Reviewing Jeffrey Johnson’s The Fatal Flaw, Part 2

Posted by R. Fowler White

In part 1 of our series of posts reviewing Jeffrey Johnson’s The Fatal Flaw, we covered his introductory survey and the two subsequent chapters on the absence of a NT command to baptize infants and on the analogy between circumcision and baptism. In this post—part 2—we’ll cover chs. 3-6, in which Johnson begins to present, in deliberately crafted increments, his exposé of the fundamental flaw of paedobaptist covenant theology. The four chapters of our present focus are devoted respectively to the continuity between the old and new covenants and to the nature of the old covenant.

Chapters 3-4 set the course for chs. 5-6 (and, in fact, the rest of part one of Johnson’s book). So, in chs. 3-4, we find Johnson intent on showing that the legitimacy of infant baptism hangs especially on the continuity between the old and new covenants: that is, it hangs on the belief that the covenants and the communities formed under those covenants remain essentially the same. Johnson identifies the principles that governed membership under the old covenant as 1) racial distinction, 2) national affiliation, 3) racial perpetuity, and 4) the federal headship of parents—all signified by circumcision. Throughout his discussion, however, Johnson emphasizes that something even more fundamental than those principles is at work: the old covenant did not secure (guarantee) a saving relationship with God to anyone participating under its terms (p. 63; cf. pp. 63-64). In that emphasis we get our most explicit clue into what Johnson believes is the trait that distinguishes the old covenant from a covenant of grace (i.e., that differentiates it from an administration of the covenant of grace). Any covenant that does not guarantee salvation for all its members is no covenant of grace. With that trait in mind, Johnson goes on in chs. 5-6 to offer observations to support his conclusion that the old covenant was a covenant of [based on] works and not of grace. In ch. 5 he lays out four such points: the old covenant 1) made its promised blessings contingent on Israel’s obedience; 2) threatened Israel with curses for their disobedience; 3) was breakable and broken by Israel; and 4) is described in Scripture with terms that identify it as a covenant of works (e.g., law, commandments, ministry of death). To close out his argument in ch. 5, Johnson anticipates the objection that, if the old covenant made its promises contingent on Israel’s obedience, then their identity as God’s elect people must also have been contingent on their obedience. Johnson answers the objection by urging that election applies only to a remnant within the nation, and the ground of the remnant’s election was according to grace. Capping off his contention that the old covenant was a covenant of works and not grace, Johnson devotes ch. 6 to a consideration of Gal 4.21-31. In that passage he finds what he calls a ‘singular refutation’ of the paedobaptist claim that the old and new covenants were each covenants of grace (i.e., were essentially the same covenant). No, says Johnson, in Gal 4 Paul denies the continuity between the old and new covenants and thus denies the continuity of the communities formed under them.

What can we say about Johnson’s arguments in chs. 3-6? First, with regard to chs. 3-4, Johnson’s point that covenant continuity is foundational for infant baptism is certainly relevant. Even so, Johnson’s agenda is driven fundamentally by the fact that the old covenant did not secure salvation for all its participants and so is no covenant of grace. In response, we have to observe that no covenant before the new covenant (as Johnson defines it) guaranteed salvation for all its participants, and no covenant community before the new covenant was coextensive with the elect in Christ. Hence, on Johnson’s terms, no covenant before the new covenant qualified as a covenant of grace. Observations such as these highlight a key question for us to answer: when, if ever, are we to reduce divine covenant to an administration of election in Christ and guarantees of salvation? More specifically, are we to identify the new covenant (i.e., the new covenant administration of the covenant of grace) with God’s eternal purpose in Christ (i.e., the covenant of redemption), or are we to distinguish the two? Once more: are we to identify the new covenant community with the elect in Christ or to distinguish the two? Briefly, in my view, the argument of Heb 7.20-22, 28; 8.6 is decisively in favor of distinguishing the two. Since the new covenant has been enacted on the oath-promises of the Father to the Son—since the oath is the basis of the new covenant, we must distinguish the one from the other. In addition, we must also distinguish the people given to the Son by oath (Ps 110.3; Isa 53.10; John 6.37, 39) from the community formed under the new covenant. Thus, the new covenant is not reducible to an administration of salvation to the elect; the new covenant is also an administration of judgment to the reprobate. In other words, Christ, as Lord of the new covenant church, is both its Savior and its Judge. We’ll have occasion to come back to this topic, but for now it looks to me that in all preconsummate historical covenants (i.e., administrations of the covenant of grace), covenant is broader than individual election according to grace.

Second, respecting chs. 5-6, Johnson’s attempt to support his conclusion that the old covenant was a covenant of works and not of grace fails to convince. It does so because he omits from consideration the role of God’s designated sureties of grace in the election of both the nation and the remnant under the old covenant. To God’s designation of sureties, even those born under the old covenant, Scripture gives careful attention, as when God gave certain of His servants as sureties in the promises, prophecies, ordinances, and other types (“shadows”) of the old covenant, especially those related to the messianic-mediatorial offices. This is not to say that the pre-Christ designees were sufficient and efficacious to prevent the nation’s loss of election and temporal blessing for its disobedience, much less to secure the nation’s election to eternal blessing with their exemplary obedience. To the contrary, their failings made the nation’s election revocable. The remnant’s election to eternal blessing was a different matter, however. It was irrevocable because of the perfections of the Surety to come. In fact, God’s designation of sureties under the old covenant was sufficient and efficacious, through the work of the Spirit, to instruct and build up the remnant in faith in the promised Surety. Thus, Johnson is mistaken not to recognize that by setting forth the promised Surety in shadow and type, the old covenant was a covenant of grace. This is not to deny that the old covenant spoke of conditions, curses, and covenant-breakers. Nor is it to deny the discontinuity between the old and new covenants. It is to say that the folly of the Galatian churches (Gal 3.1) was to consider the works of the law apart from God’s promises of a Surety. In doing so they would have to regard their own works as adequate to qualify them (or their children) as true heirs of Abraham, as adequate to secure their justification and eternal salvation. In doing so members of the Galatian churches would fail to listen to the law, would break its conditions, and would subject themselves to God’s curse, all because they had severed themselves from the Surety God had promised.