God relates to His people through covenant, seen in the covenants He made with Noah, Abraham, Moses, and David, as well as the new covenant. If a Christian wants to understand how the Bible fits together, he should thus study covenant theology. And there is a great new resource available on the subject, Covenant Theology: Biblical, Theological, and Historical Perspectives, edited by Guy Waters, Nicholas Reid, and John Muether.
This 600-page hardcover book covers a wide variety of topics within covenant theology. While a book with multiple authors sacrifices unity (in both writing style and theology), Covenant Theology also gains the knowledge of 20+ authors (all associated with Reformed Theological Seminary) that would be near-impossible for one man to produce. (In full disclosure, I studied under several of these men in seminary, though I can assure you my critiques are impartial.) This book is worth it alone for the footnotes and references for further study. It is a resource every pastor and serious student of the Bible should have on his shelf.
At 27 chapters, an interaction with every chapter would take up a full book of its own. So instead I will highlight some chapters and subjects of interest and offer my overall thoughts. I focus on on the biblical chapters and where there is diversity of thought within Reformed covenant theology (sometimes even among the authors themselves).
The Covenant of Works
Since this is a Reformed work, it advocates the bi-covenantal (dichotomist) view that contrasts (1) God’s “covenant of works” with Adam in the garden and (2) God’s “covenant of grace” made with Christ (and His elect) where God redeems His people. The covenant of grace is an overarching covenant of different administrations (Adamic, Noahic, Abrahamic, Davidic, new), and this covenant was needed to save people because Adam broke the initial covenant of works with God that promised eternal life for obedience and death for disobedience.
Covenant Theology contains two chapters on the covenant of works, one by Richard Belcher on the covenant of works in the Old Testament (ch. 2) and a second by Guy Waters on the covenant of works in the New Testament (ch. 3). Belcher argues that, though the word “covenant” is not used in Genesis 1–3, all the elements of a covenant exist in this passage, including two parties, conditions, blessings and curses, and a covenant sign (64-66). He provides helpful discussions of Hosea 6:7, differences over the name of the covenant, and the role of grace in the covenant (66-69). Waters supports this case by explaining Paul’s contrast of Adam and Christ in 1 Corinthians 15 and Romans 5 (81-90).
The Covenant of Works, the Mosaic Covenant, and Christ
I want to explore how the authors of Covenant Theology (primarily Belcher, Waters, and Reid) think the covenant of works applies today, as well as its relationship to the Mosaic covenant and the work of Christ. Note that this section can get a little technical, as the Mosaic covenant is the most difficult area of covenant theology.
Belcher holds that Adam’s punishment is not only passed on to his descendants, but also that there is a “continuing obligation to fulfill the covenant requirements” of the covenant of works, which is affirmed later in Scripture under the principle “Do this and live,” citing Leviticus 18:5 (“if a person does them, he shall live by them”); Romans 10:5; and Matthew 19:16-17 (70, 73). Belcher says Christ fulfills the “obligations” of the covenant of works on behalf of His people by perfectly keeping the law (70). In this way, “Salvation is by works”—Christ’s works—and thus “the covenant of works is foundational to the work of Christ as the basis for our salvation” (71). Belcher does not explain all the details of how Christ fulfilled the obligations of the covenant of works, though he presumably refers to Christ keeping the moral requirement of the Mosaic law. Belcher mentions a “principle of works” (73), but he uses the phrase “works principle” in his recent book on covenant theology (Belcher, The Fulfillment of the Promises of God, 177-180).
While Belcher does not specify his interpretation of Leviticus 18:5 as quoted in Romans 10:5, he appears to follow Bryan Estelle’s republication view (argued in the 2009 book The Law Is Not of Faith) that “Paul is pointing to an actual works principle contained within the Mosaic administration of the covenant of grace” (169). Contrary to Estelle’s view, Waters in Covenant Theology (containing content also found in his chapter in The Law Is Not of Faith) argues that Paul in Romans 10:5 only uses Leviticus 18:5 to respond to his opponents who erroneously sought to rely on the Mosaic law apart from God’s grace (90-92, 96, 169). Thus, Waters denies the republication of the covenant of works in the Mosaic covenant—“Paul did not understand God to promulgate the Mosaic covenant to Israel as a covenant of works” (92). Rather, “When one removes the Mosaic legislation from the covenant in which it was promulgated, then one is left with the covenant of works . . . God has built into his gracious Mosaic covenant a testimony or warning against its misuse (Lev. 18:5)” (96). Though Belcher seems to adopt Estelle’s position, he in fact only cites Waters’ work on Romans 10:5 in his book on covenant theology, and he makes clear there that he rejects Kline’s republication view (Belcher, The Fulfillment of the Promises of God, 34, 92, 186-188).
How the covenant of works applies to us today gets into the issue of the imputation of Christ’s righteousness. Belcher only speaks generally of the imputation of Christ’s righteousness to His people, but he likely has the imputation of Christ’s “active obedience” specifically in view, which he later mentions (77) and which authors Nicholas Reid and Michael Kruger embrace (154, 216). Some modern Reformed theologians follow Meredith Kline in making a direct connection between the covenant of works and the imputation of Jesus’ active obedience based on a “republication” of the covenant of works in the Mosaic covenant. In his chapter on the Mosaic covenant (ch. 7), Nicholas Reid explains how Kline’s republication view holds that God only required an imperfect obedience from Israel in order to retain the land He had given them. whereas the original covenant of works required perfect obedience (168). Yet Kline still thinks the Mosaic covenant was part of the covenant of grace. Thus, Kline holds that the Mosaic covenant was part of the covenant of grace at the individual level, but at the national level it served as having typological merit for Israel (167). And this typological merit points to Christ’s meritorious active obedience that is imputed to the believer.
Several authors in Covenant Theology clearly reject Klinean republication. Belcher says so in his other work, and Waters and John Scott Redd say so in Covenant Theology (92, 143). However, Nicholas Reid’s position is not entirely clear. He considers the Mosaic covenant to be part of the one covenant of grace, though “a weak administration that results in a curse” (155). He says that “it is incorrect to assume that without a republication of a covenant of works, the active obedience of Christ cannot be foreshadowed in the old covenant” (169), implying he rejects Kline’s republication view. However, Reid also affirms that “the Mosaic administration of the covenant of grace was a covenant of works for Christ” (169). Christ “fulfilled the covenant of works for his people” and justified His people “through his active and passive obedience” (154).
Assuming Reid rejects Klinean republication, along with Belcher, it is not fully explained how the Mosaic covenant is (1) gracious and not a republication of the covenant of works for Israel, (2) but is at the same time a covenant of works for Christ. Neither Belcher nor Reid explain this point in Covenant Theology. I was only able to find clarification after consulting Belcher’s recent book on covenant theology, where he explicitly denies republication but affirms that there is still a “works principle” in the Mosaic covenant that Jesus kept for His people. There Belcher says:
The Mosaic Covenant should not be identified with the Covenant of Works as a covenantal administration, but the requirement of perfect obedience set forth in the Covenant of Works is an abiding requirement . . . there is a works principle still operative in the Covenant of Grace in a secondary sense related to the second use of the law which is particularly evident in the Mosaic Covenant . . . There is continuity with the Covenant of Works in the second use of the law showing that the requirements of the law still need to be met for there to be salvation. But the Mosaic Covenant is not the Covenant of Works because there was no provision in the Covenant of Works for salvation if the law was broken (Belcher, The Fulfillment of the Promises of God, 92-93).
Thus, Belcher’s position is that, based on Leviticus 18:5 (and its use in Romans 10:5), the works principle of the covenant of works requiring perfect obedience for salvation is operative in the Mosaic covenant, and Jesus fulfilled this for the salvation of all who believe in Him. This seems to be in line with the historic Reformed position that the moral law (not the administration of the covenant of works) given to Adam was republished in the Mosaic covenant (WCF 19.2), which Jesus fulfilled. (To add to the difficulty, some refer to this as material republication, in distinction from Kline’s formal republication.) However, Belcher’s explanation is confusing because he uses the language of a “works principle,” which is associated with Kline’s republication view. In fact, Cornelis Venema says he is not aware of writers of the orthodox Reformed period using the phrase (Venema, Christ and Covenant Theology, 72). Thus, Belcher uses Kline’s language of a “works principle” but rejects its application to an imperfect national typology, which oddly is the only way Kline said such a works principle applied in the Mosaic covenant—“The works principle in the Mosaic order was confined to the typological sphere of the provisional earthly kingdom” (Kline, Kingdom Prologue, 321).
In summary, the relationship of the Mosaic covenant to the covenant of works in not clearly explained in Covenant Theology, and the fact that I had to consult other books for clarification suggests that readers will likely find this subject to be addressed inadequately. This issue should have received more attention, if not its own chapter. This is a challenging theological issue, but the lack of clarity may also stem in part from disagreement among the authors.
Variations of the Covenant of Works
To his credit, Belcher does give attention to variations of the covenant of works within Reformed theology. He discusses John Murray’s rejection of both the terms “covenant” and “works” for God’s relationship with Adam, since the word “covenant” is not used in Genesis 1–3 and elsewhere in Scripture is only used in situations requiring redemption, and since the word “works” does not allow for God’s grace that is present in the garden. Murray prefers the phrase “the Adamic Administration” (72). Belcher criticizes Murray for failing to recognize the elements of a covenant in Genesis 1–3 and for rejecting the “works principle” operating throughout redemptive history, including Leviticus 18:5 and Romans 10:5 (73).
Belcher also discusses James Jordan’s view that Adam needed to mature in the garden, not merit eternal life (75-77). Belcher notes that Jordan’s rejection of merit in the covenant of works leads him to also reject merit in Christ’s work, and thus he rejects the imputation of Christ’s active and passive obedience. Instead, Jordan holds that Christians receive Christ’s glorification, not His merit. Belcher argues this is out of line with the Westminster Standards and the Reformed tradition—“This approach flattens the confessional understanding of the relationship between the covenants by denying the principle of works” (77). Belcher says Jordan’s approach downplays the law, redefines imputation, and changes how we understand the basis of justification by faith.
Two Noahic Covenants?
In addition to the Mosaic covenant, there is also considerable disagreement within covenant theology over the Noahic covenant. Miles Van Pelt surveys the Noahic covenant in ch. 5. His opening line explains his view:
The Noahic covenant recorded in Genesis 9 is a universal, unilateral, nonredemptive administration of the covenant of grace restoring and securing the principle of common grace in this world that was suspended during the judgment ordeal of the flood (111).
Thus, Van Pelt takes the view that there are two different Noahic covenants—one covenant in Genesis 6:18 with Noah for salvation, and a second non-redemptive, common grace covenant in Genesis 9 with all creation. This follows the likes of Meredith Kline and, more recently, David VanDrunen in his Reformed two kingdoms writings (such as Living in God’s Two Kingdoms). (For the view that there is one unified Noahic covenant, Van Pelt cites Paul Williamson’s Sealed with an Oath, 59-60.) This Noahic dichotomy is significant, as VanDrunen bases his two kingdom dichotomy on it—the redemptive kingdom of the church is rooted in the covenant with Noah in Genesis 6:18 and the common kingdom of the world is rooted in the covenant with all creation in Genesis 9.
Van Pelt notes that God makes a covenant with the individual Noah in Genesis 6:18-21, while the covenant in Genesis 9 is with the plural “you,” which includes Noah’s offspring and the animals (118). While the Genesis 6 covenant guarantees Noah and his family would survive the flood, the Genesis 9 covenant guarantees a worldwide flood will not happen again. Van Pelt also takes the Hebrew verb קוּם (qum) in 6:18 as to “confirm” a prior covenant rather than to “establish” a new covenant (119). Seeing that the word “covenant” is first used in the Bible in Genesis 6:18, what prior covenant could this confirm? Van Pelt thinks Genesis 6:18 confirms the redemptive judgment of Genesis 3:14-19 (while others argue it confirms a creational covenant of Genesis 1–2).
However, while the translation of “confirm” for קוּם (qum) in Genesis 6:18 in reference to a prior covenant is a possible understanding, it is far from certain. Moreover, the division between the covenant in Genesis 6 and Genesis 9 is arbitrary and ignores important connections throughout the narrative. As I argued in my critique of VanDrunen’s two kingdom view on this point, the Noahic covenant of Genesis 6 and Genesis 9 cannot be so neatly separated. In Genesis 6, God instructed Noah to take animals (all creation) on the ark with him, two of every unclean animal (Genesis 6:19-21) and seven of every clean animal (Genesis 7:2-3). Thus, the covenant in Genesis 6 has all creation in view, just like Genesis 9.
The covenant in Genesis 9 (vv. 9, 11, 12, 13, 15, 16, 17) also involves sacrificial animal offerings, similar to that of the Levitical system. Yahweh’s promise not to strike down living creatures was in response to Noah’s sacrifice—“And when the LORD smelled the pleasing aroma, the LORD said in his heart, ‘I will never again curse the ground because of man, for the intention of man’s heart is evil from his youth. Neither will I ever again strike down every living creature as I have done’” (Genesis 8:20-21). Sacrifice is intricately connected with God’s promise to never again destroy creation by flood.
Therefore, Genesis 9 is not a mere “common grace” covenant but involves sacrifices related to redemption. These are not two separate covenants, but the covenant in Genesis 9 is a continuation and extension of the promise in Genesis 6:18. It is noteworthy that the author in the very next chapter, John Scott Redd, shares my position—“The Noahic covenant is singular and complete even though it is administered at different points and with different emphases before and after the flood” (134). This is not just tension among the authors, but outright disagreement on the structure of the Noahic covenant.
Historical and Theological Chapters
The second section of Covenant Theology includes seven historical chapters, surveying covenant theology in the early church to the present day. These chapters provide helpful historical information and interact with prominent writers of old on covenant theology. The medieval, Reformation, and post-Reformation chapters are all a wealth of information.
I was very pleased to find an entire chapter devoted to “Covenant Theology in the Dutch Reformed Tradition” (ch. 18) by Bruce Baugus. This is often a neglected area among Presbyterians, in part because the Dutch Reformed confessional standards, the Three Forms of Unity, have a less-developed covenant theology than the Westminster Standards. However, Baugus interacts with important Dutch theologians such as Johannes Cocceius, Franciscus Junius, Franciscus Gomarus, Gilbertus Voetius, Herman Witsius, and Welhelmus à Brakel. He also briefly mentions modern Dutch theologians such as Geerhardus Vos, Herman Hoeksema, Klaas Schilder, Abraham Kuyper, Herman Bavinck, and Louis Berkhof. Several of these men are well-known today among the Reformed churches.
The third section of the book is called “Collateral and Theological Studies.” This includes chapters on ancient Near Eastern covenants (ch. 21), covenant in second temple Judaism (ch. 22), and the sacraments (ch. 27). There are also two helpful chapters on non-Reformed approaches to the covenants—one on dispensationalism (ch. 25) and one on “new covenant theologies” (ch. 26). In the latter chapter, Scott Swain interacts with “new covenant theology” and “progressive covenantalism” (e.g. Gentry and Wellum’s Kingdom through Covenant), which are attempted middle ways between dispensationalism and covenant theology. Swain contrasts these views with Reformed covenantal theology and discusses the differing approaches to the new covenant in Jeremiah 31:31-34.
A Glaring Omission (Dual Aspect of the Covenant)
A glaring omission in my opinion is the lack of a chapter on the dual aspect of the covenant (legal vs. vital aspect). As I have argued previously, dual aspect theology, including the distinction between covenant and election, is necessary for grounding the practice of infant baptism and for understanding biblical warnings against apostasy/covenant-breaking (such as Hebrews 6 and 10). Ch. 12 is entirely devoted to the covenant theology of Hebrews, yet it astonishingly has no discussion of the apostasy passages.
There are only scattered references to these issues throughout the book. In ch. 16, Howard Griffith says that “election and covenant are not identical,” and he then cites Calvin as saying that covenant breaking and apostasy are real possibilities in the new covenant, just as they were in Israel (346-347). In ch. 18, Baugus makes brief reference to Dutch Reformed theologians who affirmed presumptive regeneration, grounding it in the “infant’s membership in the (external) covenant of grace” (399). In ch. 26, Scott Swain speaks of the dangers of falling away from the new covenant community based on the warnings of Hebrews 6 and 10. He cites 1 John 2:19 for the language of being “in” the covenant but not “of” the covenant, and he distinguishes between the “external” and “internal” embrace of Christ (568). This is the most substantive paragraph in the book about dual aspect, yet instead of elaborating, Swain points readers in a footnote to Geerhardus Vos’ discussion of this subject in his Reformed Dogmatics (vol 2).
In ch. 27, Derek Thomas cites Scripture passages showing the conditionality of the covenant, meaning “faith and repentance are covenantal requirements” (580-581). John Scott Redd makes a similar point in ch. 6, where he rejects Kline’s typological works principle (republication) because it “ignores the clear conditional elements in that [Abrahamic] covenant, elements that are present in the new covenant as well (Heb. 12:4-11).” Redd argues that if “the interest in obedience to the law is primarily a matter of ‘appropriateness of expression’ of faith and not of legal merit [citing Vos], then it follows that the same is true of the Abrahamic covenant requirements of blamelessness and covenant keeping” (143).
Thus, dual aspect theology is referenced throughout the book, but there is no development of these essential issues. The Dutch theologians have tended to give these matters more attention, with Louis Berkhof including an entire chapter devoted to dual aspect in his Systematic Theology. It is unclear why a book of this size could not include a full chapter on the topic. My best guess is that many modern Reformed theologians are uncomfortable speaking about apostasy/covenant-breaking and the distinction between covenant and election. Many have adopted the Baptist view that equates covenant and election, and they often attribute language of covenant-breaking to the Federal Vision rather than historic Reformed theology. Moreover, even the above citations hint at disagreement among the authors. This is an area ripe for further work.
Other Omissions (Federal Vision, Theonomy, Sabbath)
This leads to my final point that Covenant Theology fails to engage much with modern debates within Reformed covenant theology. While there is a chapter titled “Covenant in Recent Theology” (ch. 20), it interacts only with Michael Horton and John Webster. The only mention of the Federal Vision (a variation of covenant theology) in the entire book was Belcher’s interesting discussion of James Jordan’s view of the covenant of works (ch. 2). Though there is an entire chapter on the sacraments (ch. 27), it only makes a brief mention of paedocommunion and uses a footnote to cite two books critical of the practice (577).
Benjamin Gladd briefly mentions theonomy in his chapter on “Covenant in Contemporary New Testament Scholarship” (ch. 23), noting that the dominant Reformed view is that Israel’s civil laws are no longer in effect (490). However, Gladd notes that within the mainstream Reformed camp there is still disagreement as to how the Mosaic laws apply today, giving the example of the Sabbath commandment (the Fourth Commandment). Thus, a discussion of the different approaches to the Sabbath would have been helpful in surveying different approaches to applying the Mosaic law.
Reid in his Mosaic covenant chapter (ch. 7) devotes almost five pages to the threefold division of moral, ceremonial, and civil (judicial) law (155-160), but he provides little discussion of how to apply the Mosaic civil laws today. He simply says we can “learn” from Israel’s civil laws, but “they are not directly repeatable since they are historically conditioned with respect to their application” (157). He then follows this with a footnote citing a paper by Ligon Duncan on the intellectual origins of Christian Reconstruction. Thus, Reid discusses the civil/judicial law without mentioning theonomy (though it is implied in the footnote on Reconstruction), and he discusses the application of the Ten Commandments today as moral law without even mentioning debate over the Sabbath. This question of how to apply the Mosaic law to today is another area ripe for further work.
These omissions would be understandable in a shorter work, but this large book made room for more obscure subjects, such as covenant theology in Barth and the Torrances (ch. 19). It seems the editors consciously chose to avoid such intra-Reformed debates, with footnotes pointing readers to other books on the topics. However, as noted above, there are strong treatments of non-Reformed views, including dispensationalism and attempted middle-way views known as “new covenant theologies.”
Conclusion
Covenant Theology will not be my top book on the subject, in part because it is a multi-author work rather than a unified theology of one author. It also has significant omissions for a book of its size. That being said, it is a book I will regularly consult when teaching on covenant theology, both for its treatment of specific biblical passages and its historical theology. This is a goldmine of information, pointing readers to other resources for further study (including an annotated bibliography at the end). It meets a significant need in the literature and is highly recommended for those interested in covenant theology.
