B. B. Warfield gave a helpfully succinct summary of the Reformed case for paedobaptism:1

The argument in a nutshell is simply this: God established His Church in the days of Abraham and put children into it. They must remain there until he puts them out. He has nowhere put them out. They are still then members of His Church and as such entitled to its ordinances. Among these ordinances is baptism, which standing in similar place in the New Dispensation to circumcision in the Old, is like it to be given to children.2

This article probes these words standing in a similar place. How similar, precisely? Typically, Reformed paedobaptists describe this similarity as one of identity, or near identity. John Calvin, for example, asserted that “whatever belongs to circumcision pertains likewise to baptism … baptism has taken the place of circumcision to fulfill the same office among us.”3 John Murray claimed that “there is an essential identity of meaning” between circumcision and baptism.4 Geoffrey Bromiley used the word “equation” to describe the relationship between the two rites.5

This view of the circumcision-baptism relationship is grounded in the larger historical interpretation that one rite has taken the place of the other: “circumcision … was replaced in the New Testament by baptism” (the Heidelberg Catechism);6 “Christ established in [circumcision’s] place the sacrament of baptism” (the Belgic Confession).7 As a result, Reformed paedobaptists argue for the baptism of infants on the basis of the lack of any abrogation of this Old Testament practice. As Pierre Marcel reasoned, “if children ought to be debarred from the birthright which they enjoyed ever since there was a Church on earth, for thousands of years in fact, then there is need for a positive commandment which enjoins their exclusion.”8 Similarly, Louis Berkhof argued, “for twenty centuries children had been formally initiated into the Church, and the New Testament does not say that this must now cease.”9 Indeed, in typical Reformed paedobaptism argumentation, infant baptism has a sufficient “similarity” (Warfield’s term) to infant circumcision that the two institutions stand or fall together: “any argument against infant baptism is necessarily an argument against infant circumcision.”10

Much of the contemporary response to his argument among credobaptists11 comes in a more dispensational vein, emphasizing discontinuity between circumcision and baptism. But it has not always been so. Older traditions of credobaptist argumentation, like that of Nehemiah Coxe in the 17th century, built their case on the basis of covenant theology.12 Similarly, the best of more recent credobaptist argumentation affirm an overarching unity from the Abrahamic covenant up through the new covenant, but insist on certain points of development and crescendo within that unity, such that the meaning of circumcision cannot be wholesale carried over into the meaning of baptism. Thus Karl Barth put it: “recognition of the unity [between the old and new covenants] … does not include an immediate transfer of what is said about Old Testament circumcision to what must be said about New Testament baptism, as though the definitions and meaning of the two were interchangeable.”13

Following on Barth’s heels, Paul Jewett affirmed an essential continuity throughout the biblical covenants, but objected that paedobaptism tends to over-stretch this continuity, flattening out developments across redemptive history. According to Jewett, there is unity between circumcision and baptism, but it is the kind of unity that one finds between the Old and New Testaments more generally: a unity that embraces movement and typological fulfillment.14 Thus, the meaning of baptism overlaps with but does not exhaust the meaning of circumcision: “circumcision is neither unlike baptism nor identical with it.”15 Other credobaptists have said similarly.16 Thus, a credobaptist can agree with Warfield that baptism is “similar” to circumcision, but will take this similarity as most paedobaptists take the relation of the Passover meal and the Lord’s Supper, in that one fulfills the other but not in so identical a manner as to justify paedocommunion.

In this article I offer a further objection to the Reformed paedobaptist argument. It is situated within the Barth-Jewett “similar but not identical” tradition of argumentation, agreeing with them that baptism and circumcision, while “standing in a similar place,” should nonetheless not be strictly equated.17 It seeks to go one step further, however, suggesting that the practice of baptizing the infant children of believers is not, in fact, identical with the practice of circumcision throughout the Old Testament. Thus, even if we did want to identify the meaning of circumcision and baptism, the result would not be paedobaptism as practiced by the majority of Reformed paedobaptist churches today. In other words, the problem is not simply with the equation of baptism and circumcision, but with how circumcision has been construed in the first place.

In what follows are three steps: (1) a brief sketch of the argument, (2) an engagement with several possible responses, and (3) an exploration of historical antecedents in the Reformed tradition that anticipate the concerns reflected in the argument.

1. Why Not Grandchildren? A Brief Statement of the Argument

The Reformed paedobaptist argument does not claim that all infants indiscriminately are proper subjects of Christian baptism. Certain infants are to be included. Others are not. Thus, when Warfield stipulates that “God established his Church in the days of Abraham and put children into it,” the question immediately arises, which children?

The proper subjects of circumcision are identified in Genesis 17:9 as “you (Abraham) and your seed after you, for the generations to come.”18 The “children” in view here (Hebrew זֶרַע , “seed,” [KJV], or “offspring” [ESV], or “descendants” [NIV, NASB, RSV]) are the inter-generational descendants of Abraham that would comprise the nation of Israel. Hence the words “for the generations to come” here, and “every male throughout your generations” a few verses later in Genesis 17:12. It is to this national and intergenerational body that God commands: “every male among you shall be circumcised” (Genesis 17:10).

Thus, the “children” to whom circumcision was pledged are something less than identical to the “children” envisioned by paedobaptism. The lines of covenant throughout the Old Testament were not drawn around particular believing households within Israel, but around the national family of Abraham that comprised an intergenerational people. It was not the spiritual or covenantal status of the mother and/or father of an infant Israelite boy, in itself, that established their right to circumcision, but rather their identification as the “offspring” (זֶרַע ) of Abraham.

This raises the question: if the basis for infant baptism is infant circumcision, and infant circumcision was practiced intergenerationally, should not infant baptism be practiced intergenerationally as well? In other words, why should the grandchildren of believers not be considered eligible for baptism? Consider the following scenario: John Sr. is a devout believer in a particular paedobaptist church; John Jr. attends the church semi-regularly but has never personally professed faith in Christ or become a member, though he attends church nominally; John III is one week old. Should John III be considered a member of the church and a proper candidate for baptism? Most contemporary paedobaptists say no to this question, while historically, more have said yes (more on that a bit later).

But if the argument for baptizing infants arises from continuity with circumcision, why not baptize grandchildren (and great grandchildren, etc.) of believers? On what basis do we differentiate the covenantal status of John Jr. and John III?

To get from “you and your seed after you for the generations to come” to “those who believe and their children” is not the continuation of an established practice. It is a movement, a development, a change. There is a sense in which credobaptists can affirm a stricter continuity between the old and new covenants than paedobaptists insofar as they designate the church as the “children of Abraham”—defined in the Genesis 17:9 sense during the old covenant, and defined in the new covenant in the sense of Galatians 3:7: “it is those of faith who are the sons of Abraham” (cf. Philippians 3:3, Romans 2:29). This is a continuity that also embraces typological movement and fulfillment. In such a schema, the children of believers are in a position of enormous blessing, but they are not necessarily the “children of Abraham” in the sense of either Genesis 17:9 or Galatians 3:7.19

2. A Rejoinder to Possible Replies

To unpack the argument further we will consider three possible replies. In what follows, “children of Abraham” refers to the intergenerational offspring of Abraham that comprised the nation of Israel, while “children of believers” refers to the sons and daughters of professing believers.

2.1. Response #1: The “Children of Abraham” and the “Children of Believers” Are Identical Because Unfaithful Israelites Were Excommunicated

First, and most simply, some might simply deny any discontinuity between the “children of Abraham” and the “children of believers” by affirming that all unbelieving Israelites were (or should have been) excommunicated from the believing community and considered of non-covenantal status. Each generation, in this view, had to profess faith in order to perpetuate their inclusion in the covenant. If an Israelite person failed to personally embrace the covenant promises and evidence a saving relationship to the God of Israel, their children lost (or should have lost) the right to circumcision.20

But it is difficult to regard eligibility for circumcision as conditional on the faithfulness of one’s parents. It may be worth noting, first of all, that such a view is out of alignment with traditional Reformed thought on the nature of the Abrahamic covenant. Calvin, for example, in his commentary on Genesis 17:7–14, drawing from the repeated phrase “throughout their generations” and the description of circumcision as “everlasting,” emphasized that while the outward rite signified the inward reality, it was not conditional on how the inward reality had been received. Rather, the sign of circumcision was a perpetual ordinance for all the offspring of Abraham, from generation to generation.21 Calvin maintained this view because the text specifies the intended recipients of circumcision to be the seed of Abraham, not the seed of Abraham whose parents believe.

Calvin’s view lines up with what we read throughout the Old Testament. It is difficult to imagine, for example, the parents in Israel being lined up at Gilgal in Joshua 5:2–8 to be examined concerning whether they professed faith, in order to determine whether their children were eligible for circumcision. No, Joshua 5:8 records that the entire nation was circumcised because—as specified by Genesis 17—circumcision was for the entire nation, not just for believers and their children within the nation. Thus, throughout the Old Testament, apostate, unbelieving Israelites still fall under the appellation “my people.” The rite continues generation after generation, at times so far apart from inward appropriation that the prophet laments, “all the house of Israel are uncircumcised in heart” (Jeremiah 9:27).22

As the Baptist Nehemiah Coxe put it back in the 17th century:

The right of the remotest generation was as much derived from Abraham and the covenant made with him, as was that of his immediate seed, and did not at all depend on the faithfulness of their immediate parents. Thus, the immediate seed of those Israelites that fell in the wilderness under the displeasure of God were made to inherit the land of Canaan by virtue of this covenant with Abraham.23

This is not to say that exclusion from the covenant community was impossible, or that Gentiles could not be grafted in. Unfaithful Israelites could be killed or banished from the land of Israel for a variety of sins. Similarly, non-Israelites sojourning among the people could enter into the covenant community, partaking of the nation’s laws and ordinances, eating the Passover meal, etc.—and in such cases they and their offspring were circumcised (e.g., Exod 12:48). Thus, it is not exactly right to say the offspring of Abraham simpliciter received circumcision—it was the nation that this offspring comprised, into which people enter, and from which people could be excluded.

But excommunication or extermination from Israel occurred in response to specific and high-handed acts of rebellion like witchcraft, sorcery, blasphemy, particularly egregious forms of idolatry, etc. (e.g., Lev 20:27, 24:16, Deut 17:2–5). It is unwarranted to infer the necessity of personal faith simply because someone has avoided the specific sins for which one will be stoned or banished. Membership in the nation of Israel had cultural, national, economic, and social dimensions, and huge numbers of Israelites remained Israelites without any evidence of personal faith in the God of Israel (think how many wicked kings throughout Samuel–Kings, for example, remained kings over God’s people despite their rejection of God’s laws).

Thus, the conditions of excommunication introduced by the Mosaic law hundreds of years after the institution of circumcision did not redefine the Abrahamic covenant as “those who believe and their children,” as would be necessary to establish continuity with contemporary paedobaptist practice. Rather, God’s people were a national and inter-generational body, in line with Genesis 17:9–15; this was the entity from which one was excommunicated, or into which one was grafted. “Stone the sorcerer among you” is a far cry from “examine the credibility of their profession.” With regard to membership and excommunication among the people of God in the old and new covenants, we must say as we have said with circumcision and baptism: similar but not identical.

2.2. Response #2: The “Children of Abraham” Entails the “Children of Believers” Even If It Is Not Identical with It

Others, however, could respond by admitting that having parents who did not possess faith did not disqualify an Israelite from covenant status, but argue that there is still enough overlap of meaning between “children of Abraham” and “children of believers” to establish precedent for paedobaptism. In other words, perhaps “those who believe and their children” is not the same thing as “you and your seed after you for the generations to come,” but is still entailed by it.

But it is not easy to construe how “children of Abraham” might entail “children of believers.” Such an argument would seem to amount to a deviation from the stronger and more standard paedobaptist claim that paedobaptism is a continuation (not merely an implication) of Old Testament practice, as represented by the assertions of Warfield, Calvin, and the like quoted at the beginning of this article. Moreover, we might once again ask how such an implication would not justify the baptism of grandchildren. If “children of Abraham” entails “children of believers,” on what grounds does it not simultaneously also entail “grandchildren of believers”?

It could be argued, perhaps, that the “children of believers” and “children of Abraham” are sufficiently similar to be practically the same. For instance, one might suggest that although circumcision was given intergenerationally, the parents were responsible to bring the child to circumcision, and therefore the spiritual body envisioned in Genesis 17 is inclusive of “those who believe and their children.”

But we must distinguish between circumstantial questions concerning what role an Israelites’ parents played in the circumcision of their child, on the one hand, and the theological basis for the child’s right to circumcision, on the other. Godly Israelite parents doubtless played an important role in their child’s spiritual development, including their circumcision, but it does not follow that a child with ungodly parents was excluded from the right to circumcision. If an Israelite’s parents declined to present the child for circumcision, and an uncle or neighbor instead brought them on the 8th day, the presiding priest would have no grounds on the basis of Genesis 17:7–14 to reject the child’s legitimacy for the rite. Regarding the role of parents with respect to circumcision and baptism, we again say similar but not identical.

Pointing out these differences is not nitpicky. The Reformed paedobaptist case rests upon the claim of continuity with circumcision, so discovering a breach in the alleged continuity presents a vital challenge, not a trivial one. The argument wants to affirms practice B on the grounds that it is the continuation of practice A, and practice A has not been explicitly abrogated. If it turns out that practices A and B are not the same, then it must be clearly demonstrated why one should entail the other. This is not easy to see, because “those who believe and their children” is not just a slightly different ecclesiology from “you and your seed after you, for the generations to come.” It is not as though the two were thought to be brothers, but turned out on closer inspection to be merely cousins. These two formulae are utterly distinct. Their similarity lies in the fact that they both involve infants. But the infants involved in each system are fundamentally different: in one it is the inter-generational lineage of one man which comprised a nation, while in the other it is families in the modern sense of that term (no slaves, one generation). It is not clear why the former should entail the latter.

2.3. Response #3: New Testament Texts Delimit the “Children of Abraham” So As to Constitute the “Children of Believers”

A final response could be to concede that the argument from continuity with circumcision may not, in itself, meet the required burden of proof for paedobaptism—but that it nonetheless establishes a more general principle of God’s intergenerational ways of working through families in some broader sense, which is then further specified by New Testament texts so as to constitute a proof of paedobaptism. Thus, Genesis 17 might not teach a paedobaptist ecclesiology, but it establishes a framework of thought within which texts like Acts 2:39, 1 Corinthians 7:14, etc. give a more precise and explicit delimitation of baptism to families.

Once again we note that such a concession would amount to a different kind of argument than the typical Reformed paedobaptist claim (though, in my judgment, a more plausible one). To provide a full response to this distinct kind of paedobaptist argumentation is beyond the scope of this article, but it may be useful to briefly engage a few relevant texts.

A passage likely to be involved in this context is Peter’s assertion to his Jewish contemporaries in Acts 2:39 that God’s promise is “for you and your children.”24 To be sure, the word “children” here is more in line with the target of paedobaptist practice (the first generation offspring of believers). There are three difficulties, however, with taking Peter’s language as a warrant to delimit the recipients of the Abrahamic covenant to families. First, in context, the “promise” in view here is not covenant membership per se, but the Holy Spirit who has just been poured out, as identified in the immediately preceding clause: “you will receive the gift of the Holy Spirit” (Acts 2:38; cf. Luke 24:49). Second, the recipients of this promise are not just “you and your children,” but also “all who are far off” (v. 39). Syntactically, both these classes of people are in the same position. It is not clear how we might regard children as recipients of the promised Holy Spirit in a different sense than those far off are recipients of this promise.

Third, Peter indicates that the conditions of receiving this promise are repentance (v. 38) and calling (v. 39). He enjoins them in verse 38, “repent … and you will receive,” and in verse 39 further qualifies the recipients of this promise with the phrase, “everyone whom the Lord our God calls to himself.” The overall import of Peter’s claim is that anyone—from those under your very roof to those in distant lands—can receive the Holy Spirit upon repentance, in response to God’s calling. This passage thus falls short of establishing the covenant membership of the children of believers exclusively (but not for those “far off”), apart from and prior to repentance.

Another relevant passage is Jesus’s blessing of children in Mark 10:13–16 (// Matt 19:13–15; Luke 18:15–17), and in particular Jesus’s statement, “to such belongs the kingdom of God” (Mark 10:14). Once again, however, there are several difficulties with seeing this assertion as a sufficient warrant to delimit Genesis 17 to a one-generation ecclesiology. First, Jesus says that the kingdom of God belongs to such as these, not to children per se. He is commending childlikeness, with all it entails, as a prerequisite to entering the kingdom. That is why he immediately follows by saying, ““Truly, I say to you, whoever does not receive the kingdom of God like a little child shall not enter it” (Mark 10:15 // Luke 18:17, italics added). So with Matthew 18:3–4, where anyone (of any age) can enter, if they become childlike: “Truly, I say to you, unless you turn and become like little children, you will never enter the kingdom of heaven. Whoever humbles himself like this child is the greatest in the kingdom of heaven” (italics added).

Second, Jesus’s statement is not made with respect to the “children of believers,” but children in general. Thus, if the blessing Jesus pronounces upon children here is to be correlated with covenant membership, it is difficult to know why it would constitute grounds for the baptism specifically of “those who believe and their children.” Third, Jesus refers to children entering the kingdom, which is not precisely what covenantal paedobaptists believe is happening at the baptism of an infant. The point of Jesus’s teaching can be summarized as: kingdom entrance requires childlikeness. Such a teaching can enrich our theology of children in many ways, but it falls short of meeting the paedobaptist need at this point.

A final passage worth engaging here is 1 Corinthians 7:14, and particularly Paul’s claim that the children of one or more believing parents are “holy.” In context, Paul is addressing questions of marriage and divorce in the Corinthian church. In verses 12–16, his particular burden is with the circumstance of a Christian who is married to an unbeliever. Verse 14 functions as the rationale supporting his appeal to remain married in the preceding verses: “for the unbelieving husband is made holy because of his wife, and the unbelieving wife is made holy because of her husband. Otherwise your children would be unclean, but as it is, they are holy.”

What is noteworthy for our purposes is that the unbelieving spouse is said to be in the same position as the child—the children are “holy” (ἅγιος) precisely because the unbelieving spouse is “made holy” (ἁγιάζω). This suggests that “holiness” here may refer to something other than covenant membership. Throughout the Bible the term holy is used in a wide variety of ways—for the sabbath (Exod 20:11), priestly garments (Exod 28:2), objects in the Tabernacle (Exod 30:25–29), materials involved in various offerings (Lev 6:25; 14:13), those under a Nazirite vow (Num 6:5), times of covenant renewal (Neh 8:9–12), angels (Acts 10:22), the law (Rom 7:12), and on and on we might go. The term has no intrinsic association with covenant membership, and that would not be the first or most natural association in Paul’s context here, where he is concerned that believers not feel they must divorce their unbelieving spouse when they have become a Christian (see especially verses 12–13). It would seem more plausible and more relevant to interpret the “holiness” of the unbelieving family members in this passage in relation to the sanctity and legitimacy of the marriage (and consequently, the resulting offspring), such that, as Paul stipulates, divorce is not required. After all, if “holy” means “in the covenant” here, then presumably the unbelieving spouse must also be considered a covenant member.

This passage may, however, draw attention to the complexity of the different relationships that those outside the church might have to those inside it. It is unnecessary and rather cumbersome to require that every person outside the church membership is equally distant from the church in every other way: as though one is either baptized or a pagan. One can be blessed by the church in a variety of ways without yet being a member of it (as is, for example, the non-Christian spouse of a Christian). Similarly, it is possible to regard the children of believers in a position of spiritual privilege in relation to those children born without any contact with the gospel, even if they are not covenant members automatically and from birth.

3. Historical Antecedents

There is, of course, yet another way that advocates for paedobaptism could respond. They could say, yes, the blessing of circumcision was intergenerational and so should baptism be as well. Therefore, we should baptize inter-generationally (including, for instance, the grandchildren of believers).

Strikingly, this was the dominant practice of the early Reformed tradition. For example, in a letter dated 27 August 1559, the Scottish Reformer John Knox wrote to John Calvin to ask him several questions concerning the administration of baptism and ecclesiastical property.25 In his reply of 7 November of the same year, Calvin referenced Knox’s question as to “whether it be lawful to admit to the sacrament of baptism the children of idolaters and excommunicated persons before their parents have testified their repentance.”26 Calvin indicated that he had brought this question before his Genevan colleagues, and that the answer he is supplying to Knox is their unanimous verdict.27

Calvin began his response by noting the sanctity of baptism depends upon it being administered to the proper recipients:

We ought always to be carefully on our guard that the sanctity of this mystery be not profaned, which it certainly should be if it were promiscuously administered to aliens, or if any one received it without having such sponsors as may be counted among the legitimate members of the church. But as in the proper use of baptism the authority of God is to be considered, and his institution ought to derive its authority from certain conditions, one of the first things to be considered is who are the persons that God by his own voice invites to be baptized.28

We will return to Calvin’s reference to “sponsors” in a moment. For now, we note Calvin’s interest in “who are the persons that God by his own voice invites to be baptized.” This is the question animating this article: who are the proper subject of Christian baptism?

In answer this question, Calvin stipulated that the proper recipients of baptism are the intergenerational offspring of believers, extending to “thousands of generations.” He then used this claim to maintain the legitimacy of the sacrament during the corruptions of the Roman Catholic church:

Now God’s promise comprehends not only the offspring of every believer in the first line of descent, but extends to thousands of generations. Whence it has happened that the interruption of piety which has prevailed in Popery has not taken away from baptism its force and efficacy. For we must look to its origin, and the very reason and nature of baptism is to be esteemed as arising from the promise of God.29

Because God’s promise in baptism extends intergenerationally, Calvin concluded that “it is by no means doubtful that an offspring descended from holy and pious ancestors, belong to the body of the church, though their fathers and grandfathers may have been apostates.”30 Calvin went so far as to claim that to withhold the sacrament from children in such a circumstance is to unjustly defraud them of their rights:

Wherever the profession of Christianity has not been altogether interrupted or destroyed, children are defrauded of their privileges if they are excluded from the common symbol; because it is unjust when God, three hundred years ago or more, has thought them worthy of his adoption, that the subsequent impiety of some of their progenitors should interrupt the course of heavenly grace. In fine, as each person is not admitted to baptism from respect or regard to one of his parents alone, but on account of the perpetual covenant of God; so in like manner, no just reason suffers children to be debarred from their initiation into the church in consequence of the bad conduct of only one parent.31

Nonetheless, Calvin maintains that such children must have a “sponsor,” by which he evidently has in mind some member of the child’s extended family who is a legitimate member of the church and willing to instruct the child in the meaning of their baptism:

In the mean time we confess that it is indispensable for them to have sponsors. For nothing is more preposterous than that persons should be incorporated with Christ, of whom we have no hopes of their ever becoming his disciples. Wherefore if none of its relations present himself to pledge his faith to the church that he will undertake the task of instructing the infant, the rite is a mockery and baptism is prostituted. But we see no reason for rejecting any child for whom a due pledge has been given.32

Calvin also makes a distinction between a church that is in the midst of reform and a healthy church that has been established for some time, evidently seeing a greater need to involve parents in the latter situation.33

The Genevan view commended here by Calvin to the Reformed effort in Scotland came to dominate in the early Reformed tradition. The Italian Reformer Girolamo Zanchi (1516–1590), for example, argued that it is the piety of a child’s church and ancestors, not their immediate parents, that establishes their right to baptism:

The children of those that are indeed in the church, but, because of their unclean way of living, declare that they are not indeed of the church; if they be offered to baptism, they cannot be debarred therefrom, nor ought they. The reason is, because though the parents be wicked, yet their impiety ought not to prejudge their children which are born within the church. But if you say, only the children of the faithful are to be baptized, because those infants only are judged to be within the covenant, and they only holy; I answer, the impiety of their nearest parents is not to be considered here, but the piety of the church in which they are born;—as also their ancestors who have lived godly and holily.34

William Bucanus, the Swiss-French Reformed professor of theology at Lausanne from 1591 to 1603, made a similar appeal:

Are the children of those which are in the Church, but by the uncleanness of their life declare themselves indeed not to be of the Church, to be baptized? They are, because 1. the iniquity of the parents ought not to defraud the children born in the Church…. 2. Neither is the impiety of the next parents to be considered so much as the piety of the Church in which they are born.35

In the Church of England as well, Richard Hooker approved of “the answer of the ecclesiastical college of Geneva unto Knox,” referencing that Knox himself “did not think it lawful to baptize bastards or the children of idolaters (he meaneth Papists) or of persons excommunicate, till either the parents had by repentance submitted themselves unto the Church, or else their children being grown unto the years of understanding should come and sue for their own baptism.”36 Hooker then quoted the Genevan response to Knox and affirmed its conclusion as “sound,” indicating that by their letter “Knox’s oversight herein they controlled.”37 He did, however, question the manner in which Calvin had arrived on this conclusion, since it could imply that “all the world may be baptized, inasmuch as no man living is a thousand descents removed from Adam himself.”38

In the development of the Reformed tradition, Calvin’s affirmation of intergenerational baptism was broadly taken up by those who affirmed the national organization of the church, while Congregationalists frequently challenged this practice as insufficiently reformed.39 Discussion came to be centered around the baptism of the children of “adherents”—i.e., those who had been baptized but who were not, for one reason or another, participants of the Lord’s Supper. Adherents might have grown up in the church but never have professed personal faith; or they might be under church discipline; or they might be in a spiritually precarious position that called into question their fitness for the Lord’s Supper. Whether the children of such individuals were to be baptized was one of the issues debated among Presbyterians and Independents in Great Britain in the period leading up to and during the Westminster Assembly in the 1640s.40

Samuel Rutherford, responding to such Congregationalist criticisms and representing the mainstream Reformed view, developed nine arguments to establish the thesis that “all the infants born within the visible church, whatever be the wickedness of their nearest parents are to be received within the church by baptism.”41 Rutherford based this argument on an appeal to continuity with circumcision, in the same manner that the whole argument for Reformed paedobaptism is made: “if the children of wicked parents were circumcised, all without exception, notwithstanding the wickedness of their parents, then the children of these who are born in the visible church of Christians, are to receive that same seal in nature and substance of that same covenant of grace, which is baptism.”42 And again later: “there was no more required of the circumcised but that they were Abraham’s seed according to the flesh, and by that same reason, there is no more required of infants that they may be baptized but that they be born in the Christian church.”43 These appeals from Rutherford have the same thrust as the paedobaptist quotes at the start of this article: as circumcision was administered, so should baptism likewise be.

Rutherford appealed to a range of Old Testament passages to establish his intergenerational view of the covenant, many of which we have drawn attention to in this article. At one point, having referenced texts like Joshua 5 and Genesis 17, he reasoned:

I prove that the proposition “I will be thy God and the God of thy seed” extends the covenant to the seed of the faithful to many generations downward until it please the Lord to translate his Son’s Kingdom and remove the candlestick from a people. Neither can the meaning be: ‘I will be thy God and the God of thy seed, except the nearest parents of thy seed be unbelievers,’ for that is contrary to the Scriptures above cited.44

This last sentence accords with how we have interpreted Genesis 17 (and Joshua 5) above.

Congregational churches in Puritan New England faced a similar debate to this one among the Westminster divines. The solution adopted at the 1662 Synod of Boston, the so-called “halfway covenant,” embraced as eligible for baptism the children of those who were baptized but had not had a conversion experience themselves. This view, like its predecessors among European Reformed traditions, was grounded in the same appeal to continuity with circumcision that drove the whole paedobaptist framework. For instance, Cotton Mather’s reference to the distinction between a “visible expression” of the covenant and an “explicit personal expression” of engagement with the covenant, which was used to support intergenerational baptism, drew from Old Testament precedent:

an implicit covenant preserves the being of a true church, and so of true church-membership. We also say, the second generation, continuing in a visible profession of the covenant, faith and religion of their fathers, are a true church of Christ, though they have not yet made any explicit personal expression of their engagement, as their fathers did. Even, as the Israelites, that were numbered in the plains of Moab, were a true church.45

Eventually this view fell out of favor, and today it is so scarce that there seems to be very little awareness of it, even among Reformed paedobaptists who otherwise identify their understanding of paedobaptism with the likes of Calvin, Knox, and Rutherford.

4. Conclusion

This historical survey, brief as it has been, furthers the legitimacy of our question: “why not grandchildren?” This question is not bizarre or unwarranted. It springs naturally from the logic of Reformed paedobaptism. Indeed, it follows the same intuition as that of the early Reformed paedobaptists, which can be summarized by Calvin’s claim that in baptism, “God’s promise comprehends not only the offspring of every believer in the first line of descent, but extends to thousands of generations.”

Now, to be clear, I am not advocating for this older Reformed view of covenant transmission. It is not clear to me that intergenerational baptism is in continuity with the administration of circumcision any more than contemporary paedobaptist practice, because circumcision was administered along national lines as well as intergenerationally—it was for the “descendants of Abraham,” not the “descendants of every believer.” In the practice advocated by Calvin, Rutherford, etc., the originator of the covenant line has simply switched from Abraham to any believer.

Nonetheless, the prevalence of this intergenerational view of baptism in the early generations of the Reformed tradition draws attention to a problem embedded in traditional covenantal paedobaptist argumentation. If the rationale for infant baptism lies in its continuity with infant circumcision, what is the justification for its limitation to the first-generation children of believers? If children, why not grandchildren?46

---

[1] For my purposes here I refer to “covenantal paedobaptism” and “Reformed paedobaptism” interchangeably to refer to that species of paedobaptism which emerged in Reformed, Presbyterian, and Congregational traditions and which is developed on the basis of covenant theology, in distinction from other (frequently more sacramentalist) expressions of paedobaptism in, e.g., Roman Catholic, Orthodox, Lutheran, or Anglican traditions.

[2] B. B. Warfield, “The Polemics of Infant Baptism,” in Studies in Theology, reprint ed. (Grand Rapids: Baker, 1981), 9:408, italics added.

[3] John Calvin, Institutes of the Christian Religion 4.16.4, ed. John T. McNeill, trans. Ford Lewis Battles (Louisville: Westminster John Knox Press, 2006).

[4] John Murray, Christian Baptism (Philadelphia: The Committee on Christian Education, The Orthodox Presbyterian Church, 1952), 75.

[5] Geoffrey W. Bromiley, Children of Promise: The Case for Baptizing Infants, reprint ed. (Eugene, OR: Wipf & Stock, 1998), 18.

[6] The Heidelberg Catechism, Question 74.

[7] The Belgic Confession, Article 34.

[8] Pierre Marcel, The Biblical Doctrine of Infant Baptism: Sacrament of the Covenant of Grace (Cambridge: Lutterworth, 2002), 122.

[9] Louis Berkhof, Systematic Theology, reprint ed. (Grand Rapids: Eerdmans, 1996), 636.

[10] Robert R. Booth, Children of the Promise: The Biblical Case for Infant Baptism (Phillipsburg, NJ: P&R Publishing, 1995), 109.

[11] By “credobaptist” I simply mean those who affirm that the proper subjects of Christian baptism are those who make a credible profession of faith in Christ. I use this more generic theological term rather than the denominational term “Baptist” to avoid confusion, because many who affirm this view come from other traditions, such as Congregational, Quaker, Pentecostal, free church, non-denominational, etc.

[12] Nehemiah Coxe and John Owen, Covenant Theology: From Adam to Christ, ed. Ronald D. Miller, James M. Renihan, and Francisco Orozco (Palmdale, CA: Reformed Baptist Academic, 2005).

[13] Karl Barth, Church Dogmatics, ed. G. W. Bromiley and T. F. Torrance, study ed. (New York: T&T Clark, 2009), IV/4:173.

[14] For Jewett, the Abrahamic covenant, of which circumcision functioned as a sign, is similar but not identical to the new covenant. On the one hand, God’s covenant with Abraham had trans-dispensational elements: for example, the promise “to be God to you and to your offspring after you” (Gen 17:7) is never abrogated in the New Testament but rather fulfilled in the church age (Gal 3:7) and ultimately in eternity (Rev 21:3). At the same time, aspects of this covenant were unique to its own dispensation—for example, the promise of the land of Canaan—and these were symbolized in circumcision as well. Thus, the meaning of the Abrahamic covenant cannot be strictly equated with the meaning of the new covenant.

[15] Paul K. Jewett, Infant Baptism and the Covenant of Grace (Grand Rapids: Eerdmans, 1978), 240.

[16] E.g., David Kingdon, Children of Abraham: A Reformed Baptist View of Baptism, the Covenant, and Children (Haywards Heath: Carey Publications, 1973); Fred A. Malone, The Baptism of Disciples Alone: A Covenantal Argument for Credobaptism Versus Paedobaptism (Cape Coral, FL: Founders Press, 2003); Stephen J. Wellum, “Baptism and the Relationship between the Covenants,” in Believer’s Baptism: Sign of the New Covenant in Christ, ed. Thomas R. Schreiner and Shawn D. Wright (Nashville: B&H Academic, 2006), 97–161.

[17] Jewett, Infant Baptism and the Covenant of Grace, 101–3, anticipates my argument a bit, though he does not make much of it. He also writes: “If God has children (believers) and grandchildren (believers’ children), why may he not have great grandchildren (believers’ children’s children)?” (p. 116).

[18] All translations ESV unless otherwise noted.

[19] Some of the concerns reflected here I have articulated earlier: “Why I Changed My Mind About Baptism,” The Gospel Coalition, 8 March 2013, https://www.thegospelcoalition.org/article/why-i-changed-my-mind-about-baptism/. There I point out that such questions can be applied to various other paedobaptist claims. For example, paedobaptists often query whether the blessings associated with the new covenant are, for credobaptists, less generous than those of the old. The question may be put just the same to paedobaptists, however: if the grandchildren of believers are not included “in the covenant,” then is a paedobaptist ecclesiology not less generous than its Abrahamic precursor?

[20] Drew Trammel raised this concern in response to an earlier articulation of my argument, and some of my response to it here draws from our subsequent interaction. See Drew Trammell, “Reformed Paedobaptist Response to Gavin Ortlund’s Article—‘Why I Changed My Mind About Baptism,’” Draw Up for Battle, 22 April 2013, https://tinyurl.com/y8ylaryp.

[21] John Calvin, Genesis, Crossway Classics Commentaries (Wheaton, IL: Crossway, 2001), 163–69.

[22] This statement illumines the difference between God’s people as a mixed community under the terms of the old covenant and God’s people as the continuation of the remnant of Israel under the terms of new covenant. Where Jeremiah could lament that all the house of Israel is “circumcised in flesh only” (Jer 9:25), Paul can assert to a new covenant community, “as many of you as were baptized into Christ have put on Christ” (Gal 3:27). Of course, as paedobaptists frequently point out, churches which practice credobaptism are also mixed communities, for not all professions of faith are sincere, and no one but God can discern the heart infallibly. But there is a difference between an error of discernment and an error of practice. Imperfections in a church’s pursuit of regenerate church membership do not in themselves constitute an argument against that ideal.

[23] Coxe, Covenant Theology, 97. Coxe is one of the earlier Baptists that I have been able to identify who (briefly) anticipates my “why not grandchildren?” argument. At one point later he wrote: “if I may conclude my concern in this covenant is such that by one of its promises I am assured that God has taken my immediate seed into covenant with himself, I must on the same ground conclude also that my seed in remote generations will be no less in covenant with him, since the promise extends to the seed in their generations” (p. 107). I am grateful to Brandon Adams for directing me to this quote.

[24] It is difficult to imagine a more frequently utilized New Testament text among proponents of paedobaptism. For one example, see Joel R. Beeke and Ray B. Lanning, “Unto You, and to Your Children,” in The Case for Covenantal Infant Baptism, ed. Gregg Strawbridge (Philipsburg, NJ: P&R Publishing, 2003), 49–69.

[25] John Calvin, “Letter 549,” in Letters of John Calvin: Tracts and Letters, ed. Jules Bonnet, trans. Marcus Robert Gilchrist (Philadelphia: Presbyterian Board of Publication, 1858), 4:73. I was directed to this letter by David and Tim Bayly, “Godfathers, Calvin, and Knox…,” BaylyBlog, 19 September 2006, http://baylyblog.com/blog/2006/09/godfathers-calvin-and-knox.

[26] Calvin, “Letter 549,” 4:74.

[27] Calvin, “Letter 549,” 4:74.

[28] Calvin, “Letter 549,” 4:74.

[29] Calvin, “Letter 549,” 4:74.

[30] Calvin, “Letter 549,” 4:74.

[31] Calvin, “Letter 549,” 4:74–75.

[32] Calvin, “Letter 549,” 4:75.

[33] Calvin, “Letter 549,” 4:75.

[34] As cited by Thomas Boston, “Miscellaneous Questions,” in The Whole Works of Thomas Boston, Vol. 6: Sermons and Discourses on Several Important Subjects in Divinity (Aberdeen: George and Robert King, 1682), 139–40 (Question 6). I am grateful to Brandon Adams for initially drawing my attention to several of these early Reformed views.

[35] William Bucanus, Theological Institutes (Geneva, 1602), 713–14. Further summary of this argument can be found in “The Baptism of the Children of Adherents,” Reformed Books Online, https://reformedbooksonline.com/topics/topics-by-subject/baptism/the-baptism-of-the-children-of-adherents.

[36] Richard Hooker, The Works of Richard Hooker, ed. John Keble, 7th ed. (Oxford: Clarendon, 1888), 1:349 (3.1.13). I am grateful to Sam Ashton for directing me to Hooker’s view.

[37] Hooker, The Works of Richard Hooker, 1:349 (3.1.13).

[38] Hooker, The Works of Richard Hooker, 1:350 (3.1.13).

[39] John Macpherson, The Doctrine of the Church in Scottish Theology (Edinburgh: MacNivin & Wallace, 1903), 82–90, provides an overview of the different positions, using Rutherford and Thomas Boston as representative examples of each side.

[40] For further reading, see Chris Coldwell, ed., The Grand Debate (Dallas: Naphtali, 2014).

[41] Samuel Rutherford, “On the Baptism of the Children of Adherents,” in A Peaceable and Temperate Plea for Paul’s Presbytery in Scotland (Austin’s Gate, London: 1642), 4, https://tinyurl.com/y7srx5z8.

[42] Rutherford, “On the Baptism of the Children of Adherents,” 4.

[43] Rutherford, “On the Baptism of the Children of Adherents, 5.

[44] Rutherford, “On The Baptism of the Children of Adherents,” 8.

[45] Cotton Mather, Magnalia Christi Americana; or, The Ecclesiastical History of New England (Hartford: Silas Andrus & Son, 1858), 2:305.

[46] I received helpful feedback to earlier versions of this article from the St. Anselm Fellowship of the Center for Pastor Theologians during our October 2019 symposium, as well as during the annual meeting of the Evangelical Theological Society in San Diego in November 2019. Nathan Chambers sent me thoughtful feedback via email. I have benefitted from countless conversations with paedobaptist friends over the last 12 years during which this argument has crystalized in my mind.