Rordorf, Willy. “Baptism according to the Didache” pp. 212-222 in Draper, Jonathan (editor). The Didache in Modern Research. Leiden: E.J. Brill, 1996.
As the earliest non-canonical instruction we have on baptism, Rordorf considers Didache 7 as crucial in reaching an understanding of the doctrine and practice of baptism in early Christianity. He notes that the passage has aroused considerable interest in recent years (Rordorf 1996, 212).
One intriguing feature is a change from a second person plural in 7:1 to a singular elsewhere. Audet takes this to be a sign of an interpolator’s insertion (Rordorf 1996, 213). On the contrary, Rordorf takes the material of 6:3 through 7:1 as a unity, despite its change between the singular and the plural (Rordorf 1996, 214). He further suggests that chapters 9-10, speaking of the eucharist, are part of the same material as 6:3-7:1.
Rordorf assumes that in 7:1 any believer could be the one baptizing. In chapter 15 he takes it that baptisms were not yet routinely conducted by bishops or deacons. Yet in 7:2-3 and 7:4b, Rordorf finds an implication that only one designated person in the community was expected to perform baptisms (Rordorf 1996, 215). He gathers that it was this custom of having one person who administered the sacrament which led to the later delegation of the task to bishops and deacons.
Rordorf goes on to say he would expect extensive catechesis to occur before baptism as early as the beginning of the second century. This is also consistent with the overall logical flow of the Didache (Rordorf 1996, 216).
Prior to the baptism the convert was expected to fast, along with those others who might join in. Rordorf considers that this may have a relationship to the growth of a custom of fasting prior to Easter. Though dates and even days of the week are unclear, it is certainly possible that this baptismal fast led up to a Sunday, and quite likely Easter itself (Rordorf 1996, 216).
The baptismal formula is a popular topic of discussion. Rordorf notes that the formula is given three times, in 7:1, 7:3, and 9:t, using slightly different words each time (Rordorf 1996, 217). Rordorf observes some variation in the New Testament as well. He sees the specific Trinitarian formula arising in common use after the mission to Gentiles has begun (Rordorf 1996, 217).
The means of baptism was important in the Didache. The norm was to use running water. However, in 7:2-3, it is clear that baptism can be performed without running water, and even with water that has been warmed. Rordorf describes this as a purposeful warming, not, for instance, the incidental case of a pool or container in the sun (Rordorf 1996, 219). If needed, it was even permissible to pour water over the head three times, if there was not enough water for immersion (Rordorf 1996, 219).
The question of anointing in relation to baptism arises, in large part due to the Coptic passage inserted after the baptismal language (Rordorf 1996, 220). Rordorf takes the anointing prayer as a later interpolation as opposed to being something original and later suppressed (Rordorf 1996, 221). The anointing itself does not seem clearly linked to the baptism.
Rordorf notes that there are four elements absent from the Didache’s teaching on baptism. There is no renunciation of Satan. The water itself is never consecrated. “There is no trace of the Pauline theology of baptism” (Rordorf 1996, 222). There is no mention of a laying on of hands or impartation of the Holy Spirit. Rodrof thinks the absence of these elements points to the baptismal language being from very early Christianity (Rordorf 1996, 222).