As a starting point we would have to start by defining the Bible as a collection of books that are inspired by God. In order to be included in the collection a book must be inspired.
In stating that the ToC is infallible we would be asserting more than that there are no errors in the list (since I would agree that we have the right books and only the right books), but that there is no possibility of error in the list.
Since the Bible is a collection of inspired books, Geisler and Nix in their book "An Introduction to the Bible" rightly state that:
- God determines which books are in the canon.
- Man discovers which books are in the canon.
We should have no problem stating that the contents (at least in the autograph) of a particular inspired book (for example Ephesians) would have authority and infallibility because God was involved in the writing (determining). The authority of the book comes from God. The recognition of the authority of that book is done by man.
In order for there to be no possibility of error in the ToC, God would have to be equally involved in the discovery process as He is in the writing process to insure that was the case. However, we first have to acknowledge that there is no "list of books" in any of the books that are accepted as inspired. There are quotes and acknowledgments of other books as being inspired, for example Paul states that Scripture is inspired (2 Tim 3:16) and Peter affirms the writings of Paul (2 Peter 3:15-16), but we could not be certain what books are included in the Scriptures Paul refers to or which writings are part of the Pauline corpus that Peter mentions.
Second when one looks at the history of the discovery of the NT canon there is no evidence for a unified NT until the 4th century. The first evidence that we see the NT Canon containing the 27 books we have today and only the 27 books that we accept today is in 367 AD (Athanasius' 39th Festal Letter). This list is confirmed in a series of councils starting with Canon 36 of the Council of Hippo in 393 AD. This is well after the Apostolic era when it is generally regarded that the inspired books are written.
Finally we must acknowledge that man is fallible. To describe any of the councils where the canon was debated and where the discovery phase was concluded (for all practical purposes) as infallible would be conferring the capability of being inerrant to people where only God possesses this ability. Since man is fallible it seems logical that the discovery phase was also a fallible process. There were no signs and wonders that confirm the process (2 Cor 12:12). Since it is these councils where we find the discovery phase completed, to acknowledge the NT ToC as infallible would also invite the possibility other proclamations made in these councils can be too. Why would the NT listed in Canon 36 of Hippo be considered infallible yet not the rest of Canon 36 which includes the OT and Apocrypha? What about another Canon by the same council? We have as much basis for concluding that Canon 1-35 are infallible as we do Canon 36.
For another view on the fallibility of the NT Canon check out Michael Patton's post. He deals with the Catholic claim for an infallible list of infallible books based on the infallible authority of the church. For a contrasting view check out this post.