloped fetus. In the Samaritan Pentateuch, the translation eliminates the ambiguity, and translates the meaning to refer to one fetus.

The Targum Onqelos and the Masoretic Text, however, preserve the ambiguity. The ambiguity, according to Jewish oral tradition, has to do with the nuances between manslaughter, which is unintentional, and murder, which is intentional. For example, in the case of murder, financial remuneration is not allowable (Numbers 35:31); however, in the case of manslaughter, when the "cool off" period in the cities of refuge has passed, the accused is not prevented from returning home (Numbers 35:25-28). In this regard, the passage at hand (Exod 21:22-25) allows for the accused to make restitution and pay civil damages (whether the mother and/or fetus died), because the accused did not commit murder, but had harmed them and/or committed manslaughter (since there never was any original intent to kill either one or both of them).

Gesenius, H.F.W. (1922). Gesenius' Hebrew Grammar (2nd Ed.). Oxford: Clarendon Press, 400.

Keil, C. F., & Delitzsch, F. (2006). Commentary on the Old Testament (Vol. 1). Peabody: Hendrickson, 409.

Neusner, Jacob (2007). The Babylonian Talmud: A Translation and Commentary (Vol. 16). Peabody: Hendrickson, 416-418.


The Complete Jewish Bible provides an interlinear translation of the Masoretic (MT) text of Exodus 21:22-23, enabling us to see the Hebrew and a direct translation:

Exodus 21:22-23 (The Complete Jewish Bible): And should men quarrel and hit a pregnant woman, and she miscarries but there is no fatality [with the woman], he shall surely be punished, when the woman's husband makes demands of him, and he shall give [restitution] according to the judges' [orders]. But if there is a fatality, you shall give a life for a life

Commenting on this passage, the great medieval Jewish scholar, Rashi, says the required restitution consists of the value of the foetuses to the husband. This comment suggests that Rashi understood the value of a foetus as depending on its stage of development, apparently taking much the same meaning out of the MT text as we can read from the LXX version.

The ESV makes no reference to the stage of miscarriage. By implication it actually seems to reverse the order of responsibility that we see in the Greek Septuagint (LXX), making it appear that a mere fine applies if the baby is born well, but a greater penalty applies if the baby is lost.

Thomas F. McDaniel analyses the MT Hebrew and LXX Greek texts and finds two textual inconsistencies, where he believes the LXX probably reflects the original Hebrew (although it remains possible it may be the Septuagint that is not a literal translation):

  1. LXX refers to the loss of a child (singular), whereas MT refers to children (in the plural). This has no theological significance.
  2. From an originally unpointed word that with vowel indicators could have formed either of two possible Hebrew words, the LXX translation gives 'fully formed' whereas MT says 'harm'.

On page 7 of his paper, 'The Septuagint Has The Correct Translation Of Exodus 21:22-23', McDaniel says in Mosaic law a woman’s fertilised egg or an imperfectly formed foetus was not considered to be a person. Only a foetus that was “fully formed” was recognised as a person. With reference to the Mosaic law, he cites Joe Sprinkle ('The Interpretation of Exodus 21:22–25 (lex talionis) and Abortion')

The penalty paid is assessed on the basis of the stage of the development of the dead fetus. The rationale for this view is that the later the stage of pregnancy, the more time has been lost to the woman, the greater the grief for the loss of a child, and the more difficult. This may have been the view of the LXX.

McDaniel concludes (page 10) the Septuagint translators understood correctly the meaning of Exod 21:22–23, which states quite clearly that a fully developed foetus was a person protected by the lex talionis, but a foetus which was not fully formed was not a person but was a property properly protected by the lex pensitationis. A lifeless, fully formed baby justifies a "life for a life."

  • 6
    How is it that the Complete Jewish Bible is reckoned to be a “direct translation of the Hebrew”, apparently in contradistinction to the ESV? The latter is considerably more “literal” here (...so that her children come out but there is no harm...) by most metrics I can imagine. (Perhaps at the expense of clarity. It seems to me that ESV has suspended an interpretive decision where none is certain.)
    – Susan
    Oct 16, 2015 at 1:19
  • I have found that it is useful in approaching this passage to keep in mind that it is part of the passage of regulations specific to the property rights of a slave owner with no regard given for the rights of the slave. So for example "a life for a life" refers to "if a slave owner loses a slave in the process he is owed a slave by the reckless quarreler". Also I agree with @Susan that the translation provided by Dick needs to be defended if it is to declared superior. No translation [of a religious text] is ever without controversy (which is why the NET Bible's approach of footnotes is grt).
    – user10231
    Oct 16, 2015 at 10:20
  • @Susan In the case of the ESV I can not know whether its translators were influenced in any way by their knowledge of the LXX or of modern cultural concerns (for example regarding abortion). I called the Jewish Bible a 'direct' translation of the Hebrew version both because it is interlinear and therefore the Hebrew is there for comparison, and because its translators are presumably fluent in (modern) Hebrew. This does not necessarily make it an 'accurate' or even 'literal' translation: something that is not for me to judge. My purpose was to go back close to the Hebrew before answering. Oct 16, 2015 at 19:11
  • 2
    Maybe stop by Biblical Hermeneutics Chat sometime if interested in discussing this. There are too many things for me to object to here. :-) But importantly: I think you’ve misunderstood McDaniel’s paper. As I understand it, he’s emending the MT vocalization of אסון according to the LXX, positing that the LXX has it correct against the MT (i.e. ʾeswon or ʾeswān [rather than MT ʾāsôn] from root סוה with a prosthetic א, an affixed ן, and ו as consonant rather than the MT vowel, an otherwise unattested form...), not consistent with it. For the purpose of the OP’s question, this is an important distinction.
    – Susan
    Oct 16, 2015 at 20:55
  • @Susan As discussed, I have edited to clarify Mcdaniel's proposed emendation. Oct 17, 2015 at 2:38

The explanation for the apparent difference between the Hebrew and Greek texts is that the latter uses an interpretive translation to express what was (and currently still is) the underlying meaning of the texts. I believe part of your confusion is coming from using an English translation which does not reflect the true meaning of the underlying Hebrew text. Here is the English translation of the Hebrew text which can found in the Artscroll Chumash, a standard Jewish version:

22 If men shall fight and they collide with a pregnant woman and she miscarries, but there will be no fatality, he shall surely be punished as the husband of the woman shall cause to be accessed against him, and he shall pay it by order of judges. 23 But if there shall be a fatality, then you shall award a life for a life, 24 an eye for an eye, a tooth for a tooth, a hand for a hand, a foot for a foot.

There are two scenarios being described here. The first one (verse 22) is where two men are fighting and they accidentally hit a woman and she miscarries, although the woman herself is not killed. In this case, the Torah prescribes that the man who accidentally hit the woman will be punished according to a court. Specifically, he will pay the monetary equivalent of what the fetus is worth. The Septuagint uses the language "imperfectly formed," which describes well a fetus which was still born and had suffered damages.

The second scenario (verse 23) is one where a man accidentally strikes a woman and she dies. This is a case of involuntary manslaughter and the Talmud has a debate about how exactly it should be handled. In any case, the language of the Septuagint which you provided us is very interesting. Here is verse 23 taken from your question:

23 But if it be perfectly formed (ἐξεικονισμένον), he shall give life for life

The Septuagint does not seem to explicitly state that a fatality occurs, but instead mentions that the fetus is still born without any damage. This makes it clear that this verse is not recurring to miscarriage, but rather to the person and life of the mother.

  • This doesn't really answer the question asked, which is a (technical) question about what caused the difference in wording between the MT and LXX, not a question about what the correct understanding of the verse is.
    – ThaddeusB
    Oct 17, 2015 at 23:38
  • @ThaddeusB I updated my answer to explicitly point out the difference between the Hebrew and the Greek translation. I wish I were a Septuagint scholar, but my expertise is limited to only Hebrew, Aramaic, and a bit of Latin. Oct 18, 2015 at 6:19

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.