Michael Horton details his distinct view of the Mosaic covenant in his book God of Promise: Introducing Covenant Theology. According to Horton, “two very different types of covenantal arrangements” exist in the Old Testament (35). Horton believes that there are two kinds of covenants: covenants of law and covenants of promise. Yet immediately after speaking of two kinds of covenants, Horton speaks to two different principles, law and promise (35). It is unclear whether covenant and principle are the same or whether certain covenants can contain both principles.

Horton is insistent that salvation can only come from a covenant of grace (36). In other words, salvation is never found in the “covenant of laws” which is based on works, but it is only found in the covenant of promise which is based on God’s bestowal of grace upon the sinner. Horton believes that all biblical covenants can be split up and designated as either conditional (works-based) covenant or unconditional (promise-based) covenants. Essentially, Horton is arguing that the covenant with Adam and Moses are conditional covenants while the covenants with Noah, Abraham, David, and new covenant are unconditional covenants.

Furthermore, Horton believes that the Mosaic covenant was given only to administrate temporal blessings to the people (38). In essence, the Mosaic covenant was a “carrier” for the eternal law of God in which all people are to render perfect obedience. Horton writes, “[The Judaizers] confused the relative fidelity required in the national covenant and thus they remain in the typological land with the absolute faithfulness required of every person in order to fulfill all righteousness and thus appear safely in God’s heavenly presence” (38).

Unfortunately, Horton displays some lack of consistency in speaking about the covenants and how to categorize them. Horton argues that “not all biblical covenants fit the suzerainty treaty pattern” (36). In other words, some covenants are “promise” covenants, or to use ANE parallel terminology, “royal grant covenants.” In particular, the Abrahamic covenant is a “royal grant covenant” (41). Yet, on the very same page, Horton speaks of a “suzerain-vassal” type being enacted in Genesis 15! Horton does not deny that responsibilities are included in the Abrahamic covenant but that in Genesis 15, God “assumes all of the responsibility for carrying the promise through to the end and bearing all the curses of its breach” (41). So is the covenant a suzerain-vassal treaty or royal grant? Horton seems to think it’s both.

Horton seeing Abraham’s obedience in Genesis 22 not as the ground of justification, but as the means by which God’s promise to come to the heirs. Abraham’s obedience essentially is “typological” of Christ’s obedience who wins eternal life through his obedience (45). So while the covenant with Abraham is unconditional and Abraham’s obedience is merely a way for passing on the promise, the covenant God makes with Israel is “strictly conditional” (47). Horton believes the Mosaic covenant is of a different “structure” on the other biblical covenants (47). Horton hammers home the idea that there are two covenant traditions that run through the Old Testament (48). The Mosaic covenant is one of works, whereby the people of Israel would keep themselves in the Promise Land, while the Abrahamic covenant is one of grace, guaranteeing the salvation of God’s people.

Despite most of the chapter defending that Israel’s relationship with God is based on law, Horton goes on to argue that some aspects of Israel’s life to be based on grace: “The relationship between God and Israel is not based entirely on law” (50). Horton points to Israel’s redemption from Egypt and the exodus as acts of grace.