The Westminster Confession of Faith, Chapter 7:2 states:
The first covenant made with man was a covenant of works, wherein life was promised to Adam; and in him to his posterity, upon the condition of perfect obedience.
The term covenant does not appear in Scripture before Genesis 6:18.1 Because of this, some scholars have questioned whether the original relationship between God and man should be considered covenantal. For example, John Murray prefers to call the period the “Adamic administration.” He writes:
This administration has often been denoted ‘The Covenant of Works.’ There are two observations. (1) The term is not felicitous, for the reason that the elements of grace entering into the administration are not properly provided for by the term ‘works’. (2) It is not designated a covenant in Scripture. Hosea 6:7 may be interpreted otherwise and does not provide the basis for such a construction of the Adamic economy. Besides, Scripture always uses the term covenant, when applied to God’s administration to men, in reference to a provision that is redemptive or closely related to redemptive design.2
In response to Murray, it is not clear whether “elements of grace” are present at the focal point of the “administration,” namely the prohibition not to eat of the tree of the knowledge of good and evil and Adam’s decision to break God’s command. It is also argued below that Hosea 6:7 is best interpreted as referring to a covenant with Adam. John Bolt observes that “the mere lack of an explicit use of the Hebrew word berith in Genesis 1-2 is a weak argument from silence and insufficient reason to deny the covenantal character of the passage.”3 Passages in Jeremiah also point to a covenant with Adam as does the parallel between Adam and Christ as legal representatives in Romans 5:12-19 and 1 Corinthians 15:22, 45-47.
W. Wilson Benton, Jr. argues that the two-covenant schema of the covenant of works and covenant of grace found in the Westminster Confession of Faith is not biblical, but is primarily the result of the influences of political thought of the period, a renewed interest in the history of revelation, and the influence of Ramist logic (Pierre de la Ramee – Latin name of Petrus Ramus). He argues that the Ramist focus on the deductive method and the dichotomizing of ideas led to the federalist theological construction. He argues that federalist theology produces horrible theological effects. Benton writes of covenant or federal theology:
. . . it must be asked whether the system used to unify the whole of biblical teaching, and the categories used to make it historical, are faithful to the nature of the truth which they present or whether some alien systematic principle has been employed. At these points federal theology “is weighed in the balance and found wanting.”4
Robert Reymond argues against Benton’s position and says that Benton’s “influences alone cannot account for federal theology or show how federal theology produces the dire effects Benton sees federal theology producing.”5 In spite of these arguments against the traditional covenant of works/covenant of grace distinction, there are strong exegetical reasons for viewing the pre-fall relationship between God and man as covenantal. Historically, many Reformed theologians have argued for a pre-fall covenant. The arguments are usually framed along these lines:
First, all the elements of a covenant are present. In Genesis 2, there are parties, stipulations, promises, and threats for disobedience (Gen. 2:16-17). Since all the ingredients for a covenant are present, the relationship with God and man prior to the fall as well as prior to Noah may be considered covenantal.
Second, the Hebrew word for covenant, b’rith, does not have to be present at the time a covenant is made. For example, 2 Samuel 7 (cf. 2 Chron. 17) sets forth the establishment of the Davidic covenant, but the term covenant is not used in that context. Psalm 89:1-3, 19-37 and 2 Sam. 23:5 both state that God established a covenant with David and promised that his dynastic house would rule over Israel. The same pattern is possible in regard to God’s original relationship with man.6
Just as in the case of the Davidic covenant, subsequent Scriptures refer back to God’s relationship prior to the fall as covenantal. Hosea 6:7 states: “But like Adam they have transgressed the covenant; there they have dealt treacherously against Me” (NASB). Three possible interpretations of this passage have been suggested. First, the traditional interpretation is that the phrase, “like Adam,” refers to the first man and, therefore, refers back to a pre-fall covenant with Adam. A. Cohen points out that Jewish commentators traditionally have applied this phrase “to the disobedience of Adam in the Garden of Eden.”7
Second, “Adam” could designate a place where Israel broke the covenant. This is the most tenuous of the three possibilities. The place “Adam” is located on the Jordan River about twelve miles north of Jericho. The narrative of the Jordan rolling back to Adam (Josh. 3:16) does not refer to any national sin of Israel. An emendation of the massoretic text is necessary to support his interpretation. For example, James Luther Mays argues for a substitution of b’adam (at Adam) for c’adam (like Adam).8 However, the text as it stands reads “as or like Adam.” Louis Berkhof writes concerning Hosea 6:7:
Attempts have been made to discredit this reading. Some have suggested the reading “at Adam,” which would imply that some well-known transgression occurred at a place called Adam. But the preposition forbids this rendering. Moreover, the Bible makes no mention whatever of such a well-known historical transgression at Adam.9
Third, it is possible to interpret the text with the meaning that Israel has broken the covenant “like man” or “like mankind.”10 The point of this comparison is that Israel has transgressed the covenant just like men or non-Israelites have broken the covenant. O. Palmer Robertson comments on this idea:
In what sense may it be affirmed that non-Israelite man stands in a covenantal relationship with God that may be broken? No specific covenant with man outside Israel finds any mention in Scripture other than God’s covenant with Noah, which lacks adequate emphasis on specifics of covenantal obligation for Hosea to say with convincing clarity that man has “broken” the covenant.
Apparently Hosea intends to suggest that God has established a covenant relationship with man outside Israel through creation. If “Adam” is taken individually, the term would refer to the original representative man.11
Therefore, whether the phrase is interpreted as “like Adam” or “like man” both point back to an original covenant with man in the Garden of Eden established at the time of creation. However, it is best to understand the phase as pointing to Adam, the first man and his violation of the original creational covenant.12
O. Palemer Robertson finds evidence for the pre-fall covenant in a comparison of passages from Jeremiah to the creation narrative.13 The following analysis is a summary of his argument. Jeremiah 33:20-21, 25-26 indicates that the original relationship between God and man was covenantal:
20. “Thus says the Lord, ‘If you can break my covenant for the day, and My covenant for the night, so that day and night will not be at their appointed time, 21. then My covenant may also be broken with David My servant that he shall not have a son to reign on this throne, and with the Levitical priests, My ministers.
25. “Thus says the Lord, ‘If My covenant for day and night stand not, and the fixed patterns of heaven and earth I have not established, 26. then I would reject the descendants of Jacob and David My servant, not taking from his descendants rulers over the descendants of Abraham, Isaac, and Jacob. But I will restore their fortunes and will have mercy on them’” (NASB).
In verse 20, the Lord through Jeremiah speaks of “my covenant for the day and my covenant for the night.” Similarly, in verse 25, a covenant for day and night is mentioned. There are two possibilities for this reference to a covenant of day and night. The first possibility is God’s original ordinances at the time of creation. The second possibility is God’s covenant with Noah. In the Noahic covenant, similar language is employed in Genesis 8:22: “‘While the earth remains, seedtime and harvest, and cold and heat, and summer and winter, and day and night shall not cease.” Since day and night are mentioned in the covenant with Noah, Jeremiah could have been referring back to that covenant. It is also possible that Jeremiah is referring to the third day of creation when God created the Sun, moon, and stars and set an order for them in creation (Gen. 1:14). A closely related passage in Jeremiah provides insight into whether Jeremiah is referring to the time of Noah or creation. Jeremiah 31:35-36 has basically the same structure as the Jeremiah 33 passage:
35. Thus says the Lord,
Who gives the sun for light by day,
And the fixed order of the moon and the stars for light by night,
Who stirs up the sea so that its waves roar;
The Lord of hosts is His name:
36. “If this fixed order departs from before Me,” declares the Lord,
“Then the offspring of Israel also shall cease
From being a nation before Me forever” (NASB).
The term translated as “fixed order” or “statute” in this passage is the Hebrew word, chok, which is used elsewhere in Scripture as a parallel word for covenant (cf. 1 Kings 11:11; 2 Kings 17:15; Psa. 50:16; 105:10). Therefore, it is possible to understand the term “fixed order” in Jeremiah 31 as a reference to the covenant idea expressed in Jeremiah 33.
Jeremiah 31 has a similar structure and argumentation as Jeremiah 33. Both talk about the sun’s rule over the day and the moon’s rule over the night. However, Jeremiah 31:35 refers to the sun and the moon and stars respectively as light bearers for the day and night. This idea is found in the creation account, but not in the Noahic covenant. Genesis 1:16 refers to the stars as well as the moon as light-bearers for the night as does Jeremiah 31:35. The Genesis 8:22 passage, dealing with the Noahic covenant, does not mention the stars. Therefore, it is more probable that Jeremiah 31 is referring to the creation account rather than to the account of the Noahic covenant. Jeremiah 31 and 33 closely parallel each other and use similar language and argumentation. Since Jeremiah 31 most likely refers to the creation narrative, Jeremiah 33 also most likely refers to the same period. Therefore, since Jeremiah 33 employs the term covenant in its referencing of the creation account, it is a subsequent Scripture that affirms the original divine-human relationship at creation as covenantal. It is also important to note that the Noahic covenant echoes the creational covenant indicating a continuation of concepts that were already in place from the time of creation.
Having affirmed the idea of a pre-fall covenantal relationship between God and man, it is important to examine the contents of this covenant. This original covenant, traditionally called the covenant of works, has strong bearing on how one understands the work of the Messiah. Therefore, it is important to understand the meaning of this covenant and the relationship it has to the work of the Messiah. Robert Reymond gives this description of the covenant of works:
The covenant which God originally made with Adam was a divinely arranged suzerainty pact wherein, on the divine side, God bound himself to both promise and threat while, on the human side, Adam was expected to obey the covenantal stipulations which were accompanied by God’s promise of blessing for obedience and threat of sanction for disobedience.14
Louis Berkhof sets forth the elements of this covenant:
(1) Adam was constituted the representative head of the human race, so that he could act for all his descendants. (2) He was temporarily put on probation, in order to determine whether he would willingly subject his will to the will of God. (3) He was given the promise of eternal life in the way of obedience, and thus by the gracious disposition of God acquired certain conditional rights. The covenant enabled Adam to obtain eternal life for himself and for his descendants in the way of obedience.15
Some scholars do not accept the idea that, upon the condition of obedience, eternal life was promised to Adam. It is argued there is no Scriptural evidence of such a promise. While it is true there is no explicit promise of eternal life as a result of obedience, the promise is implied in the alternative threat of death in the case of disobedience. The implication is that if Adam was obedient, then death would not be present. If Adam would have been obedient then he would have continued in a life of communion with God. It is at this point where the covenant of works contributes to understanding the work of the Messiah. In Romans 5:12-21, Paul presents a parallel between Christ and Adam in the context of his discussion of justification. Paul argues that just as Adam’s sin was imputed to the race, all those in Christ receive the imputation of Christ’s righteousness. This leads to two concepts. First, this demonstrates a representative idea in that Adam stood as a federal or covenant head to his descendants just as Christ stood as a federal head of those who are in him (Rom. 5:14-17). Just as Adam is a legal representative so also is Christ a legal representative of all who are in him. Second, Christ’s perfect obedience to the law fulfilled all the original demands of obedience in the covenant of works. Those in Christ now receive all the benefits of the original covenant through Christ’s perfect fulfillment of it (Rom. 5:18-19). The covenant of grace is, therefore, an extension and gracious fulfillment by Christ of the covenant of works for sinners who are in Christ. Robert Reymond writes:
. . . the covenant of grace should be seen as providing the requisite redemptive provision as a second-level “covenantal overlay” upon the covenant of works. What this means is that Christ the “second Man” stepped forward, representing certain sinners who could not themselves keep the covenant (it is in his representation of these undeserving sinners and in all that this entails for them that the grace of the covenant of grace is exhibited), and as the “last Adam” he kept (where Adam had not) all of the requirements of the covenant in their behalf by meeting both the preceptive and penal demands of the covenant of works.16
Therefore, the concept of the covenant of works has direct bearing on the work of the Messiah who was promised just after Adam’s disobedience to this original covenant.
Robert Reymond notes that some biblical scholars are attacking the concept of works and justice in regard to God’s original covenantal relationship with Adam. He notes that Daniel P. Fuller argues that whatever Adam received from God must be understood in terms of God’s grace. This grace is not viewed in terms of redemption or salvation, but simply in terms of God’s goodness. Fuller denies any concept of a works/grace contrast in God’s relationship with man.17 Fuller argues that even God’s pre-fall relationship with Adam should be viewed exclusively in terms of grace. The result of this is that Fuller, in denying any concept of works or justice in man’s relationship with God, destroys any concept of true grace. Fuller argues that many passages present works as an instrumental cause of justification. He says that these works are not meritorious, but he essentially denies justification by faith alone. Fuller claims that “grace” performs justification through the “work or obedience of faith.”18 However, the genitive in the Pauline phrase, “obedience of faith” (Rom. 1:5; 16:26) should be understood as a genitive of source meaning “the obedience that flows from faith” or as an appositional genitive meaning “obedience that consists of faith.”19 After examining this view, Robert Reymond writes:
Accordingly, a view that insists upon “grace” everywhere winds up with true grace nowhere and a kind of works principle everywhere, with his representation of the relation of works to justification coming perilously close to what late medieval theologians would have called works having not condign but congruent merit. One thing is certainly clear from Fuller’s representation of this whole matter: he has departed from the sola fide principle of the Protestant Reformation.20
W. Robert Godfrey comments on Fuller’s position:
Fuller’s position is clear, explicit rejection of the Reformed doctrine that “Faith, thus receiving and resting on Christ and His righteousness, is the alone instrument of justification. . .” (Westminster Confession of Faith, XI, 2).
Fuller’s revision affects the basic understanding of the application of redemption. Historic Protestantism insisted that justification was by faith alone because faith alone looked outside of itself to rely on the perfect work of Christ. Faith justifies not because it is a virtue that pleases God, but because faith abandons all self-confidence and rests in Christ and his finished work. Faith trusts that Christ has fulfilled all righteousness and borne God’s wrath for sin. Fuller, by contrast, changes faith’s whole relation to justification. He defines faith in terms of obedience. Faith is work. Justifying faith then is not an exclusively extraspective resting in the work of another.21
This demonstrates how important the concept of the covenant of works is in regard to understanding the work of the Messiah. A rejection of the covenant of works principle ultimately denies the legal representative parallel between Adam and Christ (Rom. 5:12-21; 1 Cor. 15:21-22). This would deny the principle of the imputation of Christ’s righteousness because if Christ’s obedience is not meritorious, then his preceptive obedience is not imputed to believers. Meredith G. Kline argues that justice, not grace, is the main element of continuity between the pre-fall covenant and the covenant of grace. He writes:
The necessity of affirming the traditional works principle [in Genesis 2] becomes clear if we concentrate on the subject of justification in God’s covenantal dealings with Adam and Christ. If the first Adam had obediently fulfilled the stipulations of God’s covenant with him, then assuredly he would have been worthy of being declared righteous by his Lord. Adam’s justification would have been on the grounds of his works and would have been precisely what those good works deserved. God’s declaring Adam righteous would have been an act of justice, pure and simple. In fact, any other verdict would have been injustice. There is absolutely no warrant for obscuring the works character of such an achievement of justification by introducing the idea of grace into the theological analysis of it. Indeed, to do so would be in effect to suggest that God has the capricious capability to behaving like the Devil, declaring evil what is good and good what is evil. . . .
Rejection of the works principle [with reference to Adam] extends in the logic of the Fuller-Shepherd theology to the Second Adam. This can be shown from the argument Shepherd uses against the recognition of that principle in covenant administration. He notes that the covenantal relationship is a father-son relationship and from this concludes that parental grace, not any claim of strict justice, accounts for any favorable treatment man receives from God, his Father. But if the elimination of simple justice as the governing principle is thus due to the presence of a father-son relationship, mere justice could no more explain God’s response to the obedience of his Son, the second Adam, than it could his dealings with the first Adam. This means that in the Fuller-Shepherd theology, consistently developed, the work of obedience performed by Jesus Christ did not merit a verdict of justification from his Father. The justification of the second Adam was not then according to the principle of works in contrast to grace, but rather found its explanation in the operation of a principle involving some sort of grace – a grace required because of the inadequacy of Christ’s work to satisfy the claims of justice.22
The covenant of works affirms the principle of justice in regard to the work of the Messiah and is, therefore, crucial to understanding the redemptive work of the Messiah. Mark W. Karlberg notes the current debates in Reformed theology concerning the covenant of works and the importance of the justice principle in the covenant of works:
Among recent detractors of traditional Reformed teaching on the Covenant of Works two proposals have appeared: (1) that we abandon altogether the federalist system of interpretation or (2) that we undertake a thoroughgoing revision of the doctrine. Common to all these critics is denial of the validity of the Covenant of Works idea. They claim that the idea of merit does not find support in Scripture. In our view, however, it is a matter of justice for God to grant eternal life to his obedient image-bearers. Failure to recognize this element of the system of truth contained in the Scriptures leads to a defective understanding of the atonement, specifically the necessity of Christ’s atoning death as means of satisfying divine justice.23
Michael Horton concurs and writes:
It is therefore premature to insert into the creation covenant an element of divine graciousness, strictly speaking. To be sure, God’s decision and act to create constitute a “voluntary condescension” (Westminster Confession 7.1), as is his entrance into a covenantal relationship with creatures. Nevertheless, if “grace” is to retain its force as divine clemency toward those who deserve condemnation, we should speak of divine freedom, love, wisdom, goodness, justice, and righteousness as the governing characteristics of creation. Grace and mercy are shown to covenant-breakers and reflect the divine commitment to restore that which is fallen.
It is within this framework, then, that Reformed theology understood the active obedience of Jesus Christ, emphasizing the significance of his humanity in achieving redemption for his covenant heirs. The priority of law in the covenant of creation establishes that God cannot acquit the guilty nor simply forgive sinners. In the context of the covenant of creation, the law must be perfectly satisfied, either personally or representatively.24
Another objection that is raised to the concept of justice in the pre-fall covenant is that any obedience of Adam would not contain enough value to receive the promised blessing of eternal life. Therefore, Fuller and advocates of his approach argue that the pre-Fall covenant can only be understood in terms of grace (as they define it). Robert Reymond responds to this that “this alleged disparity in value between the obedience to be rendered and the reward to be bestowed is very debatable, . . . since insofar as Adam’s obedience would have glorified God and given him pleasure, it would have had infinite value.”25 It is also important to maintain the parallel between the First Adam and the Second Adam. Meredith Kline writes concerning this:
What was true in the covenant arrangement with the Second Adam will also have been true in the covenant with the First Adam, for the first was a type of the second (Rom. 5:14) precisely with respect to his role as a federal head in the divine government. Accordingly, the pre-Fall covenant was also a covenant of works, and there, too, Adam would have fully deserved the blessings promised in the covenant, had he obediently performed the duty stipulated in it. Great as the blessings were to be which the good Lord committed himself, granting of them would not have involved a gram of grace. Judged by the stipulated terms of the covenant, they would have been merited in simple justice.26
This chapter has presented the idea that the original divine-human relationship should be understood as covenantal. The classic terminology for the covenant relationship emphasizes the focal point of the covenant in that man is commanded not to eat of the tree of the knowledge of good and evil. Disobedience is threatened with death and the implication is that obedience would ensue in the continuation of life in a relationship with God. The principle of justice is set forth in this covenant and Adam serves as a type of the Messiah’s work of obedience (Romans 5:12-21). This concept of justice is crucial for understanding the Messiah’s work of obedience and the imputation of that righteousness to the elect of God. Wilhelmus a Brakel summarizes this well:
Acquaintance with this covenant [of works] is of the greaterst importance, for whoever errs here or denies the existence of the covenant of works, will not understand the covenant of grace, and will readily err concerning the mediatorship of the Lord Jesus. Such a person will readily deny that Christ by His active obedience has merited a right to eternal life for the elect. . . . Whoever denies the covenant of works, must rightly be suspected to be in error concerning the covenant of grace as well.27
1 Even though the word “covenant” (b’rith) occurs for the first time in Scripture in Gen. 6:18, it occurs with the pronominal suffix and the Hiphil form of the verb qum, “establish” rather than karath, “cut, make.” This indicates that this covenant was not initially made in Noah’s time, but was a covenant already in existence which was being extended into the time of Noah (Robert L. Reymond, A New Systematic Theology of the Christian Faith, [Nashville: Thomas Nelson, 1997], 512, n. 17.). The author is indebted to Robert Reymond for class lectures and his work, A New Systematic Theology of the Christian Faith from which many of the concepts of this chapter are drawn.
2 John Murray, “The Adamic Administration,” in Collected Writings of John Murray 4 Vols. (Edinburgh: Banner of Truth, 1977), 2:49.
3 Josh Bolt, “Why The Covenant of Works Is A Necessary Doctrine” in By Faith Alone: Answering the Challenges to the Doctrine of Justification, ed. Gary L. W. Johnson and Guy P. Waters (Wheaton, IL: Crossway Books, 2006), 181.
4 W. Wilson Benton, Jr., “Federal Theology: Review for Revision” in Through Christ’s Word, ed. W. Robert Godfrey and Jesse L Boyd III (Phillipsburg, N. J.: Presbyterian and Reformed, 1985), 201. See: “Federal Theology and the Westminster Standards” in The Covenant: God’s Voluntary Condescension, ed. Joseph A. Pipa, Jr. and C. N. Willborn (Taylors, SC: Presbyterian Press, 2005) for an excellent defense of the covenant of works.
5 Robert L. Reymond, A New Systematic Theology of the Christian Faith (Nashville: Thomas Nelson, 1997), 405, n. 23.
6 Reymond, A New Systematic Theology of the Christian Faith, 430. Robertson, The Christ of the Covenants, 18-19.
7 Abraham Cohen, The Twelve Prophets, Hebrew Text, English Translation and Commentary. The Soncino Books of the Bible (London, Socino Press, 1948), 23.
8 James Luther Mays, Hosea. A Commentary. The Old Testament Library (Philadelphia: Westminster, 1969), 100.
9 Louis Berkhof, Systematic Theology, 4th ed. (Grand Rapids: Eerdmans, 1939; reprint, Grand Rapids: Eerdmans, 1981), 214 (page references are to reprint edition).
10 John Calvin, Commentaries on the Twelve Minor Prophets, vol. 1, Hosea (Edinburgh, Scotland: Calvin Translation Society, n.d.; reprint, Grand Rapids: Baker, reprint, 1984), 233 (page references are to reprint edition).
11 Robertson, The Christ of the Covenants, 23-24.
12 See also Benjamin B. Warfield, “Hosea vi:7: Adam or Man?”, Selected Shorter Writings of Benjamin B. Warfield, ed. J. E. Meeter (Nutley, N. J.: Presbyterian and Reformed, 1970), 1:116-129.
13 Robertson, The Christ of the Covenants, 19-21.
14 Reymond, A New Systematic Theology of the Christian Faith, 430.
15 Berkhof, Systematic Theology, 215.
16 Reymond, A New Systematic Theology of the Christian Faith, 440. Louis Berkhof notes these same points in his discussion of the points of difference between the covenant of works and the covenant of grace (Systematic Theology, 272).
17 Daniel P. Fuller, Gospel and Law: Contrast or Continuum? (Grand Rapids: Eerdmans, 1980), 103, 109, 118-120. See Robert Reymond’s examination of this position in A New Systematic Theology of the Christian Faith, 431.
18 Daniel P. Fuller, “A Response on the Subjects of Works and Grace,” Presbuterion 9, no. 1-2 (1983): 79. Cited by Reymond in A New Systematic Theology of the Christian Faith, 431.
19 Reymond, A New Systematic Theology of the Christian Faith, 431, n. 19.
20 Reymond, A New Systematic Theology of the Christian Faith, 431-432.
21 W. Robert Godfrey, “Back to Basics: A Response to the Robertson-Fuller Dialog,” Presbuterion, 9, no. 1-2 (1983), 81.
22 Meredith G. Kline, “Of Works and Grace,” Presbuterion, 9, no. 1-2 (1983), 88-89.
23 Mark W. Karlberg, Covenant Theology In Reformed Perspective (Eugene, Oregon: Wipf and Stock Publishers, 2000), 104.
24 Michael S. Horton, Lord and Servant (Louisville: Westminster John Knox Press, 2005), 132-133.
25 Reymond, A New Systematic Theology of the Christian Faith, 433.
26 Meredith G. Kline, “Covenant Theology Under Attack,” New Horizons (Feb. 1994), 4. Cited by Reymond in A New Systematic Theology of the Christian Faith, 433.
27 Brakel, Our Reasonable Service, 1:355. Cited in Josh Bolt, “Why The Covenant of Works Is A Necessary Doctrine” in By Faith Alone: Answering the Challenges to the Doctrine of Justification, ed. Gary L. W. Johnson and Guy P. Waters, 184.