The Authority And Meaning Of The Christian Canon

A Response To Gerald Sheppard On Canon Criticism

Published in The Journal of the Evangelical Theological Society 19/2 (1976), pp. 87-96.

In the October 1974 issue of Studia Biblica et Theologica, an article by Gerald Sheppard appeared entitled “Canon Criticism: The Proposal of Brevard Childs and an Assessment for Evangelical Hermeneutics.” It has been a helpful stimulus to my own thinking. I found especially provocative his critique of the “historical-grammatical” method of exegesis. But if I understand Childs and Sheppard, I do not agree with the direction in which they are moving. In this response I want to show in what respects and why I cannot agree.

“The dissolution of the Biblical Theology Movement stems from the atomizing effects of critical studies on the Biblical text and from a loss of consensus regarding wherein lies the authority of Scripture for doing theology.”1 With this sentence Sheppard, following Childs,2 pinpoints the two problem areas to which he addresses himself: (1) the problem of authority—that is, where does our authority lie and on what basis is Scripture authoritative; (2) the problem of interpretation—that is, given an authoritative source, how shall we discover its meaning so that its authoritative message becomes effective in the Church? The road to which Sheppard points for solving these two problems is in my judgment misdirected.

I. The Problem Of Authority

The burden of Sheppard and Childs is to set forth and defend the Christian canon as the authority on which a new and fruitful Biblical theology can be based. They do not find authority in “the basic historicity of the Bible ..., the intention of Jesus, or the gospel kerygma.... In conscious opposition to this approach to hermeneutics, the confession of the Christian canon as the context for Biblical theology makes the claim that the ‘theological data’ of the Bible does not lie in some form of positivity behind the text, such as Heilsgeschichte, language phenomenology, or in a mode of consciousness illustrated by the text, such as authentic existence or the like.”3 The canon in its final form is normative for the Christian faith. What this actually means can only be seen when we pose the question of how we determine the meaning of the canon (see section II).

The basic problem of authority for the Christian, then, is this: On what basis do we claim that the Christian canon is authoritative? If I understand the following sentence, Childs poses this very problem: “The theological issue at stake is the rightness of the claim for divine authority to which the church responded in setting apart certain writings as Scripture.”4 But neither Childs nor Sheppard undertakes to show the “rightness” of this claim. Childs says, “The status of canonicity is not an objectively demonstrable claim but a statement of Christian belief.”5 One does not confirm the rightness of the canon; rather, one “confesses” it as the believing community has done for centuries.6 As a member of that community one does not test the legitimacy of the canon; one receives it joyfully as a “sacred inheritance of the church.”7

Sheppard’s view is essentially the same. For him the canonical context is a “theological a priori” (p. 9). That the Scriptures “witness to religious and spiritual reality” is the “presupposition” of genuine theological exegesis (p. 5; italics mine). It is the Church that supplies this presupposition: “The canonical context defines the authoritative reception of sacred tradition in the life and theology of the church” (p. 7). I suppose the phrase “authoritative reception” means a reception which gives authority to the canon. This seems to be what Sheppard means when he says, “What holds the Scripture together in its dialectical tension … is the affirmation of the believing community of the normative status of a given tradition shaped and contained in a set of books” (p. 5). As I understand them, both Childs and Sheppard are saying simply, “We accept the canon as authoritative because the Church has done so for centuries with profit.” If Childs and Sheppard have other reasons for accepting the canon as authoritative, I have not noticed them.8

A basic problem with this view of the canon’s authority is that if we exalt the “believing community” and depend on its decision, then we must say that each generation of believers before us should have believed similarly. But we soon come back to those first generations during whose time the canon was being compiled. How was the question of authority to be settled for them before there was a long tradition that accepted the canon as authoritative? Should our understanding of authority be different from the understanding of those early generations who could not speak of a “sacred inheritance of the church”? Should we not accept the canon on grounds similar to those used by the first generations who formed the canon?

I do not mean that all the criteria used by the early Church in compiling the canon are acceptable for us today.9 I appeal only to the fact that the early Church put great stock in the content of the books vying for canonicity.10 They did not start with a set of books as their “presupposition” or their “theological a priori.” Since the canon did not yet exist but was emerging, grounds of authority were sought within the writings, and the argument that “they are accepted in the Catholic Church”11 was secondary. When trying to ground our acceptance of the Christian canon, I do not see why we today should abandon the early Church’s interest in the content and origin of the canonical books and turn their decisions into our “theological a priori.

Sheppard’s approach is all the more surprising in view of his and Childs’ deep desire to stand in continuity with the “so-called precritical” exegetes of the Reformation, such as Calvin (pp. 5, 17).12 Calvin attacked vehemently “the pernicious error … that Scripture is of importance only in so far as conceded to it by the suffrage of the Church; as if the eternal inviolable truth of God could depend on the will of men.”13 He argues, “If the doctrine of the apostles and prophets is the foundation of the Church, the former must have had its certainty before the latter began to exist.... Nothing, therefore, can be more absurd than the fiction that the power of judging Scripture is in the Church, and that on her nod its certainty depends.”14

Calvin does not “presuppose” the authority of the canon. He seeks a ground for that authority in the content and origin of the books, not in the “authoritative reception of sacred tradition.” “As to the question, How shall we be persuaded that it came from God without recurring to a decree of the Church? it is just the same as if it were asked, How shall we learn to distinguish light from darkness, white from black, sweet from bitter?”15 Later, when arguing from the authorship of the epistles, he says, “Paul, moreover, who had not only been an avowed but a cruel and bloody foe, being changed into a new man, shows, by the sudden and unhoped for change, that a heavenly power had compelled him to preach the doctrine which once he destroyed.”16 Thus Calvin, like the early Church before him, sought to ground the authority of the canon not in its affirmation by the believing community but in the qualities of its content and the nature of its origin. I see very little continuity between Calvin’s approach to the authority of the canon and the Childs-Sheppard approach.

One of the basic discoveries of the Reformers was that neither the authority nor the interpretation of the Scriptures derived from the Church. “With that the Bible was all of a sudden set on its own” with no external guarantees.17 The logical consequence of this was that the criteria for determining what is authoritative had to be found within the canon, and this in turn opened the possibility that parts of the canon might be rejected on the basis of their content. As everyone knows, Luther made this possibility a reality with his stinging criticisms of Hebrews, James, Jude and Revelation.18 After Luther a long train of Lutheran scholars has taken the same view so that today W. G. Kümmel can say, “What Luther observed in individual examples has been irrefutably demonstrated by the scientific work with the conceptual material of the New Testament since the beginning of the nineteenth century.”19 Kümmel and many like him thus seek a “canon within the canon” as the locus of authority for doing theology.20

Childs and Sheppard eschew the “canon within the canon” approach as a “failure to take the Biblical text seriously in its canonical form.”21 This criticism is not well put, for it misses the point. Kümmel, in the footsteps of Luther, has considered the canonical form of the Scriptures seriously but has found that form deficient. What Childs means, I think, is simply that Kümmel and those like him do not start with the “presupposition” of the authority of the entire canon. That is true. Instead they hold the Reformation conviction that the authority of the writings of the Bible stands or falls on its own merits and cannot be secured by arbitrary human presuppositions or ecclesiastical affirmations. In this conviction I think they are right,22 and it seems to me that Childs and Sheppard are coming dangerously close, in principle at least, to an external view of authority that the Reformers tried very hard to throw off.

II. The Problem Of Interpretation

The question that Sheppard raises after he deems the "canonical context" to be a “theological a priori” is: How shall we discover the meaning of the canonical text so that its authoritative message becomes effective in the Church? This is a hermeneutical question, and hence the second half of the title of Sheppard’s article: “An Assessment for Evangelical Hermeneutics.”

According to Sheppard, the specific example of “evangelical hermeneutics” with which he is concerned is “the historical-grammatical method as popularly distinguished by some evangelical conservatives from that of the historical-critical method” (p. 10). But in fact a great deal of confusion is caused by his failure to distinguish between the historical-grammatical and the historical-critical methods. He lumps them together regularly (pp. 13, 15-17) as having the same faults. This lack of precision weakens his critique of evangelical hermeneutics substantially.

On one occasion he seems to contradict himself because of the imprecise association of these two hermeneutical methods. He defines the historical-grammatical method as “an approach that presupposes a relatively conservative assessment in matters of authorship and redaction history of biblical books, such that exegesis can on critical grounds be content to operate grammatically and historically only with the present text” (p. 10, fn. 35; italics mine). “Historical criticism” on the other hand, he says, “uses modern tools of literary and historical analysis to reexamine the basic questions of authorship and the literary history of books. Its interest is as much in the source, form and redactional history as in the present shape of the text” (id.). These definitions are helpful and accurate, and after reading them I expected to find a critique of the former method that is “content to operate grammatically and historically only with the present text.” But instead Sheppard continually treats the two methods together, arguing that “both methods of study are presently oriented to reconstructing an historical situation behind the text rather than determining the canonical shape of the traditions within the text of Scripture” (p. 15; italics mine). On the one hand, Sheppard says the historical-grammatical method is content to operate “only with the present text.” On the other hand, he says its aim is to reconstruct “an historical situation behind the text.”

This apparent contradiction is in reality, however, not so much an inconsistency in Sheppard’s thinking as it is a reflection of his conviction that the historical-grammatical method, along with the historical-critical method, is guilty of the “historical repristination of the sacred traditions” (p. 12). In other words, even though many evangelicals operate “only with the present text” of Scripture, they still aim to recover the Biblical writer’s intention when he wrote it (id.). Exponents of this method argue that the meaning of a text which alone has authority today is what the Biblical author (or redactor) willed to communicate. Since this “willing” or this “intention” was an historical event (at the time of the writing or redacting), to make it the authoritative norm of our own theological thinking is, according to Sheppard, tantamount to the sin of “historical repristination of the sacred traditions.”

“Historical repristination”—or as he calls it in another place, “rehistoricizing” (p. 16)—is bad, Sheppard argues, because it “decanonizes the literature by putting it in some other context than the canonical” (p. 13). Let us examine this argument more closely.

Sheppard says, “To the degree that historical-grammatical or historical-critical exegesis is successful in reviving a ‘lost’ historical context, it effectively decanonizes the literature by putting it in some other context than the canonical” (id.). I regard this sentence and others like it in the essay as very misleading. The criticism is fully justified in the case of a radical historical criticism which rejects the witness of the present text and reconstructs a different message or person (as in the case of the old “Quest”) behind the text. But historical-grammatical criticism disavows such a procedure from the outset (as Sheppard has said). It is only confusing to continually lump these two methods together.

The historical-grammatical method does not seek after what is “lost.” The “intention” of the ancient author/redactor is sought precisely in the language of the text as it stands before us, and the exhilarating event of discovering this intention is called “understanding,” not “decanonization.” An intention, once well-formulated in writing, never changes and is eternal. An author may change his mind but the intention, once expressed in writing, is unalterable. This means that the ancient author’s intention is with us today in his writing. The task of the grammatical-historical method is simply to think the author’s thoughts—that is, to understand him. It does not aim to get at something behind the text; it aims to get at something intrinsic to the text, namely its meaning.

But what is “meaning”? One of the chief weaknesses of Sheppard’s essay is its failure to ask this question. Until we can agree on the definition of “the meaning of a text,” all our arguments concerning how best to find that meaning will never be persuasive. I would suppose that Sheppard rejects my view that one grasps the meaning of a text only in so far as one grasps the intention willed by the author/redactor and the implications that follow necessarily from that intention.23 When I ask, “What does a text mean?”, I am asking, “What did the author/redactor intend to communicate when he wrote/compiled the text?”24

But what does Sheppard mean by “the meaning of a text”? Since he does not tell us explicitly, I will try to read between some suggestive lines. Sheppard says that “the very way tradition is presented in Scripture defines how it is to be heard” (p. 16). He offers the following examples:

The present shape of the Sinai narratives is first concerned about position and order. These narratives define the theological status of the Decalogue and Book of the Covenant vis a vis an elect people and their leaders, rather than draw an objective picture of the event at Sinai. Likewise the Gospels are not first of all attempts to write theologically neutral “historical accounts” of Jesus, but by their use of detail, to provide a context of appropriation for the significance and meaning of Jesus for the Christian community (italics mine).

Sheppard is trying very hard to avoid the language of the enemy here, namely the (for him) pejorative term “redactor’s intention.” This leads him to almost unintelligible incongruities of expression. What sense does it make to say that the “shape” of a narrative is concerned about the “position and order” of that narrative? That sounds to me like saying that “the shape is concerned about the shape.” Does he not in reality mean, “The redactor was concerned to shape and order his traditions so as to communicate the meaning of the Decalogue, etc.”? But that sounds too much like the old “author’s intention.”

Furthermore, when he says that the gospels are not “attempts to write” but are attempts “to provide,” who is the attempter? Is not the phrase “attempt to provide a context” synonymous with the phrase “intend to provide a context”? Whose intention then can be in view except the author’s?

Even though Sheppard has developed a new, awkward way of expressing himself about the meaning of Scripture, he cannot escape the compelling fact that meaning lies in an author’s intention. His conclusion, therefore, is that “God is the author of Scripture, for whose intention otherwise does the canonical shape represent?” (p. 16). Not the intention of any author or redactor constitutes the meaning of a text, but rather the intention of God: “For example, what a prophet’s oracles originally or at various redactional levels meant is different than what they—after careful collection and arrangement in a prophetic book canonized among others—mean as the authoritative Word of God to the later believing community” (pp. 8 f.). This canonized collection of traditions has a “trans-historical capacity … to confront each new generation directly with an imperishable Word of God, held freshly relevant by its shape and status within a canon” (p. 13).

I do not want to deny that ultimately God is the author of Scripture. What I reject is the implication that this divine authorship lessens the importance of finding the writer/redactor’s intention and the importance of defining the meaning of a text as the intention of its human author. The reason I reject this implication is that it contradicts the historical particularity of divine revelation. I will try to justify this criticism in the rest of this study.

It is clear from Sheppard’s essay that he has little taste for historical particularity: “Biblical theology in the context of the canon does not depend first upon … an attempt by ‘historical-grammatical’ means to recover a writer’s ‘intention’ in all of its full historically conditioned particularity” (p. 12). Christian Scripture is “theologically alive even before the assured results of an historically verified ‘intention’ can be ‘reconstructed’” (p. 13). I view with serious misgivings this tendency to depreciate the once-for-all, historical particularity of divine revelation. Something very crucial is at stake here: Will we in our theological thinking orient ourselves to the “midpoint” of history25 with its once-for-all character, or will we orient ourselves to a “canonical shape” which is esteemed not for its faithful preservation of a revelation made once for all concretely in history but for its uniqueness as the place where God now speaks to man?

Sheppard’s definition of inspiration illustrates which way he is moving between these two alternatives. He quotes Childs to the effect that the inspiration of Scripture should be defined as “the uniqueness of the canonical context of the church through which the Holy Spirit works” (p. 17). This definition accords perfectly with Sheppard’s view that the meaning of Scripture consists in God’s intention heard afresh each time the believer reads the Scripture (p. 16). In other words, just as the “meaning” of Scripture refers to the intention of God communicated through the Scripture to each new generation, so also the “inspiration” of Scripture refers to the present work of the Holy Spirit through the Scripture.

Here is a significant shift away from the historical particularity of divine revelation. The alternative position is that inspiration is a past, once-for-all event that resulted in concrete historical meanings in written documents that have been faithfully preserved in the Christian canon. All our theological thinking should be oriented to and judged by those meanings as they are contained in the text of Scripture. In my opinion this latter alternative accords more closely with the principle of the incarnation and with God’s “scandalous” decision to reveal himself uniquely in a very particular and limited history.

That is a theological argument against Sheppard’s depreciation of the historical particularity of divine revelation. There is also a simple linguistic argument. Sheppard says that the Scripture is “theologically alive,” that Biblical theology can go about its business and that a Word of God can be heard through the Scriptures before the historically-conditioned original intention of the authors is recovered (pp. 12 f.). The clear implication here is that the canonical shape of the Christian Scriptures has a “trans-historical capacity” to confront every generation with the Word of God apart from the rigorous historical research of Biblical scholars. While this will have wide appeal to laymen, it is surely not true.

The voice of God cannot be heard in the Scriptures if the Scriptures are written in a language that one cannot even read or understand. But this is in fact our situation: The vast majority of contemporary believers cannot read Greek and Hebrew. They are utterly dependent on translations, whether written or oral. The Scripture is not “theologically alive,” Biblical theology cannot do its business, and God cannot speak through the Scriptures until they exist in a language that people can understand. For almost all believers, this means that confrontation with the Word of God is contingent on oral or written translation.

But what does translation involve? I see basically three steps: (1) There must be a careful historical investigation of the usage of Greek and Hebrew words, idioms, and grammatical constructions—that is, lexicographers and grammarians must seek to determine the specific meanings (or intentions) that specific language conventions carried in those cultures; (2) there must be a careful investigation of the usage of words, idioms and grammatical constructions in the modern language one is translating into—that is, one must know the specific meanings (or intentions) that are conveyed by modern language conventions; (3) transferal of meaning must be made from one set of language conventions to the other. The measure of success is how well the modern reader is enabled to think the thoughts that were conveyed in the original language.

This phenomenon of translation, which is necessitated by the once-for-allness of historical revelation, contradicts Sheppard’s claim that theology can proceed without depending on the recovery of a writer’s intention (p. 12). This can be illustrated in more detail by reflecting on the translation process. Suppose a translator comes upon the word zelos in the New Testament. Should he translate it “jealousy” with a negative connotation or “zeal” with a positive connotation? There is only one way to decide, and that is by determining from the context how the author intended it to be taken. If the translator chooses wrongly, the twentieth-century reader will be inhibited from hearing the Word of God. If scholars offered us an English Bible in which the Greek ναι were translated “no” and ουκ were translated “yes” because the scholars did not know what the ancient authors intended when they wrote those words, then at many points we would not hear the Word of God in Scripture no matter what the “canonical shape” of that Scripture was.

The point is this: Hearing the Word of God in the oral or written proclamation of the Scriptures is absolutely dependent on hearing the Scriptures in an understandable language. Hearing the Word of God is thus dependent on a faithful translation of the Greek and Hebrew. But translation is only possible and successful when the specific meanings of the ancient documents are understood. Most of those meanings can be determined only by an analysis of the grammatical and historical context that displays the author’s intention. Therefore, it is wrong to say that theology and devotion do not depend on the recovery of the historically-verified intention of the Biblical writer/redactor. There would be no intelligible or faithful canon at all if thousands of scholars and translators had not labored in this grammatical and historical effort. And there is no reason to think that their work is finished, because the ongoing task of theological exegesis is simply an extension of the task of translation.

III. Conclusion

My purpose has been to show in what respects and why I think the Childs-Sheppard approach to authority and hermeneutics is misdirected. With regard to authority, their conviction that the canon is a “theological a priori” whose status is achieved by an “authoritative reception in the life of the church” (p. 7) is in my judgment inconsistent with the view of authority in the earliest Church, is out of harmony with the Reformers, and is dangerously close to an external ecclesiastical authoritarianism. With regard to hermeneutics, I have tried to show that (1) it is illegitimate and misleading to lump the grammatical-historical and historical-critical methods together as Sheppard has done, and (2) to define the meaning of Scripture not in terms of the human author/redactor’s intention but rather in terms of God’s intention communicated afresh to each new generation through the Scriptures has resulted in a depreciation of the once-for-all historical particularity of divine revelation which on theological and linguistic grounds is unsound.

I want to mention, finally, that I know my friend Gerald Sheppard is trying to break new ground in his essay. As in the case of his teacher, he is no doubt eager to refine his emerging hermeneutic. My criticisms, therefore, should in no way be construed as a final condemnation of their effort. My hope is that through this interchange we may both move closer to a sound and fruitful method of interpreting and proclaiming the Word of God for our day.



1 G. Sheppard, “Canon Criticism: The Proposal of Brevard Childs and an Assessment for Evangelical Hermeneutics,” in Studia Biblica et Theologica 4/2 (1974), p. 7. All page references in the text will refer to this article.

2 B. Childs, Biblical Theology in Crisis (Philadelphia: The Westminster Press, 1970), Chapter 6.

3 Ibid., p. 102.

4 Ibid., p. 105.

5 Ibid., p. 99.

6 Ibid., p. 99, 107.

7 B. Childs, The Book of Exodus (Philadelphia: The Westminster Press, 1974), p. xiii.

8 We could cite Childs’ argument that to accept the whole canon as normative guards against the arbitrariness and subjectivity of those who pick and choose the parts of the canon that are normative for Christians (Biblical Theology in Crisis, p. 102). But this is merely a hermeneutical expedient, not a ground for the canon’s authority.

9 For example, the boost given to Hebrews by its alleged Pauline authorship in some circles. But Childs goes too far when he asks, “How can one place any significance on the canon when, from a critical point of view, the criteria used in its compilation are no longer accepted as valid?” (Biblical Theology in Crisis, p. 104).

10 Hans von Campenhausen, The Formation of the Christian Bible, tr. J. A. Baker (Philadelphia: Fortress Press), p. 330: “It is the content of the prophetic or apostolic testimony which is decisive.”

11 This is a quote from the Muratorian Canon concerning two epistles of John and the epistle of Jude. A New Eusebius, ed. J. Stevenson (London: S. P. C. K., 1968), p. 146.

12 Biblical Theology in Crisis, p. 144: “Much of what has been characterized as ‘precritical’ is in fact highly sophisticated theological exposition…. Calvin’s commentaries are classic examples of an exegesis that works consistently from a theological context.”

13 John Calvin, Institutes of the Christian Religion, tr. Henry Beveridge (Grand Rapids: Wm. B. Eerdmans Pub. Co., 1966), Vol. I, 7, 1.

14 Ibid., I, 7, 2. As with the Muratorian Canon cited in fn. 11, Calvin also gives a secondary role to the consent of the Church. “For it is not to be accounted of no consequence, that, from the first publication of Scripture, so many ages have uniformly concurred in yielding obedience to it” (ibid., I, 8, 12).

15 Ibid., I, 7, 2.

16 Ibid., I, 8, 11.

17 W. G. Kümmel, Das Neue Testament, Geschichte der Forschung seiner Probleme, 2nd ed. (Freiburg: Verlag Karl Alber, 1970), p. 14.

18 Kümmel quotes the relevant texts from Luther in ibid., pp. 17-20.

19 W. G. Kümmel, The Theology of the New Testament, tr. J. E. Steely (Nashville: Abingdon Press, 1973), p. 16.

20 W. G. Kümmel, Introduction to the New Testament, tr. A. J. Mattill (Nashville: Abingdon Press, 1965), p. 358. Another notable spokesman for this view is E. Käsemann, “The Canon of the New Testament and the Unity of the Church,” in Essays on New Testament Themes, tr. W. J. Montague, 2nd ed. (London: S. C. M. Press, 1960), pp. 95 ff.

21 Childs, Biblical Theology in Crisis, p. 102.

22 My criticism of Kümmel et al. is not in their hermeneutic, which opens the possibility that there may be theological contradictions in the Scriptures that necessitate the search for a “canon within the canon.” What I question is that there actually are such theologically irreconcilable ideas in the Scriptures. My critique is primarily exegetical rather than hermeneutical.

23 All the true and necessary implications of an author’s intentions do not have to be a part of his consciousness in order to be a part of his meaning. This helps to account for the fact that according to 1 Peter 1:10-12 the prophets were not fully aware of all that they were implying when they wrote of “the sufferings of Christ and his subsequent glories.” For a full treatment of the relation between meaning and implications see E. D. Hirsch, Validity in Interpretation (New Haven: Yale University Press, 1967), pp. 24-67 (especially pp. 61-67).

24 I thus distinguish between the verbal meaning of a text and its so-called “meaning for me.” With Hirsch (ibid., p. 63), I think we do far better to use the word “significance” for the relationship between an author’s intention and my condition. His meaning may have any number of good or bad effects on me, but that does not change his meaning at all. The failure to make this distinction between “meaning” as an author’s willed intention and “significance” as the impact that meaning has on me causes no end of confusion in hermeneutical discussions.

25 This phrase is borrowed from Oscar Cullmann who uses it in his criticism of the Roman Catholic view of Scripture and tradition, namely that “it does not sufficiently observe the necessity of constant orientation to the event at the midpoint.” Christ and Time, tr. F. Filson (Philadelphia: The Westminster Press, 1964), p. 147.

©2014 Desiring God Foundation. Used by Permission.

Permissions: You are permitted and encouraged to reproduce and distribute this material in its entirety or in unaltered excerpts, as long as you do not charge a fee. For Internet posting, please use only unaltered excerpts (not the content in its entirety) and provide a hyperlink to this page. Any exceptions to the above must be approved by Desiring God.

Please include the following statement on any distributed copy: By John Piper. ©2014 Desiring God Foundation. Website: