Proselyte baptism

As an Anglican, Roger Beckwith naturally supports infant baptism. However, he raises some interesting questions regarding the extent of infant baptism in the early church. If Christian baptism is a modification of John's baptism, which is, in turn, a modification of proselyte baptism, and Christians originally followed that paradigm, then paedobaptism may not have been a universal practice at first. The practice of postponing baptism was another drag factor. 

There are potential strategies for blocking his inferences, but Beckwith's argument is interesting because it concedes certain restrictions on the traditional scope of infant baptism even if you grant the apostolicity of infant baptism. 

The point of contact between the two pairs of ceremonies is that John's baptism, like proselyte baptism, is an initiation rite, performed once only, at the time of conversion. John reinterpreted proselyte baptism as a washing away of sin, and therefore applied it to Jews as much as Gentiles, but likewise in an initiatory fashion.  

Jeremias makes much of the fact of proselyte baptism, and also of Jewish household baptism, in establishing his case, since these present close parallels to the missionary methods used by the apostles. Peter's actions on the Day of Pentecost established a precedent, for it is clear that the apostles and their fellow-evangelists had their converts baptized (Acts 8:12-16,36-39; 19:5; Rom 6:3; 1 Cor 12:13; Gal 3:27; Col 2:12) and that they often baptized them by households. Some of these converts were Jews, some were Gentiles, and some were that middle-category of God-fearers or God-worshippers–Gentiles who attended the synagogue but had not submitted to circumcision. They are mentioned a good number of times in the NT (Acts 10:2,22,35; 13:16,26,50; 16:14; 17:4,17; 18:7).  

The NT contains five individual examples of converts whose households or families were baptized with themselves. The first of these is apparently the God-fearer Cornelius at Caesarea (Acts 10:1f.,46-8; 11:14). Then there are various instances in Macedonia and Greece: the God-fearer Lydia (Acts 16:14f.), the Philippian jailor (Acts 16:33), the Jew Crispus (Acts 18:8) and Stephanas (1 Cor 1:16). It is hard to think that, in the case of Jews and God-fearers, when their households were baptized, this would not have included any children they had, in accordance with Jewish practice. In the cases of Gentiles converts, there might be more room for doubt, but in these cases too, the apostles supervising their baptism were of course Jews, even though they themselves were Gentiles. It is sometimes suggested that none of these five households need have included children, but if it was the practice of the apostles to baptize households, as it evidently was, many such households would have been bound to include children.  

We have more or less explicit evidence that circumcision and proselyte baptism were given to infants as well as adults. In the case of John's baptism and Christian baptism, the NT evidence is not explicit and we are dependent upon inference. But if John's baptism was an adaptation of proselyte baptism, one may assume that he too would probably have admitted the infant children of his converts, and the more so as he was baptizing in immediate expectation of divine judgment on those who did not respond to his message (Mt 3:7-12; Lk 3:7-17). A similar inference could be made in the case of Christian baptism, and here we have more to go on, because there are the records of household baptisms, at which we have just been looking, and which strongly imply baptism of infants as well as adults. 

There is one interesting difference between circumcision and proselyte baptism which may have affected early Christian practice. Circumcision was given in every generation. Proselyte baptism, however, was given only in the first generation, after which the proselyte and his family would observe the laws of ceremonial cleanness, and so would not need to repeat it. On this model, a Christian family might be baptized in every generation or only in the first generation, and it is possible that there was for a time a variety of practice, even among Jewish Christians.  

But though we may have confidence that infant baptism began in the apostolic age, we cannot be sure that it was at first universal. We have noted the possibility that in some Jewish families it was not practised except in the first generation, and in Gentile families there may have been less readiness for it than in Jewish. If this is so, there was probably a variety of practice until agreement was reached that every Christian needed to be baptized. This would still leave open the question, at what age he needed to be baptized. Infant baptism was, before very long, widely practiced, as the evidence from Irenaeus, Tertullian and Hippolytus (from the late second and early third centuries) indicates, but the numerous converts from paganism were still baptized as adults, which made it possible for Tertullian to raise the question of whether this was not a better age for baptism. He did not claim that infant baptism was a novelty, but he argued that the forgiveness of sins was less needed in infancy: it was more suitable at an age when there were actual sins to be repented of and washed away (On Baptism 18). Others thought similarly, to judge from 3C inscriptions which show baptism sometimes delayed until there was a danger of death. And after the conversion of the Empire in the fourth century, such delay became for a time, though only for a time, common. For as long as a delay was being practiced, a fruitless desire to avoid postbaptismal sin seems to have been the reason. 

At the conversion of the Empire there was naturally a flood of adult converts, but it was not long before almost all adults were at least nominal Christians, and were baptized as such, and the only remaining candidates for baptism were the children who might be born to them, in that or succeeding generations. As long as the practice of delaying baptism still existed, this did not make infant baptism universal, but it was largely a matter of time before it did. Only in missionary work in new lands or among new peoples would the situation be any different. R. Beckwith, Calendar, Chronology and Worship (Brill 2005), 220-225.