The Book of the Wars of the LORD

What it was, what survives, whether it can be recovered, and what it may have contained

The Book of the Wars of the LORD—Hebrew sēfer milḥămōt YHWH (ספר מלחמת/מלחמות יהוה)—is one of the most intriguing lost works mentioned inside the Hebrew Bible itself. Unlike texts such as 1 Enoch or Jubilees, which disappeared from some traditions but survived elsewhere, no manuscript presently known can be securely identified as this book.

What we possess is essentially one explicit citation in Numbers 21:14–15, embedded in a chapter containing several pieces of archaic-looking poetry associated with Israel's movement through Transjordan. That tiny citation nevertheless opens a surprisingly large research problem involving lost Israelite literature, ancient war poetry, the composition of the Pentateuch, geography east of the Jordan, and the development of Israelite ideas about YHWH as divine warrior.

The most defensible starting conclusion is therefore paradoxical:

We probably possess a few words from the Book of the Wars of YHWH, while possessing no identifiable copy of the book itself.

Everything beyond that must be graded carefully.

Overview

What exactly is the Book of the Wars of the LORD?

The sole unambiguous biblical reference occurs during the description of Israel's approach to Moab and the Arnon River:

“Therefore it is said in the Book of the Wars of the LORD…”

The Hebrew then continues with the extremely difficult phrase:

ʾet wāhēb bəsûpâ wəʾet hannəḥālîm ʾarnôn…

followed by a reference to the slopes or tributary valleys associated with the Arnon and the territory of Ar/Moab. (MGKeter)

This creates our first documented fact:

Numbers explicitly presents material as something “said in” a written work called the Book of the Wars of YHWH.

The Hebrew word sēfer need not mean a modern codex-style “book”; in the ancient setting it can mean a written document, scroll, record, or collection. Consequently, we should initially think of something like:

a written anthology, record, scroll, or compilation associated with wars attributed to YHWH.

We cannot yet determine its physical size or literary form.

Why it matters

The reference matters disproportionately because it appears to provide direct evidence that biblical authors or editors sometimes used earlier written sources that no longer survive independently.

The Hebrew Bible contains several comparable references, most famously the Book of Jashar, quoted in Joshua 10:13 and 2 Samuel 1:18. The latter identifies David's lament over Saul and Jonathan with that collection. (Bible Gateway)

Numbers 21 is especially interesting because several passages around the Book of Wars citation look like independent poetic material.

Within roughly twenty verses we encounter:

• the Book of the Wars of YHWH citation, Numbers 21:14–15;
• the Song of the Well, Numbers 21:17–18;
• another itinerary fragment;
• and the song introduced by “therefore the poets/ballad singers say” in Numbers 21:27, concerning Sihon and Heshbon. (Bible Gateway)

This raises the possibility that Numbers 21 preserves fragments from an older corpus of songs or traditions.

That is probable enough to investigate seriously.

It is not, however, proof that all these poems belonged to the Book of Wars.

The boundaries of the problem

Several things commonly conflated under this topic must be kept separate.

The biblical Book of the Wars of YHWH is one thing.

The Qumran War Scroll (1QM) is another.

The Book of Jashar is another.

The unnamed written memorial of the Amalekite battle in Exodus 17:14 is another.

And much later books bearing similar titles are unrelated unless evidence demonstrates otherwise.

This separation becomes crucial because internet discussions frequently merge them.

Core Element 1: The actual surviving quotation

Numbers 21 describes Israel moving through the territory east of the Dead Sea:

Oboth → Iye-Abarim → Zered → Arnon → Beer → Mattanah → Nahaliel → Bamoth → the vicinity of Pisgah.

The narrator explains that:

“Arnon is the boundary of Moab, between Moab and the Amorites.”

Then comes the Book of Wars citation.

A fairly literal representation of the surviving Hebrew is something like:

“…Waheb in Suphah, and the wadis of Arnon, and the slope of the wadis that extends toward the dwelling of Ar and leans upon the border of Moab.”

But even this reconstruction conceals severe textual difficulty. (Bible Gateway)

The problem of Waheb in Suphah

The phrase ʾet wāhēb bəsûpâ is obscure.

Modern translations disagree substantially.

Some render Waheb and Suphah as place names.

Older translations sometimes connect Suphah with the Sea of Reeds/Red Sea, producing something like “what he did in the Red Sea.”

The ancient textual witnesses themselves vary.

The NET textual notes observe that the Samaritan Pentateuch, Greek tradition, Vulgate-derived interpretations, and modern reconstructions differ considerably here. (Bible Gateway)

That tells us something important.

Strong inference

The quotation was already linguistically or geographically obscure at a very early stage of transmission.

That may indicate antiquity.

But caution is required:

difficult Hebrew does not automatically equal extremely ancient Hebrew.

Corruption, unfamiliar geography, dialect, poetic compression, or scribal misunderstanding can produce the same effect.

A fascinating modern proposal

Richard C. Steiner offered a substantial new reading in a 2020 study devoted specifically to Numbers 21:14–20.

Instead of treating Hebrew ʾet as the ordinary object marker, Steiner proposes reading it as an archaic imperative related to “come,” resulting approximately in:

“Come to Waheb…”

This creates a striking parallel with Numbers 21:27:

“Come to Heshbon…”

Steiner argues that this produces a more coherent relationship between the Book of Wars fragment and the geographical/hydrological material surrounding it. (Lockwood Online Journals)

This is an excellent example of our evidence categories.

Documented fact: the Hebrew reads the transmitted consonants.

Scholarly proposal: Steiner provides a philological explanation.

Not established fact: that reconstruction has not somehow recovered the original wording beyond dispute.

Nevertheless, it deserves serious attention in later investigation.

Core Element 2: What does “Wars of YHWH” mean?

The title itself is revealing.

It does not straightforwardly mean:

“Israel's wars.”

It says:

“Wars of YHWH.”

That wording fits a major biblical theological concept: YHWH as warrior.

Examples appear throughout early Hebrew poetic traditions.

In Exodus 15, after Egypt's defeat:

YHWH is described explicitly as a man of war/warrior.

Judges 5 portrays divine intervention in Israel's battle.

Joshua repeatedly attributes military victory to YHWH.

Later poetry depicts cosmic and terrestrial forces fighting on God's behalf.

Consequently, “Wars of YHWH” could mean wars in which:

YHWH fights for Israel,

rather than merely military annals recording Israelite campaigns.

That distinction makes the hypothesis that the work contained victory songs or heroic poetry especially plausible.

Core Element 3: Was it a historical chronicle or poetry anthology?

There are at least two main possibilities.

Historical record

The title could refer to something analogous to annals:

campaigns, locations, enemies, victories, territorial changes.

The fragment's dense geographical language could support this.

Poetic anthology

Many scholars have instead suspected a collection of battle poems or victory songs.

The surrounding Numbers material strongly encourages this reading.

Modern discussions by scholars including Jacob Milgrom, Philip Budd, and others have treated it as plausibly containing poetic accounts of divine victories; the surviving quotation itself is frequently classified as archaic or traditional poetic material. (TheTorah.com)

These possibilities are not mutually exclusive.

Ancient historical memory was often preserved through poetry.

A collection might easily have resembled:

battle narrative + song + geography + heroic tradition.

Historical Background

The Transjordan setting

The immediate context concerns the land east of the Jordan.

Three political-geographical entities matter particularly:

Moab

Amorite territory

Sihon's kingdom

Numbers presents the Arnon River as a significant boundary.

The Arnon is generally identified with modern Wadi Mujib in Jordan, flowing westward into the Dead Sea.

Numbers subsequently describes Sihon as having taken territory from Moab.

That becomes significant because Numbers 21:27 introduces another old poem:

“Come to Heshbon…”

followed by imagery of fire coming from Heshbon and destruction overtaking Moab.

The poem appears to celebrate Sihon's conquest of Moab before Israel subsequently defeats Sihon.

Thus Numbers may preserve a remarkable chain of remembered military history:

Moab → conquered partly by Sihon → Sihon defeated by Israel.

Who were the “poets” or mōšəlîm?

Numbers 21:27 introduces its second poetic citation with:

“Therefore the mōšəlîm say…”

Translations include:

• poets
• ballad singers
• proverb makers
• those who speak in parables

The root is associated with māšāl, which can mean proverb, saying, taunt, or poetic discourse.

This matters because medieval Jewish interpretation connected these poet-historians with the Book of Wars.

Nachmanides/Ramban, for example, envisioned learned writers composing accounts of great wars and celebrating extraordinary victories through poetic expressions. He interpreted the Sihon material in that context. (Wikisource)

That cannot tell us directly what existed centuries earlier, but it demonstrates that careful premodern readers independently recognized the literary relationship between war records and poetry in this chapter.

Core Element 4: Could Exodus 17:14 refer to the same book?

After Israel defeats Amalek, Exodus says:

“Write this as a memorial in a book/scroll and recite it in the hearing of Joshua…”

(Bible Gateway)

This is highly suggestive.

It produces a possible reconstruction:

Hypothesis

Moses—or an early Israelite scribal tradition—began preserving accounts of battles involving YHWH, eventually producing the Book of the Wars of YHWH.

Medieval commentator Ibn Ezra entertained a connection of this kind.

He apparently regarded the Book of Wars as a separate work and even entertained the possibility that it contained older material reaching back much further. Later commentators repeated versions of this idea. (Wikisource)

But the evidence does not establish that Exodus 17:14 and Numbers 21:14 refer to the same document.

The phrase in Exodus simply says to record the Amalekite victory “in a book/scroll.”

Therefore:

Possible connection: reasonable.

Demonstrated identification: no.

Core Element 5: Was Moses the author?

This is another point where tradition and evidence must be separated.

There is no surviving attribution in Numbers 21:14 saying:

“Moses wrote the Book of the Wars of YHWH.”

Traditional models may reasonably connect it with Mosaic-era material because its subject matter concerns Israel's wilderness-period conflicts.

But historically several models remain possible.

The work could have originated:

during the purported events;

shortly afterward as oral songs were collected;

during the early monarchy;

or

as a later compilation incorporating older poetry.

The only secure chronological statement is:

Whatever Numbers' author/editor understood this source to be, it was treated as an already existing authority when this passage was formulated.

Dating that editorial stage is itself part of the much larger and highly disputed question of Pentateuchal composition.

We should therefore avoid using the citation to “prove” either early Mosaic authorship or a late Pentateuchal date.

It does neither by itself.

Core Element 6: How large was the quotation?

This is surprisingly uncertain.

The most conservative reading identifies only Numbers 21:14–15 as the quotation.

Others have proposed that material through verse 16 belongs to it.

Still broader reconstructions associate the Song of the Well in verses 17–18 with the same collection.

And some extend the literary relationship all the way to the Sihon-Heshbon poem of verses 27–30.

Traditional and modern interpreters have disagreed about these boundaries. (Hamichlol)

Consequently we might have:

Minimum surviving fragment

Numbers 21:14–15.

Possible related material

Numbers 21:17–18.

Potentially related war-poetry tradition

Numbers 21:27–30.

But only the first passage is explicitly attributed to the Book of Wars.

Major Interpretation 1: A lost Israelite war-song anthology

This is probably the most straightforward modern reconstruction.

The work contained songs commemorating battles in which YHWH's intervention was celebrated.

It would resemble the Book of Jashar, which apparently preserved poetic material including David's lament and Joshua's famous sun-and-moon passage. (Bible Gateway)

Under this model, ancient Israel possessed multiple poetic anthologies.

One could imagine something roughly analogous to:

Book of the Wars of YHWH → divine victories and battles

Book of Jashar → heroic/national poetry

These categories remain reconstructed, not known catalog descriptions.

Major Interpretation 2: A historical military chronicle

Another possibility is that it was essentially an early historical record.

Ancient Near Eastern states routinely recorded:

campaigns, victories, kings, territory, tribute and conquered cities.

The geographical nature of the surviving fragment could fit such a source.

But unlike Assyrian royal annals or Egyptian victory inscriptions, we possess no material artifact bearing this title.

Thus the comparison is generic rather than evidentially direct.

Major Interpretation 3: A mixed prose-and-poetry national archive

This may ultimately be more realistic than forcing a choice between “chronicle” and “songbook.”

Ancient textual collections frequently accumulated material.

The Book of Jashar itself appears, from its biblical citations, to have contained poetry from different historical settings.

If the Book of Wars developed similarly, it could have accumulated songs surrounding:

Amalek,

Transjordan,

Sihon,

Og,

Canaanite conflicts,

and perhaps other victories.

But except for the Numbers citation, these proposed contents are speculative.

Major Interpretation 4: It wasn't actually a separate book

Some Jewish interpretive traditions took “Book of the Wars of YHWH” not as the title of an independent lost composition but effectively as a designation for material within the Torah itself.

Later Palestinian/Jerusalem Targum traditions move in this direction, interpreting the phrase in connection with the Torah containing accounts of God's battles and mighty deeds. (Hamichlol)

This interpretation should not simply be dismissed.

However, the natural grammatical construction:

“therefore it is said in the book of…”

strongly resembles normal source citation.

Consequently, many modern scholars regard a separate source as the simpler explanation.

Major Interpretation 5: The Book of Wars and the Book of Jashar were the same work

This idea has periodically appeared.

It is attractive because both appear to involve old poetic material connected with warfare.

But there is no ancient text explicitly identifying them.

Moreover, the titles differ.

The safest classification is therefore:

possible historical relationship, but unsupported identity.

Could the Dead Sea Scrolls have preserved it?

This is one of the most important questions.

The answer currently is:

No manuscript discovered at Qumran has been accepted by scholarship as the Book of the Wars of YHWH.

The temptation arises because Qumran produced the famous War Scroll.

But these are almost certainly different works.

The Qumran text commonly called the War Scroll, catalogued as 1QM, dates paleographically to roughly the late first century BCE or early first century CE. It describes an apocalyptic final war between the Sons of Light and Sons of Darkness. (DSS Collections)

Its surviving text belongs to the religious world of Second Temple Judaism and contains detailed rules for eschatological warfare.

That is very different from the Transjordan geography appearing in Numbers.

Even the conventional scholarly designation “War Scroll” is modern; research on its ancient title shows that only part of its war-related heading survives. (DOI)

There are modern claims on the internet that the Qumran War Scroll is the lost Book of the Wars of YHWH. One such recent website explicitly makes that identification. (Book of the Wars of the Lord)

But this should be classified as:

Speculation / fringe identification

It is not supported by mainstream Qumran scholarship.

Chronology, genre, contents, terminology and historical setting give us no persuasive reason to identify the two works.

Could fragments nevertheless be sitting unidentified among the Dead Sea Scrolls?

This question is more interesting.

Approximately 950 different scrolls or scroll groups were recovered from the Qumran caves, many extremely fragmentary. (DSS Collections)

In principle, therefore, a tiny unidentified fragment could preserve a work whose title has disappeared.

But:

there is presently no positive evidence that this happened.

So the correct statement is not:

“The Book of Wars definitely wasn't at Qumran.”

It is:

No known Qumran manuscript has been identified convincingly as it.

That distinction preserves the evidence properly.

Can we find the Book today?

At present:

No known complete or partial manuscript outside Numbers can be authenticated as the Book of the Wars of YHWH.

We possess no recognized copy in:

the Dead Sea Scrolls,

the Cairo Geniza,

the Septuagint manuscript tradition,

the Samaritan textual tradition,

rabbinic manuscript collections,

Christian manuscript libraries,

or known archaeological archives.

That does not prove that no fragment will ever be discovered.

Ancient texts have resurfaced unexpectedly before.

But any modern book claiming simply to be the Book of the Wars of the LORD must establish a manuscript provenance, paleographic date, linguistic antiquity and historical transmission.

No current claimant meets that standard.

What happened to it?

We simply do not know.

But its disappearance would not be surprising.

Most ancient literature is lost.

Even the Hebrew Bible names numerous works that have not survived as independent texts.

The Book of Jashar is one.

Biblical Kings and Chronicles also refer repeatedly to royal annals and prophetic records no longer extant.

The Book of Jashar itself appears to have been a known written collection containing national poetry, yet no authenticated ancient manuscript survives. (Bible Gateway)

Thus the Book of Wars belongs within a wider phenomenon:

the lost literary library behind the Bible.

That may ultimately be the most important research context.

What does the surviving fragment actually tell us?

Very little directly—but what it tells us is significant.

First: ancient biblical writers knew written source collections.

The citation formula strongly indicates reliance upon pre-existing material.

Second: warfare could be conceptualized as YHWH's warfare.

This reinforces the divine-warrior theology found elsewhere in early biblical poetry.

Third: geographical memory mattered.

The fragment appears concerned with specific places around the Arnon and Moabite frontier.

Fourth: poetic accounts may have preserved historical memories.

Numbers 21 contains several poetic insertions associated with specific military and geographical episodes.

Fifth: some biblical literature existed outside what later became the biblical canon.

This is perhaps the most important implication.

The existence of a source referenced approvingly in Scripture does not mean it was itself regarded as canonical scripture.

Ancient Israel clearly possessed more literature than the books that survived inside the Tanakh.

How old might the Book have been?

We cannot responsibly assign a precise date.

A very early date is possible if the underlying poems originated near the events they portray.

Some characteristics of Numbers 21 have traditionally been regarded as archaic.

But dating Hebrew poetry purely by linguistic style has become methodologically controversial.

The strongest statement is therefore:

The material may preserve very old traditions, perhaps considerably older than the final literary form of Numbers, but neither the date of the quoted poem nor the date of the collection can presently be established securely.

The book itself could also have been compiled later from earlier songs.

For example:

oral song: early

↓

written poem: later

↓

anthology: later still

↓

citation incorporated into Numbers

Such layered transmission is entirely plausible.

A potentially important analogy: Homeric-style preservation

Ramban made an observation surprisingly useful for modern thinking about the problem.

He imagined ancient learned people recording major battles and composing poetic accounts of extraordinary victories. (Wikisource)

We should not turn that medieval interpretation into historical proof.

But the basic cultural model is reasonable.

Across antiquity, battles were remembered through:

oral poetry,

victory hymns,

royal inscriptions,

heroic narratives,

and eventually written collections.

The Book of Wars may represent an Israelite version of that broader phenomenon.

The biggest unresolved mystery: whose wars were originally being celebrated?

Here Numbers 21 becomes particularly intriguing.

The nearby Sihon poem celebrates the Amorite defeat of Moab:

“A fire went out from Heshbon…”

If this material belongs to the same poetic environment as the Book of Wars fragment, then the anthology may preserve traditions that were not originally straightforward Israelite victory songs.

That opens several possibilities.

Perhaps Israelite writers appropriated older Amorite or Moabite poetry.

Perhaps the passage preserves an old regional battle song that was reinterpreted as evidence explaining Israel's later territorial situation.

Perhaps the Book of Wars contained wars involving foreign peoples because all major geopolitical victories were ultimately attributed to YHWH.

Ramban actually moved toward something like this interpretation, connecting the source with Sihon's earlier conquest of Moab. (Wikisource)

This deserves much deeper investigation.

The theological dimension: YHWH as warrior

The title also intersects with one of the oldest and most powerful patterns in biblical theology.

YHWH does not merely authorize battles.

In many poetic traditions, YHWH himself fights.

Storms, waters, celestial forces and natural phenomena become divine weapons.

This appears prominently in:

Exodus 15,

Judges 5,

Deuteronomy 33,

Habakkuk 3,

Psalms 18, 68 and 74,

Isaiah,

and related biblical traditions.

That raises a fascinating possibility:

perhaps Wars of YHWH originally referred to a broad category of divine victory traditions, not simply military campaign records.

If so, the lost book could belong to the same literary world that eventually produced biblical descriptions of:

YHWH defeating Egypt,

YHWH defeating Israel's terrestrial enemies,

and possibly even YHWH defeating cosmic waters or sea monsters.

That connection remains a research hypothesis—not a demonstrated reconstruction of the book—but it is potentially extremely important.

What we can say with different confidence levels

Open Questions

Several questions now stand out as genuinely important rather than merely curious.

How much of Numbers 21 derives from the lost collection? The boundaries of the quotation may completely change our understanding of the book's contents.

What exactly does ʾet wāhēb bəsûpâ mean? Solving those few words might reveal whether the fragment describes places, movement, conquest, divine action, or something else.

Who composed the underlying poetry? Israelite, Amorite and Moabite cultural materials may potentially intersect here.

Was the book primarily historical or liturgical? A war archive and a collection celebrating divine victories are not quite the same thing.

How early is the poetry? Linguistic, geographical and historical arguments must be tested separately rather than assuming “archaic sounding” means early second millennium or Mosaic.

What is its relationship to the Book of Jashar? The two may preserve evidence for a forgotten Israelite tradition of national song collections.

Could archaeological evidence corroborate the geography? Arnon, Ar, Heshbon, Dibon and the Moabite-Amorite frontier provide potentially testable historical anchors.

How does the Moabite evidence fit? The later Mesha Stele, although centuries removed from the events described in Numbers, independently demonstrates a Moabite political world in which Chemosh, warfare, territorial possession and divine intervention are tightly connected. Comparing these traditions could be highly illuminating.

And perhaps most importantly:

What exactly did ancient Israelites mean by “the wars of YHWH”?

A military category?

A theological category?

A literary genre?

Or all three?

Initial Research Map

The strongest next-stage investigation has several interconnected branches.

1. Reconstruct Numbers 21:10–30 line by line. Examine Hebrew, Septuagint, Samaritan Pentateuch, Targums, textual variants, poetic structure and possible quotation boundaries.
2. Investigate the historical geography of the Arnon/Heshbon region. Identify Waheb, Suphah, Ar, Heshbon, Dibon and the Moabite-Amorite frontier against archaeology and ancient inscriptions.
3. Investigate Israel's lost literary sources. Put the Book of Wars beside the Book of Jashar, royal annals, prophetic chronicles and other books referenced but not preserved.
4. Investigate ancient Near Eastern war-song traditions. Compare Egyptian victory hymns, Mesopotamian royal war texts, Canaanite/Ugaritic poetry and Moabite inscriptions with Numbers 21.
5. Investigate YHWH as divine warrior. Trace the motif from Exodus 15 and Judges 5 through Psalms, prophetic literature and later apocalyptic warfare.
6. Test the proposed relationship to Exodus 17:14. Determine whether linguistic and compositional evidence supports the idea of an early military memorial collection.
7. Investigate the Qumran possibility rigorously. Compare the Book of Wars reference with 1QM and related Qumran Milḥamah texts to establish exactly why identity is or is not plausible.
8. Trace Jewish and Christian reception. Ibn Ezra, Ramban, Targums, medieval commentators, early Christian interpreters and modern scholarship each reconstructed the lost text differently.

Where the investigation currently stands

At this first stage, the most defensible reconstruction is modest but fascinating.

There really was—or at minimum Numbers genuinely presents itself as citing—a written work called the Book of the Wars of YHWH. The citation is not a modern legend. It is embedded directly in the Hebrew text.

The work itself has not been recovered.

The strongest hypothesis is that it was some type of early collection of battle traditions, probably containing poetic material celebrating or recording wars understood as YHWH's wars.

Numbers may preserve one tiny fragment from it.

Possibly more.

And that is where the investigation becomes particularly valuable: Numbers 21 may function as a small surviving window into an entire layer of ancient Israelite literature that otherwise disappeared.

The highest-value next branch, in my assessment, would be to take Numbers 21:10–30 apart almost archaeologically—Hebrew word by Hebrew word, source by source—and determine exactly how much lost poetry may be embedded there. That would give us the strongest foundation before attempting any larger reconstruction of what the Book of the Wars of YHWH once contained.

The Book of the Wars of YHWH: Deeper Investigation

A deeper pass changes the center of gravity of the investigation. The main question is no longer simply “What lost book is Numbers quoting?” It becomes:

What exactly is the literary object behind Numbers 21:14, how much of Numbers 21 belongs to it, how old are its component traditions, and what historical function did the compiler of Numbers assign to them?

That shift matters because several things that initially looked straightforward become much less certain under closer scrutiny.

1. First correction: the title may tell us less than we assumed

In the first investigation I treated “Book of the Wars of YHWH” as strong evidence that the work was specifically a collection of war accounts or victory songs.

That remains plausible, but it needs to be downgraded.

Edward Greenstein points out an important ancient literary convention: ancient Near Eastern works were frequently identified not by descriptive titles summarizing their contents, but by an opening word, phrase, or especially memorable line. Babylonian Enūma Eliš, for example, is named from its opening words, rather than being anciently titled something equivalent to “The Babylonian Creation Epic.” Greenstein therefore cautions that “Wars of YHWH” could conceivably have been the opening or prominent wording of one poem in a broader collection rather than a catalogue description of everything in the scroll. (TheTorah.com)

That creates two distinct hypotheses:

Traditional reconstruction: ספר מלחמת יהוה = a book whose subject was YHWH's wars.

Alternative reconstruction: מלחמת יהוה may have functioned more like the remembered name of a composition or opening poem inside a miscellaneous anthology.

The second possibility helps explain something otherwise awkward: if the Song of the Well also came from the collection, it contains no warfare at all. Greenstein accordingly proposes that the collection may have been a broader compendium of poems. (TheTorah.com)

This is not settled.

But it materially changes our reconstruction.

Revised confidence

There was a written source identified as sēfer milḥămōt YHWH. High confidence.

It contained exclusively war literature. Low-to-moderate confidence.

It contained poetry. Moderately strong, because the identifiable extract is poetic or poetically compressed.

It was an anthology. Plausible, but reconstructed.

2. The citation formula itself deserves much more attention

The Hebrew introduction is:

עַל־כֵּן יֵאָמַר בְּסֵפֶר מִלְחֲמֹת יְהוָה

Approximately:

“Therefore/for this reason it is said in the sefer of the Wars of YHWH…”

This is unusually valuable because the narrator is not merely incorporating traditional material anonymously.

He identifies a written locus.

Greenstein emphasizes that sēfer can designate a written document rather than necessarily something resembling a modern bound book. (TheTorah.com)

So we should imagine possibilities such as:

a scroll,

a scribal compilation,

a collection of individual poetic texts,

a court or temple archive,

a pedagogical anthology,

or perhaps even a relatively short written composition.

What we cannot infer is:

• how many columns it had;
• how many poems it contained;
• whether one author wrote it;
• whether it was continuously copied;
• whether Numbers had the complete work before it;
• or whether its compiler knew it indirectly through another source.

That last possibility is often overlooked.

A hidden transmission problem

When Numbers says “it is said in the Book…,” at least three transmission scenarios are possible:

Direct citation: the author/editor physically possessed the work.

Indirect citation: the author used another document already quoting it.

Traditional citation: the attribution had become attached to the excerpt even though the compiler no longer possessed the full book.

Those scenarios produce very different implications for dating.

The biblical wording does not let us distinguish them.

3. The quotation may begin with a verb that everyone missed

The great syntactical problem is:

אֶת־וָהֵב בְּסוּפָה

Traditionally, ʾet is interpreted as Hebrew's direct-object marker.

That produces something resembling:

“… Waheb in Suphah…”

But then the sentence has objects without a surviving verb.

This explains why translations differ so radically.

The textual traditions show that the passage was already troublesome in antiquity. The Samaritan tradition connects the expression with the Sea of Reeds; the Greek renders something resembling a reference to Zoob being burned, while Vulgate-based interpretation understands an action at the Red Sea. NET's textual discussion explicitly notes this wide ancient variation. (NET Bible)

That is significant.

It means the difficulty is not primarily a modern translator's problem.

Richard Steiner's solution

Richard C. Steiner's 2020 Journal of the American Oriental Society article offers one of the most consequential modern proposals.

He argues that אֶת is not the direct-object marker at all.

Instead, it is an archaic shortened imperative from the Semitic verb ʾ-t-y, “come.”

Thus:

אֶת־וָהֵב

would mean approximately:

“Come to Waheb.”

This requires no consonantal emendation of the Masoretic text. Steiner argues that the form is an archaic/biliteral imperative whose grammatical identity became obscured. (Scribd)

That yields something remarkable.

Only thirteen verses later, the Heshbon poem begins:

בֹּאוּ חֶשְׁבּוֹן

“Come to Heshbon!”

So Numbers 21 potentially contains two poems employing almost the same rhetorical opening:

Come to Waheb…

and

Come to Heshbon…

Greenstein regards this parallel as especially attractive evidence for Steiner's interpretation. (TheTorah.com)

This does not prove that both songs came from the same collection.

But it makes their literary relationship more interesting than mere proximity.

4. “Suphah” may not be a place at all

Earlier we treated “Waheb in Suphah” as possibly two place names.

That needs refinement.

There are several competing readings.

One reads:

Waheb in Suphah = place + place.

Another interprets sûpâ as:

“storm” / “tempest.”

Steiner favors this.

His resulting reconstruction is approximately:

“Come to Waheb in a storm…”

(Scribd)

This would radically alter the scene.

Instead of an obscure itinerary:

Waheb → Suphah → Arnon

we might have:

a journey toward Waheb during a rainstorm.

And that matters because Steiner connects the storm to the hydrology of the region.

5. A startling possibility: the poem may describe flash-water geography

The Arnon is not simply a “river” in the European sense.

The biblical Arnon corresponds broadly to the modern Wadi Mujib system, a complex network of deep ravines and tributaries cutting through central Jordan.

Among tributaries proposed in discussions of Numbers 21 are:

• Wadi al-Haydan;
• Wadi an-Nukhayla/el-Meheires;
• Wadi as-Saide.

(TheTorah.com)

This makes the plural expression:

נחלים ארנון

potentially quite literal:

“the Arnon wadis/streams.”

One alternative, however, treats the final -m not as ordinary plural morphology but as an enclitic, meaning that even the number of streams is grammatically disputed. (TheTorah.com)

That is the kind of hidden complexity easy to lose in translation.

6. Steiner's much larger proposal: the quotation may extend to verse 20

This is one of the biggest developments in our investigation.

Most treatments assume the citation from the Book of Wars occupies only:

Numbers 21:14b–15.

Steiner argues that it may continue all the way through verse 20.

Under his reconstruction, the sequence becomes a coherent geographical itinerary:

Waheb → Arnon tributaries → their confluence → a well → Mattanah → Nahaliel → Bamoth → valley in Moab → Pisgah.

He further interprets some seemingly separate words and place-names through this hydrological-geographic framework. (Scribd)

That would mean the Book of Wars extract contains the Song of the Well itself.

This is dramatically more material than normally assumed.

But caution is essential

Steiner's reconstruction is ingenious, coherent, and published in a major specialist journal.

It is nevertheless a specific scholarly reconstruction, not a consensus reading.

It depends on a chain of arguments:

1. ʾet is an imperative.
2. Waheb is a geographic feature.
3. sûpâ means storm.
4. several obscure geographic terms can be mapped coherently;
5. mattanah participates simultaneously in the geography and the semantic field of “gift”;
6. the source quotation continues across material traditionally treated separately.

If one link fails, the larger reconstruction weakens.

So this is exactly the kind of argument where cumulative coherence should not be mistaken for independent confirmation.

7. What exactly was Waheb?

We previously treated Waheb rather casually as an unknown site.

There are at least three substantial possibilities.

Possibility A: Waheb is simply an unknown ancient place-name

This remains the conservative interpretation.

Neither Waheb nor Suphah occurs elsewhere securely in the Bible.

Possibility B: Waheb was a town

David Ben-Gad HaCohen proposes identifying Waheb with a settlement in the Arnon tributary system and links Suphah with a preserved Arabic wadi name. His geographical model seeks to identify the passage with the area of Wadi as-Saide/Wadi Sefai. (TheTorah.com)

He associates one candidate site with Iron II occupation and proposes that the ancient place-name disappeared after abandonment. (TheTorah.com)

This is intriguing but remains an identification hypothesis.

Toponymic survival across millennia is real, but resemblance alone cannot prove identity.

Possibility C: Waheb was a mountain or hydrological landmark

Steiner pushes in a different direction.

Drawing on Arabic linguistic parallels involving the root w-h-b, “give,” he suggests Waheb may have referred to a mountain characterized as a giver/provider of water.

He connects that idea with runoff hydrology: rainfall on the mountain feeds streams and perhaps the well mentioned immediately afterward. (Scribd)

The semantic network would then be elegant:

Waheb = giver

↓

rain/storm

↓

wadi runoff

↓

well

↓

Mattanah = gift

This is intellectually attractive.

But it is also precisely the type of pattern requiring caution.

Some elements may reflect genuine wordplay; others may be retrospective harmonization.

8. “Ar” is more complicated than merely a city

The first pass called Ar a city.

That is too simple.

Biblical ʿAr can apparently denote a specific place, but the expression in Numbers may also involve a settled region or district associated with Moab.

HaCohen interprets šebet ʿAr along the lines of the inhabited/settled territory of Ar, associated with pastoral or sparsely settled land east of one of the wadi systems. (TheTorah.com)

This matters because Numbers 21:14–15 may be performing boundary description, not simply recounting battle.

The quotation's function within Numbers 21 is therefore possibly juridical or territorial:

Why does Israel's occupation north of the Arnon not constitute seizure of Moabite territory?

Because, according to the embedded tradition, the Arnon marked Moab's boundary.

Greenstein specifically interprets the quotation as documentary support for the boundary statement in Numbers 21:13. (TheTorah.com)

That opens an important possibility:

The compiler may be using poetry as territorial evidence.

This is much more specific than saying merely that old songs were preserved.

9. The Book of Wars quotation may serve a legal-political purpose

Numbers and Deuteronomy insist that Israel was not to seize Moabite territory.

Yet Israel ends up possessing territory that was apparently once Moabite.

The solution presented in the biblical narrative is:

Moab formerly controlled land north of the Arnon.

↓

Sihon conquered that territory from Moab.

↓

Israel later defeated Sihon.

↓

Therefore Israel took Amorite, not Moabite, territory.

The Heshbon poem becomes highly useful in supporting this territorial genealogy.

Recent archaeological scholarship describes exactly this literary function as one plausible reason the poem was embedded in Numbers: it certifies that Moab had already lost the plateau north of the Arnon before Israel arrived. (Cambridge University Press)

So the poems may not merely ornament the narrative.

They may act almost like ancient evidentiary exhibits.

That is a major conceptual advance.

10. The Heshbon poem is stranger than it first appears

Numbers 21:27–30 is introduced:

עַל־כֵּן יֹאמְרוּ הַמֹּשְׁלִים

“Therefore the mōšəlîm say…”

The poem celebrates fire going out from Heshbon and consuming Moabite territory.

The crucial historical oddity is:

Israel is not the victor in the original battle celebrated by the poem.

Sihon is.

The poem appears to celebrate the conquest of Moab by the Amorites.

That creates competing interpretations.

Amorite-origin hypothesis

The poem originated among Sihon's people as a victory song celebrating their conquest of Moab.

This has an obvious contextual logic: the poem praises Heshbon and Sihon's expansion.

Modern archaeological discussion acknowledges this as a serious possibility. (Cambridge University Press)

Israelite taunt hypothesis

Israelite poets could instead have appropriated the language ironically.

The message would effectively be:

“Sihon, who once devastated Moab, has now himself been devastated by us.”

Ancient victory poetry frequently reuses enemy rhetoric, so the possibility is perfectly credible.

Later cultural-memory hypothesis

The poem might instead have been composed much later, when Sihon had become a legendary archetype of Israel's primordial enemy.

That possibility becomes important when we consider dating.

11. Archaeology does not currently verify Sihon's kingdom

This is an area where older biblical archaeology sometimes claimed more than the evidence supports.

Modern excavation at Hisban, conventionally identified with biblical Heshbon, has found material indicating activity in the late second millennium BCE.

But later construction badly disturbed the earlier levels.

Consequently:

archaeology cannot currently determine the settlement's political significance or demonstrate that it was Sihon's capital.

(Cambridge University Press)

More broadly, the archaeological pattern of west-central Jordan does not neatly reproduce the picture of clearly bounded Moabite and Amorite kingdoms implied by a surface reading of Numbers.

Recent archaeological synthesis stresses that evidence from the late second millennium is often sparse, mobile, regionally variable, and difficult to assign ethnically. (Cambridge University Press)

This does not prove the biblical political geography is fictional.

But it prevents us from claiming archaeological confirmation.

That distinction is important.

12. The chronology of Moab is one of our hardest problems

The Song of Heshbon is sometimes treated as if an early date is self-evident because it recounts events before Israel's arrival.

It isn't.

There are at least three layers of date that must be distinguished:

Date of the event remembered

Date of the poem

Date of insertion into Numbers

Those could differ by centuries.

A poem about a twelfth-century event could be:

• composed contemporaneously;
• preserved orally and written later;
• composed centuries later from traditional memories;
• or composed for a new political situation while projecting itself backward.

Archaeologist Benjamin Porter notes that some scholars still argue for preservation of late-second-millennium memories, but others place the literary composition significantly later, possibly around or after Israelite expansion into Transjordan under the Omride monarchy in the ninth century BCE. (Cambridge University Press)

That possibility deserves serious attention because we have an extraordinary external source from precisely that region and century.

13. The Mesha Stele transforms the comparative landscape

The Mesha Inscription, erected by King Mesha of Moab in the ninth century BCE, is not evidence for the Book of Wars itself.

But culturally it is tremendously illuminating.

Mesha describes territorial conflict between Israel and Moab and attributes victory and defeat to the activity of Chemosh, Moab's national deity.

He says that Israel had oppressed Moab because Chemosh was angry with his land, then reports Chemosh commanding him to retake territory. The inscription also refers to cultic objects associated with YHWH captured from an Israelite site and brought before Chemosh. (Archaeology)

This gives us a striking regional parallel.

Moabite ideology

Chemosh:

• becomes angry with his people;
• permits foreign domination;
• later restores his people;
• commands military action;
• grants conquest.

Israelite ideology

YHWH:

• fights for Israel;
• delivers enemies;
• grants territory;
• can also punish Israel militarily.

We therefore have direct epigraphic evidence that the conceptual world of a national deity actively conducting wars was not uniquely Israelite.

That makes the phrase:

“Wars of YHWH”

far more culturally intelligible.

It belongs to a broader Levantine political theology in which military outcomes could be narrated as actions of national gods.

14. But “YHWH as warrior” needs finer distinctions

Earlier I connected the title with Exodus 15, Judges 5, Psalms and cosmic-war traditions.

That connection remains important, but we need to separate at least three forms of divine warfare.

Historical warfare

YHWH fights against terrestrial armies.

Example:

Egypt in Exodus 15.

Meteorological warfare

Storm, rain, flood, stars and rivers become instruments of divine victory.

Judges 5 is a particularly good example: YHWH's appearance and natural phenomena frame Sisera's defeat.

Scholars have long noted the warrior imagery in Judges 5 and Exodus 15. (BSW)

Cosmic warfare

Later biblical poetry depicts YHWH defeating personified Sea, Rahab, Leviathan, or chaotic waters.

These traditions overlap rhetorically with ancient Northwest Semitic storm-god mythology.

But:

overlap does not mean every Israelite war song derived from a cosmic battle myth.

Exodus 15, for example, uses sea and chaos imagery but directs the combat primarily against Egypt, not against Sea as an independent deity. Mark Leuchter emphasizes this distinction. (BSW)

That distinction is relevant because one scholar, Naphtali Tur-Sinai, proposed emending Waheb to Rahab, thereby turning Numbers 21 into cosmic-combat imagery.

HaCohen reports that proposal. (TheTorah.com)

It is ingenious—but highly conjectural.

Without manuscript support, it should remain a minority textual reconstruction.

15. Ancient readers had already lost the meaning of the line

The history of interpretation tells us something about textual transmission.

Samaritan interpretation

The Samaritan Pentateuch apparently associates the phrase with the Sea of Reeds.

Greek Septuagint

The Greek text produces something like:

YHWH's war “set Zoob ablaze…”

showing a radically different parsing.

Onkelos

The Aramaic Targum interprets the phrase as YHWH performing battles at the Sea of Reeds and mighty acts at the Arnon.

Vulgate

Jerome similarly interprets a comparison with divine action at the Red Sea.

Peshitta

The Syriac tradition apparently read something like:

“a flame in a storm.”

HaCohen surveys this extraordinary range. (TheTorah.com)

What does this variation tell us?

Not which interpretation is correct.

Rather:

By late antiquity, the original wording was sufficiently obscure that multiple traditions independently reconstructed it.

That suggests one of several things:

• genuine linguistic archaism;
• corrupt transmission;
• obscure proper names;
• loss of geographical knowledge;
• or some combination.

This is much stronger evidence for the antiquity of the difficulty than for the antiquity of the composition itself.

That methodological distinction is crucial.

16. The phrase may have been archaic without the poem being “Mosaic”

Older scholarship often treated difficult or archaic poetry as automatically very early.

Modern linguistic dating is considerably more cautious.

Language can look archaic because:

it genuinely is old;

a later poet deliberately imitates archaic style;

formulaic poetry preserves old constructions;

textual corruption creates apparently ancient forms;

or

regional dialect differs from standard Biblical Hebrew.

The Heshbon Song itself illustrates the problem. Some scholars date its memories very early, yet recent scholarship notes that it lacks some archaic linguistic features that might be expected if it were straightforwardly a late-second-millennium composition. (Cambridge University Press)

So we should separate:

“this may contain old material”

from

“this poem was written in Moses' lifetime.”

The first is defensible.

The second is currently unprovable.

17. Exodus 17:14 becomes more interesting—but not more certain

Exodus 17, after Amalek's attack, contains the command:

write this as a memorial “in the sēfer.”

The superficial connection is obvious.

A battle involving YHWH is ordered to be recorded.

Could this have become part of YHWH's Battles?

Possibly.

But three problems remain.

First, Exodus does not name the Book of Wars.

Second, “write this in a sēfer” need not mean “put this in an already existing anthology.”

Third, literary dependence can run in several directions: the Exodus passage could reflect a general scribal-memory motif rather than an actual reference to the same artifact.

Ibn Ezra's suggestion that the Book of Wars was independent and extremely ancient—even possibly reaching back before Moses—is itself evidence for medieval interpretation, not evidence for its actual date. Greenstein records Ibn Ezra's view explicitly. (TheTorah.com)

Therefore:

connection to Exodus 17: plausible thematic hypothesis.

identity: unproven.

18. The Book of Jashar comparison becomes more useful if we stop trying to identify them

The strongest similarity is not necessarily that they were the same book.

It is that both provide evidence for a literary ecosystem mostly lost to us.

The Book of Jashar is cited in:

Joshua 10 — celestial/battle poetry;

2 Samuel 1 — David's lament for Saul and Jonathan.

Those are radically different kinds of material.

That strongly suggests either:

• a miscellaneous poetic anthology;
• an anthology organized according to principles now unknown to us;
• or a collection whose remembered name did not summarize its contents.

The Joshua and Samuel references securely attest poetic material; Greenstein additionally discusses the possibility that the Greek tradition of 1 Kings 8 preserves another citation, though that identification depends on reconstructing “Book of Song” as a corruption of “Book of Jashar.” (TheTorah.com)

This analogy strengthens—not proves—the possibility that the Book of Wars was similarly diverse.

19. Numbers 21 may preserve something like “archival montage”

Now the structure of the chapter becomes particularly interesting.

Within a short span we encounter:

itinerary prose

↓

named textual citation

↓

geographic poetry

↓

well song

↓

more itinerary

↓

battle narrative

↓

balladeers' poem

That pattern resembles what we might call archival montage: prose narrative incorporating fragments from earlier documentary and oral-poetic traditions.

This need not mean the editor mechanically cut-and-pasted physical scrolls.

Ancient literature could preserve inherited materials orally, scribally, liturgically, or through mixed transmission.

But Numbers 21 certainly creates the literary impression that the narrator knows traditions external to his immediate prose narrative.

Greenstein observes that three distinct songs occur in the chapter and that they lend documentary authority to the travel narrative. (TheTorah.com)

This opens a new question:

Was Numbers 21 deliberately constructed as a documentary dossier?

If so, the difficult fragments may not be editorial accidents.

Their antiquarian quality could itself create authority.

20. “Therefore it is said…” may be evidentiary language

Notice the repeated phrase:

עַל־כֵּן

“therefore / for that reason…”

It occurs introducing both:

the Book of Wars citation

and

the Heshbon poets.

This parallel is potentially important.

The prose makes a claim.

Then older verse is invoked.

Effectively:

Narrative claim → inherited quotation as confirmation.

For the Arnon:

“The Arnon is the boundary…”

→

“Therefore it is said in the Book…”

For Sihon's conquest:

“Sihon had conquered this territory from Moab…”

→

“Therefore the poets say…”

This looks less like random poetic ornamentation and more like citation as corroboration.

Greenstein and recent archaeological discussion both identify this documentary function. (TheTorah.com)

That is one of the strongest new insights from this deeper pass.

21. The Song of the Well might not originally be about miraculous water

Numbers 21:17–18 says Israel sang:

“Spring up, O well…”

It then says princes or nobles dug/excavated it with symbols translated as staffs or scepters.

The curious feature is that the song itself does not explicitly mention YHWH.

The prose introduction does.

That creates a possible compositional distinction:

Older song

A secular or ceremonial well-opening song.

Biblical framework

The editor links the well to divine provision.

Eryl Davies discusses an old proposal that the leaders did not literally dig with ceremonial staffs, but performed some kind of formal opening ritual at a newly discovered water source. (Biblical Studies)

If correct, the song could preserve social ritual, not miracle tradition.

That would strengthen the idea that a collection containing it was broader than battle poetry.

22. The mōšəlîm may represent a social profession—or simply a literary category

Earlier I called them poets or ballad singers.

That needs qualification.

The Hebrew root m-š-l produces māšāl, a broad literary category that can include:

proverb,

taunt,

parable,

comparison,

aphorism,

or crafted saying.

So mōšəlîm could signify:

proverb makers,

poets,

satirists,

reciters,

or perhaps professional composers of traditional sayings.

Ramban interpreted them as learned composers who recorded great battles and fashioned poetic discourse. Greenstein preserves his interpretation and notes that Ramban explicitly connects the Heshbon poem to the Book of Wars. (TheTorah.com)

That is important reception history.

But we should not convert Ramban's medieval social reconstruction into direct knowledge of Bronze- or Iron-Age professions.

23. The absence of the book from Qumran is less surprising than it sounds

The first pass said no Qumran manuscript has been identified as the Book of Wars.

That remains correct.

But there is a deeper archival issue.

For a lost book to appear at Qumran, it would have had to survive through perhaps many centuries of copying.

A work cited in an early stage of biblical composition could disappear long before the Second Temple period.

Therefore:

Qumran's failure to preserve it does not imply it was already fictional when Numbers was written.

Nor does Qumran's enormous manuscript diversity make its survival particularly likely.

Most ancient works still vanished.

The famous War Scroll is especially misleading because its modern English title sounds suggestive, yet it is an apocalyptic Second Temple composition concerned with the final conflict of the Sons of Light and Darkness.

There is no textual evidence making it the Numbers source.

So identifying the two requires not merely speculation, but overcoming a massive chronological and literary gap.

24. Could the lost book still exist unidentified?

Here we need three categories.

Existing but unidentified manuscript

Theoretically possible.

An extremely fragmentary Hebrew manuscript might contain material once belonging to the collection without preserving its title.

Existing as incorporated fragments

Much more plausible.

Numbers may itself be the surviving manuscript witness.

If so, asking:

“Can we find the Book of Wars?”

may partly have the wrong shape.

The more fruitful question is:

Can we identify additional fragments copied from it into canonical books?

That is methodologically analogous to reconstructing lost Greek historians from quotations preserved by later writers.

Complete independent manuscript

Presently no evidence.

Claims to possess the entire Book of Wars require manuscript provenance, palaeography, language analysis and demonstrable transmission history.

No authenticated ancient manuscript presently meets those criteria.

25. The most promising route to “recovering” the book is therefore reconstructive, not archaeological

Physical discovery remains possible.

But historically, the highest-yield strategy may be textual archaeology.

We would search biblical poetry for clusters of features associated with Numbers 21:

• citation formulas;
• archaizing poetic grammar;
• Transjordanian geography;
• YHWH-war language;
• boundary descriptions;
• Sihon/Og traditions;
• identical poetic openings;
• unusual lexemes;
• stylistic correspondences;
• possible shared scribal framing.

Potential candidates would then have to survive a falsification test.

Similarity alone is not enough.

For example, Exodus 15 is war poetry involving YHWH, but that does not mean it belonged to the Book of Wars.

Otherwise almost every victory text becomes a candidate by definition.

26. A potentially major chronological clue: the ninth century BCE

A new research thread emerges from combining the literary and archaeological evidence.

The Mesha Inscription proves that in the ninth century BCE:

• Israel and Moab contested precisely this Transjordanian zone;
• territorial memory mattered politically;
• national deities were credited with conquest;
• places such as Madaba, Atarot and Nebo mattered;
• Israelite presence north of the Arnon was historically real by that period.

(Cambridge University Press)

That creates an alternative context for Numbers 21's old-looking geography.

Perhaps the traditions genuinely preserve Bronze-/early Iron-Age memories.

But perhaps some literary forms were shaped during later Israel-Moab territorial disputes, when proving ancient possession would have political value.

These hypotheses are not mutually exclusive:

old territorial tradition

↓

later poetic preservation

↓

ninth-century reinterpretation

↓

still later Pentateuchal incorporation

could all be true simultaneously.

This layered model deserves considerably more investigation.

27. We now have a much sharper evidence hierarchy

The investigation can now be stratified.

Hidden Complexity: the key issue may be territorial legitimacy

The deepest new insight is that Numbers 21's poetic fragments may function politically.

The chapter is concerned not simply with “Israel won battles.”

It is concerned with who legitimately possessed land north of the Arnon.

The logic is:

Moab once possessed territory.

Sihon removed it from Moab.

Israel defeated Sihon.

Therefore Israel acquired Sihon's land without violating Moab.

The Book of Wars fragment establishes the boundary.

The Heshbon song establishes the prior conquest.

Together they form an argument.

That means these fragments could operate almost like ancient land-title evidence.

Whether that argument reflects late-second-millennium realities, later Iron-Age territorial disputes, or layers of both is now one of the most important questions in the entire investigation.

New Threads Now Visible

Several research branches have become significantly more valuable.

The first is textual criticism. We should compare the Masoretic Text, Samaritan Pentateuch, Septuagint, Peshitta, Vulgate, Onkelos and later Jewish interpretations word-by-word. The extreme variation may let us reconstruct when understanding of the phrase was lost.

The second is hydrology. Steiner's theory is testable against actual geology, rainfall patterns, subterranean water, wadi behavior and proposed site locations around the Arnon.

The third is ancient territorial law and memory. We should investigate whether old poems or inscriptions were elsewhere used to legitimize territorial claims.

The fourth is Moabite comparison. The Mesha Inscription should be placed line-by-line beside Numbers 21, Deuteronomy 2–3 and Judges 11. The geographic and theological overlap is potentially profound.

The fifth is literary archaeology. Instead of searching only for a lost manuscript, we can search the Hebrew Bible for possible fragments of the lost anthology using controlled stylistic and linguistic criteria.

The sixth is compilation history. Numbers 21 may preserve multiple historical stages rather than one source: old geography, songs, prose explanation and later theological framing.

The seventh is the social world of poets and scribes. Who maintained national songs? Court singers? temple personnel? clan poets? professional scribes? Oral performers? The answer affects what a sēfer like this actually was.

Revised provisional reconstruction

The evidence now supports a more nuanced model than the one we began with.

The Book of the Wars of YHWH was almost certainly a real written source from the perspective of the author/editor who cites it, but nearly everything about its original form remains uncertain.

It probably contained poetic or poetically compressed traditional material.

It may have been specifically a collection of warfare traditions—but we should no longer assume that merely from its title.

Most importantly, its surviving fragment seems to have been useful not simply because it described a battle, but because it preserved geographical memory about the Arnon frontier.

And Numbers 21 as a whole increasingly looks like a carefully constructed assemblage in which old songs function as witnesses to territorial history.

That moves the investigation toward a much bigger possibility:

The Book of the Wars of YHWH may have belonged to an otherwise vanished Israelite documentary-poetic tradition in which geography, collective memory, divine action, and territorial claims were preserved together in verse.

That is not yet a conclusion.

But it is now a much more precise hypothesis to test.

The next especially high-value investigation would be a forensic reconstruction of Numbers 21:13–20 across every major textual witness, because if Steiner's proposal that the Book of Wars quotation continues through verse 20 survives that test, we would suddenly possess substantially more of the lost work than has traditionally been recognized.

High-Value Research Questions: The Book of the Wars of YHWH

The investigation has now reached the point where not all unanswered questions are equally valuable. Some questions are interesting but would barely alter the larger reconstruction. Others could radically change what we think the Book of the Wars of YHWH was, when it existed, how much of it survives, and why Numbers cites it.

The central research problem can now be framed as:

Can we move from knowing that Numbers 21 cites a lost written source to reconstructing, with defensible probabilities, its textual extent, literary character, historical setting, and function?

The highest-value questions are those that discriminate between competing models rather than merely accumulate parallels.

Decisive Questions

1. Exactly where does the quotation from the Book of the Wars of YHWH end?

This is probably the single most consequential textual question.

The traditional minimum reconstruction assigns roughly Numbers 21:14b–15 to the lost work. Richard Steiner's proposal potentially extends the material considerably farther, perhaps through verse 20.

Those alternatives produce radically different lost books.

If only vv. 14–15 belong to it, the work may have contained something resembling geographical or military poetry concerning the Arnon frontier.

If vv. 14–20 belong together, then the source may also include the Song of the Well and an extended itinerary, suggesting a substantially broader genre.

The decisive subquestions are whether vv. 14–20 exhibit continuous syntax, meter, vocabulary, geography, discourse structure, and source markers; whether ancient versions understood a continuous passage; and whether the prose transitions in the Masoretic Text actually mark source boundaries or merely later editorial segmentation.

A secure answer could transform our reconstruction of the entire book.

2. What did the original Hebrew of Numbers 21:14–15 actually mean?

The key expression:

אֶת־וָהֵב בְּסוּפָה

is so obscure that different ancient translators already understood it differently.

The possibilities include some combination of:

Waheb as a place, Waheb as another geographical feature, Suphah as a place, sûpâ as “storm/tempest,” or Steiner's interpretation of ʾet as an archaic imperative meaning “Come!”

Until this is resolved—or at least probabilistically narrowed—almost every reconstruction rests on unstable foundations.

If Steiner is correct, the passage may be an intelligible piece of travel or landscape poetry.

If the traditional object-marker reading is correct, a verb may have disappeared or become corrupted.

If the ancient versions preserve remnants of a different Hebrew Vorlage, the Masoretic wording may itself not reproduce the earliest recoverable text.

This question therefore sits upstream of geography, genre, and dating.

3. Was sēfer milḥămōt YHWH actually the title of a war anthology?

This assumption underlies almost everything commonly said about the work.

But it has not been demonstrated.

The research must distinguish among at least four possibilities:

a military chronicle;

a collection of victory songs;

a miscellaneous poetic anthology whose remembered name derived from one component;

a broader collection of traditional historical-geographical material.

If the Song of the Well genuinely belonged to it, the narrow “war-song anthology” model becomes harder to maintain.

Conversely, if the quotation ends at verse 15 and can be shown to recount a military event, the traditional model strengthens substantially.

This is decisive because it determines what kinds of biblical passages we should even consider as possible surviving fragments.

4. What historical period best explains the geographical and political world presupposed by the material?

The material could theoretically preserve:

a late-second-millennium BCE political landscape,

early Iron Age memories,

monarchic-period territorial traditions,

ninth-century Israel-Moab ideological disputes,

or a later literary reconstruction incorporating older names and traditions.

The key is to avoid equating:

date of purported event = date of poem = date of anthology = date of incorporation into Numbers.

All four could differ.

If the geographical description accurately reflects a demonstrably early political landscape later forgotten, that would strongly favor genuinely old source material.

If it instead reflects Iron Age territorial realities recognizable from the Mesha period, later composition or reshaping becomes more probable.

5. What function does the citation perform inside Numbers 21?

This may ultimately matter more than identifying the book's genre.

Does the compiler invoke the source primarily to:

preserve historical memory,

authenticate geography,

establish a territorial boundary,

justify Israel's possession of former Moabite territory,

celebrate YHWH's military power,

or merely decorate the narrative with inherited poetry?

If the quotation and the Heshbon Song function as evidence for territorial legitimacy, Numbers 21 becomes something closer to a historical-legal argument supported by traditional texts.

That would substantially reframe both the chapter and the use of earlier literature within the Pentateuch.

Foundational Questions

Before larger historical reconstructions can be trusted, several lower-level problems have to be solved.

The first is the textual history of Numbers 21:14–20. We need a systematic comparison of the Masoretic Text, Samaritan Pentateuch, Septuagint, Peshitta, Vulgate, Targum Onkelos and other early interpretive witnesses. The goal is not merely to catalogue differences, but to reconstruct what Hebrew readings may stand behind them and identify when comprehension of the passage began to break down.

The second is lexical and grammatical. Every anomalous word—particularly ʾet, Waheb, sûpâ, the Arnon expressions and the description of the boundary near Ar—needs evaluation against Biblical Hebrew, epigraphic Hebrew, Northwest Semitic languages, Arabic cognates and possible archaic constructions. A rare cognate should not be accepted merely because it makes one attractive reconstruction work.

The third is geographical. Waheb, Suphah, Ar, Mattanah, Nahaliel, Bamoth and Pisgah must be classified individually as secure, probable, possible or unknown locations. Attempts to reconstruct an itinerary should then be tested against actual topography, travel corridors, wadis and water sources.

The fourth concerns literary segmentation. We need to establish whether Numbers 21 contains one extended source, several independent quotations, or quotations embedded within later prose. Parallelism, meter, syntax, discourse markers and shifts in narrative voice may help.

The fifth is comparative source citation. How does “therefore it is said in the Book…” compare with biblical references to the Book of Jashar, royal annals, prophetic collections and other written sources? If the citation formulas behave consistently, they may reveal how literally we should understand Numbers' reference to a sēfer.

Discriminating Questions

Here the objective is not simply to learn more, but to devise tests that force competing explanations apart.

One especially valuable discriminating test concerns the Heshbon Song. If its earliest linguistic form genuinely praises Sihon rather than Israel, then Israelite transmission of a foreign victory poem becomes plausible. That would provide direct evidence that biblical compilers could appropriate and repurpose non-Israelite historical poetry.

That would be a major finding.

Missing-Evidence Questions

The ideal evidence is clearer now.

The greatest prize would be an ancient Hebrew or Northwest Semitic manuscript containing the phrase “Book of the Wars of YHWH” together with substantial additional text. Even a few columns could resolve questions of genre, date and contents.

Short of that, a manuscript containing Numbers 21:14–20 in a significantly earlier textual form would be enormously valuable. A pre-Masoretic witness could reveal whether the puzzling wording is ancient or the result of transmission.

An inscription mentioning Waheb, Suphah, or another disputed site in the appropriate geographic region would dramatically improve the geographical reconstruction.

Archaeologically, we would want secure late-second-millennium through early-Iron-Age occupation sequences at candidate sites around Heshbon and the Arnon, especially evidence capable of identifying political control rather than merely habitation.

Epigraphically, additional Moabite, Ammonite, Israelite or other Transjordanian inscriptions describing territorial boundaries or battles would be exceptionally valuable. The Mesha Stele demonstrates what such evidence can contribute; another major inscription from the same cultural environment could transform the discussion.

We would also ideally want external evidence for a scribal genre of battle anthologies in Iron Age Israel or neighboring societies. A physical archive containing collections of war poems, territorial records or heroic songs would provide a cultural mechanism for the biblical source.

Finally, we lack evidence showing how long the Book of Wars continued to circulate. A later quotation explicitly attributed to it—even outside the Hebrew Bible—would help establish its lifespan and potentially its contents.

Ranked Research Agenda

1. Reconstruct Numbers 21:14–20 textually and philologically

Expected informational value: Exceptional.

Everything else depends upon knowing what the surviving passage actually says and where the quotation ends.

This investigation should collate every important textual witness, reconstruct possible Hebrew Vorlagen, analyze every disputed lexeme and then test Steiner's extended-reading hypothesis against competing reconstructions.

Possible outcomes:

If Steiner's interpretation largely survives, we may possess a much longer extract than usually recognized.

If it fails, the traditional short-fragment model gains strength.

If textual corruption proves substantial, reconstruction of the original book may be much less secure than either camp assumes.

2. Reconstruct the historical geography of Numbers 21

Expected informational value: Very high.

Map every proposed location and hydrological feature around the Arnon against archaeological surveys, ancient place-name evidence and actual terrain.

This could determine whether the text reflects genuine regional knowledge.

If the itinerary accurately maps a coherent ancient route, confidence that the material preserves authentic geographical tradition rises considerably.

If proposed identifications require repeated linguistic or geographic special pleading, elaborate reconstructions such as Steiner's weaken.

3. Determine the chronological horizon of the underlying poems

Expected informational value: Very high.

Rather than trying to “date the Book,” separately date:

the remembered events,

the language of the poem,

the formation of the collection,

and its incorporation into Numbers.

The principal alternatives should include late-second-millennium, early Iron Age, early monarchy, Omride/ninth-century, and later literary settings.

A secure early date would radically strengthen the value of the source as historical memory.

A monarchic date would instead make it extremely important evidence for Israel's later construction of Transjordanian territorial history.

Either result is significant.

4. Determine whether Numbers 21 is making a territorial-legitimacy argument

Expected informational value: Very high.

This should compare Numbers 21 with Deuteronomy 2–3, Judges 11, Moabite territorial claims and the Mesha Inscription.

If the embedded poems consistently authenticate the claim that Israel acquired land from Sihon rather than Moab, then their preservation may have had a political-juridical purpose.

That would explain why these particular pieces of old poetry survived.

5. Establish whether the Heshbon Song originated outside Israel

Expected informational value: High.

Analyze voice, rhetoric, theology, geographic perspective and linguistic features.

If an Amorite or other non-Israelite origin becomes probable, the passage becomes rare evidence for Israelite appropriation of foreign historical literature.

If it is demonstrably Israelite, the poem instead tells us how Israel constructed memories of Sihon's earlier victories.

6. Determine what kind of literary collection the Book of Wars actually was

Expected informational value: High.

Only after quotation boundaries are addressed should we return to genre.

Test:

military chronicle,

war-song anthology,

national poetry anthology,

geographical-historical collection,

or mixed scribal compilation.

A mixed anthology would make comparison with the Book of Jashar especially valuable.

7. Compare the Book of Wars with the Book of Jashar and other lost biblical sources

Expected informational value: Moderate-to-high.

The main question is not whether they were identical.

It is whether they reveal a larger Israelite archival culture now mostly lost.

Patterns in citation formulas, poetry, subject matter and textual reuse could reveal how biblical compilers interacted with external literature.

This would substantially broaden the investigation beyond a single mysterious book.

8. Reconstruct the regional ideology of divine warfare

Expected informational value: Moderate-to-high.

Compare YHWH's wars with Chemosh in the Mesha Inscription and, carefully, with Northwest Semitic and broader ancient Near Eastern divine-war traditions.

The crucial question is whether “Wars of YHWH” represents:

a specifically Israelite theological concept,

a regional Levantine conception of national deity and warfare,

or an Israelite adaptation of a widespread political theology.

This would explain the title's conceptual world even if the book itself remains lost.

9. Search systematically for additional incorporated fragments

Expected informational value: Potentially enormous, but lower probability.

Once a defensible linguistic and literary profile has been established, examine candidate texts such as:

Exodus 17,

Exodus 15,

Deuteronomy 2–3,

Deuteronomy 33,

Joshua's battle traditions,

Judges 5,

and Sihon/Og traditions elsewhere.

The methodology must be conservative.

A passage should not become a “fragment of the Book of Wars” merely because it contains warfare and YHWH.

We would need several independent correspondences.

Finding even one credible additional fragment would dramatically expand our knowledge.

10. Investigate whether any unidentified manuscript fragments could belong to the work

Expected informational value: Enormous if successful; low prior probability.

Only after establishing a textual fingerprint should Qumran fragments, Geniza material and other unidentified Hebrew texts be reconsidered.

Otherwise this becomes pattern-hunting.

The correct sequence is:

construct diagnostic criteria first → search fragments second.

Not the reverse.

The Three Questions That Could Change Everything

If we had to reduce the entire project to only three research problems, they would be:

First: How much of Numbers 21:14–20 actually comes from the Book of the Wars of YHWH?

Because that determines how much of the lost text we already possess.

Second: When and where did that material originate?

Because that determines whether we are looking at near-contemporary historical memory, later preservation of older tradition, or retrospective territorial literature.

Third: Why did the compiler of Numbers cite it?

Because that determines whether the Book of Wars is being used as poetry, history, geography, theology—or evidence supporting Israel's territorial claim.

Together, those three questions could turn the investigation from a study of a mysterious lost title into a reconstruction of an actual ancient Israelite documentary tradition.

And one methodological rule should govern everything that follows:

We should not search for the rest of the Book of the Wars until we know what a genuine fragment of it would look like.

That makes the textual reconstruction of Numbers 21:14–20 the clear highest-priority next investigation.

If useful, I can keep a future checkpoint for returning to that textual reconstruction.

Evidentiary Audit: The Book of the Wars of YHWH

This audit materially tightens several conclusions from the investigation. The strongest evidence remains concentrated in Numbers 21 itself. Once we move beyond the existence of the citation and the basic wording surrounding it, confidence drops rapidly.

A useful rule for the rest of the project is:

We have one explicit biblical citation, several nearby poetic units, multiple ancient attempts to understand an obscure text, and a number of modern reconstructions. We do not have the lost book itself.

That distinction needs to govern everything that follows.

Major Claims and Evidence Classification

Source Quality

The evidentiary hierarchy is now much clearer.

Tier 1: Primary textual evidence

The Masoretic Hebrew text of Numbers 21 is our principal evidence. It establishes the source's name, the citation formula, the Arnon context, the Song of the Well, and the Heshbon poem. Modern JPS translations themselves flag Numbers 21:14 as fragmentary and uncertain. (Sefaria)

For reconstructing an earlier Hebrew form, the Septuagint, Samaritan Pentateuch, Peshitta, Targums and Vulgate are essential textual witnesses. They are not independent witnesses to the lost book itself; they are witnesses to ancient forms or interpretations of Numbers. That distinction matters enormously.

The Mesha Stele is exceptionally valuable independent primary evidence for ninth-century Moabite-Israelite relations and the theology of Chemosh. Its relevance is comparative, not direct: it never mentions the Book of Wars. (Cambridge University Press)

The Qumran War Scroll itself is primary evidence about Second Temple Jewish warfare ideology, but essentially negative comparative evidence for our question because its identifiable contents belong to a much later apocalyptic-military context. (DSS Collections)

Tier 2: Specialist philological and archaeological scholarship

Richard Steiner's 2020 JAOS article is one of the strongest specialist contributions we have used. It is published scholarship specifically devoted to Numbers 21:14–20 and presents an integrated philological, geographic, hydrological and ethnographic argument. Its strength is specificity; its limitation is that much of the broader reconstruction depends on accepting his initial grammatical innovation. (Lockwood Online Journals)

Recent Cambridge archaeological synthesis on Moab and west-central Jordan is much stronger evidence for archaeological limits than older harmonizing treatments. It explicitly warns that archaeological remains do not neatly confirm the political landscape portrayed in Numbers. (Cambridge University Press)

Edward Greenstein's essay is specialist secondary interpretation rather than primary evidence, but it is useful because it distinguishes the observable citation from hypotheses about anthology structure and compares it with other lost biblical works. (TheTorah.com)

Tier 3: Premodern Jewish interpretation

Ibn Ezra, Ramban and later interpreters are highly valuable for reception history.

They show that intelligent readers centuries ago independently recognized many of the same problems: lost sources, battle poetry, and possible connections between the chapter's poems. Ibn Ezra explicitly entertained a lost independent book; Ramban associated the Heshbon material with it. (Voices)

But their interpretations are separated from the proposed historical events by well over two millennia.

They are therefore not evidence for authorship, date, original contents, or physical transmission of the Book itself unless they preserve an independently demonstrable older tradition.

That earlier discussion occasionally blurred this distinction.

Source Criticism

Numbers itself

Numbers 21 is indispensable but not neutral evidence.

It is a literary composition whose final form postdates the events it narrates under virtually every mainstream model of Pentateuchal composition, although precisely how much later remains disputed.

The chapter therefore supplies two kinds of evidence simultaneously:

evidence that its compiler knew or claimed to know a source called the Book of Wars, and

claims about much earlier geography and events.

The first is far firmer than the second.

A source citation inside an ancient text does not automatically authenticate every historical claim attributed to that source.

Ancient translations

The Septuagint, Samaritan, Syriac, Aramaic and Latin traditions are especially important because their disagreement demonstrates that Numbers 21:14 was difficult early.

But they cannot simply be averaged together.

A divergent translation might reflect:

a different Hebrew Vorlage;

an interpretive paraphrase;

a misunderstood proper name;

deliberate harmonization;

or ordinary translation technique.

For example, divergent ancient readings around Waheb/Suphah show ancient uncertainty, but they do not by themselves prove textual corruption. (TheTorah.com)

That earlier formulation should therefore be tightened from:

“the text became corrupt”

to:

“the text was already interpreted divergently in antiquity, for reasons that may include obscurity, different Vorlagen, or corruption.”

Steiner's reconstruction

Steiner's proposal deserves substantial weight because it solves a genuine problem: under the ordinary analysis of ʾet, the line appears syntactically defective. His imperative “come!” requires no change to the consonantal text and produces an attractive parallel with “Come to Heshbon.” (Lockwood Online Journals)

But the larger theory is cumulative.

Once ʾet becomes “come,” Steiner can reinterpret Waheb, storm imagery, hydrology, the Well Song and the passage's geographic movement. The growing coherence is impressive, but the individual conclusions are partly interdependent.

Thus we cannot count each supporting feature as wholly independent evidence.

That is a classic danger of elegant reconstruction.

Archaeology

Archaeology is strong at answering:

Was a site occupied?

What material culture existed?

What broad settlement patterns occurred?

It is much weaker at directly identifying:

ethnicity,

specific biblical persons,

political borders,

or

the composition date of poems.

Recent work on west-central Jordan specifically cautions against reading the biblical categories “Moabite,” “Amorite,” and centralized kingdoms too directly into sparse late-second-millennium archaeological evidence. (Cambridge University Press)

So absence of archaeological confirmation of Sihon should neither be turned into confirmation nor refutation.

Evidentiary Gaps

Several earlier statements deserve explicit downgrading.

“The Book almost certainly existed as a separate ancient book”

This remains probable, but “almost certainly” was slightly too strong.

The text certainly invokes a sēfer. The separate-document interpretation is natural and widely held. But because the precise nature of the referent cannot be independently checked, probable is more defensible than virtually certain.

“We possess a few words from the Book”

This is rhetorically useful but slightly stronger than the evidence permits.

More precisely:

Numbers 21 presents at least some of the following wording as material said in the Book of Wars.

That does not prove our surviving wording exactly reproduces the source's original wording. Copying, editing and textual transmission could have altered it.

“The quotation is archaic”

This was too loose.

The text is obscure and has features that can be interpreted as archaic.

That is not identical to demonstrating that the poem belongs to archaic Hebrew linguistically.

Future work should not use “archaic” as a dating conclusion until multiple independent linguistic criteria converge.

“The Book probably contained victory songs”

This remains plausible-to-probable, but the evidence is thinner than our repeated use of the formulation suggested.

We securely know only the title and whatever portion of Numbers follows the citation.

The precise genre of that surviving piece is itself debated.

“Numbers 21 is a documentary dossier/archival montage”

This was our synthesis, not an established scholarly classification.

The repeated pattern of narrative claim → ʿal kēn → quotation makes the idea attractive, but “archive,” “dossier” and especially “land-title evidence” should remain interpretive metaphors unless parallels demonstrate an actual ancient genre.

“The Book may preserve Amorite poetry”

This is currently only plausible.

The Song of Heshbon's apparent celebration of Sihon's conquest creates that possibility, but we have not established that:

1. the poem is foreign;
2. it was contemporary with Sihon;
3. it belonged to the Book of Wars.

Those are three separate propositions that should never again be collapsed.

“The Book could be hidden among Qumran fragments”

This is technically possible but provides almost no evidentiary value.

There are hundreds of highly fragmentary Qumran manuscripts, but possibility created by incompleteness is not positive evidence. The known War Scroll belongs to a clearly different Second Temple literary setting. (DSS Collections)

Therefore this idea should remain peripheral unless a specific fragment displays diagnostic correspondence.

Upgrade / Downgrade Conditions

The most important claims can be made genuinely testable.

Separate lost document. Confidence would rise sharply if an ancient text outside Numbers quoted material explicitly attributed to sēfer milḥămōt YHWH. Confidence would decrease if early textual evidence showed ba-sēfer originally referred unmistakably to the Torah or another known composition.

Extended quotation through verse 20. Confidence would rise if an early manuscript, ancient translation or demonstrable poetic structure showed vv. 14–20 functioning continuously. It would fall if textual witnesses consistently marked a break after verse 15 or if Steiner's key grammatical analysis proves philologically untenable.

Steiner's “Come to Waheb.” Confidence would increase if the same imperative form of ʾ-t-y appeared in another securely interpreted Northwest Semitic text. It would decrease if early manuscript evidence revealed a different consonantal reading or a superior grammatical solution to the missing-verb problem.

Waheb's geographic identification. Confidence would increase dramatically with an inscription containing the place-name in a securely excavated Arnon-region context. Mere Arabic name resemblance should raise confidence only slightly. A candidate site geographically inconsistent with the poem would weaken the proposal sharply.

Early date of the underlying tradition. Confidence would increase if the poem contains a bundle of independently early linguistic features together with geography demonstrably obsolete by later periods. Confidence would decrease if linguistic features consistently fit monarchic or later Hebrew or if the geography maps better onto Iron Age political conditions.

Original Amorite Heshbon Song. Confidence would rise if linguistic or theological features proved non-Israelite and matched an independently attested Amorite cultural environment. It would fall if unmistakably Israelite theological vocabulary or later Hebrew idiom dominates the earliest recoverable form.

Territorial-legitimacy function. Confidence would increase if Numbers 21, Deuteronomy 2–3, Judges 11 and external Transjordanian evidence reveal a consistent tradition of invoking prior conquest to adjudicate legitimate ownership. It would decrease if the poems can be removed without affecting the territorial logic or if their original subjects prove unrelated to boundary claims.

Broader Israelite poetic anthology. Confidence would rise if the Well Song or Heshbon Song could independently be tied to the Book of Wars. It would fall if the quotation can be securely bounded at vv. 14–15 and that unit proves specifically martial.

Connection with Exodus 17:14. Confidence would require an actual textual bridge—shared formulae, attribution, manuscript evidence, or ancient tradition demonstrably older than the extant biblical texts. Thematic similarity by itself should no longer move the probability much.

Survival outside Numbers. Confidence would require a manuscript or identifiable quotation with distinctive wording and provenance. Until then, alleged modern “Books of the Wars of the Lord” should receive effectively zero historical weight.

What survives the audit most strongly

After deliberately stripping away accumulated inference, only a fairly small nucleus remains highly secure:

1. Numbers 21:14 explicitly refers to a written thing called sēfer milḥămōt YHWH. (Sefaria)
2. The wording immediately associated with that citation is unusually difficult, and ancient as well as modern interpreters disagree about its meaning. (TheTorah.com)
3. The passage concerns the Arnon/Moab frontier in its present literary context. (Cambridge University Press)
4. Numbers 21 contains additional poetic material, especially the Well Song and Heshbon Song, but their membership in the Book of Wars is unproven. (TheTorah.com)
5. The Mesha Stele independently demonstrates that Israel-Moab territorial conflict and divine attribution of military fortunes were real features of ninth-century BCE Transjordanian political culture. It does not authenticate the Book of Wars or date it. (Cambridge University Press)
6. Current archaeology does not independently verify the Sihon polity described by Numbers, but neither does it conclusively disprove it. (Cambridge University Press)
7. The Qumran War Scroll is a much later work with a substantially different setting and cannot responsibly be identified with the Numbers source on present evidence. (DSS Collections)

Everything beyond that is reconstruction at differing confidence levels.

The audit therefore makes the research project more interesting, not less. The secure foundation is narrow, but the problem is unusually tractable because several hypotheses make different predictions. The next research stage should not attempt to reconstruct the lost book wholesale. It should attack the narrowest decisive issue first: the textual history and exact quotation boundaries of Numbers 21:14–20. Until those are established more firmly, arguments about authorship, age, contents, Waheb, the Well Song, or additional surviving fragments remain downstream of unresolved textual questions.

Chronological Reconstruction: The Book of the Wars of YHWH

The chronology is more complicated than a normal “history of a book,” because at least five different timelines have to be kept apart:

1. the events or places remembered in Numbers 21;
2. the composition of the poems or traditions;
3. the compilation of whatever was called the Book of the Wars of YHWH;
4. the incorporation of material from that source into Numbers;
5. the later transmission and interpretation of Numbers 21.

Those stages may have been separated by centuries. The largest chronological mistake would be to assume that because a poem describes an event in Israel's wilderness-era narrative, the poem—or the Book of Wars itself—must have been written at that time.

Timeline

The central fact is immediately visible: only the later portion of this timeline is securely dated. The farther backward we move toward the supposed events behind the poem, the greater the uncertainty.

I. The Earliest Possible Layer: The Events Remembered

Numbers places the material during Israel's movement through Transjordan, shortly before entry into Canaan.

Its narrative world contains Moab, the Arnon boundary, Sihon and Heshbon, and conflicts involving Amorite control of territory formerly associated with Moab.

But there is no securely datable contemporary inscription saying:

“Sihon king of Heshbon fought Moab at this date.”

Nor has archaeology established a polity corresponding straightforwardly to the biblical Sihon kingdom.

Recent archaeology of west-central Jordan instead reveals a complicated late-second-millennium landscape of relatively small settlements and mobile/agropastoral populations whose political and ethnic identities are difficult to recover. (Cambridge University Press)

Therefore the earliest chronological layer should be formulated carefully:

There may be genuine late-second-millennium or early-Iron-Age memories behind Numbers 21, but their date cannot presently be established independently.

That is considerably weaker than saying that the Book of Wars originated “in Moses' time.”

II. Possible Early Oral or Poetic Tradition

A historically plausible mechanism would be:

event

→ oral commemoration

→ repeated performance

→ poetic stabilization

→ writing

→ anthology

→ incorporation into Numbers.

But that sequence is a model, not something directly documented for this particular poem.

Ancient societies certainly preserved victories, borders, rulers and disasters through poetry, monuments and oral traditions. The question is whether this specific material traveled that path.

One reason scholars have suspected older material is the unusual and compressed language of Numbers 21.

But our evidentiary audit showed why this cannot itself date the poem.

Obscurity may result from:

• antiquity;
• dialect;
• poetic compression;
• damaged transmission;
• unfamiliar geography;
• or later archaism.

Therefore we should resist a simple equation:

difficult Hebrew = early Hebrew = Mosaic date.

That chronology is not demonstrated.

III. A Potential Missing Stage: When Was the Book of Wars Compiled?

This is arguably the largest chronological blank in the entire investigation.

The book itself must, under the most natural reading of Numbers 21:14, precede the literary stage in which Numbers quotes it.

But how much earlier?

Potentially only decades.

Potentially centuries.

Suppose, hypothetically:

a battle song originated around 1150 BCE;

was transmitted orally;

was written around 950 BCE;

entered an anthology around 850 BCE;

and was quoted by an editor around 450 BCE.

All five statements could simultaneously be true.

Thus asking:

“When was the Book of Wars written?”

may be too simplistic.

A compilation might have had a growth history.

That possibility should become central going forward.

IV. The Ninth Century BCE: Our First Hard External Anchor

Here the chronology becomes dramatically firmer.

During the Omride period, roughly the first half of the ninth century BCE, Israel exercised political control over territory east of the Jordan and north of the Arnon/Wadi Mujib.

The Mesha Stele, composed in the ninth century, provides independent Moabite testimony.

Mesha says that Omri, king of Israel, oppressed Moab; that Omri controlled the land of Madaba; and that Mesha later recovered territory through the favor of Chemosh. (Cambridge University Press)

The Louvre dates the stele broadly to approximately 830–805 BCE, associating it with Mesha's reign and the recovery of Moabite territory from Israel. (Louvre Collections)

This artifact is critical because it independently places several ingredients of Numbers 21's conceptual world into a securely historical context:

Israel versus Moab

territorial possession north of the Arnon

national identity attached to territory

war understood as the activity of a national god

poetic/monumental commemoration of territorial victories

The inscription reports that Chemosh had been angry with his land during Israelite domination and later enabled Mesha's recovery. (Cambridge University Press)

This does not mean the Book of Wars was composed in the ninth century.

But it demonstrates that the sort of political theology implied by “YHWH's wars” fits securely within the cultural world of Iron Age Transjordan.

V. Could the Ninth Century Explain the Heshbon Tradition?

This now becomes one of our strongest chronological hypotheses.

Some scholars have proposed that the Heshbon/Sihon traditions were composed or reshaped during the period when Israel was actually competing with Moab over the plateau north of the Arnon.

The mechanism would be political memory.

If ninth-century Israelites controlled disputed Transjordanian territory, a tradition claiming:

Moab lost this territory to Sihon before Israel acquired it

could provide a powerful ancestral explanation for Israelite possession.

Recent archaeological-historical discussion explicitly notes the possibility that the Numbers material took literary form during or after Omride expansion into the mišor, the Transjordanian plateau. (Cambridge University Press)

The resulting ideological chronology could be:

older memory of conflict

→ ninth-century Israelite territorial expansion

→ renewed need to explain possession

→ Sihon tradition becomes politically useful

→ old song or newly composed archaizing song incorporated into Israelite memory.

This mechanism is plausible.

But chronology alone does not prove causation.

We would need linguistic or literary evidence connecting the poem specifically to that period.

VI. An Alternative: The Poems Are Older Than the Ninth-Century Conflict

The competing model must be treated equally seriously.

An old poem could have survived from the late second millennium or early Iron Age.

Later Israelites may then have reused genuine old poetry because it supported their territorial claims.

The causal mechanism would therefore be:

ancient poem survives

↓

later political problem arises

↓

scribes discover that inherited tradition legitimates the present situation

↓

poem gains renewed relevance

↓

eventually incorporated into Numbers.

This model has an attractive feature:

it explains why the wording became obscure.

But that is circumstantial reasoning.

We currently lack the linguistic chronological resolution necessary to choose confidently between:

old composition reused later

and

later composition written to sound old.

VII. The Formation of Numbers

We then enter another uncertain interval.

Modern Pentateuchal scholarship does not generally treat Numbers as a transcript produced during the wilderness events it narrates.

Its literary formation is widely understood as layered.

Recent scholarship continues to place substantial composition or final shaping of Numbers in the Persian period, broadly the late sixth through fourth centuries BCE, although scholars disagree intensely about individual sources, editorial stages, and how much older material was incorporated. (Cambridge University Press)

Baruch Levine likewise argued that Priestly material in its final form belongs to the post-exilic period. (TheTorah.com)

This gives us an important chronological ceiling.

By the time Numbers 21 reached the form in which it explicitly says:

“Therefore it is said in the Book of the Wars of YHWH…”

the source—or at least the tradition that this material came from that source—already existed.

Therefore:

Book of Wars tradition

must precede

the Numbers citation at that literary stage.

But how far it precedes it remains unknown.

VIII. What Might the Compiler of Numbers Have Been Doing?

If the final literary shaping is Persian-period or thereabouts, the editor may have been working with traditions already several centuries old.

This creates a potentially important analogy to historians quoting earlier sources.

The compiler may have had:

an actual scroll,

a copied excerpt,

an existing anthology,

an oral attribution,

or another text that itself quoted the Book of Wars.

We cannot tell which.

That makes direct transmission only one of several possibilities.

The chain could have been:

Book of Wars manuscript

→ Numbers compiler

or:

Book of Wars

→ intermediate anthology

→ Numbers compiler

or even:

Book of Wars

→ traditional quotation attribution

→ Numbers compiler.

Every additional intermediary increases the possibility of textual alteration.

IX. By the Hellenistic Period, the Torah Is Clearly Circulating

A major chronological benchmark arrives around 200 BCE.

The Jewish community of Alexandria possessed a Greek translation of the Torah by roughly this period; the Torah translation associated with the Septuagint arose in the Hellenistic Egyptian Jewish environment. (Cambridge University Press)

That means Numbers 21—and its Book of Wars citation—had entered a widely circulating Pentateuchal textual tradition by then.

Crucially, the Greek rendering of the difficult phrase does not straightforwardly preserve the later Masoretic interpretation.

This tells us something chronologically significant:

The difficulty in understanding Numbers 21:14 is ancient, not a problem first created by medieval or modern scholars.

By the Hellenistic period, translators may already have been dealing with words or place-names whose original meaning was obscure.

That provides a terminus ante quem for obscurity, though not for composition.

X. The Second Temple Period: The Lost Book Has Already Vanished from View

Qumran is chronologically important because it gives us actual ancient biblical manuscripts.

The Israel Antiquities Authority identifies 4Q27 / 4QNumbersᵇ as a Hebrew manuscript of Numbers written in a Herodian-period hand. (The Dead Sea Scrolls)

This pushes direct manuscript evidence for Numbers many centuries earlier than the great medieval Masoretic codices.

But no Qumran manuscript is presently identified as the Book of the Wars of YHWH itself.

That suggests—without proving—that the independent book may already have stopped circulating.

There is an important survivorship problem here.

A document could be:

written in 900 BCE,

cited in 500 BCE,

lost by 300 BCE,

yet survive indirectly forever because Numbers copied part of it.

That phenomenon is completely normal in ancient literary history.

XI. The War Scroll Appears—But Centuries Too Late

Qumran also produced the famous War Scroll, conventionally designated 1QM.

Its existence can produce false chronological associations.

It belongs to the Second Temple period and describes an eschatological conflict between the Sons of Light and Sons of Darkness.

It therefore comes from a world perhaps half a millennium or more after the likely formation of the relevant Numbers traditions.

The similarity is thematic:

God + warfare.

The chronology, genre and specific contents are radically different.

Consequently, any theory saying:

“The Qumran War Scroll is the Book of the Wars of YHWH”

faces an enormous chronological transmission problem.

One would need evidence that:

the ancient Book survived for centuries,

was transformed into apocalyptic literature,

retained some identifiable textual continuity,

and eventually produced 1QM.

We possess no such chain.

So chronology actively weakens that identification.

XII. Early Jewish and Christian Translators Preserve Increasing Divergence

During late antiquity, additional versions and interpretations appear.

The Syriac Peshitta, Aramaic Targum traditions and Latin Vulgate represent different attempts to make sense of Numbers 21.

Several associate the difficult wording with the Sea of Reeds, divine miracles, flames, or other interpretations that differ from modern geographical reconstructions.

Chronologically, the most important conclusion is not that one of these must preserve the original interpretation.

It is:

knowledge of Waheb/Suphah and the original context had probably ceased to be transparent long before medieval exegesis.

Transmission continued.

Contextual memory apparently did not.

That is a major distinction.

XIII. The Masoretic Stage

The surviving Masoretic codices belong much later.

The Masoretic tradition preserves an older consonantal text, but its vowels, accents, and transmission apparatus were systematized by medieval Masoretes. Cambridge's overview notes major surviving Masoretic codices from roughly the ninth and tenth centuries CE while emphasizing that their consonantal tradition rests on much earlier manuscripts. (Cambridge University Press)

Therefore the vowel pointing of:

אֶת־וָהֵב בְּסוּפָה

is itself part of a later interpretive tradition.

This matters enormously for Steiner.

If ʾet originally functioned as an archaic verb “come,” later scribes could naturally have vocalized it as the familiar object marker after the old grammatical construction became incomprehensible.

That is exactly the sort of mechanism Steiner's theory requires.

It remains hypothetical—but chronologically possible.

XIV. Ibn Ezra: Twelfth-Century Recognition That the Book Was Lost

Abraham ibn Ezra, writing in the twelfth century, explicitly states that the Book of Wars:

was a separate book.

He further observes that many ancient works had disappeared and speculates that it might even have existed as early as Abraham. (Sefaria)

The first statement is a grammatical/exegetical judgment.

The Abrahamic dating is speculation.

That distinction matters.

Ibn Ezra lived roughly 1,500 years or more after the likely biblical editorial processes.

He therefore does not supply independent testimony that he had seen the lost book.

Instead, his commentary demonstrates that by the medieval period a careful Hebrew scholar could read Numbers and independently conclude:

“This refers to a lost separate work.”

That is valuable reception history, not manuscript evidence.

XV. Ramban: Thirteenth-Century Reconstruction of the Social World

Nahmanides/Ramban (1194–c.1270) took the reconstruction further.

He envisioned learned persons in ancient generations composing books of major wars, using poetic sayings to memorialize remarkable victories. (Sefaria)

He also connected this environment with the Heshbon poem.

Chronologically, Ramban is still far too late to function as evidence for how Bronze- or Iron-Age scribes actually operated.

But his interpretation anticipates a modern anthropological question:

What sort of people created and maintained battle poetry?

His answer—professional or learned composers—remains plausible but historically unverified.

XVI. 1868–1869: The Mesha Stele Changes the Investigation

The modern phase of the story begins dramatically.

The Mesha Stele was encountered at Dhiban in 1868; surviving fragments and records became central to modern study after the stone was broken. Contemporary scholarly notices describe the discovery and fragmentation, while Louvre records date associated fragments to discovery in 1869. (Louvre Collections)

Suddenly scholars possessed a ninth-century Moabite voice.

This was transformative because Moab was no longer known only through biblical literature.

The inscription independently mentioned:

Mesha,

Omri,

Israel,

Gad,

territories north of the Arnon,

and Chemosh's role in war.

The causal influence on scholarship is straightforward:

new primary evidence

→ better knowledge of Moabite language and political theology

→ improved comparative framework for biblical Transjordan traditions.

But the stele did not reveal the Book of Wars.

Its contribution is contextual.

XVII. Twentieth-Century Archaeology Makes the Chronology Harder, Not Easier

Early biblical archaeology sometimes hoped specific sites would straightforwardly verify particular biblical episodes.

More recent work is considerably more cautious.

Hisban/Heshbon and the broader region show occupation histories, but archaeology has not securely established the political kingdom of Sihon described in Numbers.

Modern synthesis emphasizes that the late-second-millennium settlement record is insufficient to assign confident political or ethnic identities to communities around the mišor. (Cambridge University Press)

This creates a chronological asymmetry:

Ninth century BCE

We have excellent external textual evidence: Mesha.

Earlier periods

We have archaeology but much less political identification.

Therefore ironically, the later period is much clearer historically than the supposed period of Sihon.

XVIII. 1947 onward: The Dead Sea Scrolls Transform Textual History

The Qumran discoveries provided ancient copies of biblical books, including Numbers.

They demonstrated directly that multiple biblical textual traditions circulated before the medieval Masoretic codices.

For this investigation, however, the most important negative result remains:

no recognized scroll entitled the Book of the Wars of YHWH emerged.

That does not prove the book was nonexistent.

If anything, it reminds us how drastically ancient literature disappeared.

The Book could have vanished centuries before the Qumran library existed.

XIX. 2020: Steiner Reopens the Chronology of Interpretation

Richard C. Steiner's 2020 article represents the most important recent philological development in our research.

He argues that the mysterious אֶת in Numbers 21:14 should be understood not as the familiar object marker but as an archaic imperative of the verb “come.”

Thus:

“Come to Waheb…”

He then argues that the lost-source quotation continues through verse 20, rather than ending around verse 15. (Scribd)

This theory has chronological implications.

If Steiner is right, the textual history might look something like:

archaic verbal form intelligible to original audience

↓

construction becomes obsolete

↓

later copyists preserve consonants faithfully

↓

meaning is forgotten

↓

Masoretic vocalization treats ʾet as familiar object marker

↓

ancient and medieval translations struggle with syntax

↓

modern comparative Semitic philology recovers the lost grammatical possibility.

That is an elegant historical model.

But it remains a reconstruction, not a demonstrated chain.

Earlier and Later Layers

We can now identify at least six possible strata.

Layer A — Events

Some underlying conflicts, boundaries, settlements or regional memories involving Moab and the northern plateau.

Date: uncertain; possibly late second millennium/early Iron Age.

Layer B — Poetry

Songs or sayings memorializing geography, travel or conflict.

Date: uncertain; could be early or significantly later.

Layer C — Compilation

One or more poems incorporated into something called sēfer milḥămōt YHWH.

Date: unknown.

Layer D — Biblical reuse

A Numbers compiler cites the source and embeds the material into a larger narrative about Israel's movement and territorial acquisition.

Date: final stages probably considerably later than the narrative setting; Persian-period editorial activity is a major scholarly model. (Cambridge University Press)

Layer E — Ancient reinterpretation

Greek, Aramaic, Syriac and Latin readers attempt to interpret increasingly opaque wording.

Layer F — Modern reconstruction

Epigraphy, archaeology and comparative Semitic linguistics reopen questions that ancient and medieval interpreters could not resolve.

This layered model is much more defensible than imagining a single author writing the Book of Wars and Numbers in one historical moment.

Transmission and Intermediaries

Three transmission mechanisms now appear historically plausible.

Oral-to-written transmission could explain archaic poetic features embedded in later prose. Songs could survive through repeated performance before being written.

Scribal anthology transmission could explain the Book of Wars itself. Existing poems may have been collected rather than composed specifically for the anthology.

Canonical incorporation then becomes a third preservation mechanism. Once an excerpt entered Numbers, the anthology could disappear while its quotation survived indefinitely.

The potential transmission chain therefore looks like:

historical memory → oral poetry → written poem → anthology → excerpt → Numbers → ancient translations → medieval Hebrew tradition → modern critical editions.

But only the latter portion of that chain is directly observable.

The earlier links remain hypotheses.

Chronological Problems

Several chronology tests now eliminate or weaken attractive theories.

The Book of Wars cannot be dated simply from the events it describes

A poem can be centuries younger than its purported subject.

The Mesha Stele cannot directly date the Book

It supplies a secure ninth-century analogy, not a quotation from the Book.

Linguistic difficulty cannot independently prove antiquity

Later archaism, corruption and dialect remain alternative explanations.

Numbers' final date cannot date its sources automatically

A Persian-period editor could quote a centuries-old text.

Qumran's War Scroll is chronologically remote

Without an intermediary textual chain, identification with the Book of Wars is extremely difficult.

Ibn Ezra and Ramban cannot supply ancient historical information merely because their interpretations are insightful

They reveal medieval reception.

The disappearance date is completely unknown

The Book might have vanished before the Septuagint, or centuries later. No surviving testimony tells us when the last independent copy existed.

That last gap is particularly striking.

We cannot presently construct:

Book written → copied until year X → disappears.

The middle of that history is entirely missing.

Revised Chronological Narrative

The most defensible narrative now looks like this.

At some uncertain point in the late second millennium or early first millennium BCE, communities east of the Jordan developed traditions about territory, battles, water routes and boundaries involving Moab, the Arnon region and Heshbon. Some of these traditions may preserve genuine early historical memories, but we cannot securely date them.

Over time, at least some material was cast into poetry or traditional sayings. Whether those compositions were contemporary with the events, later poetic memorializations, or still later political constructions is unresolved.

At another unknown stage, one or more such compositions became part of a written object remembered as:

ספר מלחמות יהוה — the Book of the Wars/Battles of YHWH.

Its date, compiler and institutional setting are unknown.

During the Iron Age—especially by the ninth century BCE—we can independently demonstrate that Israel and Moab contested the territory north of the Arnon and that both peoples interpreted military fortune through national deities. The Mesha Stele proves this cultural and political environment existed, although it does not establish the Book's date. (Cambridge University Press)

At some later point, probably within the long compositional history culminating in the Pentateuch's post-monarchic form, a compiler of Numbers embedded material attributed to the Book of Wars into Numbers 21. The quotation apparently helped explain or authenticate the Arnon boundary and may have participated in a broader argument concerning why Israel possessed territory formerly connected with Moab.

By the Hellenistic era, Numbers circulated as part of the Torah, and its Greek translators were already confronting a difficult text. By the Second Temple period, manuscript copies of Numbers existed at Qumran, while the independent Book of Wars is no longer visibly attested.

The original collection then disappears entirely from identifiable history.

Late-antique translators and medieval commentators continued trying to decipher the quotation. Ibn Ezra explicitly concluded that it came from a lost independent book; Ramban reconstructed a world of ancient battle-poets.

Modern discovery of the Mesha Stele in 1868 supplied the first major external window into the actual political theology of ancient Moab. Twentieth-century archaeology and the Dead Sea Scrolls added enormous contextual and textual data but still produced no identifiable Book of Wars.

Finally, modern philology has reopened the surviving fragment itself. Steiner's 2020 proposal raises the possibility that what scholars had long regarded as a short damaged fragment may actually be a coherent excerpt extending through Numbers 21:20.

So the chronological history is not:

Moses wrote book → Numbers quoted book → book was lost.

That is possible only as a traditional reconstruction and is unsupported by the presently available chronology.

The evidence-sensitive reconstruction is instead:

uncertain Transjordanian events/memories → uncertain poetic development → unknown stage of written collection → later biblical incorporation → early loss of the independent work → increasingly obscure textual transmission → medieval recognition of the lost source → modern archaeological and philological reconstruction.

And the largest chronological mystery remains surprisingly basic:

Between the earliest plausible origin of the traditions and the point at which Numbers cites them, we have almost no securely dated evidence at all.

That roughly several-century “dark zone” is where the identity and history of the Book of the Wars of YHWH must ultimately be recovered—or where the limits of recovery may prove permanent.

Meaningful Connections Across the Investigation

A connection audit produces several genuinely important links that deserve to be separated from looser resemblance. The strongest new pattern is not a hidden identity between texts, but a regional literary-cultural network in Transjordan: Numbers 21–24 contains old-looking poetry, named or quasi-named textual authorities, boundary traditions, foreign diviners, and divine-war ideology; external Iron Age inscriptions from the same broader region independently attest several of those cultural ingredients.

The key is to distinguish three levels:

• Direct connections: explicit textual, geographic, or epigraphic links.
• Structural connections: recurring mechanisms that may reflect shared cultural practice.
• Speculative connections: patterns intriguing enough to test, but not strong enough to infer common origin.

Direct Connections

1. “The Wars of YHWH” is not an isolated biblical phrase

One of the most important connections not previously emphasized is linguistic.

Numbers 21:14 contains:

מִלְחֲמֹת יְהוָה — milḥămōt YHWH “the wars/battles of YHWH.”

Very similar language occurs twice in the David narratives.

Saul tells David:

“fight the battles of YHWH”

in 1 Samuel 18:17, and Abigail tells David that YHWH will establish his house because he is fighting “the battles of YHWH” in 1 Samuel 25:28. The Hebrew in the latter explicitly contains milḥămōt YHWH. (Bible Hub)

This is a direct lexical connection.

It matters because it shows that milḥămōt YHWH was an intelligible Israelite expression independent of Numbers 21. It could denote ordinary human warfare interpreted as warfare undertaken on YHWH's behalf.

That slightly changes our understanding of the title.

The phrase need not imply that the lost book contained supernatural battles in which YHWH personally appeared as combatant. It could also mean:

accounts or songs about wars understood as YHWH's cause.

What it does not prove

It does not prove that 1 Samuel knew the Book of Wars.

It does not prove a Davidic date for the collection.

And it does not prove literary dependence in either direction.

Three explanations remain:

1. the phrase was a conventional Israelite idiom;
2. Samuel and the Book of Wars drew on the same older ideological vocabulary;
3. one literary tradition influenced the other.

At present, shared conventional vocabulary is the safest explanation.

2. Numbers 21 and 1 Samuel reveal the same political theology

There is more than vocabulary here.

In 1 Samuel, David's military activity becomes legitimate precisely because it constitutes YHWH's battles. His military success is then linked with his future dynasty: Abigail connects his fighting YHWH's battles with YHWH establishing an enduring house for him. (Bible Gateway)

That produces a structure:

divine cause

→ human warrior

→ victory

→ political legitimacy.

Numbers 21 potentially operates similarly:

YHWH-authorized territorial movement

→ victory over Sihon

→ acquisition of land

→ territorial legitimacy.

This suggests that “wars of YHWH” may participate in an Israelite ideological mechanism for converting successful warfare into legitimate political possession.

That is stronger than merely calling YHWH a “warrior god.”

3. The Heshbon Song and the Mesha Stele independently identify Chemosh with Moab

Numbers 21:29 describes the Moabites as:

“the people of Chemosh.”

The ninth-century BCE Mesha inscription independently presents Chemosh as Moab's national deity, repeatedly crediting him with Moab's defeat, restoration, territorial commands, and victories over Israel. (Bible Gateway)

This is a direct and significant cultural connection.

The Heshbon poem says, in effect:

Chemosh failed to protect his people against Sihon.

The Mesha Stele says:

Chemosh permitted Israel to oppress Moab because he was angry, then ordered Mesha to retake Moabite territory.

The shared theology is striking:

military fortunes reveal the condition of the relationship between a people and its god.

That conceptual structure is externally documented.

What this connection establishes

It strongly supports the claim that Numbers 21's description of Moab as the people of Chemosh reflects authentic knowledge of Moabite religion.

What it cannot establish

It cannot date the Heshbon poem to the purported Sihon period.

An Israelite author in the ninth, eighth, seventh, or later centuries could also know perfectly well that Chemosh was Moab's god.

So this is strong evidence for cultural accuracy, but weak evidence for early chronology.

4. The Book of Wars and the Heshbon Song may participate in the same boundary argument

Numbers 21 contains two unusually explicit acts of citation.

First:

the Arnon is the boundary between Moab and the Amorites;

therefore the Book of YHWH's Battles is quoted.

Later:

Sihon had captured this land from Moab;

therefore the mōšəlîm are quoted.

Greenstein specifically interprets the first quotation as attesting the Arnon boundary and the second as corroborating Sihon's earlier conquest. (TheTorah.com)

Recent archaeological-historical analysis likewise notes that the Heshbon Song could function to certify that Moab had lost territory north of the Arnon before Israel obtained it. (Cambridge University Press)

This is a strong structural connection because the rhetorical pattern repeats:

prose territorial assertion

→ “therefore”

→ traditional poetic authority.

That makes the idea that Numbers 21 uses inherited poetry as evidentiary support considerably stronger than it initially appeared.

5. Judges 11 preserves essentially the same territorial logic

A major connection not yet adequately developed is Judges 11.

In Jephthah's dispute with the Ammonite king, the argument concerns who legitimately owns Transjordanian territory.

Jephthah's defense depends on the same essential historical sequence:

Israel did not seize Moabite or Ammonite land directly;

Israel encountered Sihon;

Sihon fought Israel;

Israel defeated him;

Israel therefore took possession of the territory Sihon controlled.

This is closely related to the logic of Numbers 21 and Deuteronomy 2.

That means we do not have an isolated territorial explanation.

We have a biblical tradition complex.

Numbers 21 ↕ Deuteronomy 2 ↕ Judges 11

all preserve versions of the Sihon intermediary argument.

This substantially increases the importance of the tradition.

Alternative explanations

This could represent:

Common historical memory — several texts preserve an old remembered territorial sequence.

Literary dependence — one biblical tradition borrows from another.

Shared scribal archive — multiple authors use a common historical tradition.

Later ideological harmonization — traditions were shaped to justify Israelite claims retrospectively.

Chronology and literary dependence must be investigated before choosing among them.

6. The geography is not incidental: it is a contested frontier system

The lands between the Arnon/Wadi Mujib and Jabbok/Wadi Zarqa repeatedly change political affiliation in biblical and external evidence.

The relevant actors eventually include:

• Moabites;
• Israelites;
• Ammonites;
• Arameans;
• Assyrians;
• and, in the biblical remembered past, Amorites.

Frontier studies emphasize exactly this characteristic: multiple groups compete over the same region and generate competing historical narratives of legitimate possession. (Bible Interp.)

This geographical fact helps explain why old songs and boundary traditions would acquire unusual importance.

In a stable heartland, an old poem about a river boundary may be antiquarian.

In a contested frontier, it can become politically useful.

Thus geography provides a plausible mechanism linking poetry to territorial ideology.

Structural Parallels

7. The Heshbon Song and Mesha Stele share the structure of victory literature

Consider the Heshbon poem:

Heshbon is exalted.

Fire goes out from the victor's city.

Enemy territory is consumed.

The enemy god fails.

Captives are taken.

Territorial control changes.

Now consider the Mesha Stele:

Mesha identifies himself and Chemosh.

Israel's domination is explained.

Chemosh authorizes retaliation.

Cities are attacked and taken.

Populations and sacred objects are captured.

Territory is restored.

The two texts are not the same genre exactly—one is poetry embedded in narrative, the other monumental royal inscription—but they share the broader Levantine grammar of victory.

It consists roughly of:

deity → chosen military agent → enemy defeat → territorial transfer → ideological explanation.

The Mesha inscription provides independent evidence that this mechanism was historically real in the region. (Archaeology)

This makes a collection called YHWH's Battles culturally quite unsurprising.

8. Defeat of a nation is also defeat—or withdrawal—of its god

The Heshbon poem says:

“Woe to you, Moab… people of Chemosh.”

Chemosh has effectively allowed his people to become fugitives and captives. (Bible Gateway)

Mesha uses almost the inverse logic:

Israel dominated Moab because Chemosh was angry with his land; Moab recovered because Chemosh restored favor and issued military commands. (Archaeology)

This suggests an important structural rule:

Ancient territorial theology could explain both victory and defeat without abandoning the national god.

Victory:

our god defeated theirs.

Defeat:

our god was angry and temporarily permitted defeat.

That theological flexibility helps explain how national gods survived military catastrophe.

It is relevant to YHWH traditions as well: biblical texts frequently explain Israelite defeat not as YHWH's weakness but as YHWH's judgment.

This is therefore likely regional convergence/common cultural grammar, not proof of direct borrowing.

9. “Book” culture in Transjordan: the Deir ʿAlla connection

This may be the most unexpected new connection of the entire investigation.

Immediately after Numbers 21, with its:

“Book (sēfer) of YHWH's Battles”

Numbers 22–24 introduces Balaam son of Beor.

Centuries later, an inscription discovered at Tell Deir ʿAlla in the eastern Jordan Valley names:

Balaam son of Beor

and apparently opens by referring to a “book/account (spr) of Balaam son of Beor.”

The inscription dates around the late ninth/eighth century BCE and reflects a Transjordanian literary tradition independent of Numbers. (The BAS Library)

Even more interestingly, Martti Nissinen notes that the plaster composition apparently presents itself as derived from a “Book of Balaam son of Beor,” with spr referring to a scroll rather than the plaster surface itself. (Peach Foundation)

This gives us an actual Iron Age regional example of something remarkably close in form to what Numbers 21 implies:

a named literary figure/tradition

↓

a spr / written source

↓

material reproduced in another physical setting.

This does NOT mean

the Book of Balaam and Book of Wars were related;

the same scribal school produced them;

Balaam authored the Book of Wars;

or Numbers copied from Deir ʿAlla.

There is no evidence for any of those claims.

But it does demonstrate something important

The hypothetical cultural world needed for the Book of Wars—a world in which named literary collections or scrolls circulated in Transjordanian culture—is not purely theoretical.

We have archaeological evidence for a roughly analogous phenomenon.

That upgrades the general scribal-cultural plausibility of the Book of Wars.

10. Numbers 21–24 may be a remarkable Transjordanian literary cluster

Once the Deir ʿAlla connection is noticed, the structure of Numbers becomes intriguing.

Numbers 21 contains:

• Book of YHWH's Battles;
• Arnon geography;
• Well Song;
• mōšəlîm;
• Heshbon poem;
• Moab;
• Chemosh.

Numbers 22–24 immediately contains:

• Moab;
• Balak;
• Balaam;
• repeated poetic oracles;
• Transjordanian geography;
• a foreign religious specialist;
• unusual archaizing poetry.

And the Balaam poems themselves are repeatedly called māšāl—the same root m-š-l underlying the mōšəlîm, the “poets/balladeers,” in Numbers 21:27. The Deir ʿAlla inscription independently attests a Balaam son of Beor as a celebrated Transjordanian seer. (Zenodo)

This raises a highly valuable new question:

Do Numbers 21–24 preserve a consciously assembled cluster of Transjordanian traditional literature?

That could include:

battle poetry,

boundary traditions,

well poetry,

foreign oracles,

and regional prophetic lore.

This would be far more interesting than imagining one “Book of Wars” behind everything.

Evidence status

Structural connection: strong.

Common source: unverified.

The contiguous placement could result simply from narrative geography: these stories all occur as Israel approaches Moab.

So two explanations compete:

Narrative geography explains the cluster naturally.

versus

The compiler also deliberately gathered inherited Transjordanian literary materials here.

Both may be true.

11. Mōšəlîm and Balaam's māšāl: possible literary-social connection

This deserves its own treatment.

Numbers 21:27 introduces:

הַמֹּשְׁלִים — hammōšəlîm

usually translated poets, ballad singers, or proverb makers.

In Numbers 23–24, Balaam repeatedly:

וַיִּשָּׂא מְשָׁלוֹ — “takes up his māšāl.”

The words derive from the same Hebrew root.

This could indicate a shared category of elevated traditional speech:

• crafted proverb;
• taunt;
• poetic oracle;
• formal saying.

It suggests that mōšəlîm need not mean “authors” in our modern sense. They may have been performers or specialists in authoritative poetic speech.

Balaam gives us an example—within the very next narrative section—of a person producing precisely this kind of māšāl.

That doesn't prove the mōšəlîm were prophets like Balaam.

But it provides a better internal biblical model for what they might do.

12. Deir ʿAlla suggests poetry/oracle traditions could cross ethnic boundaries

The Balaam inscription's language is notoriously difficult to classify: scholars have proposed Aramaic, Canaanite, a local dialect, or another Northwest Semitic classification. The site also sat in a politically fluid region influenced by surrounding kingdoms. (Zenodo)

This is highly relevant to the Heshbon Song problem.

We have often asked:

“Is this poem Israelite or Amorite?”

But the regional evidence suggests the question may be too binary.

Frontier literature could circulate among:

scribes,

sanctuaries,

courts,

travelers,

tribal groups,

and multilingual populations.

A poem could originate in one political setting and be transmitted, adapted, and linguistically reshaped by another.

Thus:

foreign origin

and

Israelite textual form

are not mutually exclusive.

This provides an actual transmission mechanism for the otherwise puzzling idea that Israel preserved an enemy victory song.

Geographic and Network Connections

13. The King's Highway and Transjordanian movement

Numbers' Sihon narrative revolves around Israel requesting passage through his territory along a recognized route. Deuteronomy contains a parallel request and also places Heshbon on the Transjordanian movement corridor. (Bible Hub)

This matters beyond military strategy.

Major routes move more than armies.

They move:

• merchants;
• scribes;
• diplomatic messengers;
• religious specialists;
• songs;
• place names;
• political rumors;
• ritual practices.

Thus a regional literary tradition does not require direct state-to-state literary borrowing.

A mobile border society provides numerous low-level transmission channels.

That makes cross-cultural preservation of songs and figures much more plausible.

14. Water routes may connect the geography and literary preservation

If Steiner's hydrological interpretation is even partly correct, Numbers 21:14–20 may encode unusually specific knowledge of:

storms,

wadis,

watersheds,

wells,

and travel routes.

That would make the poem not merely a battle song but a kind of mnemonic landscape text.

In oral or semi-literate societies, verse can function as durable storage for:

routes,

water sources,

boundaries,

genealogies,

and landmarks.

This offers a possible mechanism explaining why geography appears in poetry.

Rather than asking:

“Why would a war book contain a description of streams?”

we might ask:

Did early traditional poetry preserve strategic geography precisely because routes and water sources mattered to warfare and travel?

That possibility deserves testing ethnographically and comparatively.

It could reconcile the “war collection” hypothesis with the apparently geographical content.

Linguistic / Conceptual Connections

15. Sēfer in Numbers and spr at Deir ʿAlla

Hebrew ספר (sēfer) and the cognate Northwest Semitic spr belong to the same widely attested scribal semantic field.

Numbers says:

bə-sēfer milḥămōt YHWH “in the book/document of YHWH's battles.”

The Deir ʿAlla plaster text apparently refers to:

spr blʿm br bʿr a “book/account of Balaam son of Beor.” (Peach Foundation)

This is a genuine linguistic-cultural connection.

It demonstrates that spr terminology was used for literary documents in precisely the broader Northwest Semitic cultural world relevant to Numbers.

It does not show dependence.

But it makes the notion of identifiable written collections considerably more concrete.

16. “Wars of YHWH” may be an institutional category rather than only theology

Because 1 Samuel uses the precise phrase for David's military service, milḥămōt YHWH may have functioned as a recognized ideological category:

legitimate national warfare undertaken in the deity's cause.

If so, Book of the Wars of YHWH might mean something closer to:

“Record/collection of YHWH's campaigns”

than:

“Book describing supernatural battles fought personally by God.”

This semantic connection pushes slightly toward a historical-national interpretation of the title.

But the title problem remains: ancient texts may be known from opening words rather than descriptive titles, so this cannot settle the genre. (TheTorah.com)

17. “Come to Waheb” and “Come to Heshbon”

Steiner's interpretation creates a potentially meaningful poetic formula.

Numbers 21:14:

“Come to Waheb…”

under Steiner's reading.

Numbers 21:27:

“Come to Heshbon…”

The closeness of these formulas within the same chapter is unlikely to be meaningless if Steiner's reading is correct. Greenstein notes this explicitly. (TheTorah.com)

Possible explanations include:

Genre formula: ancient travel/victory songs commonly began with invitations to approach a place.

Shared composer or collection: both poems came from the same literary environment.

Editorial imitation: one passage influenced the formulation of another.

Coincidence: imperative “come” is common enough that similarity alone cannot establish dependence.

The first explanation currently requires the least unsupported inference.

18. Book of Wars and Book of Jashar: the connection is institutional, not identity

The productive question is no longer:

“Were they the same book?”

There is insufficient evidence for that.

The better connection is that both appear as named written authorities embedded inside later biblical prose.

Book of Wars:

→ Numbers 21.

Book of Jashar:

→ Joshua 10 and 2 Samuel 1.

The Book of Jashar quotations are associated with poetic material, and scholarship has long inferred that it may have been a gradually accumulated national poetic collection. (Bible Gateway)

Together they imply something larger:

The canonical Hebrew Bible may preserve excerpts from a broader pre-canonical or extra-canonical literary ecosystem.

This is one of the strongest structural conclusions of the entire investigation.

The Hebrew Bible may represent only the surviving visible part of a much larger textual tradition.

Unexpected Connections

19. The “Book of Balaam” may provide an archaeological analogue for how the Book of Wars disappeared

This deserves emphasis.

At Deir ʿAlla, a literary tradition associated with Balaam appears physically on plaster, yet scholars infer that spr refers to a prior scroll/source rather than the wall itself. (Peach Foundation)

Imagine that the underlying scroll vanished.

We would still possess:

an excerpt/copy embedded on another surface.

That is remarkably similar to the situation we hypothesize for Numbers:

Book of Wars disappears

but

a quotation survives because another composition copied it.

This gives us a concrete ancient mechanism for textual survival by secondary embedding.

It does not prove that this happened to the Book of Wars.

But it demonstrates that such a preservation pathway was historically possible in the region.

That is one of the most evidentially useful analogies we've encountered.

20. The lost book may be less “lost” than redistributed

This follows from the previous point.

When we ask whether the Book of Wars can be “found,” we instinctively imagine discovering a complete scroll.

But ancient collections can disappear while their constituent material survives elsewhere.

Possible transmission:

Poem A Poem B Poem C

↓

compiled into Book of Wars

↓

later biblical authors extract A

↓

original collection disappears

↓

Poem A survives inside Numbers.

Poems B and C might likewise survive elsewhere without attribution.

Thus the research problem becomes computationally and philologically interesting:

Can we identify an intertextual fingerprint strongly enough to detect unlabelled sister fragments?

Possible features might include:

• rare formulas;
• distinctive geographic vocabulary;
• particular poetic syntax;
• divine-war terminology;
• unusual māšāl structures;
• Transjordanian toponyms.

But this must be done only after diagnostic traits are established independently, otherwise it degenerates into circular pattern hunting.

21. Numbers 21–24 and Deir ʿAlla suggest a “Transjordanian memory zone”

Several otherwise separate observations now converge geographically:

Arnon boundary poetry

Heshbon/Moab traditions

Chemosh

Balaam

māšāl poetry

Deir ʿAlla Balaam tradition

fluid Northwest Semitic dialects

contested political frontiers

written spr traditions.

It may therefore be fruitful to treat central Transjordan not merely as a backdrop but as a cultural memory zone.

Such a zone could preserve traditions shared across later national boundaries.

That could explain why biblical texts contain material that appears simultaneously:

Israelite,

Moabite-aware,

foreign in voice,

and difficult to assign ethnically.

This is not evidence for a single archive.

It is a regional-network hypothesis.

But unlike many speculative connections, it has a plausible mechanism:

geographic proximity + political turnover + travel routes + multilingual populations + cultic centers + scribal activity.

22. The frontier may explain contradictory ownership traditions

Numbers depicts the land as:

Moabite → Amorite → Israelite.

Judges introduces Ammonite claims.

Mesha later describes Moab recovering territory from Israel.

Archaeological frontier research emphasizes repeated political turnover in this region. (Bible Interp.)

This means contradictory territorial traditions need not imply that one side simply fabricated everything.

Different communities can retain accurate memories from different chronological snapshots.

For example:

At time A: Moab controls area.

At time B: another polity controls it.

At time C: Israel controls it.

At time D: Moab reclaims it.

Centuries later, each group's tradition says:

“This land is ours.”

All can appeal to genuine historical periods.

This is especially important for interpreting Numbers 21's territorial logic.

23. The most surprising theological inversion: Numbers preserves a song about another god's failure

If the Heshbon poem originated outside Israel, an Israelite text has preserved a foreign political theology in which:

Chemosh's people are defeated because Chemosh gives them over to Sihon.

The Mesha Stele later presents essentially the same theological mechanism from the Moabite side:

Chemosh allows foreign oppression when angry, then reverses it.

This raises the possibility that the Heshbon Song preserves genuinely Moabite-compatible religious thinking even if the surviving text passed through Israelite hands.

That would make the poem more than a source about Sihon.

It could preserve a fragment of West Semitic interstate theological rhetoric.

Again, that does not prove Moabite authorship.

Shared religious assumptions could explain the same structure independently.

Alternative Explanations for the Major Patterns

The emerging network must be tested against competing explanations.

Connections That Should Be Resisted

Several attractive ideas remain too weak.

Book of Wars = Book of Jashar. No direct evidence.

Book of Wars = Deir ʿAlla Book of Balaam. No evidence whatsoever.

Balaam wrote the Book of Wars. Unsupported.

The Heshbon poem proves an Amorite archive existed. No.

The Mesha Stele proves Numbers 21 dates to the ninth century BCE. No; it proves cultural and geopolitical comparability.

The Deir ʿAlla Balaam proves the biblical Balaam narrative happened as written. No. It independently establishes a regional tradition concerning Balaam son of Beor, not the events of Numbers 22–24. (The BAS Library)

Shared phrases prove direct literary borrowing. Not without chronology and a plausible transmission route.

The Most Important New Connection

After comparing everything, the strongest newly visible pattern is this:

Numbers 21–24 may preserve a concentrated block of Transjordanian traditional literature situated within a real Iron Age cultural environment where named written sources, poetic specialists, prophetic figures, contested territorial memories, national gods, and cross-border literary transmission are independently plausible—and in several cases independently attested.

That is much more defensible than claiming we have identified the lost Book.

And it suggests a new research direction that may be more fruitful than continuing to study Numbers 21:14 in isolation.

The key comparison should now become:

Numbers 21–24 as an entire Transjordanian literary complex

against

Deir ʿAlla + Mesha + other Transjordanian inscriptions + Deuteronomy 2–3 + Judges 11.

If that comparison reveals distinctive shared vocabulary, geographic perspective, poetic forms, or ideological structures that cannot easily be explained by ordinary biblical Hebrew, we may begin to reconstruct something much larger than the lost book itself:

the literary world from which the Book of the Wars of YHWH emerged.

That may ultimately be the more recoverable object.

Anomaly Audit: The Book of the Wars of YHWH

The strongest way to test the emerging reconstruction is to look precisely at what resists it.

The leading working model at this point is roughly:

Numbers 21 preserves one or more older Transjordanian poetic traditions, at least one of which was attributed to a written source called the Book of the Wars of YHWH. Those traditions were later incorporated into Numbers, possibly because they supported geographical or territorial claims.

That model is plausible. But several anomalies could materially alter it.

1. The opening quotation is almost syntactically broken

The most serious textual anomaly is still the phrase:

אֶת־וָהֵב בְּסוּפָה

On the ordinary grammatical reading, ʾet is the direct-object marker. But then there is no obvious verb governing “Waheb” and the following geographical expressions.

This is not a minor stylistic oddity. Major twentieth-century commentators described the passage as fragmentary, corrupt, or essentially unintelligible. Steiner's 2020 article was motivated specifically by this problem. (Lockwood Online Journals)

Why this does not fit neatly

If Numbers is deliberately citing a useful authoritative source, it is strange that the cited proof begins with wording that appears grammatically defective.

That raises competing possibilities.

The quotation was already fragmentary when incorporated. The compiler may have possessed only an excerpt.

The text later suffered corruption. The original quotation made sense, but one or more words were lost or altered.

The grammar is archaic. Steiner's interpretation of ʾet as an archaic imperative “Come!” would remove the anomaly without changing the consonants.

The place names became unintelligible. Grammatically normal poetry may have become opaque once its geography was forgotten.

The passage was never supposed to be ordinary prose syntax. Very compressed poetic diction might explain some—but probably not all—of the problem.

Importance

Rank: #1.

This anomaly can change both the extent of the lost quotation and our judgment about its age and transmission.

2. The ancient translators disagree dramatically

The Masoretic reading is only one ancient trajectory.

The Samaritan tradition, Septuagint, Vulgate and other ancient versions produce sharply different understandings around Waheb/Suphah. The NET textual notes summarize the variation: the Samaritan Pentateuch connects the phrase with the Sea of Reeds, Greek presents something like a burning of Zoob, while Latin-influenced readings produced still other interpretations. (NETBible)

Why this matters

If the text were still transparent in the Hellenistic and early Roman periods, we would expect ancient translators who knew Hebrew and regional traditions to converge more closely.

Instead, they diverge.

That is negative evidence for continued comprehension.

It suggests that by the time our earliest translations were made, at least one of these things had happened:

• the geographical knowledge was gone;
• the grammar was archaic;
• their Hebrew exemplars differed;
• the text was damaged;
• or interpreters were already guessing.

An especially important implication

Steiner's theory explains this elegantly:

an old verbal form ceased to be understood,

but the consonants survived.

However, that elegance is itself a danger. A theory should not be preferred simply because it explains multiple anomalies neatly.

Importance

Rank: #2.

Ancient disagreement is one of the strongest constraints on any reconstruction.

3. Why would a “Book of Wars” contain what looks like a hydrological itinerary?

This is a conceptual anomaly.

The secure quotation appears to concern:

• Waheb;
• streams/wadis;
• the Arnon;
• slopes or drainage;
• Moab's boundary.

Under Steiner's larger reconstruction, it may continue through:

• a well;
• Mattanah;
• Nahaliel;
• Bamoth;
• Pisgah.

That looks surprisingly little like a conventional collection of heroic battles. (Lockwood Online Journals)

Possible resolutions

“Wars of YHWH” was not a descriptive title. It may have been the conventional name of a broader anthology.

Military literature naturally contained strategic geography. Water sources, wadis, approaches, boundaries and routes are directly relevant to campaigning.

The passage was extracted for geographical relevance from a longer battle narrative. The actual combat context may have been omitted.

Steiner's extended quotation is wrong. If the source ends at verse 15, the genre problem becomes smaller.

The source was not primarily martial at all. Our inference from the title may simply be mistaken.

Why it matters

This anomaly bears directly on the central assumption that the lost work was a war-song anthology.

Importance

Rank: #4.

4. The quotation does not actually describe a recognizable war

This is even more specific.

The text explicitly says Book of the Wars of YHWH, yet the portion immediately following the citation does not clearly contain:

• an enemy;
• a battle;
• a king;
• an Israelite army;
• a victory;
• or even an explicit act of YHWH.

That absence is surprising.

Negative evidence

If we possessed an ordinary excerpt from a document devoted exclusively to “YHWH's wars,” we might reasonably expect some explicit military content.

We apparently do not.

Possible explanations

The quotation begins or ends in the middle of a military poem.

The named places encoded campaigns recognizable to the original audience.

The source's title was broader or conventional.

The words “wars of YHWH” referred to Israel's wilderness progress broadly, not individual combats.

Or the lost citation boundary has been reconstructed incorrectly.

Importance

Rank: #5.

It reinforces the genre problem independently of the Song of the Well.

5. The Heshbon poem celebrates Sihon more naturally than Israel

This is one of the most important historical anomalies.

Numbers 21:27–30 introduces a poem beginning:

“Come to Heshbon…”

and describes fire coming from Sihon's city and consuming Moab.

The poem's natural historical subject seems to be Sihon's conquest of Moab, not Israel's conquest of Sihon. (Posen Library)

Why this is awkward

The surrounding biblical narrative is about Israel defeating Sihon.

Yet the text pauses to preserve poetry celebrating Sihon's earlier victory.

Why preserve enemy propaganda?

Possible resolutions

Original Amorite victory song. Israel inherited or appropriated foreign poetry.

Israelite ironic reuse. The song is quoted because Sihon's former glory makes his subsequent defeat more humiliating.

Historical proof. The compiler cares less about celebrating Sihon than about establishing that Sihon had already taken the disputed territory from Moab.

Later literary composition. An Israelite poet wrote retrospectively in Sihon's voice to provide narrative background.

The third solution currently fits the surrounding territorial logic especially well, but it does not determine the poem's origin.

Importance

Rank: #3.

If foreign origin could be established, it would be extraordinarily significant evidence for cross-cultural textual preservation.

6. Verse 30 may not fit neatly with the rest of the Heshbon Song

The poem's voice is not entirely simple.

Verses 27–29 celebrate Sihon's destruction of Moab.

Verse 30 shifts toward language of devastation:

“We have shot them down...” / “We devastated…”

depending on textual reconstruction.

This raises an old problem:

Who is “we”?

If the poem is Amorite, “we” might be Sihon's forces.

If it has been reused by Israelites, verse 30 might represent an Israelite adaptation or continuation.

If the text suffered corruption, original voice boundaries may no longer be recoverable.

Possible layered solution

An older Amorite poem may have been updated by Israelite singers.

That would give us:

original foreign victory poem

→ Israelite continuation/reversal

→ biblical embedding.

This possibility has major implications for how ancient poetry was transmitted: compositions may not have remained fixed.

Importance

Rank: #7.

Potential evidence of textual layering, but difficult to establish securely.

7. The archaeology does not give us the Sihon kingdom we might expect

If Numbers' account were straightforward political reportage of a substantial late-second-millennium kingdom centered on Heshbon, we might hope to find a corresponding archaeological signature.

We do not.

Recent synthesis notes that earlier attempts to identify defensive installations with Sihon's polity failed; those structures proved much later. Excavations at Hisban do show late-second-millennium material, but later construction badly damaged the relevant strata, preventing archaeologists from determining the settlement's political or economic role. (Cambridge University Press)

This is genuine negative evidence—but limited

It weakens the strong claim:

“Archaeology confirms Sihon's kingdom.”

It does not establish:

“There was no Sihon.”

The relevant archaeological record is badly preserved.

Possible resolutions

Biblical political scale is exaggerated.

A relatively small local ruler was retrospectively called “king.”

Heshbon's exact biblical identification is wrong or incomplete.

The biblical tradition reflects later political geography projected backward.

The archaeology simply cannot recover the necessary evidence because the levels were destroyed.

Importance

Rank: #6.

Potentially damaging to literal historical reconstruction, but weak against more modest models of old regional memory.

8. The strongest archaeological-political parallel is centuries later

This creates a chronological anomaly.

The Mesha Stele, securely dated to roughly 830–805 BCE, gives us excellent evidence for Israel-Moab territorial struggle and Chemosh-based war theology. (Louvre Collections)

But the biblical narrative locates Sihon and Israel's wilderness movement much earlier.

The evidence that fits Numbers' political concerns best is therefore later than the purported events.

Two very different interpretations are possible

Continuity model: ninth-century realities preserve an older territorial conflict pattern.

Retrojection model: ninth-century or later territorial politics shaped stories projected into Israel's foundational past.

Both explain the correspondence.

Chronology alone cannot decide.

What would discriminate

A securely dated earlier inscription mentioning Heshbon, Moab, Sihon or the Arnon boundary would enormously strengthen the continuity model.

Without it, the later political context remains suspiciously relevant.

Importance

Rank: #8.

Not fatal, but potentially important for dating the tradition.

9. Chemosh fits historically—but too broadly to date anything

The Heshbon poem calls Moab the people of Chemosh.

The Mesha inscription independently confirms Chemosh as Moab's chief national deity. (Louvre Collections)

At first glance this looks like remarkable historical confirmation.

But there is a hidden complication.

Chemosh remained associated with Moab long enough that a much later Israelite author could know the same fact.

Therefore:

Chemosh confirms cultural accuracy.

It does not establish early composition.

This is an evidentiary outlier because a datum that looks chronologically powerful turns out to have little dating resolution.

Importance

Rank: #11.

Useful cultural evidence, weak chronological evidence.

10. The text invokes a written authority—but does not tell us who wrote it

If Numbers deliberately names a source, we might expect something about:

• author;
• compiler;
• date;
• provenance;
• ownership;
• institutional origin.

We get none of that.

The title stands alone.

Why this is strange

Other ancient documentary traditions often tie texts to kings, prophets, scribes or dynasties.

Here:

“Book of the Wars of YHWH”

contains no human attribution.

Possible explanations

The work was so well known that no author needed naming.

It was anonymous by nature.

It was a growing anthology rather than a single-authored work.

The citation formula preserved only a shortened conventional reference.

Or the “title” itself was not really a formal title.

Importance

Rank: #10.

This does not challenge existence, but it makes authorship reconstructions—especially Mosaic ones—extremely weak.

11. The Book completely disappears after being cited

This is a major negative-evidence issue.

If the collection were important enough for Numbers to cite by name, we might expect some later evidence:

• another biblical citation;
• quotation in rabbinic literature;
• a manuscript;
• a catalog reference;
• a translation;
• a Second Temple mention.

So far, none is securely known.

Possible resolutions

Ancient textual survival rates were extremely low.

The Book may already have been obsolete when Numbers quoted it.

Numbers may have preserved the only useful surviving fragment.

It may have circulated under another title.

Its contents may survive elsewhere without attribution.

The citation itself could conceivably derive from literary convention rather than an accessible contemporary document—though there is no need to invoke that explanation unless other evidence pushes us toward it.

What the silence really means

It is weak evidence against long-lived canonical or institutional importance.

It is not strong evidence against earlier existence.

Importance

Rank: #9.

The disappearance requires explanation but is historically unsurprising for ancient literature.

12. No identified Qumran copy appears despite extensive manuscript survival

Qumran preserves a Hebrew manuscript of Numbers, 4Q27, written in a Herodian-period hand. (The Dead Sea Scrolls)

Yet no recognized Qumran manuscript has been identified as the Book of Wars.

Why this is an outlier

Qumran preserved an extraordinarily broad range of biblical and nonbiblical Jewish literature.

If the Book of Wars remained important in Second Temple Judaism, we might expect some trace.

We apparently do not have one.

Best interpretation

This is negative evidence against the hypothesis that the Book remained a widely copied Second Temple work.

It tells us little about whether it existed hundreds of years earlier.

The more cautious inference is:

If it was once an independent composition, it had probably ceased to occupy a major place in Jewish literary culture by the Qumran period.

Importance

Rank: #12.

Important for transmission history; minor for original existence.

13. The Book of Wars and Book of Jashar behave similarly—but no ancient source connects them

We have two named lost sources apparently associated with poetic material.

That structural parallel is real.

But if they belonged to the same institutional milieu—or were actually the same anthology—we might hope for:

• shared attribution;
• identical title;
• cross-reference;
• overlapping quotation;
• ancient identification.

We have none.

Negative evidence

The absence of any explicit ancient identification should substantially suppress theories equating the two books.

Possible resolutions

They were separate anthologies.

One replaced the other.

Their titles reflected different sections of the same collection.

Later compilers no longer knew their relationship.

But the simplest explanation remains:

they were different works.

Importance

Rank: #14.

Mostly useful as a guard against overconnection.

14. The Deir ʿAlla “Book of Balaam” connection is weaker than it initially looked

This needs explicit correction from the previous stage.

We noted that some reconstructions read the beginning of the Deir ʿAlla inscription as something like a spr—an account/book of Balaam son of Beor.

However, recent scholarship explicitly cautions that Baruch Levine's reconstruction of the opening as “this is the account/book of Balaam” is far from certain. (Brill)

That matters.

The Deir ʿAlla inscription itself securely mentions Balaam son of Beor, and its archaeological context is late ninth to early eighth century BCE. (Brill)

But the stronger claim:

“we have an archaeological example of a named Book of Balaam analogous to the Book of Wars”

must be downgraded.

Revised status

Balaam son of Beor at Deir ʿAlla: strong.

Explicit Book of Balaam title: disputed reconstruction.

Analogy with Book of Wars: intriguing but not secure.

This is exactly the sort of anomaly/evidentiary correction this stage is designed to expose.

Importance

Rank: #13, but highly important methodologically because it prevents cumulative inference.

15. Numbers 21 contains too many distinct literary forms to assume one source casually

Within a very short stretch we have:

itinerary prose,

Book of Wars quotation,

well song,

travel list,

battle narrative,

Heshbon poem.

The temptation is to explain this diversity by reconstructing one large source behind everything.

But diversity itself could imply the opposite:

Numbers 21 may be a composite assembled from multiple independent traditions.

The conventional textual presentation explicitly introduces the Well Song separately in verse 17 and the Heshbon poem separately in verse 27. (JW.ORG)

This challenges Steiner's expansion

If verses 14–20 are really one Book of Wars excerpt, why does verse 17 contain a fresh formula:

“Then Israel sang this song”?

That can be explained as material already embedded in the source, but it creates another layer that must be accounted for.

Possible models

Single source with embedded song.

Book of Wars excerpt ends before the Well Song.

Editor combined multiple old traditions.

Later redaction inserted prose transitions into an older continuous text.

No model currently solves all features decisively.

Importance

Rank: #4, tied with the genre anomaly.

16. “Come to Waheb” solves one problem while creating another

Steiner's ʾet = “Come!” is attractive because it restores a verb and parallels “Come to Heshbon.”

But then another question appears:

Who is being told to come to Waheb, and why?

The solution clarifies grammar but does not fully clarify discourse.

(TheTorah.com)

This is important because a successful philological emendation or reinterpretation should ideally improve the passage globally, not just locally.

Possible explanations

A military host is being summoned.

Travelers are being invited.

The phrase is a conventional opening whose original pragmatic context is lost.

The subject was specified in earlier lines not preserved by Numbers.

Or ʾet is not “come” after all.

Importance

Rank: #15.

A real residual problem for Steiner, but not enough by itself to defeat the proposal.

17. If the quotation really preserves an ancient route, why can we not confidently map it?

A major claim of hydrological/geographical reconstructions is that the difficult terms make sense in the actual Arnon landscape.

Yet candidate identifications of Waheb, Suphah and some subsequent locations remain disputed.

Negative evidence

If the poem encoded a clear itinerary, we might hope that:

• place-name sequences;
• distances;
• watershed direction;
• archaeological sites;
• preserved toponyms

would converge.

They do not yet converge strongly enough.

Possible explanations

Ancient place names disappeared.

Later landscape naming changed.

The poem was metaphorical rather than itinerary-like.

Our modern reconstruction is wrong.

Or the relevant places lie at sites that have not been identified.

Importance

Rank: #16.

Potentially important but dependent on further regional fieldwork.

18. The “Wars of YHWH” title never explicitly appears again as a book title

The phrase “wars of YHWH” is intelligible elsewhere in Hebrew, particularly in Samuel, but we do not find clear later references like:

“as written in the Book of the Wars of YHWH…”

That creates an unusual split:

the ideological expression survives,