Hosea, Ezekiel, and the Original Register of the Covenant

A study in the relationship between prophetic tradition and Pentateuchal composition

I. The Question

It is said that Hosea predates the final redaction of the Pentateuch. The question that arises is the following: to what extent does Hosea’s statement — ‘Therefore, I will allure her, I will lead her into the wilderness and speak tenderly to her’ (Hosea 2:14 [Hebrew 2:16]) — explain the Exodus, or influence the final composition of the Exodus narrative? Which shapes the other? Similarly, in Ezekiel, God declares: ‘I spread my cloak over you and covered your nakedness; I made a covenant with you and entered into an alliance with you’ (Ezekiel 16:8). What are the oldest textual strata behind these traditions? What are their possible datings? Which scholars take the most modern positions, and which the most conservative?

It is said almost unanimously that the prophets already presuppose the tradition of the Exodus. From this perspective, Hosea does not create the image of the wilderness; he reinterprets a foundational tradition already known. Gerhard von Rad, whose formulation has set the terms of the discussion for generations, argued that the credo of Israel’s saving history — including the Exodus — was the bedrock upon which the prophets built their proclamations, not the other way round. But this consensus, however well established, leaves open a question that has not received the attention it deserves: was this foundational tradition nuptial from the very outset?

In Exodus 19:5, it is written: ‘Now therefore, if you obey my voice and keep my covenant, you shall be my segullāh among all peoples.’ The nuptial dimension, as I shall argue, is already structurally present in this text — not as a later prophetic colouring, but as the original register of the covenant relationship. If this is so, then the question of who shapes whom must be reconsidered, and the initial form of the Exodus narrative must be read in a new light.

II. The Textual Strata and Their Datings

The Pentateuch: the classical hypothesis and its revisions

The question of the sources of the Pentateuch has undergone considerable revision over the past two centuries. The classical documentary hypothesis of Wellhausen, formulated in the 1870s and 1880s, postulated four principal sources: the Yahwist (J), the Elohist (E), the Deuteronomist (D), and the Priestly source (P). Their approximate datings, both in their traditional form and as revised by more recent scholarship, may be set out as follows.

The Yahwist was traditionally placed in the tenth to ninth centuries before the common era, representing a southern, Judahite perspective. More recent scholarship — notably that of John Van Seters and Konrad Schmid — has tended to date J considerably later, to the seventh or even sixth century, or has questioned whether J existed as a coherent independent source at all. The Elohist, associated with northern traditions, was placed in the ninth to eighth centuries, though many contemporary scholars regard E as either a fragmentary source or as inseparable from J. The Deuteronomist — broadly associated with the reform of Josiah around 621 — is the most widely accepted strand of the hypothesis and remains largely intact in current scholarship. The Priestly source, associated with the exilic and post-exilic period of the sixth to fifth centuries, is equally well attested and widely accepted.

Exodus 19:5 — the verse containing the segullāh — is assigned by different scholars to E, to a pre-Deuteronomistic stratum, or to Deuteronomistic redaction. This very uncertainty is itself significant. The vocabulary of the segullāh does not sit comfortably within the stylistic markers of any single source. It appears to belong to a tradition that the great documentary categories have difficulty in capturing — a tradition older, perhaps, than the literary strands through which it reaches us.

Hosea: the earliest writing prophet

Hosea is dated with relative confidence to the eighth century, between approximately 760 and 720 before the common era, making him a contemporary of Amos and Isaiah of Jerusalem, and a generation before the fall of the Northern Kingdom to Assyria in 722. He is the first writing prophet to develop an explicitly nuptial interpretation of the covenant relationship between YHWH and Israel. The foundational text is Hosea 2:14 in English versification (2:16 in Hebrew):

לָכֵן הִנֵּה אָנֹכִי מְפַתֶּיהָ וְהֹלַכְתִּיהָ הַמִּדְבָּר וְדִבַּרְתִּי עַל־לִבָּהּ  lākēn hinnēh ʾānōkî mefatteyhā wehōlakhtîhā hammidbar wedibbartī ʾal-libbāh
‘Therefore, behold, I will allure her, and bring her into the wilderness, and speak tenderly to her’

Three elements of this verse demand close attention. The verb mefatteyhā — from pittāh — carries a nuance of seduction, of persuasive wooing. It is not the language of compulsion but of desire seeking to awaken desire in return. The wilderness — hammidbar, with the definite article — is not a symbolic or imaginary landscape: it is the wilderness, the one the audience is expected to recognise immediately. This is an explicit reference to the Exodus tradition, and its use with the definite article presupposes that the tradition is already well known to Hosea’s hearers. Finally, the expression dibbartī ʾal-libbāh — ‘speaking to the heart’ — is in the Hebrew Bible consistently the language of tender, intimate address, amorous or consoling in register. It appears in Genesis 34:3, where Shechem speaks to Dinah’s heart, and in Ruth 2:13. It is not juridical discourse. It is the speech of a lover.

Hosea, then, in returning the people to the wilderness, is not constructing a new narrative. He is recalling a tradition already charged with meaning and reinterpreting it through the register of nuptial love. The question that immediately arises is whether this nuptial register is his own invention, his own hermeneutical contribution — or whether he is drawing out what was already latent, already present, in the tradition itself.

Ezekiel: the great nuptial allegory

Ezekiel is dated with confidence to the sixth century, between 593 and 571 before the common era, placing him firmly in the Babylonian exile. His sixteenth chapter is the most sustained and theologically elaborate nuptial allegory in all of the Hebrew Bible. God’s address to Jerusalem reconstructs the entire history of the relationship in erotic and matrimonial terms:

וָאֶעֱבֹר עָלַיִךְ וָאֶרְאֵךְ וְהִנֵּה עִתֵּךְ עֵת דּוֹדִים וָאֶפְרֹשׂ כְּנָפִי עָלַיִךְ  wāʾeʾebōr ʾalāyikh wāʾerʾēkh wehinnēh ʾittēkh ʾêt dôdīm wāʾefrōs kenāfī ʾalāyikh
‘I passed by you and saw you, and behold, your time was the time of love; I spread the edge of my garment over you’

The gesture of spreading the edge of one’s garment — kānāf — is in the Hebrew Bible the gesture of marriage. It appears with full clarity in Ruth 3:9, where Ruth asks Boaz to spread his garment over her, that is, to take her as his wife. In Ezekiel 16, God performs this identical gesture over Israel. The Exodus, in Ezekiel’s reading, is not primarily a political liberation or a juridical transaction: it is a nuptial act. God sees his people in their abandonment and destitution, and he weds them.

This reading of Ezekiel is often treated as a bold prophetic reinterpretation of an older tradition — a creative theological elaboration of what was, in its original form, a narrative of historical deliverance. But this characterisation, whilst defensible on the level of literary history, may be misleading on the level of theological structure. For Ezekiel does not present his allegory as a novelty. He presents it as a revelation of what the relationship always was. The question, again, is whether he is constructing nuptiality or uncovering it.

III. The Scholarly Landscape

The critical consensus: prophets presuppose Exodus

The position that has been dominant in critical scholarship since the nineteenth century holds that the prophets presuppose the tradition of the Exodus and do not create it. Gerhard von Rad argued that the Exodus belonged to an ancient confessional tradition of Israel, already crystallised before the earliest prophets. Martin Noth, in his foundational work on the traditions of the Pentateuch, similarly identified the Exodus as one of the oldest and most stable themes in Israelite memory, antedating any of the literary sources.

In this view, Hosea’s reference to the wilderness carries the definite article precisely because the tradition is already known; Ezekiel’s nuptial allegory builds upon a foundational narrative that his audience already inhabits. The prophets are interpreters, not inventors, of the Exodus. This remains the consensus position in mainline critical scholarship.

The most critical contemporary positions

Contemporary scholarship has moved beyond Wellhausen’s documentary hypothesis whilst broadly retaining its scepticism about early dating. Thomas Römer, currently Professor at the Collège de France and one of the most influential living specialists on the Pentateuch, argues that the narrative form of the Exodus as we now possess it is in large measure a Deuteronomistic construction of the seventh and sixth centuries. This does not mean that no earlier tradition existed; but it does mean that the theological shaping we read in the final text is late, and that the relationship between the prophets and the Pentateuch is more reciprocal than the classical hypothesis allowed. On Römer’s reading, there is genuine interaction between prophetic proclamation and Pentateuchal redaction, with each tradition shaping the other in a long process of literary formation.

Konrad Schmid of the University of Zurich has argued persuasively that the narrative link between the patriarchal traditions and the Exodus is itself a Deuteronomistic construction, and that these two bodies of tradition circulated independently for a considerable period. Jan Christian Gertz has similarly demonstrated that the Exodus traditions existed in multiple independent forms before their integration into the unified narrative of the Pentateuch.

What these positions share is a tendency to view the theological grandeur of the Exodus narrative as a product of later reflection rather than early composition. The nuptial register, on this reading, could easily be understood as a late development — introduced either by the prophets or by Deuteronomistic editors influenced by prophetic tradition.

The conservative positions

Against this critical tendency, a number of scholars have defended the antiquity and substantial historicity of the Exodus traditions. Kenneth Kitchen of the University of Liverpool, working from Egyptological evidence and comparative ancient Near Eastern data, has consistently argued for a dating of the Exodus traditions in the Late Bronze Age, approximately the fourteenth to thirteenth centuries before the common era. On Kitchen’s account, the vocabulary of Exodus 19:5 — including the segullāh — reflects genuinely ancient usage with close parallels in second-millennium texts.

John Bright and Roland de Vaux, whilst accepting a degree of literary development in the Pentateuchal traditions, maintained the substantial historicity of the Exodus event and the antiquity of the theological reflection that accompanied it. On their reading, the covenant theology of Sinai — including its vocabulary of possession and election — belongs to Israel’s earliest experience of YHWH, not to a later editorial stratum.

William Moran’s landmark 1963 article on the love of God in Deuteronomy added a further dimension to this discussion. Moran demonstrated that the vocabulary of love in the covenant context has close parallels in ancient Near Eastern treaty formulae, particularly Assyrian vassal treaties, as well as in Egyptian love poetry. His argument implies that the emotive and affective language of the covenant — including the language that we are identifying as nuptial — is not a late Israelite invention but belongs to the broader semantic field of ancient Near Eastern covenant discourse. If Moran is right, the nuptial register was available to the earliest formulators of Israelite covenant tradition, not only to eighth-century prophets or seventh-century Deuteronomists.

IV. The Central Question: Was the Tradition Nuptial from the Outset?

The weight of Exodus 19:5

It is at this point that the question I am posing acquires its full force. The argument that the nuptial register was introduced into the Exodus tradition by Hosea, or by Deuteronomistic editors influenced by Hosea, rests on an assumption that the tradition, in its original form, was not nuptial. But this assumption is placed under considerable pressure by a careful examination of Exodus 19:5.

The text reads: ‘Now therefore, if you obey my voice and keep my covenant, you shall be my segullāh among all peoples, for all the earth is mine.’ The word segullāh has, as I have demonstrated in an earlier study, a semantic field that extends far beyond mere juridical possession. In its royal usage — attested in Qohelet 2:8, where the Preacher speaks of the segullat melākhīm, the treasured possession of kings — the segullāh designates not the general revenues of the crown but the king’s private and intimate treasury: the objects he keeps in his own chamber, not in his storehouses, because they express his personal glory and his intimate attachment. This is not the language of administration. It is the language of cherishing.

More decisively, the grammatical structure of the phrase is lî segullāh — ‘for me a treasure’, ‘mine as a treasure’. This lî is structurally identical to the lî of the Song of Songs’ most fundamental formula: dôdī lî waʾanī lô — ‘my beloved is mine and I am his.’ The possessive lî in both cases does not designate domination or juridical ownership. It designates the intimate mutual belonging of those who love one another. When God says lî segullāh, the semantic register is the same as when the beloved of the Song says dôdī lî. The structure of nuptial love is already present in the vocabulary of Exodus 19:5.

Two possibilities

Two distinct possibilities present themselves in light of this observation, and it is worth stating them clearly before adjudicating between them.

The first possibility — the standard critical reading — holds that the segullāh was originally a royal economic metaphor, neutral in emotional register, which was subsequently reinterpreted in a nuptial direction by Hosea and the Deuteronomists. On this reading, the nuptial colouring of Exodus 19:5 is a product of the redactional process, not of the earliest tradition.

The second possibility — the one that I am proposing — holds that the segullāh was nuptially charged from the very beginning, because the nuptial register is the original and primary register of the covenant relationship. On this reading, Hosea does not introduce nuptiality into the Exodus tradition; he names explicitly, in response to the crisis of Israel’s infidelity, what had always been implicit in the vocabulary and structure of the covenant. He reveals rather than invents.

The evidence for the second possibility

Several converging lines of evidence support the second possibility. The first concerns the Bundesformel — the covenant formula that recurs throughout the Hebrew Bible from Exodus to Jeremiah to Ezekiel to the Apocalypse: ‘I will be their God and they shall be for me my people.’ The formula appears in so many diverse literary contexts, with such varying formulations, that it almost certainly antedates all the major documentary sources. It is one of the most archaic elements of Israelite covenant tradition. And, as I have argued elsewhere, its grammatical structure — particularly the lî, ‘for me’, ‘mine’ — is structurally identical to the nuptial formula of the Song of Songs. If the Bundesformel is genuinely ancient, then nuptial structure is genuinely ancient.

The second line of evidence concerns the verb ḥāshaq, used in Deuteronomy 7:7–8 to describe God’s attachment to Israel: ‘It was not because you were more in number than any other people that the Lord set his love on you and chose you — for you were the fewest of all peoples. But it is because the Lord loves you.’ The verb ḥāshaq denotes amorous desire, passionate attachment. It is the vocabulary of the Song of Songs, not of juridical treaty-making. Its presence in the Deuteronomistic tradition — which on any reading is among the most carefully crafted theological strata of the Pentateuch — suggests that the semantic field of nuptial desire was not introduced from outside but drawn upon from within an existing tradition.

The third line of evidence is precisely Hosea’s use of the definite article in hammidbar — ‘the’ wilderness. This detail, apparently minor, is theologically decisive. Hosea is not proposing a new wilderness experience as a pedagogical device of his own invention. He is recalling a specific, already-known wilderness — the wilderness of the Exodus — and he is recalling it in nuptial terms. If the tradition he is recalling were not already nuptial, his reinterpretation would require a much more explicit argument. The fact that he simply uses the definite article, presupposing immediate recognition, implies that the nuptial register he is applying to the wilderness tradition was not entirely foreign to it. He is not translating from one register to another; he is amplifying what was already present.

Moran’s findings on the ancient Near Eastern background of covenant love provide a fourth and more general support for this hypothesis. If the affective, desirous, even erotic language of covenant allegiance was part of the common semantic inheritance of the ancient Near East — present in treaty formulae, in royal correspondence, in love poetry — then it is entirely plausible that the earliest Israelite formulations of the Sinai covenant drew on this register from the outset, not as a late refinement but as a natural idiom for expressing the unique character of the relationship between YHWH and his people.

V. A Hypothesis of Synthesis

On the basis of these converging considerations, I would propose the following synthetic hypothesis regarding the original form of the Exodus narrative and its relationship to prophetic tradition.

The Exodus tradition, in its oldest recoverable strata — whether one dates these to the tenth century, the ninth, or earlier — already carried within it a nuptial intelligence of the covenant relationship. This nuptial intelligence was not yet systematically developed or theorised; it was present as a register, as a fundamental colour of the relationship between YHWH and Israel, inscribed in the vocabulary of the segullāh, in the structure of the Bundesformel, in the verb ḥāshaq.

Hosea, in the eighth century, facing the infidelity of the Northern Kingdom, did not introduce this nuptial register. He revealed it in its full force and made it theologically explicit, because the crisis of infidelity demanded it. One does not speak of adultery unless there was a marriage. And that marriage, Hosea insists, is not his metaphor: it is the reality of what happened at the Sinai, in the wilderness, at the moment of the covenant. His proclamation in Hosea 2:14 is not a creative theological invention but a prophetic recall — a summoning of Israel back to the truth of its own foundational experience.

Ezekiel, writing two centuries later in the context of the Babylonian exile, carries this prophetic recall to its most elaborate literary expression. His allegory in chapter 16 does not merely interpret the Exodus through a nuptial lens — it claims that the nuptial event is what the Exodus was. The gesture of the garment, the spreading of the kānāf, is not imposed upon the tradition from without; it is drawn out of a depth that was always there.

The Deuteronomistic redactors, working in the seventh and sixth centuries with prophetic tradition as a primary resource, preserved and amplified this nuptial register. They did not create it; they received it, recognised it, and gave it its most systematic theological articulation. The final form of the Pentateuch — with its segullāh, its Bundesformel, its ḥāshaq, its covenant commandments as a response of love to a love already given — is not the product of a nuptial reinterpretation of an originally juridical tradition. It is the mature theological articulation of what the Exodus tradition had always carried in its deepest structure.

VI. Consequences for the Reading of the Exodus Narrative

If this hypothesis is correct, several consequences follow for the way in which the Exodus narrative is to be read.

The first and most fundamental consequence is that Exodus 19:5 — ‘You shall be my segullāh’ — is not a juridical declaration followed by a set of legal conditions. It is, in its original register, a declaration of love: the love of a God who desires his people with the same intimacy and exclusivity that a king reserves for his most cherished possession, or that a bridegroom reserves for the bride he has chosen above all others. The commandments that follow in chapters 20 and beyond are not, in their primary logic, conditions to be fulfilled in order to merit this love. They are the form in which the beloved responds to a love already received and already given.

The second consequence concerns the relationship between the prophetic and the legal traditions of the Hebrew Bible. The standard account tends to see the prophets as critics of a legal tradition that had become too rigid, too formal, too detached from the living relationship between God and his people. On the hypothesis advanced here, the prophets are not introducing a corrective from outside: they are recalling the community to the original register of its own founding tradition. Hosea does not correct the Exodus; he remembers it more deeply than the community had remembered it.

The third consequence is exegetical and theological in the broadest sense. It concerns the face of God that the Exodus narrative presents. A God who says ‘you shall be my segullāh’, who attaches himself to his people with the ḥāshaq of erotic longing, who leads his beloved into the wilderness in order to speak tenderly to her heart, who spreads his garment over her in the gesture of marriage — this is not primarily a legislator or a judge. He is a God whose first word to his people is a word of desire, of election, of love that chooses one above all others not because of merit but because of the sovereign freedom of love itself.

The commandments, the law, the covenant obligations — all of these retain their full weight and seriousness within this vision. But their weight is not that of juridical compulsion. It is the weight of a love that calls for a response commensurate with its own depth: a response of all the heart, all the soul, and all the meʾôd — all the excess, all the measureless surplus — that the nuptial register, and only the nuptial register, can adequately name.

Conclusion

The question of whether Hosea shapes the Exodus or the Exodus shapes Hosea is, in the end, less important than the recognition that both are expressions of a single and coherent theological vision: the vision of a God whose relationship with his people is nuptial in its very structure. The wilderness tradition did not need Hosea to become nuptial. It was nuptial because the event it recalled — the encounter at Sinai, the covenant sealed in the desert — was experienced from the outset as an act of love, of sovereign and gratuitous desire, of a God who chose one people as his segullāh among all the peoples of the earth.

Hosea names this with a force and an urgency that the crisis of infidelity demanded. Ezekiel unfolds it in its most vivid and most painful dimensions. The Deuteronomists give it its most systematic theological form. But none of them invented it. They received it from a tradition whose original form was already, in its vocabulary and its structure, a declaration of love.

To recover this original register is not merely an exercise in historical exegesis. It is to recover the face of God that the Exodus narrative was always, from the beginning, attempting to show: the face not of a legislator who demands obedience, but of a bridegroom who desires a response of love — total, excessive, measureless — because the love he himself has given is precisely that.

Jean Khoury, with the help of AI Claude

Select Bibliography

Bright, J., A History of Israel, 4th edn (Louisville: Westminster John Knox Press, 2000).

Gertz, J. C., Tradition und Redaktion in der Exoduserählung (Göttingen: Vandenhoeck & Ruprecht, 2000).

Kitchen, K. A., On the Reliability of the Old Testament (Grand Rapids: Eerdmans, 2003).

Moran, W. L., ‘The Ancient Near Eastern Background of the Love of God in Deuteronomy’, Catholic Biblical Quarterly 25 (1963), pp. 77–87.

Noth, M., Überlieferungsgeschichte des Pentateuch (Stuttgart: Kohlhammer, 1948); ET: A History of Pentateuchal Traditions (Englewood Cliffs: Prentice-Hall, 1972).

Römer, T., The So-Called Deuteronomistic History (London: T&T Clark, 2005).

Schmid, K., Erzväter und Exodus (Neukirchen-Vluyn: Neukirchener, 1999); ET: Genesis and the Moses Story (Winona Lake: Eisenbrauns, 2010).

Van Seters, J., The Life of Moses: The Yahwist as Historian in Exodus–Numbers (Louisville: Westminster John Knox Press, 1994).

Von Rad, G., Old Testament Theology, vol. 1, trans. D. M. G. Stalker (Edinburgh: Oliver & Boyd, 1962).

Wellhausen, J., Prolegomena zur Geschichte Israels (Berlin: Reimer, 1883); ET: Prolegomena to the History of Israel (Edinburgh: A. & C. Black, 1885).

* * *