In a number of places in the suttas, we encounter the situation where a wanderer of another sect requests ordination. The Buddha cautions them that they must wait for four months, at the end of which time the mendicants, if they’re satisfied, will give the going forth and full ordination.
In his notes for SN 12.16, Ven Bodhi says that the going forth is the initial ordination as a novice, while full ordination is when the novice becomes a bhikkhu. This is, of course, the conventional wisdom, as practiced in all schools.
However I have long believed that this is incorrect, and this passage offers evidence for this. It seems to me that originally pabbajjā and upasampadā were the same thing, and only gradually was pabbajjā associated with novice ordination. Furthermore, it seems impossible that all mendicants would first ordain as a novice. Novice ordination is for children.
The passage on the ordination of one who formerly followed another path says nothing of novice ordination. Rather, it specifies that after the four months have passed the mendicants give the pabbajjā and upasampadā into the state of a bhikkhu.
Catunnaṃ māsānaṃ accayena āraddhacittā bhikkhū pabbājenti upasampādenti bhikkhubhāvāya.
Now, the detailed explanation of this procedure is given in Khandhaka 1 of the Vinaya. However, the situation there is different.
In the Vinaya passage, when they request ordination, they first recite the three refuges. Although the terms are not actually mentioned here, this clearly constitutes the pabbajjā as a novice. They then request the upasampadā. Here, the pabbajjā is omitted, since it has already happened. Only then does the Sangha administer the four months probation.
Thus in the suttas the pabbajjā happens after the four months, while in the Vinaya it happens before. This is a clear contradiction, and it shows that we can’t automatically read ideas from the Vinaya back into the Suttas.
The Vinaya represents a later stage of development, where the ordination procedure had become more complex. In the suttas, we should usually treat pabbajjā and upasampadā as synonyms. Of course, in such a fluid situation this might not be always true, but I think it is the normal case.