On what basis do we claim that the 27 books in our NT constitute the final and authoritative “canon” of Scripture? The best treatment of this issue and answer to our question is provided by Michael Kruger in his book, Canon Revisited (Crossway).

One view is that the canon is something determined by the Christian community, either individually or corporately. Thus, canonicity is not something inherent to any group of books but is something officially or authoritatively imposed on certain pieces of literature. Thus “a ‘canon’ does not exist until there is some sort of response from the community” (Kruger, 30). “Canon”, therefore, is a human construct that can be accounted for in purely natural terms. No direct work of the Spirit is necessarily involved in determining which books are canonical and which ones are not. Again, explains Kruger, according to the community view of canon,

“The canon has no metaphysical or intrinsic qualities that need to be accounted for – ‘canon’ is not something that describes the quality of a book, but is something that is done to books” (32).

The Roman Catholic view shares much in common with the “community-determined” approach, but locates the authority for this determination in the Magisterium of the church. According to Rome, in order to have an infallible Scripture we need to have an infallible guide, i.e., the church, to tell us what is and is not inspired and canonical Scripture. Rome ultimately must claim that its authority is self-authenticating and needs no external authority to validate it.

What is referred to as the “existential” view of the canon argues that authority is not found in the Scriptures themselves but only in the individual who engages with them and experiences them as God’s Word and responds in faith. According to this view, “Christians do not experience God in the Scriptures because they are canonical; rather they are ‘canonical’ because Christians experience God in them” (Kruger, 62).

A major problem with this view is that if the boundaries of the canon are determined solely by the experience of the individuals in a community (i.e., “I find and engage with God when I read the Scriptures and that is how I know they are inspired and authoritative”), such boundaries are fluid and always subject to change. Says Kruger, “If the canon is regarded as an entirely personal and existential ‘event,’ then there can never be a ‘right’ canon, but simply the ‘current’ canon – the canon that the church is now using” (64-65).

As over against all the above views of canonicity, there is the position that the canon is determined historically as each book is evaluated according to selected criteria. There have always been variations in the criteria that supposedly determine which books are inspired and authoritative, but in general they would include such factors as apostolicity (was a book written by an apostle or by his appointed representative), orthodoxy (does the book teach theology consistent with the OT and what the church then believed), antiquity (when was the book written? early composition is a sign of authenticity, later composition typically disqualifies a book), and usage by the church.

Thus, the boundaries of the canon are determined on the basis of accepted principles of historical method. But again, to authenticate the canon based on supposedly independent standards ultimately subjects the canon to an authority outside itself. “It allows autonomous human assessment of historical evidence to become an external authority over God’s Word” (Kruger, 80). In the final analysis, this view (widely held by evangelical scholars) isn’t significantly different from the views already stated: all of them subject the canon’s authority to some standard outside itself, be that the Roman Catholic church or autonomous historical investigation or the impact on one’s personal spiritual experience. These so-called “criteria of canonicity” unduly inflate or elevate the church’s active role in the development of the canon.

One final objection to the “criteria of canonicity” view is stated by Kruger:

“If the criteria of canonicity . . . provide some sort of norms or standards by which we determine whether a book comes from God, then where do the criteria themselves come from? How do we know that these ‘tests’ are the way a divine book is identified? Put differently, what are the criteria that determine these criteria?” (83).

The only viable and, in my opinion, correct view is one that “does not ground the New Testament canon in an external authority, but seeks to ground the canon in the only place it could be grounded, its own authority” (Kruger, 89). The canon, therefore, is self-authenticating. It isn’t the early church’s use of the 27 books nor the church’s final reception of these books that makes them canonical. “They are canonical by virtue of what they are, namely, God’s books” (Kruger, 58).

We know certain books to be inspired and authoritative because they bear the imprint of the Holy Spirit. That is, the book bears the divine qualities or divine character of a book from God.

“The books received by the church inform our understanding of which books are canonical not because the church is infallible or because it created or constituted the canon, but because the church’s reception of these books is a natural and inevitable outworking of the self-authenticating nature of Scripture” (Kruger, 106).

Again,

“In the self-authenticating model . . . the church’s reception of these books proves not to be evidence of the church’s authority to create the canon, but evidence of the opposite, namely, the authority, power, and impact of the self-authenticating Scriptures to elicit a corporate response from the church” (Kruger, 106).

It could well be said that the church didn’t choose the canon but that the canon chose itself. The 27 books are the NT canon because they imposed themselves upon the church. Thus, “books are not canonical because they are recognized; they are recognized because they are already canonical. It is this critical distinction that sets the self-authenticating model apart from many of the community-determined models” discussed above (Kruger, 108).

In summary, we can know which books are canonical “because God has provided the proper epistemic environment where belief in these books can be reliably formed. This environment includes not only providential exposure to the canonical books, but also the three attributes of canonicity that all canonical books possess – divine qualities, corporate reception, apostolic origins – and the work of the Holy Spirit to help us recognize them” (Kruger, 113).

The canonical books of the NT “imposed themselves on the church and it could not, in the end, resist them” (202). As J. I. Packer once said, “The Church no more gave us the New Testament canon than Sir Isaac Newton gave us the force of gravity. God gave us gravity . . . Newton did not create gravity but recognized it” (Packer, God Has Spoken, 109).

In the final analysis, then, there are three senses in which we may use the word “canon.” There is canon as reception, that is, the final collection of books formally received and acknowledged by the early church (probably in the late 4th century). There is canon as use, which points to the way the early church appealed to the Scriptures as authoritative. This took place in the early second century. Then there is canon as divinely given. This refers to the canon as existent and closed immediately after the writing of the book of Revelation. The first view is called the exclusive view of canonicity. The second view is called the functional view of canonicity. And the third view is the ontological view of canonicity.

The canon thus actually existed (ontologically) in its complete and final form with the inspiration of the last book of the NT, Revelation.

So, in sum, we know the canon is closed because God closed it. He ceased giving the form of revelation that possessed the self-authenticating authority that would invariably impose itself on the collective conscience of the church.