Covenants in Cloudcuckooland and the Greek Old Testament

We have seen, in our series so far, the way that the word “covenant” is used in the Old Testament. How did the Hebrew word for covenant (בְּרִית) come to be translated by the Greek word διαθήκη? A good case can be made that there is a specific use of the Greek word διαθήκη that corresponds almost exactly with the definition of the word that we have confirmed by our study of the Old Testament: “an elected relationship of obligation under oath”. This usage of διαθήκη confirms our assertion that a biblical “covenant” is not just any kind of relationship, and nor is it inherently sociological or ecclesiological. Rather, a covenant a very specific kind of relationship between two well-defined parties.

A covenant is not just a ‘will’

The fact that translators of the LXX chose to translate בְּרִית with the Greek word διαθήκη is a little surprising given that by far the most common use of διαθήκη in the extant Greek literature is not a “covenant” in the Old Testament sense (“elected relationship of obligation under oath”), but a written document drawn up to distribute property after a person’s death, a “disposition”, “testament” or “will”.[1] Prior to the 3rd century BC, there are about 240 instances of διαθήκη.[2] It occurs most abundantly in orators arguing legal cases and in Plato’s Laws (e.g. 922.c, 923.c, 923.e, 924.a, 926.b). Often the plural is used to refer to a will, since the various “dispositions” (διαθήκαι) collectively form a will (Isocrates, Aeginet. 1, 12, 15, 34; Isaeus, Cleonymus, 3). By contrast, the plural of διαθήκη never occurs in the Old Testament. The testator could make or leave a will (διατίθεμαι [Isaeus, Cleonymus 3, 11, 15, 20, 48], ποιέω [Isaeus, Cleonymus 10, 30–31], καταλείπω [Isocrates, Aeginet. 5, 15, 34]), confirm a will (βεβαίω [Isaeus, Cleonymus 18–19]), alter a will by codicil (ἐπανορθόω [Isaeus, Cleonymus 26]), and revoke a will (ἀναιρέω [Isaeus, Cleonymus 14, 18, 21; Philoctemon 30], λύω [Isaeus, Cleonymus 3, 18, 50]). After the death of the testator, a court could declare his will invalid (ποιέω ἄκυρον [Isocrates, Aeginet. 3, 15; Isaeus, Cleonymus 21, Philoctemon 4], καθίστημι ἄκυρον [Isaeus, Aristarchus 22]). By contrast, Old Testament covenants are “broken” (הֵפֵר, διασκεδάζω) by negligence or wilful disobedience (e.g. Gen 17:14; Lev 26:15; Deut 31:16, 20; Isa 24:5, 33:8; Jer 11:10, 31:32; Ezek 16:59), never simply invalidated.

One feature common to Old Testament covenants and some Greek wills is the creation of kinship bonds by “election”. A διαθήκη could be used to bestow legal rights upon people who naturally did not have these rights (e.g. Isaeus, Philoctemon 28). A διαθήκη was a possible means for adopting a son and thus allowing him to inherit property (Isaeus, Aristarchus 9; Astyphilus 5, cf. 10–11; Ciron 40). However, there is evidence that this process was fraught with complications. For example, “I was adopted by Menecles with the strictest possible legality, and [. . .] the form of adoption was not merely verbal or by will [διαθήκῃ] but by very act and deed” (Isaeus, Menecles 44 [Forster, LCL]; see also Isaeus, Hagnias 8–9). By contrast, an ot covenant is a much stronger means of creating kinship bonds.

A covenant is more than just a ‘pact’

The choice of διαθήκη is even more surprising when one considers that there was a Greek word for “pact”: συνθήκη.[3] Demosthenes (1 Steph. 41.9) mentions both συνθήκαι (articles of agreement) and διαθήκαι (articles of disposition) in parallel, showing that they are not synonyms. The translators of the LXX clearly knew the former word; they used it for political pacts between humans (1 Macc 10:26; 2 Macc 12:1, 13:25, 14:20, 14:26–27; Isa 30:1; Dan 11:6, 17) and metaphorical pacts between humans and “death” (Wis 1:16, Isa 28:15). Once, it is used of an agreement between God and humans (Wis 12:21). Nevertheless, 270 times, διαθήκη translates בְּרִית in the Septuagint.[4] Why did the Greek translators of the Old Testament consider the word commonly used for “will” to be more appropriate than the word for “pact” for translating the Old Testament word “covenant”?

The covenant in Cloudcuckooland

The answer, quite literally, may be found in Cloudcuckooland. The ancient comic Aristophanes (c. 445–385 bc), although familiar with the meaning “will” for διαθήκη (Wasps, 584, 589), and also with the word συνθήκη (“pact”, Peace, 1065; Lysistrata, 1268–69), uses διαθήκη in one passage in a way that is identical with our inductive definition of the OT word בְּרִית: “an elected, as opposed to natural, relationship of obligation under oath”.[5] In his fantastic play Birds the hero, Peisetaerus, wants to convince the bird community to establish a carefree hedonistic utopia called “Cloudcuckooland”. The birds gather and ask Peisetaerus to disclose his plan to them. But they are armed and look a little too dangerous, so Peisetaerus says (lines 438–42, my translation):

By Apollo! I will not / unless (and not otherwise) they make a covenant with me [διάθωνται, … διαθήκην ἐμοι] / the very same one that the ape made [διέθετο] with the woman[6] / (the knifemaker): that they neither bite me / nor yank [my] testicles nor dig…

After some brief innuendo, the play continues (lines 444–47, my translation):

Leader of the birds: I make [a covenant] [διατίθεμαι ’γώ]

Peisetaerus: “Now swear these things to me”

Leader of the birds: I swear to prevail in [the opinion of] these: all the judges / and all the spectators [. . .] But if I should transgress, to prevail in [the opinion of] one judge only.

The result of this sworn oath is that Peisetaerus now has a claim over the birds, and is able to order them to dispose of their arms. While the details of the oath are obscure, the meaning of διαθήκη is clear. The birds choose to enter into a new relationship with a human by oath, whereby they are obliged not to hurt him. While the covenantal obligation is unilateral, the covenant established a “truce” (τὰς σπονδάς, line 461), a relationship which enables Peisetaerus to get on with his original task of explaining his idea about Cloudcuckooland to the birds.[7]

The significance of this reference is that it shows a popular usage of διαθήκη, quite distinct from the legal usage, which overlaps significantly with the semantic range of the Hebrew word בְּרִית as we have defined it. This both strengthens our inductive definition of בְּרִית (not merely “pact” but “an elected relationship of obligation under oath”) and also enables us to allow that the Septuagint translators (who undoubtedly influenced Pauline usage of the term more than anyone else) could have easily had this concept in mind when they used διαθήκη.[8]

Although it is only one instance among hundreds, it is a much more significant instance than the statistics would appear to suggest. Firstly, the statistics are already skewed in favour of the legal usage, since there are far more extant legal texts than comic texts.[9] Secondly, Aristophanes himself was being critically studied and copied by scholars at around the same place and time as the Septuagint was being translated (i.e. Alexandria in the 3rd and 2nd century bc).[10] Hence there are good reasons to suppose that the translators of the Septuagint would have been familiar with the sort of usage we find here in Aristophanes. In any case, our definition of διαθήκη in the Septuagint, “elected relationship of obligation under oath”, has arisen from the inductive semantic study of Hugenberger and others and does not rely on Aristophanes’ usage. This instance in Aristophanes merely helps to confirm a definition arrived at independently.


[1] W. Danker, “διαθήκη”, BDAG 228–29;

[2] This is based on an exhaustive search of the Thesaurus Linguae Graecae. For details of the TLG see Luci Berkowitz and Karl A. Squitier, Thesaurus Linguae Graecae Canon of Greek Authors and Works (3rd ed.; Oxford: Oxford University Press, 1990).

[3] J. Behm and G. Quell, “διαθήκη”, TDNT 2:106–34 (126).

[4] John J. Hughes, “Hebrews IX 15ff. and Galatians III 15ff: A Study in Covenant Practice and Procedure”, Novum Testamentum 21 (1979): 27–96 (30).

[5] Hugenberger, Marriage, 11.

[6] “The various guesses in the scholia show that not even ancient scholars could explain this allusion” (Jeffrey Henderson, “Birds”, in Aristophanes [4 vols.; Loeb Classical Library 178, 488, 179, 180; Cambridge, MA.: Harvard University Press, 1998–2002], 3:1–251 [3:77 fn 30]).

[7] William L. Lane, “Covenant: The Key to Paul’s Conflict with Corinth”, Tyndale Bulletin 33 (1982): 3–29 (22).

[8] See also the definition in Behm and Quell, TDNT 2:112: “a legal fellowship under sacral guarantees”.

[9] Henderson, “Introduction”, 1:33.

[10] Henderson, “Introduction”, 1:33.

Full bibliography