Artigo Revisado por pares

Senses of Scripture in the Second Century: Irenaeus, Scripture, and Noncanonical Christian Texts

2017; University of Chicago Press; Volume: 97; Issue: 1 Linguagem: Inglês

10.1086/688994

ISSN

1549-6538

Autores

D. Jeffrey Bingham,

Tópico(s)

Theology and Canon Law Studies

Resumo

Previous articleNext article FreeSenses of Scripture in the Second Century: Irenaeus, Scripture, and Noncanonical Christian TextsD. Jeffrey BinghamD. Jeffrey BinghamSouthwestern Baptist Theological Seminary Search for more articles by this author PDFPDF PLUSFull Text Add to favoritesDownload CitationTrack CitationsPermissionsReprints Share onFacebookTwitterLinked InRedditEmailQR Code SectionsMoreIn a fourth-century prayer that praises God’s omniscience, greatness, and mercy while making several petitions, there is nestled without introduction a quotation from Hermas’s Mandate 1.1, between quotations of Isa. 60:1 and Isa. 40:16. This prayer, which we know only from a papyrus fragment of unknown provenance, is one example of a phenomenon widely attested within early Christian life and worship: the seamless integration with the prophets of Israel, or with the apostles, of texts that were not ultimately canonized.1 This phenomenon poses a problem for our understanding of the history of scripture and canon in early Christianity: if certain texts were originally believed to be scripture, why were they not ultimately canonized as scripture? Some such texts were explicitly called scripture in the second century or, as in the example just given, were identified implicitly as scripture or at least used in equal standing with it. Why later on, particularly in the fourth century, do some writings seem to lose that elevated status?Perhaps, however, they are not losing an elevated status after all. Perhaps in some cases the fourth-century community received these extrabiblical texts in a manner similar to the way they were received in the second century, despite their classification as noncanon. This article entertains that possibility. Of course, the competing notion that the history of the New Testament canon is in certain respects a history of the rising and falling of texts is neither new nor lightly held. Thus, I will begin by briefly sketching the model used by some scholars who interpret the movement of certain texts from between the second century and the canonical lists of the fourth as a decline.I will then evaluate the usefulness of this model by investigating a second-century notion of scripture, that of Irenaeus. I will seek to understand what he meant by the term, what variance there was in his usage, and how we should arrange and classify that usage. For Irenaeus, it was impossible to understand the idea of scripture without relating it to several Christian writings that ultimately were not canonized. He identifies as scripture both texts that were ultimately canonized and other texts that were not. In short, he has different senses of scripture. Using these senses, he provides a taxonomy that will prove useful in addressing the question of why certain texts that were regarded as scripture were never canonized.Finally, I will examine Irenaeus’s treatment of four extrabiblical Christian writings. There has never been a comprehensive account of how Irenaeus understood the authority of Papias’s Exposition, Hermas, 1 Clement, and Polycarp’s Epistle. These writings stand out from other extrabiblical Christian texts in Irenaeus. Each of these he either calls γραφή or carefully attributes to an author closely associated with the apostles. Although I thoroughly discuss each of these texts, Hermas, or Mandate 1.1, receives what may appear to be a disproportionate amount of space. However, Irenaeus refers to that text more frequently than he does the others and contemporary scholarship has paid more attention to the role of Hermas in Irenaeus. Thus, in order to deal with Irenaeus fairly, a more rigorous discussion of the problem of Hermas is necessary. By studying these early Christian writings I seek to answer four questions: (1) How do they relate to Irenaeus’s senses of scripture? (2) How does Irenaeus understand the authority of these nonbiblical texts that he identifies as scriptural? (3) How do such writings relate to the prophets and the apostles? (4) Do his views help anticipate how some extrabiblical texts were understood and employed in the fourth century?This study argues that for Irenaeus “scripture” includes, first and foundationally, the writings of the prophets and the apostles but also, secondarily and derivatively, the aforementioned writings of Clement, Papias, Polycarp, and perhaps Hermas. For Irenaeus, then, “scripture” was a broad term, but he maintained a clear bias toward the prophets and apostles as foundational and essential. It is this bias that he passes on and that gives rise to future fourth-century principles of canon building. In this light, scholarly conceptions of the “rise” and “fall” of writings to and from sacred stature become unnecessary. In light of this discussion, I conclude with an evaluation of the contemporary model for understanding the history of canon and scripture and suggest some adjustments.A Contemporary ModelThe fourth century began the era of extended canon catalogs. Manuscripts of the Christian Bible (Codex Vaticanus and Codex Sinaiticus), along with the letters, lectures, and poems of Christian bishops (Athanasius of Alexandria, Cyril of Jerusalem, and Gregory of Nazianzus) all set forth fixed lists of sacred texts.2 One curious practice, observable in both Athanasius of Alexandria and Codex Sinaiticus, for instance, is the classification of a text like Hermas. Given the lofty label of scripture in the second century, it had apparently been reduced in status within the fourth century. Whereas once it seemed sacred, by the latter period it was considered merely educational and was set off noticeably from the biblical texts.3 Some have seen this fourth-century development as a radical change. When Lee M. McDonald speaks of “the reality of ‘transient’ authority in antiquity,” he has in mind the books of Hermas and 1 Clement, among others: they were once authoritative, and then they were not.4 Albert C. Sundberg concurs: such books are indistinguishable, in terms of “usage and authority,” from the New Testament writings until the fourth century.5 Others have introduced the ideas of “temporary canonicity” and of a shift from an “inclusive” collection to one that was “exclusive.” They do so in order to explain how some books that had been received as scripture in the second century were not canonized in the fourth century or later.6 Some also speak of “decanonization.”7Sundberg rightly insisted on careful definitions for the terms “scripture” and “canon” and cautioned scholars against using them synonymously.8 By understanding “scripture” in the second century more precisely, we can clarify the relationship between second-century thought and fourth-century scripture collections. To this end, I offer an analysis of Irenaeus, whose model makes radical canonical concepts like “lofty and lowered,” “rise and fall,” “transient authority,” or “temporary canonicity” unnecessary, at least in part.Irenaeus and ScriptureWhile the explicit definition of the canon as a fixed list of books became widespread only in the fourth century, there was already a second-century precedent in Irenaeus of Lyons. Within his polemical environment and time, for example, he had already restricted the number of Christian gospels to four (Matthew, Mark, Luke, and John).9 But he also closely associated early Christian texts that were not ultimately canonized, like Hermas and 1 Clement, with the prophets and the apostles. He effected this association sometimes by juxtaposing texts (as does the fourth-century prayer mentioned above) and sometimes, I argue, by labeling his quotations as “scripture” (γραφή).10A recent article by Matthew Steenberg provides significant help on Irenaeus’s notion of scripture.11 He demonstrates through analysis of Irenaeus’s use of γραφή that “non-delimited use of the term is always scriptural in its implication.”12 In other words, for Irenaeus “to refer simply to ‘the writings’ (αἱ γραφαἱ) or ‘the writing’ (ἡ γραφή), without further qualification, is always and without exception to indicate a passage or concept drawn from a book of scriptural [or biblical] authority which Irenaeus regards as genuine to the Christian tradition according to the teaching of the apostles.”13 Steenberg goes on to define Irenaeus’s concept of scripture: “For Irenaeus, ‘scripture’ is that which speaks of Christ in accordance with the typological revelation of the Old Testament, in the manner set forth by the witness of the apostolic preaching.”14 For Irenaeus, then, scripture went beyond the Septuagint, the writings of the prophets, and those apostolic texts that he believed to be of theological weight equal to the prophets.15 The rule of apostolic faith provided the boundaries for identifying a text as scripture, and texts other than those of the prophets and apostles were given that appellation. On this note, Denis Farkasfalvy concurs: “There can hardly be detected in Irenaeus any clear distinction that would set apart the apostolic writings from other apostolic traditions” or “other channels transmitting apostolic doctrine.”16 He correctly notes that Irenaeus speaks of the apostolic writings—their letters and gospels—as set apart along with the prophets;17 that for Irenaeus the scriptures are primarily divided into prophets and apostles or the prophecies and gospels;18 that both proclaim the Logos and both are empowered by the Spirit of truth;19 and that the apostles, like the prophets, are the vehicles through whom the Holy Spirit speaks, and that they lay a perfect foundation of truth.20Yet Irenaeus’s listings of the various parts of scripture are even more diverse than Farkasfalvy shows. Indeed, we find the following formulas or schemes in his lists: evangelists, apostles, law and prophets; prophets, teachings of the Lord, and apostles; prophets, apostles, and words of the Lord; the words of the Lord, the law, and the prophets; prophets and the gospels; scripture, apostles, and the Lord’s teachings; the declarations of the Lord, Moses, and the rest of the prophets; law, prophets, Christ, apostles; apostles, the Lord’s teaching, prophets; prophets and apostles; gospel (apostles) and prophets; and prophets, apostles, words of the Lord.21 Irenaeus’s breakdown of the more general category of “apostles” is important for his theology of scripture. Within this more common category for the writings of the New Covenant era, he lists the specific groups “evangelists” and “the Lord” and his teachings.Irenaeus notes an explicit hierarchy in the sacred writings between prophets and apostles, where apostles are in the premier place. Yet the testimony of one is in harmony with the other.22 There is progress in the history of redemption and revelation from law to prophet to Christ and gospel; grace is present in all, just not to the same degree. There are two testaments and there is law and gospel, but one God is the source of all things brought forth in that history, and both testaments have the same God for their author.23 The aim of this article is to show that Irenaeus may have perceived another hierarchy, a more implicit one, evinced by his concept of γραφή.Steenberg’s analysis has demonstrated the breadth of the term’s meaning for the bishop.24 It can refer generically to any writing, but it also applies, as a technical term, to ecclesiastical writings of a special character. What follows proposes that Irenaeus conceived of scripture, in a special ecclesiastical sense, as a twofold reality. In the primary place, I argue, “scripture” designated the Old Testament (law and prophets) and the New Testament (evangelists, apostles, and dominical sayings; or simply the prophets and apostles). To help make sense of Irenaeus’s taxonomy, I will label this category foundational, ancestral, or primary scripture. Second, Irenaeus uses the term “scripture” to designate certain early Christian documents (like Hermas or 1 Clement) written by those closely associated with the apostles to transmit their teaching. I label this category secondary or derived scripture, using “derived” in the positive sense of drawing upon the foundational teachings of the prophets and apostles.25 Within this taxonomy, the prophets and apostles are the fountain, and certain writings authored by those closely associated with the apostles are the stream. I argue that Irenaeus labels both categories of texts as scripture but never intends that secondary scriptures be confused with primary scriptures. Secondary scriptures are distinct from those scriptures in which the apostles transmitted the gospel, “the foundation and pillar of our faith.”26 In using this language, Irenaeus echoes Eph. 2:20, where Paul identifies the apostles and prophets as the Church’s unique foundation. Thus, Irenaeus has different senses of scripture.If this proposed understanding of Irenaeus’s concept of scripture proves warranted, then an aspect of Steenberg’s and Farkasfalvy’s understanding of scripture in Irenaeus might need to be revisited. Furthermore, this twofold understanding invites us to question the appropriateness of the distinction associated with the contemporary model developed above between the second century and the fourth on the issue of the perceived authority of various writings. For instance, one who claimed to speak of “some earlier books [like Hermas and others]” that had received scriptural status but later “were excluded” would fail to grasp the nuances of Irenaeus’s idea of the scriptural.27 Explanations of canonical dynamics between the second century and the era of a fixed canon would need to take Irenaeus’s taxonomy into account. I will begin by observing this taxonomy at work in the way Irenaeus classifies certain early, noncanonical Christian texts: Papias’s Exposition, Hermas’s Mandate 1.1, Clement of Rome’s Epistle, and Polycarp’s Epistle.Papias’s ExpositionWhat did Irenaeus think of the writings of Papias, Bishop of Hierapolis? Eusebius writes that Papias left the church five volumes titled Exposition of the Words of the Lord and that Irenaeus says these are the only works from the hand of Papias.28 Eusebius then reproduces the words of Irenaeus: “To these things Papias gives witness in writing (ἐγγράφως) in the fourth of five books which he composed, he who heard John, was a companion of Polycarp, and who lived in early times.” Most do not pause to consider Irenaeus’s adverb “in writing” (ἐγγράφως) as a sign that he might regard Papias’s Exposition as scripture in a sacred sense. Y.-M. Blanchard, however, does.29 He finds the Sources chrétiennes translation wanting, inattentive to the Latin version and too strongly swayed by the Greek fragment received from Eusebius and the Armenian version. Such an approach, he warns, runs the risk of ignoring “the unity and coherence” of the Latin version and Irenaeus’s tendency, preserved in the Latin, to allow “the idea of Scriptures to spill over into the first, nonbiblical Christian literature.”30 The Latin version of Adversus haereses has per Scripturam, which Blanchard argues should be read with attention to “the semantic field [in Irenaeus] of the noun Scriptura.”31 By this reckoning, Irenaeus says: “This is what Papias gives witness to through the scripture [per Scripturam].”32If Blanchard’s argument on the basis of the Latin version seems dubious, other considerations support his thesis that Irenaeus treats Papias’s Exposition as scripture. First, Irenaeus’s Greek term (ἐγγράφως) can mean “in the scriptures”; Lampe gives several examples of this usage.33 Furthermore, the additional fragment from Papias’s preface, preserved by Eusebius, says three times that what he records is in accordance with truth.34 The fragment also emphasizes that Papias’s sources were oral. For Irenaeus, Papias wrote (as scripture) what he had heard spoken. This is identical to Irenaeus’s understanding of the composition of Mark’s Gospel. At the beginning of his third book against the heretics, he says: “Mark, Peter’s disciple and interpreter, handed down [παραδίδωμι; trado] to us in scripture [ἐγγράφως; per scripta] what Peter preached [κηρύσσω; nuntio].” Just prior to this sentence, Irenaeus writes that the church had learned its soteriology only from the gospel of the apostles, which first they had preached and later “had handed down [trado] to us in the scriptures [scriptura].”35 These two sentences in Adversus haereses 3.1.1 closely parallel each other and the one in Papias’s preface.There are similar parallels to the manner in which Irenaeus discusses other early noncanonical Christian writings that he calls scripture. As we will see with Hermas, the words of Papias are woven into a fabric of testimony along with biblical material. Immediately before and after the quotation from Papias there are quotations of and allusions to the words of Matthew (26:27–29; 19:29; 13:38), Psalms (103:30), Luke (14:12–13; 18:29–30), Genesis (1:26–28, 30; 2:2–3; 27:27–29), and Isaiah (11:6–9; 65:25).36 Furthermore, Irenaeus uses the linking term “Lord” (Dominus), the source of the sayings, to connect the words of Luke, Papias, and Isaiah—they all contain the teachings of the Lord.37 Finally, Irenaeus’s introduction to Papias anticipates an element repeated in his introductions to authors of other scriptural texts: Papias saw and heard an apostle, in his case John, “the disciple of the Lord.”38 It is difficult to deny that this John is the apostle, the author of the Apocalypse and Gospel, in favor of another John, an “elder.”39 It seems likely, then, that we are to take Irenaeus’s introduction to Papias’s Exposition as an introduction to a text that Irenaeus considered to be scripture in a secondary or derived sense. Already, then, there is reason to reconsider the common scholarly understanding that texts regarded as scriptural in the second century but not ultimately canonized were simply texts that fell from a formerly lofty place.HermasAs I mentioned earlier, Hermas (Mandate 1.1) requires extended treatment because of the attention given it by both Irenaeus and contemporary scholarship. But it requires special attention for another reason: among all the Christian noncanonical texts I will examine, Irenaeus’s classification of this text is the most obscure. He may see Hermas as a nondescript writing; he may see it as sacred scripture; he may see it as scripture in the secondary sense; or he may mistake Mandate 1.1 for a biblical text. Because this case is so complex, I will carefully explicate Irenaeus’s perspective on Hermas, along with the relationship of that perspective to contemporary scholarship. Our discussion will come in two parts. First, I will examine the biblical and theological connections Irenaeus makes through his usage of Hermas. This examination lays the groundwork for the way Hermas is ultimately classified by drawing parameters around our options. Next, I will specifically treat the question of how we should classify Hermas. Is it writing? Sacred scripture? Scripture in a secondary sense? Biblical text? As in the case of Papias’s Exposition, my questions are, in what ways does Irenaeus see Hermas as scripture, and how can his classification help us to think about a model for the process of canonization?Here is the passage in which Irenaeus treats Hermas (Mandate 1.1), most fully:40Rightly, then, [do we learn from] the scripture [γραφή; scriptura] which said, “First, above all, believe that there is only one God, who has created [κτίζω; constituo] and completed [καταρτίζω; consummo] all things, who has made [facio] all things to exist from nothing so that they might exist, who contains all things, and yet is contained by nothing.”41 Rightly, also, from among the prophets, Malachi said: “Is there not only one God who has created us? Do we not all have only one Father?”42 Also, in accordance with these witnesses, the apostle rightly said, “There is one God, the Father, who is above all, and in us all.”43 In like manner, the Lord also said: “All things have been handed over to me by my Father.”44Irenaeus introduces his citation of Hermas’s teaching on the unique creator by identifying it as scripture (γραφή; scriptura). Immediately following the words of Hermas are citations from Malachi, introduced as a prophet; Paul, introduced as an apostle; and Jesus, from Matthew, introduced as the Lord. The words of Hermas are joined to the words of a prophet, an apostle, and the Lord in a cento supporting the church’s faith that both testaments proclaim that the creator is the Father.45 The texts of Mal. 2:10, Eph. 4:6, and Matt. 11:27, and the preceding texts of Gen. 1:26 and 2:7 and John 1:1–3, are added to Mandate 1.1 because they make explicit what is only implicit in Hermas: the one creator of the church’s faith is the Father who creates by his Son and Spirit. In addition to his emphasis on one creator, Hermas’s particular contribution comes in his three verbs that describe the creator’s act: create, make, complete. Irenaeus uses the more explicit biblical witnesses to interpret Mandate 1.1 as proclaiming the common creative activity of the Three: the Father who creates, the Son who makes, and the Spirit who completes.46 He reuses three of the same biblical texts alongside other sections from Mandate 1.1 for the same theological purpose.47 In this cento, Irenaeus introduces only Hermas as scripture, while he is satisfied to identify the others by office; but the diversely labeled texts are seamlessly joined into a cohesive, complementary, authoritative witness to the unique supremacy of the one God and Father who creates by means of his Son.This is not the first time in Adversus haereses that Irenaeus employs a network of prophet, apostle, and Hermas. In book 1, he uses language from several writings to compose a confession of the rule of truth. Interwoven into a cohesive theological statement are bits of language from Hermas (Mandate 1.1), which provides most of the content, and other nonspecific biblical material, all of which occur without attribution. The statement reflects Irenaeus’s peculiar theological concerns regarding the Creator. The language of Hermas declares faith in the one God who creates, organizes, and makes all things out of nothing, echoing 2 Macc. 7:28 and Wisd. 1:14. Into this Irenaeus inserts the biblical language “by his Word” and “almighty,” typical confessional language for the one supreme Father who creates through his agent, his Word and Son.48 This allows him to read Hermas as teaching that the Father creates, while his Son makes and his Spirit organizes.Irenaeus provides immediate biblical support for this typical, confessional statement of faith with a cento of texts from Ps. 33 [32]:6 and John 1:3, introducing them with the phrase “just as scripture says.”49 Once again, together they form a unified testimony to the Father’s creation of all things through his Son (Word), who makes, and his Spirit, who organizes, as he makes the implicit faith of Hermas explicit by means of biblical witness.50 It is worth noting that although biblical language is used to compose the confessional statement, scripture is called forth to confirm it. Irenaeus, by this move, is not saying that the language of scripture—whether that language is from Hermas, a psalmist, or an evangelist—is absent from the summary statement of the rule of truth; bits of scripture are certainly there. Instead, he is saying that confessional statements, even when they contain bits of scriptural language, require the fuller witness of passages of scripture. After all, anyone, including Irenaeus’s opponents, could have put together a pastiche of disassociated bits to pervert the proper order and connection of that material, its metanarrative.51 An authentic, Christian, doctrinal pastiche or cento requires the warrant provided by the testimony of the whole scriptural witness in proper connection. Irenaeus’s use of Hermas in this context, then, seems to refute Adelin Rousseau’s fourth objection, and one of Norbert Brox’s, to the notion that Irenaeus views Hermas as scripture in any sense other than as a generic writing.52Once again, then, Irenaeus interweaves Hermas’s material with biblical material. Again, Irenaeus does not name Hermas as the source. Again, the material used is confined to the content of Mandate 1.1. Again, Hermas must be supplemented by biblical texts with Christological content. The difference between our first example of Hermas in Irenaeus and the one here is that, here, the words of Hermas are part of the confessional cento that requires the fuller scriptural witness; in the first example, the words of Hermas are included in the fuller scriptural witness that supports the confessional pastiche. We will return to consider the first occurrence below.In two other places, Irenaeus repeats this pattern and freely reproduces the language of Mandate 1.1, while introducing additional biblical and theological language into the material from Hermas.53 In one of these moments, he adds specific mention of the Father. In the other, he adds, by means of Heb. 1:3, the notion of the Father’s word as the agent of creation (“by the word of his power”) and inserts as well the role of the Father’s wisdom in creation (“by his wisdom”). Irenaeus must contribute the concepts of the Father, Son, and Spirit (wisdom) to Hermas’s statement. The results are remarkable theological statements that he expands into the church’s trinitarian faith in God and creation with the aid of additional scripture. As he develops his thought outward from his free, amended versions of Mandate 1.1, he alludes to an abundance of biblical material.54 References to Ps. 32 [33]:6; Eph. 4:6; and Matt. 11:27 are most useful.55I have described the most prominent uses of Hermas (Mandate 1.1) in Irenaeus, but smaller portions do appear a few more times. The language of “the Father who contains all things, but cannot be contained” appears in ridicule of the Marcosian division of the Father of all.56 Irenaeus prefaces his use of Hermas with an allusion to Ps. 32 [33]:6, a text he frequently joins to Mandate 1.1, when he chastises his opponents for perverting the Creator, who is “the Word of God,” and the “Lord of all things who establishes the heavens.”57 Furthermore, free allusions to Hermas’s language on the God who creates and establishes all things from nothing may also be present in three other polemical contexts in book 2 of Adversus haereses.58The teaching of Hermas works well for Irenaeus, but it does not stand sufficiently on its own. As he develops the catholic doctrine of the Father’s nature and the trinitarian means of his creative act, he must regularly supplement Hermas with biblical material on the Father, Son (Word), and Spirit (Wisdom) in creation. All instances of Hermas’s words in Irenaeus are taken from Mandate 1.1, and they receive the striking introduction that they are scripture only once.59 The theological and biblical contexts in which Irenaeus regularly places Mandate 1.1 are instructive in classifying what Irenaeus means when he labels it as scripture. They introduce an option not considered before and simultaneously lend support to the taxonomy already observed in Irenaeus’s understanding of Papias’s Exposition.So how are we to understand Irenaeus’s reference to Mandate 1.1 as scripture? Scholars are divided in their opinions. Sundberg notes that the term indicates Irenaeus’s “approval” of Hermas but objects that it is “overstatement to say that this means canonical since γραφή was not a technical term.”60 For Farkasfalvy, the terminology (γραφή) applied to Hermas “hardly permits any conclusion about its special authority.”61 Many others agree, concluding that γραφή does not identify Hermas as holy scripture but simply as a generic writing.62 Steenberg, however, argues that γραφή introduces the entire cento of quotations—the words of Malachi, Paul, and the Lord, as well as those of Hermas—as scripture.63 Philippe Henne, nine years before, made the same argument, concluding that Irenaeus cited Hermas “among the canonical books,” and “included it within Scripture,” yet does not know precisely “where to place it.”64I agree with Steenberg and Henne that Irenaeus included Hermas within the texts he knew as scripture. It was more than a generic writing and was closely associated with the texts of the apostles. Irenaeus’s concept of scripture and his introduction of Hermas with γραφή tend toward this conclusion, and Eusebius is our best early witness to this thesis.65 But I disagree that Irenaeus uses the term to introduce the whole collection of texts. Irenaeus introduces only Hermas in this manner, while the biblical texts have their own introductions: Malachi is a prophet, Paul is an apostle, and the text from Matthew is a saying of the Lord.Rousseau, Henne, and Brox have pointed out the hierarchy within Irenaeus’s cento of texts. To them it manifests the ascending order in Irenaeus’s perspective on the components of scripture: Hermas, prophet, apostle, and then the Lord.66 Rousseau and Brox think that by distinguishing Hermas from the three latter categories of writings Irenaeus excludes Hermas from scripture.67 But it is important to remember the point made above by Steenberg and Farkasfalvy concerning Irenaeus’s broader concept of scripture, and the one that will be developed below concerning Irenaeus’s views of 1 Clement, Papias’s Exposition, and Polycarp’s Letter.68 Irenaeus reckoned as scripture certain writings, produced by those who had communication with the apostles, which transmitted the apostles’ teaching. It may be that we are to place Hermas in this class. These writings may rank lower in the hierarchy of witnesses—Irenaeus knows that the words of Hermas are not from a prophet, an apostle, or the Lord—but Irenaeus does not exclude them from scripture. The point of the cento is to provide a harmonious scriptural wit

Referência(s)
Altmetric
PlumX