God and Child Sacrifice (Ezekiel 20:25-26): The Last Pieces of the Puzzle


William Blake, “The Flight of Moloch” (1809). WikiMedia


This is now, in effect, my third post—and hopefully the last one—in a series on the multifaceted and indeed counterintuitive early Israelite attitude toward child sacrifice, and how this is reflected in the Old Testament.

I’ve tried to structure all these posts to stand alone, so you won’t have to go back and read the others. Unless, of course, you want to.

Just a recap on the last post, then: this focused on the connection between Ezekiel 20:25 and 20:26, two verses that often have pride of place in the debate on Biblical child sacrifice. (I’ll quote the verses shortly.) I noted that although the overwhelming majority of scholars assume that 20:26 explicates 20:25—which would then strongly suggest that the Israelite God is claiming that there was a time when he really did legally sanction the sacrifice of firstborn children—this assumption hasn’t really been critically tested; and in the interest of fairness I wanted to see if we could come up with a plausible alternative explanation that would play more in favor of a sort of independence of the two verses (and thus less in favor of child sacrifice).

As things often go, however, I had overlooked a minor detail—though one that happened to have pretty significant ramifications for what I had argued.

Again, the main issue of contention I addressed in the last post was (what was interpreted as) the conjunction between the two verses: the Hebrew letter vav, often just meaning “and.” I looked at about a dozen of the most respected and popular Biblical translations to see how they handled this transition between the two verses, and they basically fall into two camps here: one group—including LXX, RSV, NJB, ESV, JPS, and NASB—translates “and” between the two verses, which consequently is more amenable to the independence interpretation; but the other group—NRSV, NAB(+RE), NJPS, NET, HCSB, NIV, and NLT—didn’t interpret this letter vav as a straightforward conjunction, and thus is more amenable to the explication interpretation.

To take a translation from each group to illustrate the difference, compare the translation of ESV:

Moreover, I gave them statutes that were not good and rules by which they could not have life, and I defiled them through their very gifts in their offering up all their firstborn, that I might devastate them…

…to that of NJPS:

Moreover, I gave them laws that were not good and rules by which they could not live: When they set aside every first issue of the womb, I defiled them by their very gifts—that I might render them desolate…

In short, NJPS actually uses a colon between the verses, suggesting that the firstborn sacrifice (which NJPS more delicately translates as the children’s “setting aside”) may have been a part of the “laws that were not good,” while with ESV there’s more room to say that the Israelites being “defiled . . . through their very gifts in their offering up all their firstborn” was not necessarily directly related to these laws. To paint in extremes, it’s the difference between God having sanctioned child sacrifice or not.

After pointing all this out in my last post, I ended up in favor of sort of “splitting the difference” between the two interpretations: I suggested retaining the straightforward conjunctive aspect of the Hebrew letter vav here (“and”), as a few of the translations did; yet I marshaled evidence in support of the syntax here also including the firstborn sacrifice of v. 26 among things commanded in the “not good” laws (from v. 25), or as something that following these laws entailed

…which brings me to the detail I had overlooked.

The earliest Hebrew texts of the Old Testament actually weren’t written with the vowels included—only the consonants. Naturally, this would sometimes lead to some ambiguities in interpretation, when the knowledge of the original vowels underlying the consonantal text was lost or obscure.

When I was discussing the Hebrew text of Ezekiel 20:25-26, I was relying on the earlier vowel-less text. Although I had taken a look at the text with the later vowel pointings, too, nothing had jumped out to me—but it should have.

Now, in my last post I had only mentioned the conjunctive element of the Hebrew letter vav; but simple conjunction isn’t the only grammatical function that this letter has. Indeed, the letter vav is often joined to a verb in a way that’s often thought to be more oriented toward simply affecting the “tense” or aspect of the verb itself.¹

What’s important here is that in the actual voweled Hebrew texts, there’s a small but still unambiguous way to differentiate between these different uses of the vav; and if you zoom in close enough on the first word of Ezekiel 20:26 (which apparently I didn’t), it’s clear that it in fact doesn’t have the simple and typical conjoining vav, but the one oriented toward the verb.²

Now, this idea of this other use of vav being oriented purely toward the verb itself—as opposed to having any sort of conjoining function, or similar larger grammatical function—is only qualifiedly true. (It’s beyond the scope of this post to get into the finer linguistic details here.³)

But what this does mean—if the voweled text of Ezekiel 20:26 indeed accurately reflects the original intended—is that in this case, it’s much less likely that Ezekiel 20:25 and 20:26 are simply two independent ideas.⁴ Instead, it’s far more likely that v. 26 is an explication of the previous verse (in a way that I discussed in my previous post), or perhaps even suggests a more unambiguous causal relationship between the two.

Before moving on, it’d probably be helpful to quote Ezekiel 20:23-25 all together, from a typical translation:

23 Moreover I swore to them in the wilderness that I would scatter them among the nations and disperse them through the countries, 24 because they had not executed my ordinances, but had rejected my statutes and profaned my sabbaths, and their eyes were set on their ancestors’ idols. 25 Moreover I gave them statutes that were not good and ordinances by which they could not live. (NRSV)

One of the more well-known recent challenges to the usual interpretation of vv. 25-26 here has come by way of Kelvin Friebel, in his essay “The Decrees of Yahweh That Are ‘Not Good’: Ezekiel 20:25-26.” The centerpiece of Friebel’s proposal is v. 25, and the translation of the words themselves that are usually rendered “statutes” and “ordinances.”

He differs from all previous commentators by translating/interpreting the underlying Hebrew words here (חקים and משפטים, or in normal script huqqîm and mishpātîm)

not as statutes and ordinances that [God] had given for the people to follow—that is, the Sinai law—but, rather, the newer decrees of punishment that [God] had declared would come upon the people due to their disobedience. (30, emphasis mine)

As to what these decrees of punishment refer to then, Friebel suggests that

A clue to this can be found in the structural link between vv. 23–24 and vv. 25–26. Both begin with גם אני ‘also I’, which literarily highlights their structural interconnectedness. This repetition in v. 25 can be understood in one of two ways. It can be understood as an additional, unrelated declaration of judgment, or it can be understood as a literary device indicating a resumption of the subject of v. 23, in which [God] swears to scatter and disperse the people. As a resumptive repetition, v. 25 would refer to that specific oath of judgment. Clearly that decree of judgment in v. 23 is not to the people’s benefit—in other words, it is not good. (29-30)

Here, then, Friebel thinks that the “decrees” in v. 25 can be understood in line with God’s decision “to scatter and disperse the people”—clearly something “not good,” too.

Yet right off the bat, this argument obscures the actual parallelism. Yes, obviously we can infer that this decree of judgment in v. 23 is “not good.” But it’s not explicitly said to be so here; indeed nothing is said to be this until v. 25.

The most likely counterpart to the “not good” things of v. 25 is not decrees of judgmentor, more accurately, we might say “retributions” or punishments—but actual laws. Interestingly, although the most obvious and straightforward support for this comes from the second part of v. 25 (“…by which they would not live”) being an antithetical parallel to the laws “by which they would live” (Ezekiel 20:13, 21), there’s also an antithetical parallel in that there are other places in the OT where the laws are explicitly called good: for example, Nehemiah 9:13.⁵ 

By contrast, to my knowledge nowhere else are retributions or punishments specified as “not good”—which would seem redundant anyways.

More on this later; but Friebel makes several other arguments that he claims increase the likelihood that v. 25 simply resumes or elaborates on v. 23, and thus isn’t really “independent” (and is also specifically about decrees of retribution/punishment, as opposed to actual bad laws). 

He compares vv. 23-25 to the structure of vv. 11-12, where

in v. 12 . . . וגם . . . נתתי להם ‘and also . . . I gave to them’, with respect to the Sabbaths, is a further resumptive repetition that elaborates on and specifies the preceding “giving” of the statutes and ordinances in v. 11 (ואתן להם) (Friebel, “Decrees,” 29 n. 25)

But—as Friebel pointed out earlier—both v.23 and v. 25 begin with וגם, “and also” or “moreover.” Again, he suggested that this “highlights their structural interconnectedness.” Yet his argument isn’t quite that these two verses speak of the same thing, but that v. 25 elaborates on v. 23—a “resumption of the subject of v. 23.” I can’t help but think then that, if anything, highlighting the “structural interconnectedness” of v. 23 and 25 would only go to support an argument that both verses were an elaboration on something that came before that.

And the problem here is that this would bring us back to square one: in this case it’s just as likely that God’s having given the Israelites “laws that were not good” was simply one of the punishments (in addition to that of v. 23) for the disobedience hinted at in, say, v. 21 and elsewhere. And, in fact, we should note v. 24 gives the reason for the punishment in v. 23, which just goes to further dissociate v. 23 and 25.

If up until this point these arguments have been unpersuasive, Friebel has another argument which could pretty uncontroversially be considered his strongest or most novel one. He notes that the Hebrew words usually translated as “statutes” and “ordinances” in v. 25 actually appear in a form that’s slightly different from the one that they’re in elsewhere in Ezekiel (where it’s less controversial that they really do refer to laws):

many commentators have noted the variant of the masculine plural חקים ‘decrees’ rather than the feminine plural (חקות) used elsewhere in chap. 20 for the covenant statutes (vv. 11, 13, 16, 19, 21, 24). (28)⁶

Further, in the word translated in v. 25 as “ordinances,” משפטים (mishpātîm), this rather uniquely lacks the first-person possessive suffix that usually accompanies this word elsewhere in Ezekiel. (That is, it’s usually “my ordinances.”)

Now, speaking of the masculine forms of “statues” and “ordinances” in 20:25, Friebel recognizes that there is a similar conjunction of solely masculine forms elsewhere in Ezekiel—11:12 and 36:27—where, again, these really do suggest laws. Yet here he notes that, unlike in 20:25, in these verses these words end in first-person possessive suffixes. In sum, then, Friebel suggests that

In [20:]25, the grammatical deviation of the masculine plural coupled with the variant of no possessive pronoun attached to the nouns seem to be ways of literarily distinguishing the “decrees” and “judgments” of v. 25 from the “statutes” and “ordinances” of the divine covenantal law spoken of elsewhere in the chapter. (29)

Although you can refer to my Note 6 for more on the translation of חקים (huqqîm; again, usually translated as “decrees”), Friebel writes that

The generic משפטים ‘judgments’ also conforms with this interpretation, because the only other occurrence in Ezekiel of the plural not determined by a suffix or a construct relationship is in 5:8, where it also bears the meaning of ‘judgments’ in the sense of punishments for sin. (30)

But several things here are problematic, too. For one, noting that all other occurrences of this word in Ezekiel are “determined by a suffix or a construct relationship” should hardly have much effect: for example, “construct relationship” simply refers to the word “ordinances” being used in a grammatical construction meaning “ordinances of <person or quality>.”

And an irony here is that the two of the instances where these words are indeed “determined by a suffix or a construct relationship”—11:12 and 36:27—are in a sense closer to 20:25 than other instances, in that these are the verses where both masculine forms of the words are used together, as opposed to those other occurrences where “decrees” is used in feminine form (again 20:11, etc.).

Again, Friebel wrote that the “only other occurrence in Ezekiel of the plural [משפטים] not determined by a suffix or a construct relationship is in 5:8.” Yet another sort of coincidence here is that the use of this word in Ezekiel 5:8 is itself anomalous: this is in fact the only other place in Ezekiel that uses the word משפטים for “judgments,” whereas every instance in Ezekiel uses the form of this word which lacks the first letter,⁷ שפטים.

When we take a closer look at the exact terminology for “statutes” and “ordinances” throughout the Hebrew Bible—the terminology reflected in Ezekiel—we can begin to construct an alternate hypothesis to Friebel’s.

One interesting observation is that outside of Ezekiel, the masculine form of “decrees,” חקים, is overwhelmingly dominant in the book of Deuteronomy. (Masculine forms of the same word elsewhere in the Torah include Exodus 18:20; Leviticus 10:11; 26:46; Numbers 30:16. This occurrence in Leviticus 26:46 will be very important for my arguments to come.) Outside of Ezekiel, the direct conjunction of masculine חקים, “decrees,” and משפטים, “ordinances,” is typically confined to Deuteronomy—though see Leviticus 26:46 and Malachi 4:4, as well as Nehemiah 9:13, discussed earlier.⁸

By contrast, the use of the word משפטי—”ordinances” with the first-person possessive suffix (“my ordinances”)—is virtually confined to the book of Leviticus: Leviticus 18:4, 5, 26; 19:37; 20:22; 25:18; 26:15. The same goes for the feminine form of the word “decrees“: extremely rare elsewhere, but found in Leviticus 18:4, 5, 26; 19:19, 37; 20:8, 22; 25:18; 26:3, 15, 43 (and Genesis 26:5, the only use of either word in another book of the Torah). Similarly, “my sabbaths,” while appearing many times in the surrounding area in Ezekiel—20:12, 13, 16, 20, 21, 24; 22:8; 23:38; 44:24—only appears elsewhere in the Torah in Leviticus 19:3, 30; 26:2; Exodus 31:13. (It appears once in Isaiah, too.)

The relevance of this is this will become clear in the next few paragraphs. To summarize though, the more likely explanation for Ezekiel 20:25’s anomalous use of the feminine form of “decrees” is something like the author’s reliance on different sources (or preference for the source’s terminology); and we shouldn’t read too deeply into it in the same way that Friebel and others have.⁸ᵇ

That the author of Ezekiel relied on specific literary sources and/or conventions in the surrounding area is clear. For example, the language of Ezekiel 20:15-16’s

I swore to them in the wilderness that I would not bring them into the land that I had given them . . . 16 because they rejected my ordinances and did not observe my statutes [יען במשפטי מאסו ואת־חקותי לא־הלכו בהם], and profaned my sabbaths

bears a striking resemblance to Leviticus 26:43’s

For the land will be abandoned by them, and will make up for its sabbaths while it is made desolate without them. They, meanwhile, will be making amends for their iniquity, because they rejected my ordinances and their soul abhorred my statutes [יען וביען במשפטי מאסו ואת־חקתי געלה נפשם].

This is especially obvious as the word for “reject” here, מָאַס, is extremely rare in the Torah—used just five times total, and only twice specifically in reference to rejecting laws: in Leviticus 26:15 and, indeed, 26:43.

This is doubly relevant because just three verses later, in Leviticus 26:46, we find

These are the statutes and ordinances and laws [החקים והמשפטים והתורת] that the LORD established between himself and the people of Israel on Mount Sinai through Moses.

Here, in contrast to Leviticus 26:43’s use of feminine “statutes,” we have a switch to the masculine form—the same switch that occurs in Ezekiel 20:25, in contrast to the earlier verses; and the very one on which Friebel has based his argument. In Leviticus 26:46, however, there is no ambiguity that this really is referring to the laws (as opposed to any other kind of “judgment” or decree); there is so semantic shift as Friebel and others have proposed.

There’s actually another similar example relevant to Ezekiel 20:25-26 here, too. The word for “gift” in 20:26, מתנה—recall “I defiled them in their very gifts”—is rare in the Torah, used in legal material only in Exodus 29:38; Leviticus 23:38; Numbers 18:6-7, 29; and Deuteronomy 16:17. However, among the three feminine forms of the noun used in Numbers 18, we have a stray masculine form of the word that just so happens to appear in Numbers 18:11. I originally noticed this because Numbers 18:6-7 might actually be used to support the child sacrifice interpretation of Ezekiel 20:25-26 in another way, too. (I’ve elaborated on this further in the note.⁹)

One suggestion here, which I hinted at near the beginning of this discussion, is that the process behind Ezekiel’s shift to the masculine “decrees” in 20:25 is more or less parallel to that of the same shift in Leviticus. (I have some further comments on Leviticus 26:46 in a note.¹⁰) In sum, at various places the author of Ezekiel seems to have preferred specific language more uniquely associated with Leviticus or with Deuteronomy.¹⁰ᵇ

If the last few paragraphs have focused on the shift to the masculine form of “decrees” in Ezekiel 20:25, there remains one other obvious argument to explain the shift away from first-person possessive forms that Friebel made so much of: a text reading “my statutes which were not good” would hardly seem tolerable.¹¹

Friebel concludes his essay be reiterating his argument that “the decrees and judgments of [Ezekiel 20:25] refer back to the divine judgment of exile in v. 23” (35); but he concludes with a striking claim: “The reason for the decrees of judgment and the declaration of the people as defiled are the consequences of the people’s gifts and child sacrifices” (emphasis mine). Yet in the end he translates 20:25-26, with bracketed explanations, as follows:

And also I gave them decrees [regarding their dispersion and scattering] that were not good and judgments [of punishment] by which they would not have life. And I declared them defiled on account of their gifts [to their idols], on account of offering [by fire to idols] all [their] firstborn, so that I would cause them to be horrified, so that they would know that I am Yahweh.

Friebel translates/interprets v. 25 as “also I gave them decrees,” with the bracketed clarification “regarding their dispersion and scattering.” Indeed he’s argued that v. 25 was elaborating on v. 23, “I swore to them . . . that I would scatter them among the nations and disperse them”—a punishment which, as he noted earlier, was clearly “not to the people’s benefit”; or “in other words, it is not good” (29-30). But here in the translation of v. 25, instead of these “decrees” elaborating on or referring back to the earlier judgments, these seem more like additional sorts of things.

It may be the case that at the end of the day, it’s simply the syntax of the text itself that stands as one of greatest barriers to Friebel’s arguments, noble though the effort may be.

Further, Friebel’s suggestion that the “reason for the decrees” comes from “the consequences of the people’s gifts and child sacrifices”—that is, that v. 25 outlines the consequences of v. 26—is bizarre. Although I think I’ve convincingly showed that v. 26 can be understood as an elaboration on v. 25, to suggest that the reverse is true—not that v. 26 is a consequence of the former verse, but that it’s the cause of itstretches credulity to limit.

To be sure, Friebel has tried to emphasize the connection between v. 26 and the idolatry described earlier in the chapter. But for one, this isn’t as clear as it may seem: in the typical interpretation, the child sacrifice in view here is to God himself, not to “idols.”¹² More than this, though, it seems that he’s trying to play both sides here, simultaneously severing and connecting v. 25 and 26 as it helps various facets of his arguments.¹³

There have been several other negative assessments of Friebel’s arguments. Barber, in his recent article “Jesus as the ‘Fulfillment’ of the Law and His Teaching on Divorce in Matthew,” writes that

Friebel’s solution has linguistic and contextual problems. Linguistically it overlooks how the pairing of huqqîm and mišpātîm used elsewhere refers to Torah legislation (see, for example, Neh. 9:13; 10:29). In terms of context, as others such as Choi have observed, the suggestion by Freibel that what is in view are “judgments” and not “laws” “does violence to the plain sense of the text. Admittedly, it is theologically troubling to assert that Yahweh would give bad laws, but this is what the text says in unambiguous fashion.”¹⁴

For Barber, this is at least partially premised on the interpretation of the practices in v. 26 as elaborating on (or as a consequence of) the laws in v. 25:

the reading offered by Friebel and Gile fails to explain why 20:25-26 links what God gave to the Israelites to their defilement (“I gave them statues that were not good … and I defiled them through their very gifts.”). Hahn and Bergsma’s view, however, does just that. 

Barber refers here to Scott Hahn and John Bergsma’s article “What Laws Were ‘Not Good’? A Canonical Approach to the Theological Problem of Ezekiel 20:25-26”; and in the end, Barber indeed adopts their view. (Hahn and Bergsma’s interpretation of Ezekiel 20:25-26 is ultimately that the “bad laws” are indeed intended to refer to laws from the Torah, and moreover are about sacrifice, too—yet they actually suggest they’re ultimately about animal sacrifice, not human sacrifice. I’ve critiqued Hahn and Bergsma’s article in greater detail here.)

It’s interesting, however, to look at the more fundamental rationale behind Barber’s agreement with Hahn and Bergsma:

Given the Torah’s multiple prohibitions, it is unlikely that ancient readers would have understood the sacrifices in Ezekiel 20:26 as human beings. What decisively tips the scales against such a reading is that Ezekiel 20 describes the laws as having been given by God himself. Given that we have evidence of laws that fit the language of Ezekiel 20:26 in the Torah—that is, the offering of the firstborn animals—it seems like special pleading to insist that something else is in view.

Here I can’t help but note that the claim that “God himself” having given the laws “decisively tips the scales against such a reading,” is an egregious form of begging the question. That God himself gave these laws isn’t in dispute; the dispute is what the laws were. Further, “evidence of laws that fit the language of Ezekiel 20:26 in the Torah” applies to child sacrifice, too, as has been demonstrated.

If anything, then, to insist that the interpretation must be something (anything!) other than child sacrifice, simply because one is theologically uncomfortable with the notion—which, unless I’m off base here,¹⁵ seems to me to be exactly what was done in preemptively discounting the possibility, merely because God himself was the lawgiver—appears to be the true special pleading here.

If I can wrap up on a minor theological note here: on divine command theory, no action that God performs can really threaten the idea of his ultimate justice, no matter how much it may appear to do so. To state the obvious, this collapses the distinction between actions we would deem “good” or “bad” in light of who’s doing them—or at least when a divine authority is.

The irony in our current situation (with regard to Ezekiel 20:25-26), though, is that if the typical distinction between good and bad is normally obscured in this divine command ethics, the author of Ezekiel in a way actually brings it to light again, in unhesitatingly invoking the language of “not good” in characterizing these divine laws. (Interestingly, in Hesychius, a section of Sophocles’ Andromeda is quoted where Carthaginian child sacrifice is explicitly said to be a practice οὐ καλῶς—one that’s done “not well,” “not rightfully.”¹⁶ Compare in the Greek text of Ezekiel 20:25, οὐ καλὰ.)

Yet it seems that in Ezekiel’s revisionistic strategy here—reframing what seems to be clearly bad divine action as good divine action (which is now “bad” only for humans, in the sense that they’re really the ones who bear the brunt of the bad laws)—he’s perhaps inadvertently given us a glimpse at the true man behind the curtain underlying Biblical ethics; and the man is indeed all too human. 

That is, if (at least on divine command ethics) normal human judgment is all too prone to error in hastily casting judgment as to what’s “good” or “bad” when it comes to things divine, the author of Ezekiel seems to suggest that it can actually do it all too well: Ezekiel has rewritten God’s own history—assuming his own voice—in an attempt to help him regain the moral high ground.¹⁷

⁂       ⁂       ⁂


[1] The term “tense” is not preferred, but I use it for sake of simplicity.

[2] It reads וָאֲטַמֵּא and not וַאֲטַמֵּא, as I originally read it.

[3] The dichotomy between the two is somewhat of an oversimplification, as it’s regularly recognized that verb forms with very much have a conjunctive function. Quite a few recent studies have focused on aspects of this. See especially the work of John Cook (e.g. “The Biblical Hebrew Verbal System: A Grammaticalization Approach”; “The Semantics of Verbal Pragmatics: Clarifying the Roles of Wayyiqtol and Weqatal in Biblical Hebrew Prose”) and Alexander Andrason (e.g. “Biblical Hebrew Wayyiqtol: A Dynamic Definition”); Baden, “The Wǝyiqtol and the Volitive Sequence,” etc.

For a historical-comparative perspective, see Smith, The Origins and Development of the Waw-Consecutive. As an interesting tidbit, apparently use of wayyiqtol accounts for “30% of the 50,000 cases of waw in the Hebrew Bible.”

[4] We might also look toward Ezekiel 20:27-28 here, where we have the same grammatical form as the transition between vv. 25 and 26. This is understood in a way so that v. 28 is a clear illustration of what’s mentioned in v. 27. As NJPS did for vv. 25-26 itself, it similarly has colon + “when” in 27-28: ‘Thus said the Lord God: “By this too your fathers affronted Me and committed trespass against Me: When I brought them to the land…’

Now, to be sure, there are differences between the structure of vv. 25-26 and 27-28, too: v. 27 uses the pronoun “this” (“By this too your fathers affronted Me…”) to look ahead to the next sentence.

[5] Incidentally, Nehemiah 9:13 also attests to the direct conjunction of masculine חקים, “decrees,” and משפטים, “ordinances.”

In any case, it seems like it’s not just as an antithetical counterpart of Leviticus 18:5 that we have parallels to the idea of “not good” legislation: see Isaiah 10:1, “Woe to those who enact evil statutes, and to those who constantly record unjust decisions,” etc. (Also Psalm 94:20, and maybe even things like Iliad 16.388, too. I’m still on the hunt for a nice Near Eastern parallel, if anyone knows one.)

Naturally, Jeremiah 7:22 is another parallel here. But if Ezekiel and Jeremiah tackle the same type of problem here, it’s interesting that while Ezekiel’s approach is to reframe it, Jeremiah basically just tries to erase it. In any case, however, the language that Jeremiah uses here is in fact attested elsewhere in the Near East, and also related to actual laws: a Neo-Babylonian text condemns temple officials and others that “things that their gods did not command they establish for their gods.”

[6] Osborne, “Elements of Irony: History and Rhetoric in Ezekiel 20:1-44,” expands on this that

It is not uncommon to find a feminine noun (or even a noun which can occur in both forms like חקות) with a masculine plural ending to communicate a more poetically-charged usage, such as a collective or an abstract sense. (11)

Here Osborne cites Joüon Muraoka, A Grammar of Biblical Hebrew, 249—the relevant section of this beginning “[s]ome nouns have both m. and f. plural endings, but often only one is used frequently, the other being reserved for special or poetic usage.” However, in more of the examples that they offer than not, it’s actually the feminine form that has the broader or more abstract usage; though a couple of examples of the feminine are indeed used for specific/individual items as opposed to collectives.

[7] That is, they lack the preformative mem. See Ezekiel 5:10, 15; 11:9; 16:41; 25:11; 28:22; 28:26; 30:14, 19; Ezekiel 14:21: “my judgments.”

[8] החקים והמשפטים in Leviticus 26:46. Hahn and Bergsma write that

חקים appears . . . in Ezek 20:25 paired with משפטים, and “the expression חקים ומשפטים is found exclusively in D.” This corroborates the sense that Ezekiel refers here to Deuteronomic rather than Priestly laws. (“What Laws Were ‘Not Good’?”, 207)

The uses of this compound found in Ezekiel 20:11f. all have חקותי ומשפטי, though the reverse order משפטי וחקותי appears in 20:24. See Leviticus 18:5, 18:26; 25:18.

I suppose here is as good a place as any to discuss some of the broader source critical issues at play here. Hahn and Bergsma note that

Recent commentators have also begun to recognize the influence of D language and thought patterns in Ezekiel. As Kohn remarks, Despite his apparent affinities with P, Ezekiel was also influenced by the language and concepts of D. She singles out Ezek 20 for extended analysis as one of the most striking examples of the fusion of Priestly and Deuteronomic language and theology in the book. Jacques Pons has also devoted an essay to the literary relation of Ezek 20 to Deuteronomic and Deuteronomistic material, concluding that the presence of D language is incontestable but ironically serves to subvert Deuteronomistic theology. (“What Laws Were ‘Not Good’?”, 209)

Patton writesmore relevant specifically to Ezekiel’s views on child sacrifice itself here and their legal background—that

the laws of P and the Holiness Code are often contradicted. . . . the most glaring discrepancy is associated with the law of child sacrifice, which Ezekiel 20.25-26 suggests was still an effective law for Ezekiel’s audience, as opposed to the requirement of substitution in both P and the Holiness code. (“‘I Myself Gave Them Laws That Were Not Good’: Ezekiel 20 and the Exodus Traditions,” 82)

and suggests that

Either Ezekiel, especially in ch. 18, knew one or more law codes that had already combined P, the Holiness Code, the Covenant Code and D, a problematic solution in light of Ezekiel’s attitude toward child sacrifice, or Ezekiel bases its judgements on pre-exilic law codes similar to the extant laws, but not identical with them. (84)

Halpern, from a more traditional Documentary perspective, argues

pre-seventh century BCE sources presuppose infant sacrifice, which was of course practiced in Jerusalem until Josiah’s day . . . it would appear that Jeremiah and Ezekiel, in an age of the rejection of tradition, embrace the rejection of JE, probably already combined and promulgated in the early seventh century, in favor of the traditions represented by Deuteronomy, the Deuteronomistic History, and P. Ezekiel appears to refer to reports in the Deuteronomistic History, especially Judges, in connection with his assessment of Israelite disobedience after the entry into the land. This is not, however, his main point, but merely supporting evidence for his allegations. (“The False Torah of Jeremiah 8 in the Context of Seventh Century BCE Pseudepigraphy,” 340)

Finally, after discussing Ezekiel 20:25-26 and Micah 6:6-8, Van Seters writes

This brings us to a consideration of the parallels within the Pentateuch. The closest example is the P text in Exod 13:2 . . . Taken by itself, the statement is unqualified and, as Fishbane rightly points out, “the existence of the unqualified rule . . . would be hard to explain if the qualifications were original.” This observation supports the conclusions to be drawn from Jeremiah and Ezekiel that the practice of child sacrifice existed before such qualifications were made, so that all such statements of redemption by animal sacrifice or compensation (Exod 13:13-16; 34:20; Num 3:11-13; 8:16-18; 18:15-16) must be later. Nevertheless, that does not mean that this particular law in Exod 13:2, which is without qualification in the P source, is earlier than those in Exod 13:11-13. (A Law Book for the Disapora, 148-49)

[8b] I’m also thinking of Scott Hahn and John Bergsma’s article “What Laws Were ‘Not Good’? A Canonical Approach to the Theological Problem of Ezekiel 20:25-26” herewhich is certainly on the right track in terms of recognizing the masculine form as typical Deuteronomistic language, but then uncritically uses this as the basis of a far more specific argument. As I mention later, I’ve critiqued their article in more detail in an earlier post.

[9] Here in Numbers, the Levites are themselves made a “gift”: “It is I who now take your brother Levites from among the Israelites; they are now yours as a gift, dedicated to the LORD, to perform the service of the tent of meeting” (18:6). I’ve discussed at length the Levites’ consecration to this service in the context of “substitution” rites for human sacrifice. (See also Finsterbusch, “The First-Born between Sacrifice and Redemption in the Hebrew Bible.”)

Most importantly, the language of Numbers 18:6-7 recalls the earlier “dedication” of the Levites from the 3rd chapter of Numbers, which is transparently sacrificial:

9 You shall give the Levites to Aaron and his descendants; they are unreservedly given to him from among the Israelites. . . . 11 Then the LORD spoke to Moses, saying: 12 I hereby accept the Levites from among the Israelites as substitutes for all the firstborn that open the womb among the Israelites. The Levites shall be mine, 13 for all the firstborn are mine; when I killed all the firstborn in the land of Egypt, I consecrated for my own all the firstborn in Israel, both human and animal; they shall be mine. I am the LORD.

(Also, it’s worth noting that Numbers 18:14-18 explicitly legislates for firstborn sacrifice, though here clearly including redemption clauses.)

While this word “gifts” is used on several other more general occasions in Ezekiel (46:16-17; 20:39?), the best context to understand it is the more immediate one: e.g. its use just a few verses later, in 20:31. Similarly, as for Ezekiel 20:26’s collocation of the verb form “give” and עָבַר, “pass over,” this is found in Leviticus 18:21, “You shall not give any of your offspring to sacrifice them to Molech, and so profane the name of your God: I am the LORD” (NRSV). See also 2 Kings 23:10.

[10] I relate some of the arguments of Nihan, “The Priestly Covenant, Its Reinterpretations, and the Composition of ‘P'”, here.

Speaking of the Holiness Code (a.k.a. H: chapters 17-26 of Leviticus), “H is not only post-P, but also post-D, i.e. it reinterprets and revises not only the Priestly document . . . but other, not-Priestly documents, including especially the Deuteronomic code (D)” (113-14). Some of the vocabulary of Leviticus 26 “can hardly refer to P’s covenant and are obvious references to the non-Priestly covenant.” H “was conceived as supplement not only to P, but also to the non-Priestly narrative in Exodus.” Leviticus 26:46 actually “forms the conclusion not only to the Holiness Code (Leviticus 17–26), nor even to the book of Leviticus, but actually to the entire Sinaitic account starting with Exodus 19ff.”:

This is also suggested by the terms used in 26,46 to refer to the laws revealed to Moses, namely, משפטים ,חקים, and תורת: while the combination of חקה (fem.) and משפט is distinctive of H, the author of Leviticus 17-26 otherwise never uses the masculine, חק, to refer to Yahweh’s laws. Such usage is exclusively found in the non-Priestly portions of the Pentateuch, especially (but not exclusively) in Deuteronomy. Thus, the terminology used by the author of H in 26,46 betrays his willingness to include in that colophon not only the various laws revealed at Mount Sinai according to P and H, but also also other, non-Priestly traditions. (115)

A footnote at the end of “the combination of חקה (fem.) and משפט is distinctive of H” reads

In the Pentateuch, that combination is almost exclusively found in Leviticus 17-26 or in passages that are usually ascribed to the “Holiness School”: Lev 18,4. 5.26; 19,37; 20,22; 25,18; 26,15.43; further, Num 9,3.14; 27,11; 35,29. Otherwise, it occurs in the Pentateuch only in Deut 8,11; 11,1; 30,16. It is also frequently found in Ezekiel (5,6.7; 11,20; 18,; 20,; 37,24; 44,24), whose phraseology is often closely allied to that of H. (115, n. 93)

[10b] Refer to my Note 8 here.

[11] Besides being grammatically awkwardand despite the fact that there’s no getting around the suggestion that God would indeed be giving bad laws herewouldn’t the author be interested in at least some dissociation of the two? For some reason I think of Philo of Alexandria’s comments about Dike as a sort of “henchman” for doing God’s dirty work, dissociating him from it.

[12] Patton writes that “Ezek. 20.25-26 plainly states that the problem with the law of child sacrifice was not that it was misinterpreted—that is, that it was not read in the context of substitution or redemption of the sacrificial victim—or that it was a non-Yahwistic ritual” (“‘I Myself Gave Them Laws That Were Not Good'”, 79).

[13] Instead of v. 25 being a consequence of v. 26, hasn’t Friebel’s argument all along been that v. 25 was a throwback to, or “resumptive repetition” of the earlier verse(s)? I find it very hard to reconcile Friebel’s conclusion with his earlier arguments:

The grammatical sequence should not be read as “I gave them decrees . . . so as to defile them” or “I gave them decrees . . . and thereby defiled them,” but as “I gave them decrees . . . and, in addition, I defiled them.” (31, emphasis mine)

[14] “Jesus as the ‘Fulfillment’ of the Law.” Barber quotes Choi, Traditions at Odds, here. The larger context of Choi’s quotation reads

As with Jer 7:21-26, for most interpreters such a negative stance towards scripture is inconceivable. Instead, they argue that though Ezekiel refers to various pentateuchal laws on the firstborn, what is “not good” is the interpretation of the laws. . . . While this emphasis on legal interpretation protects the authority of the Pentateuch, such an approach does violence to the plain sense of the text. Admittedly, it is theologically troubling to assert that Yahweh would give bad laws, but this is what the text says in unambiguous fashion (Traditions at Odds, 203)

However, as for the relationship to the actual laws of the Torah here, Choi puzzlingly suggests that

the reference to the sacrifice of the firstborn may not point to any pentateuchal material. I would argue, on the contrary, that the similarity in this respect is coincidental. The author of Ezek 20 explicates the content of the “not good” law by referring to an abhorrent cultural practice, and, in doing so, provides a vivid illustration of the travesties that Yahweh brings upon Israel as punishment for its disobedience. (205)

One wonders how it’s possible for the author of Ezekiel to ‘[explicate] the content of the “not good” law’ by referring specifically to an “abhorrent cultural practice”—again, child sacrifice—and yet at the same time for this law to not be referring to any of the pentateuchal laws, if these are exactly what the “not-good” laws are (as Choi suggested earlier, in criticizing attempts to “protect the authority of the Pentateuch” by interpreting the “not-good” laws as not really pentateuchal).

[15] Which is certainly a possibility, because Hahn and Bergsma’s interpretation certainly still makes God the giver of bad laws.

Thus far I really have tried to refrain from assigning motive to people here; though for a case study in how far academic apologetics seems to be willing to go in order to avoid the obvious here, see Brian Neil Peterson’s Ezekiel in Context, which follows KJV in impossibly translating 20:25 as “gave them over to (their own) bad laws” (187). Besides the fact that there’s no warrant for interpreting these as their own laws, the text clearly would have read something like נתנם לחקים if “gave them over to” had been intended.

[16] Cf. Plato, Minos

ἐπεὶ αὐτίκα ἡμῖν μὲν οὐ νόμος ἐστὶν ἀνθρώπους θύειν ἀλλ᾿ ἀνόσιον, Καρχηδόνιοι δὲ θύουσιν ὡς ὅσιον ὂν καὶ νόμιμον αὐτοῖς, καὶ ταῦτα ἔνιοι αὐτῶν καὶ τοὺς αὑτῶν υἱεῖς τῷ Κρόνῳ…

With us, for instance, human sacrifice is not legal, but unholy, whereas the Carthaginians perform it as a thing they account holy and legal, and that too when some of them sacrifice even their own sons to Cronos…

[17] See also Note 5 here on Jeremiah, dealing with a not dissimilar issue.