Volume 8 - Issue 1

Covenant, treaty, and prophecy

By Ernest Lucas

Study of the covenant traditions in the Old Testament was given a new impetus by the publication in 1955 of Law and Covenant in Israel and the Ancient Near East by G. E. Mendenhall. Basing his work on the study of Hittite vassal treaties made by V. Korošec1 he argued that the Sinai covenant seems to have been similar in form to the ancient Near Eastern vassal treaties of the fourteenth-thirteenth centuries bc. The pattern of these treaties is:

  1. Preamble, introducing the sovereign.
  2. Historical prologue outlining previous relations of the parties.
  3. Stipulations: (a) basic, (b) detailed.
  4. Document clause, providing for the deposition of a copy of the covenant in the vassal’s sanctuary, and its periodic reading.
  5. Witnesses, a long list of gods invoked to witness the covenant.
  6. Curses and blessings.

In the Old Testament this pattern is clearest in the structure of the book of Deuteronomy, but can also be discerned in the covenant ceremony in Joshua 24, and partially in Exodus 20.

1.   Dt. 1:1–5


Jos. 24:2a


Ex. 20:1


2.   Dt. 1:6–3:29


Jos. 24:2b–13


Ex. 20:2


3a. Dt. 4:1–11:32


Jos. 24:14 & 15


Ex. 20:3–17


3b. Dt. 12:1–26:19


Jos. 24:16–25




4.   Dt. 27:1–26


Jos. 24:26a


Ex. 25:16; 34:1


5.   Not appropriate


Jos. 24:26b & 27




6.   Dt. 28:1–68 (order reversed)


Jos. 24:27 (implied)




These parallels convinced many scholars that the Sinai covenant was indeed based on the vassal treaty form. This led to a blossoming of studies arguing that various features of the vassal treaty or the suzerain-vassal relationship illuminated aspects of Israel’s history and religion.2

Mendenhall’s thesis has never been without its critics, but opposition to it seems to be becoming increasingly fashionable. This opposition takes two variant forms:

  1. There are those who argue that originally the Sinai covenant was not expressed in treaty form. The use of this form was a later development, arising in the seventh century bc.3
  2. Some go even further, and argue that the very notion of a covenant was of little importance in Israel until the seventh century bc.4

The purpose of this paper is to examine the chief arguments put forward in support of the second, more radical, position, but in dealing with these we will in fact consider the evidence relevant to the less radical position.

The relevant arguments can be summarized as follows:

  1. Mendenhall compared the Sinai covenant form with that of the fourteenth-thirteenth century Hittite vassal treaties. Other scholars have argued that an equally good parallel can be drawn with the Assyrian vassal treaties of the ninth-seventh century bc.
  2. The word berîṯ (‘covenant’)5 is used only rarely by the eighth-century prophets (10 times of God-man relationship), but is much more common in the sixth-century prophets (39 times).
  3. In the Pentateuch the occurrences of the word berîṯ are nearly all in those passages that most scholars would attribute to the later sources D and P.6
  4. The important covenant terms bāḥar (‘choose’) and āhaḇ (‘love’) are common in Deuteronomy (30 times and 19 times respectively), yet rare in Genesis–Numbers (3 times and once of the God-man relationship).
  5. The clearest expression of the treaty form is found in Deuteronomy (held to be seventh century in date in its present form). The evidence for it in earlier passages (e.g. Ex. 20; Jos. 24) is debatable because the parallel here is limited.

As far as the first argument is concerned the plain fact seems to be that there are some basic differences between the vassal treaties of the first and second millennia bc. The six-fold form of the second-millennium treaty has been outlined above. Occasionally elements are omitted, but the order of elements is very rarely changed. The structure of the first-millennium treaties is simpler and more variable:7





Preamble or title.




  Stipulations and curses.




succeeded or preceded




by witnesses.


Two points are particularly worth noting. Firstly, an historical prologue is typical of the second-millennium treaties but very rare in the first-millennium treaties (only one, disputed, example is known). Secondly, in the second-millennium treaties the curses are balanced by blessings, whereas in the first-millennium treaties the blessings are very brief or, more often, non-existent. In fact the tone of the treaty has changed from being a gracious endowment to being an instrument of naked force.

M. Weinfeld8 admits these differences but argues that the lengthy series of curses in Deuteronomy is more like the curses of the seventh-century treaties than the short, generalized, curses of the Hittite treaties. On this ground alone he asserts that Deuteronomy reflects the first-millennium treaty form. However, this is to treat the evidence selectively by emphasizing only one aspect of it. At the most what he might have shown is that the form of the curses in Deuteronomy could have been influenced by the later treaty form. This point has relevance for the dating of the final form of Deuteronomy, but does not undermine the argument that the book as a whole reflects the second-millennium treaty form.

Here it should be noted that G. J. Wenham has argued9 that the Old Testament covenant form is a distinctive one that occurs in the Old Testament alone. However, it bears some resemblance both to the form of ancient Near Eastern ‘law codes’ and that of ancient Near Eastern treaties. Moreover, and this is the main point as far as the present argument is concerned, the form that he puts forward for the Old Testament covenant is very much closer to that of the second-millennium treaties than that of the first-millennium treaties.

There can be little doubt, then, that the covenant form reflected in Deuteronomy is that of the second millennium bc, and not that of the first millennium. This does not necessarily prove that Deuteronomy itself dates from the second millennium, but does indicate that the covenant form was known and used for the Sinai covenant from the time of Moses onwards.10 The incompleteness of the form in the other passages does not weaken the argument. The nearest thing to a covenant document in the Old Testament is Deuteronomy itself. The other passages are narrative accounts of covenant ceremonies (Jos. 24) or part of a covenant document incorporated into such a narrative (Ex. 20). One could not expect complete correspondence with the treaty form in these cases—though the parallel in Joshua 24 is striking.

We must now turn to the eighth-century prophets and consider whether they show evidence of knowing the Sinai covenant, and if so whether they knew it in the vassal treaty form. In the late nineteenth and early twentieth centuries there were many Old Testament scholars who argued that the ethical element in Israel’s religion did not go back to Moses but was the legacy of the eighth-century prophets, who were the real originators, on the human side, of the Ten Commandments. This view no longer finds much favour because:

  1. The ethical element in the preaching of Israel’s prophets clearly predates the eighth century, e.g. Nathan and David (2 Sam. 12), Elijah and Ahab (1 Ki. 21).
  2. The eighth-century prophets themselves do not seem to be promulgating something new, but to be indicting the nation on the basis of generally accepted ethical norms—norms that are paralleled by laws in the Book of Covenant (e.g. Am. 2:6–16; cf. Ex. 21:8–9; 22:26) and the Decalogue (Ho. 4:2).11
  3. There is nothing in the Ten Commandments that would be out of place in the time of Moses.12

It is therefore reasonable to conclude that the ethical preaching of the eighth-century prophets was based on, and a development of, an ethical code that was known and accepted in Israel from Mosaic times. But we can go further than that. Amos’ preaching of judgment is based on the fact that Israel stands in a special relationship to Yahweh and so has a particular responsibility to keep his laws (Am. 3:1–2). This relationship is related to the exodus (Am. 9:7). Hosea is also aware of a special relationship between Yahweh and Israel resulting from the exodus (Ho. 11:1; 12:9; 13:4–5)—a relationship which is pictured in terms of the marriage bond (a type of covenant) in chs. 1–3. Micah also speaks of this relationship and its ethical demands (Mi. 6:1–8).13

Here, then, we have evidence from the eighth century of the concept of a special relationship between Israel and Yahweh, originating in the exodus/wilderness period and resulting in Israel being obligated to live by certain moral norms (which coincide with stipulations found in the Pentateuch). Yet the fact is that of these prophets only Hosea speaks of the relationship as a covenant, berîṯ (Ho. 6:7; 8:1). Why is it that the other prophets (and indeed earlier sources too) do not make use of the term? Two answers are possible:

  1. They avoid the term because they wanted to avoid the popular ideas that had come to be associated with it. Their contemporaries stressed the promises of God to care for and protect his people and played down the ethical demands of the relationship. An example of this attitude might be Micah 3:11.
  2. The word became a common technical term for the Sinai covenant only with the rise of the Deuteronomic movement in the seventh century bc. This movement may have been sparked off by the finding of the long-lost book of Deuteronomy and by Josiah’s reforms; the movement may possibly have put the book of Deuteronomy into its present form. In any case the movement may have been responsible for establishing a standardized terminology for covenant ideas—ideas that were themselves not new.14

Is there any evidence of the form in which the prophets knew the covenant concept, whatever term they used for it? There are two features of their preaching which are at least consistent with the view that they knew the treaty form.

Firstly, the prophets speak of certain disasters as punishment for breaking Yahweh’s law. These are just the kind of things included in the treaty curses (e.g. Am. 4; cf. Dt. 28; Lv. 26). Thus it is possible to argue that the prophetic doom oracles are based on these curses.15

Secondly, there are in the prophets oracles which have the form of a law-suit (e.g. Ho. 4:1–3; Is. 1:2–3; 3:13–15; Mi. 6:1–5; cf. Dt. 16:22). These contain the following elements (not always all present):16

  1. A call to witnesses to listen to the proceedings.
  2. A statement of the case at issue by the divine Judge.
  3. An account of the benevolent acts of Yahweh.
  4. Indictment.
  5. Sentence.

Now J. Harvey17 has pointed out that this form finds a parallel in the letters of accusation sent by a suzerain to a vassal accused of breaking a treaty. He suggests that these oracles therefore have a background in the treaty form. Even before Harvey’s suggestion others had seen a probable link between these oracles and the covenant regarded as a vassal treaty.18

The case must not be over-stated. Neither of these features provides conclusive proof that the prophets knew the covenant in the vassal treaty form. After all, the curses in the treaties are only a special case of the more general use of curses in the ancient Near East, and the law-suit oracles could be based on general law-court procedures. However, taken with the other evidence discussed above these two features help to build up a cumulative case in favour of the view that the Sinai covenant did originate in the second millennium bc and was modelled on the vassal treaty form of that era.

There is another issue concerning the covenant which we must consider briefly since it has some relevance to the preceding discussion. In Dt. 31:9–13 there is a command that there should be a reading of the covenant law every seven years at the Feast of Booths. Jos. 24 probably records such a ceremony (cf. 2 Ki. 23:1–3). Whether this was done regularly is unclear. A. Weiser19 claims that many of the psalms come from the liturgy of such an occasion (e.g. Ps. 50, 81, 105, 111). If there was such a ceremony held at intervals it would provide an explanation for the persistence and knowledge of the treaty form in Israel. Its liturgy would surely reflect this form (cf. Jos. 24), and liturgy is notably conservative.20 The liturgy of such a festival has been suggested as the background for the prophetic law-suit and doom oracles. One would expect the liturgy to make provision for the indictment of the nation for failures to keep the covenant law in the period since the previous renewal. Psalm 50 could be such a liturgy. All this, however, is rather speculative since we have no clear evidence of how faithfully Deuteronomy 31:9–13 was obeyed.21

Finally, something needs to be said briefly about the significance of this discussion for the understanding of Old Testament history and theology. The historical significance is fairly obvious. If the evidence that the Sinai covenant was expressed in the form of the fourteenth/thirteenth-centuries bc treaties is sound, it supports the antiquity of the Mosaic covenant traditions within the Old Testament. The major, though by no means the only, theological significance of the treaty form is what it implies about the way we should view the covenant law and its curses and blessings. There is a danger (and it is one that the Israelites did not always avoid) of understanding them on a strict quid pro quo basis. Obedience will earn favour and guarantee a reward. However, this confuses covenant with contract. The Hittite treaties are not contracts. They are gifts of grace given by the overlord to define and confirm an existing relationship (hence the historical prologue). The vassal keeps the stipulations of the covenant not to earn favour but as a response of gratitude for the overlord’s benefactions. The point of the blessings and curses is that the faithful vassal continues to enjoy these benefactions, whereas persistent infidelity (which in the context of the treaty is seen as an expression of gross ingratitude) effectively puts an end to the relationship expressed by the covenant. However, the end is not necessarily definitive. The overlord could exercise mercy and renew the relationship with a repentant vassal.

1 V. Korošec, Hethitische Staatsverträge (Leipzig, 1931).

2 Some examples are: G. Mendenhall, Law and Covenant in Israel and the Ancient Near East (Pittsburgh, 1955). H. Huffmon, ‘Covenant Lawsuit and the Prophets’, JBL 78 (1959), pp. 286–295. W. Moran, ‘The Ancient Near Eastern Background of the Love of God in Deuteronomy’, CBQ 25 (1963), pp. 77–87. D. Hillers, Treaty—Curses and the Old Testament Prophets (Rome, 1964). W. Beyerlin, Origins and History of the oldest Sinai Traditions (Oxford, 1965). W. Zimmerli, The Law and the Prophets (Oxford, 1965).

3 D. J. McCarthy, Treaty and Covenant (revised edition, Rome, 1978). E. W. Nicholson, Exodus and Sinai in History and Tradition (Oxford, 1973).

4 G. Fohrer, History of Israelite Religion (London, 1973). R. E. Clements, Prophecy and Tradition (Oxford, 1975).

5 On the etymology and meaning of berîṯ see the article on it in G. J. Botterweck and H. Ringgren (eds.), Theological Dictionary of the Old Testament, vol. II (revised edition, Grand Rapids, 1977).

6 Obviously the validity of this argument stands or falls with the validity of the usual source analysis of the Pentateuch. However, note the occurrence of berîṯ in Gn. 15:18; Ex. 19:5; 24:7; 34 (5 times), which are usually assigned to J or E.

7 K. A. Kitchen, Ancient Orient and Old Testament (London, 1966), pp. 90–102. K. A. Kitchen, The Bible in its World (Exeter, 1977), pp. 79–85. A. D. H. Mayes, Deuteronomy (NCB) (London, 1979), pp. 31–33.

8 M. Weinfeld, Deuteronomy and the Deuteronomic School (Oxford, 1972), pp. 59–157.

9 G. J. Wenham, ‘The Structure and Date of Deuteronomy’ (an unpublished thesis which is discussed briefly in J. A. Thompson, Deuteronomy (TOTC) (Leicester, 1979), pp. 18–20).

10 P. C. Craigie, Deuteronomy (NICOT) (London, 1976), Appendix II, suggests that Egyptian labour contracts had the same form as the Hittite vassal treaties, and that this is why the Sinai covenant reflects the same form—the Hebrew slaves exchanged one overlord (pharaoh) for another (Yahweh).

11 For example, on Hosea, D. Stuart ‘The Old Testament Prophets self understanding of their prophecy’ (Themelios, 6:1 Sept. 1980, pp. 11–12) asserts that, ‘there is no passage in the book that does not have the Mosaic scripture as its basis’, and presents some evidence to support this. Similarly F.I. Andersen and D.N. Freedman, Hosea (Anchor Bible, Doubleday, 1980), p. 75 observe ‘Hosea’s discourses are threaded with Deuteronomic ideas in a way that shows they already were authoritative in Israel’.

12 H. H. Rowley, Men of God (London, 1963), pp. 1–36. J. P. Hyatt, Encounter 26 (1965), pp. 199–206.

13 On the authenticity of this oracle see the NICOT commentary by L. Allen (London, 1976).

14 Cf. the ‘technical’ vocabulary of the ‘charismatic movement’ re experiences of, and teaching about, the Holy Spirit which in themselves have not been unknown in Christianity before the 1950’s.

15 F. C. Fensham, ZAW 74 (1962), pp. 1–19; ZAW 75 (1963), pp. 155–175. D. Hillers, op. cit.

16 G. E. Wright in Israel’s Prophetic Heritage, eds. B. W. Anderson and W. Harrelson (London, 1962), pp. 26–67.

17 J. Harvey, Le plaidoyer prophétique contre Israel après la rupture de l’alliance (Paris-Montreal, 1967).

18 E.g. G. E. Wright, op. cit. and H. Huffmon, op. cit.

19 A. Weiser, The Psalms (London, 1962).

20 Cf. the opposition to the revision of The Book of Common Prayer in Britain!

21 M. Weinfeld, op. cit., pp. 51–58, 158–178, argues that the covenant form was preserved in a literary tradition of covenant writing associated with scribes/wise men.

Ernest Lucas

Bristol Baptist College