Hearing the New Testament: Strategies for Interpretation

1

The Challenge of Hearing the New Testament

Joel B. Green

1. Reading as Communication

The challenge of hearing texts like the Gospel of Luke or Paul’s First Letter to the Corinthians is both similar to and more exacting than that of our daily attempts at discourse with one another. Conversation over a cup of coffee about, say, rising fuel prices or world hunger or next week’s Impressionist exhibit involve an addresser, an addressee, a context for communication, a medium for communication, and a message. The same can be said for our reading of 1 Corinthians, even if the terms differ somewhat. For example, is the medium of communication in a shared language, or must we concern ourselves with translation? Verbal, nonverbal, or both? Direct face-to-face contact, or indirect?¹

Reading texts is thus participation in a form of communication and can be diagrammed simply in the same manner as other forms of communication:

In either case, oral or written, we assume that the addresser communicates with the aim of being understood and, therefore, presumably shapes her message in order to cultivate understanding. Similarly, the addressee works to make sense of the message he hears or reads—and, indeed, tends to assume that the message is coherent and meaningful. In either case, oral or written, moreover, the addressee will be influenced in the process of sense-making. This influence might take the form of more deeply held convictions, of the possibility of revising previously held views, of affective responses, of viewing the Impressionists differently in the future, of new forms of behavior, and so on.

Interactions such as conversation or reading a biblical text share a further, common feature—namely, their openness to all sorts of ambiguities and possibilities for misunderstanding. Some of these ambiguities result from the nature of language itself. Others are the consequence of the different perspectives of the addresser and addressee. Still others result from challenges (or noise) experienced at the moment of communication. If my conversation partner has an oil well in her backyard, she might experience discussion of rising fuel prices differently than I do (since my backyard sports no oil well). We might have different experiences of hunger or different backgrounds with regard to agribusiness that translate into different views of the global food supply. Our hearing might be disturbed by the siren of a passing fire engine or by the ungrammaticalities of a Pauline text or by the rise and fall of coffeehouse chatter.

And here lies a major distinction between the two sorts of communication we are comparing. Over coffee, I can ask, Did I hear you correctly? or Are you serious? Accordingly, the simple diagram of communication sketched earlier is actually overly simplistic, since it does not account for feedback or the ongoing negotiation of meaning. It needs to be revised in favor of ongoing clarification and mutual exchange:

This works well in coffeehouse exchanges, but is more difficult when it comes to engaging ancient texts like those of the NT. In the latter case, the act of sending (say, Paul’s generating the letter we know as 1 Corinthians) is distant in time and space from the act of receiving (say, reading this Pauline text or hearing it read). On the one hand, then, Paul cannot adjust his language or deploy different metaphors to accommodate our reactions and assist our understanding. On the other hand, neither can we get Paul another cup of coffee while asking him whether we have understood what he wants to say. We can interrogate the whole of Paul’s letter, or even the whole of the Pauline letter collection, as we inquire whether our understanding coheres well with the text, but we cannot directly ask the apostle, Did we get it right?

Reading Paul or Luke is an act of communication, then, but not a straightforward one. We are reminded, then, that reading Luke’s Gospel is itself already an interpretive act. Communication consists in part of generating messages via a medium for presentation to an addressee and the interpretation of that medium to surmise a message. In reading, we decipher marks on the page and attribute meaning to them, and we do so in ways that reflect what we have learned before—about other similar marks on pages, for example, about the world reflected in this Gospel, and about what Luke is seeking to accomplish.

This means that the Gospel of Luke ought itself to be a partner in conversation—not simply an object to be investigated or analyzed, but a subject who speaks and even talks back to us. The conversation might take place at multiple levels. The conversation might have its focus behind the text if we are interested in the intention of the Evangelist or the nature of his community or the character of the historical Jesus to whom the Evangelist bears witness. The conversation might have its focus in front of the text, in our own imagination or experience. Where we locate the conversation will depend largely on our interpretive aims—and, then, on the tools and approaches we put into play as we seek to achieve those aims. Wherever we locate the conversation, hearing the NT presumes that, as an act of communication, our reading can never be dispassionate. We involve ourselves in a give-and-take that transcends an approach to the text motivated by an interest in obtaining information alone. By its nature, conversation has the potential to shape us in some way.

2. Obstacles to Conversation

Let us consider more deeply some of the challenges that face us as we engage these texts in conversation. Some have to do with the nature of language itself, but others with the character of our interaction with the text.

2.1. Language Is Linear

Texts present their data in a sequence—in English (and Greek), we read from left to right. We do not see a text like the book of Revelation as a whole. In fact, we cannot see Revelation as a single whole, since our brains are simply incapable of holding in the forefront of our minds the whole of this literary work in its picturesque detail. Instead, our reading (and rereading) produces a progressively fuller grasp of the book as we add to what we have already read. One vision gives way to the next, allowing us to build the message of Revelation at the same time that element after element of what we have already read moves into the shadows of our memories. Were we to visit an art gallery we might experience the totality of a landscape portrait, all at once. Returning to the coffee shop to review that experience with someone else, though, we would again be reduced to linearity: What to describe first? Second? Third?

The linearity of NT texts is not only a challenge to communication, but also an invitation to a series of important interpretive questions based on the simple observation that James or John has arranged his material in this (and not some other) sequence. What comes first? Second? Last? Why? Luke tells us that he is concerned with order (1:1–4), but what sort of order? Chronological order? Note the parallelism of these two texts in the Acts of the Apostles—which identify the beginning of two narrative sequences at the same temporal moment, even though they are separated by some three chapters of the narrative:

That day a severe persecution began against the church in Jerusalem, and all except the apostles were scattered throughout the countryside of Judea and Samaria.… Philip went down to the city of Samaria and proclaimed the Messiah to them. (8:1–5)

Now those who were scattered because of the persecution that took place over Stephen traveled as far as Phoenicia, Cyprus, and Antioch, and they spoke the word to no one except Jews. But among them were some men of Cyprus and Cyrene who, on coming to Antioch, spoke to the Hellenists also, proclaiming the Lord Jesus. (11:19–20)

The former relates the accounts of Philip’s evangelization in Samaria and on the road to Gaza, followed by Saul/Paul’s commission as a witness to the Gentiles and the missionary work of Peter among the Gentiles. Acts 8:1–11:18 is sometimes over the top in its detail of how each step of the mission beyond Jerusalem was undertaken at God’s behest. The narrative that begins at Acts 11:19, however, picks up at the same point in time as Philip’s mission among the Samarians, but seems to take as obvious to all that the consequence of the spread of the gospel would be a single people of God comprising both Jew and Gentile. Why, we might ask, does the narrator of Acts order his account in just this way?

Similarly, we might ask, was Jesus actually baptized after John was imprisoned (Luke 3:18–22)? In Revelation 5, John paints a word-picture, describing first the one worthy to open the scroll as the Lion of Judah and only then introducing in this role a Lamb standing as if it had been slaughtered (Rev 5:5–6). What direction of interpretation of both Lion and Lamb is encouraged by this sequence and by this creative mixture of images?

2.2. Language Is Selective

No addresser can say everything. Every detail cannot be presented. This is true today, in spite of the burgeoning capacity of our computers or blogs. How much more would be it true in the case of these ancient texts which, by most any contemporary reckoning, are brief? As a result, addressees—in our case, readers of NT texts—are left with some work to do. In fact, we can trace an inverse relationship between the amount of detail and cohesiveness provided by the addresser and the degree of interpretive work required of the reader.²

Some gaps occur because of the assumptions of the writer. Paul may assume, for example, that his audience at Corinth understands his reference to the death of some within the community as a consequence of Corinthian practices at the common meal (1 Cor 11:30). For readers today, though, his remark is overly cryptic, a puzzle to be resolved. Did James share with his original readers the identity of a particular righteous person who had been condemned and murdered by wealthy landowners (Jas 5:6)? That may be so, but modern translations struggle with whether to take ὁ δίκαιος (ho dikaios) as a particular righteous person (e.g., NRSV; Jesus? James?) or as a class of righteous ones (e.g., NIV; the poor whom James addresses?). Of course, in some instances, the writer provides needed background information—e.g., the narrative asides in John 4:8 (explaining the absence of Jesus’ disciples) and 4:9b (Jews do not share things in common with Samaritans). But other gaps remain, standing as invitations for readers to participate imaginatively in the production of meaning. What, for instance, lies behind John’s indictment of the crowds as offspring of poisonous snakes (Luke 3:7)?

2.3. Language Is Ambiguous

We encounter words in texts that are capable of multiple meanings, giving rise to potential ambiguities, whether one is reading or hearing someone read. Umberto Eco illustrates the ambiguity of symbols with reference to a picture of a ranch-style house with an old wagon wheel leaning against the front exterior wall. Is this a picture of a movie set for a western film? A snapshot of a ghost town? Should we be looking for a wagon with only three wheels, awaiting repair? Is this a restaurant serving authentic, country-style food? The workshop of a blacksmith? An antique shop? An actual ranch house?

2.4. Language Is Culturally Embedded

The decision to take these marks on a page as evidence of documents written in a now-dead language by Christ-followers of the ancient Mediterranean world already implicates our hearing of the NT in the work of social history. In fact, this is one of the liabilities of our ready access to the NT writings in our native languages; we easily assume that these words are our words, available to us in our cultural contexts, and easily forget that every reading of every NT text today is an exercise in intercultural communication and understanding. Too easily, then, we conscript NT texts to serve our own interests and find in them justification for our sometimes comfortable practices in the world.

Do we too easily find the Eucharist in 1 Cor 11:17–34? How do our contemporary eucharistic practices relate to the troubled picture of the community meal to which Paul addresses himself? Do we read our baptismal practices into Luke’s portrait of John’s repentance-baptism (Luke 3)? How do our practices, sanctioned by the church and typically located within church sanctuaries, relate to John’s prophetic activity in the wilderness, far from the Jerusalem Temple? What would it mean to read James as persons who share the status of whose to whom the letter itself is addressed—a marginal, not-at-home people (1:1)?

2.5. Readers Are Located

Not only texts but readers, too, are culturally embedded. As readers, we do not come clean to NT texts, but bring with us our selves, in all of the complexity of our rootedness. Are we people who experience what it means to be born on the winning side of history, or are we people who inhabit the margins of our social worlds? Another example: Do we locate our reading of NT texts primarily as persons grounded in and serving ecclesial communities, or as persons sans any particular faith commitments? In a fascinating experiment, Mark Allan Powell documented how students in the U.S., Russia, and East Africa heard different things in the telling of the same story, the so-called Parable of the Prodigal Son (Luke 15:11–32). For the East Africans, the prodigal ended up starving in a pigpen because no one gave him anything to eat; for the Americans, because he squandered his money; for the Russians, because of the famine.³ Whose is the correct reading? In fact, not only are all three of these factors present in the text, but the Lukan parable itself documents the importance of just this sort of perspective. As Jesus tells the story, the younger son wasted all his money on wild living (15:13); when the elder son tells the tale to his father, though, he denies any familial relationship with his brother and fills in the details of his wild living: this son of yours comes back after squandering your money on prostitutes (15:30). As literary critic and novelist C. S. Lewis observed, For what you see and hear depends a good deal on where you are standing: it also depends on what sort of person you are.⁴

2.6. Readers Are More or Less Practiced

Like conversation partners, some readers of the NT are more competent than others and are therefore apt to produce readings that are more faithful and convincing. It is tempting to suggest that this is because some readers of the NT are more methodologically astute than others, or that some readers of the NT have chosen the right method (and others the wrong), but neither of these is a good measure of readerly competence. Competence is not a simple matter of good technique.

It is easy to see where our fascination with technique derives from and why we might be tempted to imagine that what is most needed is a technology of NT interpretation. We need only consider the pervasive influence of the polymath René Descartes (1596–1650) on modern ways of thinking and doing. During his intellectually formative years, Descartes was impressed by the hydraulic machines animating the figures of Neptune, Diana, and other mythical characters in the royal gardens of St. Germain, by mechanical dolls in the windows of upscale shops, and by the intricate mechanics by which tower clocks kept the time of day. Not surprisingly, then, we learn from him, erroneously, that thinking is the product of a disembodied mind (as though our own histories and contexts were not factors in understanding), but also that our work in the world takes the form of clocklike mechanisms. To the influence of Descartes we could add others, including Isaac Newton, whose mechanics emphasized cause-effect relations in a way having direct bearing on interpretive work. In a universe thus explained, is it any wonder that we gravitate so easily to interpretive methods supposedly capable of disambiguating texts in step-by-step, follow-the-instruction processes?

By tying competence to practice, however, I am urging an alternative view. On the one hand, the practiced reader of the NT is one who regularly puts into play the very sorts of perspectives, methods, and sensibilities discussed in the chapters that follow. In this sense, practice is to NT interpretation what rehearsal is to an orchestra: the behind-the-scenes work that enables faithful performance. Accordingly, we might ask, what is the end (or goal) of critical engagement with the NT? An essay? A sermon? A commentary? A people embodying its message? On the other hand, the practiced reader of the NT is one whose reading of the NT has itself been formative. This assumes that hermeneutics is concerned with more than passing on information and that NT interpretation concerns itself with more than getting the meaning right.⁵ What practices are encouraged—even put into play—through ongoing engagement with these texts?

Concerns with method are easily tied to expectations of set procedures, executable laws, robotic routines, consistent results, and repeatable outcomes. Method might conjure images of how-to manuals, programmable steps to take, guidelines to follow as if working on a conveyor line, or the interpreter as a veritable meaning-making machine. Like driving a car or learning to paint, certain rules apply, but, eventually, the rules are internalized so that they operate at a preconscious level. The best woodworking is done by those who are beyond parroting the instructions, the best surgery by those whose practiced hands seem almost to move on their own. This is not the absence of technique as much as its embodiment.

Similarly, Alasdair MacIntyre observes that the practice of a craft might involve procedures and steps. It might even build on a history of performance. Nevertheless, craft is ultimately defined by complexity and innovation that surpass those procedures and help to constitute evolving rules.

The authority of a master within a craft is both more and other than a matter of exemplifying the best standards so far. It is also and most importantly a matter of knowing how to go further and especially how to direct others towards going further, using what can be learned from the tradition afforded by the past to move towards the telos of full perfected work. It is thus knowing how to link past and future that those with authority are able to draw upon tradition, to interpret and reinterpret it, so that its directedness towards the telos of that particular craft becomes apparent in new and characteristically unexpected ways.⁶

Mastering a craft thus entails developing particular intuitions, forming particular dispositions, becoming a particular kind of person whose commitments and predilections have been shaped in relation to the activity in question.

We may speak of biblical interpretation by referring to methods, but our interest is keenly directed toward the inculcation of certain sensitivities and sensibilities, the habituation of patterns, postures, and practices that express themselves as second nature, without observable calculation. Over time, our intimacy with the NT is such that we would never imagine packing our equipment in preparation for a visit to this strange land of the NT, for it is here in the engagement with these texts that we find ourselves truly at home.

3. Orientation

Method has to do with any of a range of interpretive strategies with which one might engage a text. In recent years, many students of the Bible have learned to catalogue these in terms of where each locates meaning in relation to the biblical text. Three (rather obvious) options have emerged:

behind the text

in the text

in front of the text

Behind-the-text approaches address the text as a window through which to access and examine the deposit of meaning. These approaches, then, locate meaning especially in the history assumed by the text, the history that gave rise to the text, and/or the history to which a text gives witness. In-the-text methods recalibrate their gaze so as to bring into focus the qualities of the text itself, its architecture, consistency, and texture. Emphasis falls on the perspective contained within and transmitted by the text, apprehending the text as a kind of sealed container of meaning. In-front-of-the-text approaches orient themselves around the perspectives of various readers of the text, on readerly communities, and/or on the effects that texts (might) have on their readers. In this case, readers do not simply perceive but actually produce, or at least assist in the production of, meaning.

Let me register two immediate caveats about this typology. First, I have expressed these as ideal types that rarely, if ever, appear in such pure forms. This has always been true to some degree, but is today even more so, since growing recognition of the potential of all of these approaches has fueled attempts to integrate them. Second, at the same time, this typology continues to have utility, since our methods align themselves more with one type than another, according to one’s particular interests. Where do we locate meaning? The history behind the text? The text itself? Those persons and communities doing the reading?

(1) Behind-the-Text approaches have been central to NT study for more than two centuries. Accordingly, contemporary concern with method cannot be sketched without reference to the enormous upheavals that have shaken the foundations of biblical study in the last forty years. During these decades, we have witnessed the fall of historical criticism as the approach that previously, quite literally defined critical biblical studies. Accordingly, the two notions, usually held in tandem, that the meaning of a text is contained within its history and that the role of the interpreter is to isolate the one, single, historically intended, correct meaning of a biblical text have become increasingly marginal in biblical studies. If at the turn of the twenty-first century the foundations of historical study have been shaken rigorously, we must nonetheless recognize the incontestable importance of history for NT studies. This is true not only because a number of NT writings look like history-writing—that is, they adopt modes of discourse congruent with ancient traditions of biography and historiography (i.e., the Gospels and Acts). Additionally, every NT document situates itself in history, whether localized to a city or region or even a particular household (e.g., Phil 4:2; Philemon; Revelation 2–3) or worldwide in its scope (e.g., Luke 2:1–7; 3:1–6; 1 Pet 5:9). The historical dimension of the NT is further, and decisively, documented in this central theological declaration: The Word became flesh and lived among us, and we have seen his glory, the glory as of a father’s only son, full of grace and truth (John 1:14). Hence, we could hardly call for the suspension or rejection of historical inquiry, even if it is clear that we need more nuanced ways of construing notions of history and historical.

In different ways and to different degrees, Chapters 2–11 of this book concern themselves with issues of historical inquiry. Some are explicitly oriented toward behind-the-text questions—for example, Chapter 3 (Historical Criticism and Social-Scientific Perspectives in New Testament Study), the parallel Chapters 4 and 5 (The Relevance of Extracanonical Jewish Texts to New Testament Study and The Relevance of Greco-Roman Literature and Culture to New Testament Study), and Chapter 6 (Traditio-Historical Criticism). Establishing the text itself (Chapter 2) is a historical project essential to all other interpretive concerns. Others are concerned with locating NT texts within their literary environs (Chapter 7: The Use of the Old Testament by New Testament Writers, and Chapter 8: Genre Analysis), or with how words and modes of argument might have been perceived (Chapter 9: Rhetorical Criticism, Chapter 10: Modern Linguistics and Word Study in the New Testament, and Chapter 11: Discourse Analysis and New Testament Interpretation).

(2) In-the-Text approaches to the NT redirect the focus from the world behind the text to the text itself. For a brief period, but only a brief period, and in only selected quarters, it appeared that the hyper-concern with historical issues that characterized critical scholarship since the late eighteenth century might be replaced with a narrow emphasis on the text itself. In the case of the NT writings, we are physically incapable of checking our interpretations in any direct way against the intentions of their writers; indeed, in some cases, we do not even know to whom we should attribute the authorship of these texts. In the end, readers have before them to consult only the text itself. The effect of this truism was, for some interpreters, that the text itself could and ought to be regarded as a self-sufficient, self-contained verbal artifact. Accordingly, the text is presumed to be the unique and privileged source of meaning and interpretive value, with this available to the interpreter by means of careful attention to its language and structure.

In reality, however, NT texts are not self-sufficient verbal artifacts that contain within themselves all that is needed for their interpretation. NT studies has long recognized that these texts did not fall out of the sky; they are not facsimiles from heaven, but arose in particular times and places in response to particular situations. They are cultural products that participate in, legitimate, perpetuate, criticize, etc. the worlds within which they were generated. Texts exist in a relationship of constraint and mobility with their cultural contexts, as authors assemble and shape the forces of their worlds in fresh ways that both draw upon and point beyond those cultural elements.⁷ Hence, although nothing can make up for a reader’s lack of attention to the text, in-the-text readings are generally viewed as important points of beginning in need of additional interpretive emphases. In fact, an interest in the text as text is not only an important but, indeed, an indispensable and non-negotiable initial step.

Among the approaches that center our attention especially on the text itself are those discussed and illustrated in Chapters 7–12. Literary concerns are at work in discussions, then, of The Use of the Old Testament by New Testament Writers (Chapter 7), Genre Analysis (Chapter 8), Rhetorical Criticism (Chapter 9), Modern Linguistics and Word Study (Chapter 10), and Discourse Analysis (Chapter 11). That Chapters 7–11 also appear under the heading of behind-the-text approaches suggests the degree to which, in practice, these categories are not hermetically sealed off from one another. Indeed, many literary approaches embrace historical interests. This is true of Narrative Criticism (Chapter 12), too, in spite of the origins of narrative study in more narrowly text-based analysis of stories.

(3) In-front-of-the-Text approaches take seriously the question, Who is doing the reading? In recent decades, this has become an increasingly crucial question, since, it is increasingly acknowledged, all interpretation is influenced and conditioned by the interests and social location of the interpreter.

At a rather abstract level, awareness of the importance of the reader in biblical studies grows out of the claim that meaning is not so much repeated or reproduced in the experience of reading; instead, reading constitutes, at least in some sense, the production of meaning. My use of the phrase at least in some sense comes in recognition that there is no singular perspective on the role of the reader today. Reader-oriented study is not so much a method as a constellation of approaches that share a common, basic rejection of any portrait of the reader as a mere (potential) receptacle for meaning. Reader-oriented approaches are scattered across a continuum according to the extent of readerly involvement they presume.

Additionally, persons concerned with the role of the reader struggle to account for the perspective from which reading occurs. That is, the question is not only what one means by reader, but also the perspective from which the reader does the reading—and it is this awareness of perspective that fuels a variety of significant approaches to the hermeneutical task. For those who take interpretation as nothing more than the discovery of a meaning already resident in the text, this interest in perspective will seem strange, just as the possibility of the Bible being used as an instrument of anything, positive or negative, may seem remote. This, however, is precisely the starting point of a liberationist hermeneutic: the recognition that all textual inquiry is shaped by the reader’s context. Taking this observation seriously leads many to read NT texts in fresh ways at the same time that it raises for them a fundamental critique of the commitment to observer neutrality propagated in the practice of biblical interpretation in the modern period.

Chapters 13–18 are each in their own ways attempts to take seriously the role of the reader, or the readerly community, in NT interpretation. The focus of Chapter 13 on The Reader in New Testament Interpretation opens the way to more thoroughgoing explorations of the role of social location in the work of NT study in Chapters 14 (Feminist Criticism), 15 (African American Criticism), and 16 (Latino/a Hermeneutics). In related ways, the final two chapters—Chapter 17: Reading the New Testament in Canonical Context, and Chapter 18: The New Testament, Theology, and Ethics—continue this exploration by identifying canon and church as contexts within which to read the NT.

Today, no one interpretive method can claim to provide the one authentic meaning of a NT text. No particular method can be identified as the correct one nor can any method ensure a faithful reading of the NT. We are dealing here with a mysterious alchemy for which the NT text serves as the single stable factor in an otherwise shifting equation. The purpose of juxtaposing these various methods and approaches in this collection is not to suggest that a valid interpretation will somehow combine them all (even if that were possible). Rather, it is to sketch something of the range of questions that might be posed to particular texts and to demonstrate how different interpretive aims sponsor different interpretive protocols.

In order to display this variety, contributors to this volume have been asked to work with a handful of set texts: Luke 3:1–20; John 4:1–42; 1 Cor 11:2–34; Jas 4:13–5:6; and Revelation 5. These texts represent distinct challenges to the interpreter, have their own histories of interpretation, and draw upon a range of literary forms. No chapter treats all of these texts in detail, but each exhibits how that method might work itself out in practice with reference to one or more of these texts.

Suggestions for Further Reading are given at the close of each chapter for those interested in pursuing further the various approaches discussed.

1. On what follows, see Casare Segre with Tomaso Kemeny, Introduction to the Analysis of the Literary Text (AS; Bloomington: Indiana University Press, 1988), §1.1.

2. Gillian Brown and George Yule, Discourse Analysis (CTL; Cambridge: Cambridge University Press, 1983), 266, 269–70.

3. Mark Allan Powell, What Do They Hear? Bridging the Gap between Pulpit and Pew (Nashville: Abingdon, 2007), 11–27.

4. C. S. Lewis, The Magician’s Nephew (New York: Macmillan, 1970), 125.

5. Cf. Anthony C. Thiselton, New Horizons in Hermeneutics: The Theory and Practice of Transforming Biblical Reading (London: HarperCollins, 1992), 17: "Why in hermeneutical theory up to 1970 had so little attention been given the capacity of biblical texts to produce certain transforming effects, rather than only to transmit certain disclosures?" (emphasis original).

6. Alasdair MacIntyre, Three Rival Versions of Moral Inquiry: Encyclopaedia, Genealogy and Tradition (Notre Dame: University of Notre Dame Press, 1990), 65–66.

7. See Stephen Greenblatt, Culture, in Critical Terms for Literary Study (ed. Frank Lentricchia and Thomas McLaughlin; Chicago: University of Chicago Press, 1990), 225–32.

2

Textual Criticism of the New Testament

Bart D. Ehrman

Unlike other methods discussed in this book, textual criticism is not an option for interpreters of the NT. Whereas other approaches presuppose the wording of the text under consideration, textual criticism determines that wording. To put the matter somewhat differently, we cannot begin to explore what a text means until we know what it says. Rather than interpreting the text, textual criticism decides which words belong in the text. For this reason, textual criticism is a foundational discipline—indeed, the foundational discipline—for NT studies.

1. Why Textual Criticism Is Needed

We need the discipline of textual criticism because we do not have the original manuscripts of any of the books of the NT. What we have are copies that were made much later than the originals, in most instances many centuries later. These later copies were not themselves made from the autographs;¹ they were instead made from copies of copies of copies of the autographs. The problem with these later copies, the ones that have survived to our day, is that they all differ from one another to a greater or lesser extent. Textual criticism examines these surviving manuscripts²—and there are thousands of them—and tries to decide on the basis of established critical principles what the originals themselves must have said.

Only with the invention of moveable type five hundred years ago was it possible to mass produce literature with absolute accuracy, one copy appearing exactly like another. In antiquity, exact replications of texts were virtually impossible. When an author produced a book, he or she had to have it copied by hand; anyone else wanting a copy needed to produce it for himself or herself or hire a professional scribe to do so. Thus, to use an illustration from the NT, whoever wrote the Gospel of Matthew no doubt produced his text for his own community. If other members of the community wanted a copy for themselves, they would have to go through the laborious and painstaking process of copying it, one word at a time. Or if a Christian from a neighboring community wanted a copy, he or she would have to go through the same process. We can assume that Christians from different locations did in fact want to have copies of such valuable works. This led to a proliferation of copies, made by different Christians from different places, most of them private individuals rather than professional scribes, with differing aptitudes for the task.

The difficulty with making handwritten copies of such a long text as Matthew, or even shorter ones, is that it is nearly impossible to do so without making mistakes. (Anyone who doubts this should try to copy the Gospel of Matthew by hand, a text of about thirty-four pages in my NRSV Bible.) Moreover, anyone who makes a copy not of the original text, but of a copy of the original, will not only make new mistakes, but will also reproduce the mistakes created by the person who made the copy being used. In this way, mistakes multiply from copy to copy. Sometimes, of course, a copyist may detect a mistake in the manuscript he or she is using (for example, when the previous scribe has accidentally left out a word or an entire verse—a fairly common occurrence in our manuscripts). When this happens, the scribe producing the new copy may try to correct the error. Unfortunately, if the original is not available for checking, the correction may not restore the reading of the original, but introduce a new error, which will then be copied by the next copyist. And so it goes. Copies of copies of copies, each with its own errors and the errors of the copy from which it was produced.

Moreover, we know from a study of ancient manuscripts that errors were not always accidental. Sometimes scribes felt inclined to change the text that they read. For example, copyists who came to a verse like Matt 24:36—which indicates that no one, not even the Son of God himself, knows when the end will come—could well take offense at the idea that Christ did not know the time of his own return (this was especially a problem for scribes who were convinced that Christ was none other than God). This might lead a scribe to modify the text. Indeed, in this case it often did lead to a modification: a number of our manuscripts omit the words not even the Son from Matt 24:36. Scribes who made this change no doubt saw it as a correction or an improvement, but textual critics, concerned to know what Matthew himself wrote, would label it a corruption.³

To this point we have been speaking chiefly in the abstract about scribes inadvertently and intentionally modifying the texts of the NT that they copied. What, though, are the concrete realities? Were there in fact a large number of such errors? For us to realize the extent of the problem, some basic data may prove useful. At present, we have over 5360 manuscripts of all or part of the NT in Greek (the language in which all its books were originally written).⁴ These manuscripts range in size from tiny fragments the size of a credit card to hefty volumes that include all twentyseven books of the NT. They range in date from the early second century⁵ to the sixteenth century (some copies were made by hand even after the invention of moveable type printing). What is striking is that among these thousands of manuscripts, with the exception of the smallest fragments, no two are exactly alike in all of their particulars.

The manuscripts themselves thus leave no question that scribes made changes in their texts, and many such changes. How many differences are there among our surviving manuscripts? Although estimates typically put them in the hundreds of thousands, no one knows the real number for certain because no one has yet been able to count them all. What we can say with confidence is that there are more differences among the manuscripts than there are words in the NT.

This is not to say that we are totally at a loss when trying to decide what the original text of the NT was. In fact, the vast majority of the differences in our manuscripts are insignificant, irrelevant, and easy to explain. Far and away the most common mistakes involve differences in spelling; many others involve accidental omissions of words and phrases by careless scribes.

There are, however, a large number of cases where the wording of a passage differs significantly among our manuscripts in ways that are critical for exegesis. In these instances, textual critics have to balance the arguments for one form of the text over those for another and then render a judgment as to which appears to be the original text and which a corruption of it by a later scribe.

To illustrate the potential significance of such differences, we might consider some of the more famous and striking examples. Did the author of the Gospel of Mark end his narrative at 16:8 with the women fleeing Jesus’ empty tomb in fear, telling no one what they had seen? Or did the author write the final twelve verses found in some of our manuscripts but not in others, verses in which the resurrected Jesus appears to his disciples and tells them that those who believe in him will be able to handle deadly snakes and drink poison without suffering harm? Did the author of the Fourth Gospel write the famous story of the woman taken in adultery, or was this a later addition to the Gospel by a well-meaning scribe? The story is found in many of our later manuscripts between chs. 7 and 8, but in none of the early manuscripts. Did the voice at Jesus’ baptism in the Gospel of Luke originally declare, You are my beloved Son, in whom I am well pleased (exactly the words found in Mark’s account), or did it proclaim, You are my Son, today I have begotten you (as the text is worded in some of our earliest witnesses)? The latter version, a quotation of Psalm 2, proved acceptable to second-century Christians who denied that Jesus was God by insisting that he came to be adopted as God’s Son at his baptism. As a final example, did the author of 1 John include the famous Johannine comma (5:7–8), the only passage in the entire Bible that explicitly affirms the trinitarian views of later Christians—that the Godhead consists of three persons and that these three are one? Even though the passage is part of the Latin Bible and found its way into the King James Version, it does not occur in any Greek manuscript of the NT earlier than the fourteenth century.

2. The Theory and Practice of Textual Criticism

Since all our surviving manuscripts have mistakes, scholars must decide the original wording of the text on a case-by-case basis. The process of making this decision in view of the whole range of evidence is sometimes called the eclectic method.⁶ In rough terms, the textual evidence is classified either as external—that is, based on the kinds of manuscripts that support one reading or the other; or internal—that is, based on the likelihood of a reading going back to the original author or to an error introduced by a scribe.

2.1. External Evidence

Since our only access to the words of the NT authors comes through the flawed manuscripts of their writings, it is important to understand how critics use these witnesses when trying to reconstruct the original text. First it is necessary to know about the kinds of manuscript evidence that are available.

a. Greek Manuscripts. The more than 5360 Greek manuscripts of all or part of the NT range in date, as we have seen, from the early second century through the sixteenth. Very few, however, are from the earlier period, down to about the year 400.

Greek manuscripts are normally divided into three main categories.⁷ The papyri, that is, those written on papyrus,⁸ are the earliest available witnesses, most of them dating from the second to the fifth centuries; the majuscules were written on parchment or vellum in uncial letters—comparable to our English capitals—from roughly the fourth to the ninth centuries; the minuscules were written on parchment or vellum, but in minuscule (= small) letters—comparable to our English cursive—after the ninth century.

b. Early Versions. In addition to the Greek witnesses, we have a large number of NT manuscripts in other languages. Christians from the second century on recognized the need for translations of the Scriptures into other languages for those who did not read or speak Greek. The earliest translations were probably into Syriac and Latin, perhaps as early as the late second or early third centuries; some time thereafter the NT was translated into Coptic, and eventually into Armenian, Ethiopic, Georgian, Gothic, and other languages. Like the Greek NT, each of these versions survives in a number of manuscripts, all of which, again, appear to contain mistakes.⁹

It is possible to compare the various manuscripts of any of the versions to decide what the earliest form of the translation was, to take that form and translate it back into Greek, and on that basis to decide what form of the Greek text was available to the original translator. Needless to say, this is an arduous and technical process, but it does produce useful results for scholars concerned to learn what Greek manuscripts were available in the early period when the versions were made—a period from which, unfortunately, very few actual Greek manuscripts survive.

c. Patristic quotations. Finally, we have numerous Christian writings from the second century on; the authors of these works frequently quote the NT, making it theoretically possible to reconstruct the wording of the manuscripts that they themselves were using. This particular kind of external evidence is especially fraught with uncertainty, as it is not always easy to decide whether an author is quoting exactly or simply paraphrasing; moreover, these early writings have also been handed down to us only in manuscripts produced by scribes who sometimes changed quotations found in their texts in conformity with the wording of the Scriptures as they themselves knew them. Nonetheless, when this kind of evidence is studied carefully, it can prove quite valuable; unlike the scribes of our surviving Greek manuscripts and the early versions, the patristic writers can be fixed in time and space. We know exactly when and where most of them lived; their quotations can therefore indicate with relative certainty how the text of the NT had been changed in different times and places.

How can these thousands of bits of data be used to determine the original text wherever there are differences among our witnesses? Over the years, scholars have devised a number of principles of criticism, some of which, as we will see, are of greater use than others. These principles can be expressed in terms of questions that a critic will bring to a passage that is attested in a variety of ways among different witnesses.

a. How many witnesses support each reading? For some critics, this is the all-important question; others, however, discount it entirely. Those who support this principle argue that if a passage is worded in one way in three hundred manuscripts but differently in only three, it is more likely that the majority text is original and that the three aberrant witnesses have simply incorporated a mistake.

The problem with this logic is that it overlooks an important feature of our manuscript evidence. All of our witnesses were copied from others. Suppose that at some time during the second century there were two manuscripts with a different reading for a particular verse. One was in a remote area and came to be copied three times before being destroyed in a fire. The other was in a major metropolitan area with a large Christian population; it was copied thirty times, and each of the copies made from it was copied ten times. At the end of this process, there would be three hundred and one manuscripts with one reading and three with the other. Does that mean that the three hundred and one are correct and the three are in error? Not necessarily! In fact, the numerical difference in the manuscript support for the two readings is not three hundred and one to three, but one to one. Each reading goes back to a solitary manuscript of the same time period. For this reason, simply counting the number of manuscripts supporting a reading is not the best way to decide the original text.

b. Which reading is supported by the most ancient manuscripts? The same illustration shows the importance of knowing the date of the witnesses supporting each variant reading. If there are two hundred manuscripts from the fifth to the sixteenth centuries with a particular form of the text, but two from the third century with a different form, then the form supported by the two may well be superior to that found in the two hundred. The logic here is that since manuscripts become increasingly corrupt with the passage of time (since copyists reproduce the mistakes of their predecessors as well as create some of their own), the earlier manuscripts will as a rule be better than the later ones.

Although this argument is generally valid, it, too, is not entirely foolproof. Suppose there are two manuscripts with different forms of the text, one from the third century and the other from the fifth. A critic might naturally think that the third-century witness is superior since it is older. But we cannot always be certain about the age of each manuscript’s exemplar, that is, the copy that the scribe used to produce the manuscript. It is entirely possible, for example, that the third-century manuscript had as its exemplar a manuscript made ten years earlier, still in the third century, whereas the copyist of the fifth-century manuscript had access to a very old copy preserved in his church library, say, from the second century. In this case, ironically enough, the fifth-century manuscript would reproduce an older form of the text than the third-century manuscript! For this reason, even though the age of a manuscript can be important in determining the quality of its text, it is by no means a fail-safe guide.

c. How geographically diverse is the attestation for each reading? Less problematic is the matter of the geographic distribution of a reading. Here again the principle can be stated simply. If manuscripts support two different forms of a passage, with one of the forms restricted to witnesses produced in only one geographical area (Italy, for instance), whereas the other is found in witnesses spread throughout the Mediterranean (e.g., Northern Africa, Alexandria, Syria, Asia Minor, Gaul, and Spain), then the former is more likely a local variation reproduced by scribes in the region, and the other is more likely older since it was more widely known. If witnesses supporting a reading are both early and widespread, a strong case can be made that this reading is original.

d. What is the quality of the supporting witnesses? As in a court of law, there are some textual witnesses that are more reliable than others. The general principle of external evidence is that witnesses that are known to produce an inferior text—when the case can be decided with a high degree of certainty (on the internal grounds discussed below)—are also likely to produce an inferior text where the