Tampering with the Text: Was the New Testament Text Changed Along the Way? (part 2)

Andreas J. Kostenberger and Michael J. Kruger
Tampering with the Text: Was the New Testament Text Changed Along the Way? (part 2)

EDITOR'S NOTE: The following is part 2 of an excerpt from The Heresy of Orthodoxy by Andreas J. Kostenberger and Michael J. Kruger (Crossway).

Thesis 3: The Reliability of the Text Critical Method: of the small portion of variations that are significant, our text-critical methodology can determine, with a reasonable degree of certainty, which is the original text 

In part 1 of this article series, it was demonstrated that the vast majority of textual variations are insignificant and irrelevant to determining the original text of the New Testament. However, that leaves a small portion of textual variants that can be deemed "significant." The definition of this term has two aspects: (1) "significant" textual variants are simply those that are not included in the "insignificant" category discussed above; and (2) "significant" variants are those that in some sense affect the meaning of the passage (though the effect can range from fairly minimal to more substantial). 

Even though the quantity of these significant variants is quite small in comparison to insignificant variants, some of them can still make an impact on our understanding of New Testament passages (as we shall see below). Thus one might conclude that these sorts of changes present a real challenge to the textual integrity of the New Testament. However, such a conclusion would be built upon an assumption that we have no way to determine which of these significant variants were original and which were not. Put differently, significant variants would be a problem if we could assume that every one of them was as equally viable as every other. The problem with such an assumption, however, is that it stands in direct contradiction to the entire history of textual criticism—indeed, to the very existence of the field itself—which has consistently maintained that not all textual variants are equally viable and that our methodology can determine (with a reasonable degree of certainty) which is the original text.44 If that is the case, then these few "significant" textual variants do not materially affect the integrity of the New Testament because, put simply, we can usually spot them when they occur. 

Examples of Significant Variants 

It may be helpful for us to review some examples of significant variants, though we can only scratch the surface of the issue here. For instance, in Mark 1:14 we are told that Jesus came preaching the "gospel of God." However, some fifth-century (and later) manuscripts—such as Codex Alexandrinus (A) and Codex Bezae (D)—read the "gospel of the kingdom of God." The cause for this slight change is obvious: the phrase "kingdom of God" is quite common throughout Mark (and the other Synoptic Gospels) and the scribe was likely harmonizing 1:14 with these other passages (a very common cause of scribal variations). Is there a difference in meaning between "gospel of God" and "gospel of the kingdom of God"? Perhaps. But the difference is hardly a cause for concern. And even if the difference were substantial, it matters little because the textual evidence is clear that Mark originally wrote "gospel of God."45 Mark 1:14 is a very typical example of a "significant" variant. 

However, there are other "significant" variants that have a more substantial impact on the meaning of a text. Two examples will suffice. One of the most commonly mentioned variants is found in 1 John 5:7-8 and is known as the Comma Johanneum.46 The italicized portion of the following verses is found in only a handful of manuscripts: "For there are three that testify: in heaven: the Father, the Word, and the Holy Spirit, and these three are one. And there are three that testify on earth: the Spirit and the water and the blood; and these three agree." Out of hundreds of Greek manuscripts, only eight contain this variant reading—and four of those have the variants added by the scribe into the margin—and the earliest of these is tenth century.47 Moreover, the variant is attested by none of the Greek fathers and is absent from almost all our early versions. In the end, despite the fact that this variant found its way into the Textus Receptus (and thereby the King James translation), the text-critical evidence is decidedly against it being original to John's epistle. What, then, do we make of this variant? No one can doubt that it is "significant" in that it affects the theological understanding of this verse. However, it simply has no claim to originality and therefore does not impact our ability to recover the original text of the New Testament.48 Nor is our understanding of the Trinity in the slightest dependent on this verse—indeed, the orthodox conception of the Trinity can be derived from many other New Testament verses and was well in place for centuries before this variation would have been widely known. 

A second example is Mark 16:9-20, known as the long ending of Mark.49 Most modern English translations bracket off this portion of the text and note that two of our earliest manuscripts of Mark, Codex Sinaiticus (א) and Vaticanus (B), do not contain the long ending. Moreover, the long ending was unknown in a number of early versions (including a number of Latin, Syriac, and Armenian manuscripts) and was not mentioned by prominent Greek fathers such as Clement of Alexandria and Origen. There is also the problem of non-Markan vocabulary in the long ending, as well as the awkward transition between 16:8 and 16:9. In short, most scholars agree that the long ending of Mark was not original to his Gospel. So, what is the impact of this particular variant? There is no doubt this textual change is "significant" both in regard to its scope (twelve verses) and also its content (resurrection, drinking poison, picking up snakes). But, since we can clearly see that these verses are an addition, they bear no impact on our ability to recover the original text of Mark. There may be residual questions regarding why Mark would end his Gospel in verse 8 (which we cannot enter into here), but the textual evidence is quite clear that he did not write verses 9-20.50 

Theologically Motivated Changes 

There has been a long-standing discussion in the world of textual criticism concerning the degree to which scribes intentionally altered passages of the New Testament to better conform to their own theological preferences. Ever since the well-known statement from Westcott and Hort that "there are no signs of deliberate falsification of the text for dogmatic purposes,"51 there has been a steady chorus of scholars intending to show the opposite to be the case. The idea of theologically motivated scribal changes can be traced back to Kirsopp Lake and J. Rendel Harris and more recently to scholars like Eldon J. Epp and his well-known book The Theological Tendency of Codex Cantabrigiensis in Acts.52 Ehrman joins this chorus in a number of his recent books, but most notably The Orthodox Corruption of Scripture, where he argues that scribes in the early church were not merely disinterested copyists who mechanically transmitted the text in front of them, but, in one sense, continued "writing" the New Testament text by changing it to adapt to the theological and social challenges of the day.53 Thus, argues Ehrman, these scribal changes need to be understood within the context of the early church battles over heresy and orthodoxy—battles that not only affected the development of the New Testament canon but affected the development of the New Testament text itself. 

Because these theologically motivated changes can affect the meaning of a passage (though just how much is in doubt), they are rightly considered to be "significant" textual variants. A few examples may be helpful. In Luke 2:33, after Simeon blesses the baby Jesus, we read, "And his father and his mother marveled at what was said about him." However, a number of later manuscripts read, "And Joseph and his mother marveled at what was said about him" (K X Δ Θ Π Ψ). Ehrman argues that this scribal change is designed to bolster the doctrine of the virgin birth—an issue that was often challenged by some heretical groups like the Ebionites—by making sure no one can (mis)use this passage to argue that Jesus had a human father.54 A second example comes from 1 Timothy 3:16 which, speaking of Christ, declares, "He was manifested in the flesh." However, other manuscripts show a scribal change which then makes the verse declare, "God was manifested in the flesh" (אe A2 C2 Dc K L P Ψ). Ehrman again argues that this scribal change was intentional and designed to state the divinity of Christ in more explicit terms.55 In the midst of all the Christological debates in early Christianity, scribes may have wanted to make sure this verse expressly affirmed that Christ was God come in the flesh. A third example is found in John 19:40 where Jesus' body is being prepared for burial. We are told there that "they took the body of Jesus and bound it in linen cloths." But the fifth-century codex Alexandrinus (A) reads, "So they took the body of God and bound it in linen cloths." This very obvious Christological change again appears to have been introduced for theological reasons—perhaps to keep Docetists from arguing that since Jesus was God he could not have had a real flesh-and-blood body.56 

How should we assess Ehrman's arguments with regard to intentional scribal changes? Let it be said at the outset that Ehrman's detailed textual work in The Orthodox Corruption of Scripture is where he is at his best. Overall, this is a very impressive monograph with much to offer the scholarly community in its assessment of the history of the New Testament text. Surely Ehrman's overall thesis is correct that, on occasion, scribes did change their manuscripts for theological reasons. That being said, there are two issues that need to be raised. First, although Ehrman is correct that some changes are theologically motivated, it seems he too quickly passes over equally (if not more) plausible explanations that are not nearly as provocative. For example, in 1 Timothy 3:16 above, the scribal switch to "God was manifested in the flesh" can be naturally explained by the fact that the word for "who" (OΣ) is very close to the abbreviation for "God" (ΘΣ). A simple scribal slip would easily turn one word into the other. However, Ehrman still maintains that the change was theologically motivated because four of the uncial witnesses (א A C D) show that OΣ ("who") was actually corrected by the scribe to read ΘΣ ("God")—meaning the scribe did it consciously. But the fact that these four scribes did it consciously is not the same as saying they did it for theological reasons. These are not the same thing. These scribes may have simply thought the prior scribe got it wrong; or maybe they simply corrected it according to what was in their exemplar. Moreover, a number of other majuscules have ΘΣ ("God") but not as part of a correction (K L P Ψ), so there is no indication that they did it intentionally. In the end, the explanation for the variant in 1 Timothy 3:16 is likely a very boring one. Simply a mistake. 

A second issue with Ehrman's work has to do with the overall conclusions that can be drawn from it. Let us assume for a moment that Ehrman is correct about the motivations of the scribes in every single example he offers—they all changed the text for theological reasons. But how does this change our understanding of the original text of the New Testament? What is the real payoff here in terms of assessing the New Testament's integrity? Not much. Ehrman's study may be helpful to assess scribal habits or the nature of theological debates in early Christianity, but it has very little effect on our recovery of the original text because in each of the instances he describes we can distinguish the original text from the scribal changes that have been made. In other words, even theologically motivated changes do not threaten the integrity of the text for the simple reason that our textcritical methodology allows us to spot them when they occur.57 

It is here that Ehrman finds himself in somewhat of a conundrum. On the one hand, in Misquoting Jesus he wants the "original" text of the New Testament to remain inaccessible and obscure, forcing him to argue that text-critical methodologies cannot really produce any certain conclusions. On the other hand, in The Orthodox Corruption of Scripture he needs to argue that text-critical methodologies are reliable and can show you what was original and what was not; otherwise he would not be able to demonstrate that changes have been made for theological reasons. Moisés Silva comments: 

There is hardly a page in [The Orthodox Corruption of Scripture] where Ehrman does not employ the concept of an original text. Indeed, without such a concept, and without the confidence that we can identify what the original text is, Ehrman's book is almost unimaginable, for every one of his examples depends on his ability to identify a particular reading as a scribal corruption.58 

The essence of Ehrman's argument, then, seems self-defeating. He is using theologically motivated scribal changes as a reason for why we cannot know the original text, but then he must assume we can know the original text in order to prove these scribal changes. Which one is it? In the end, it seems that Ehrman wants to be able to have his text-critical cake and eat it, too. 

Unfortunately, it seems the agenda in Misquoting Jesus is forcing Ehrman not only to deny the overall reliability of the field of textual criticism—the very field to which he has committed his life's work—but to deny even his own prior scholarly works. 

What, then, is driving these inconsistencies in Ehrman's text-critical approach? Inevitably, it goes back to his commitment to the Bauer thesis and, in particular, his application of the Bauer thesis to the field of textual criticism. Even though the field of textual criticism has historically argued that some variants really are more original than others, the Bauer thesis implies that, in one sense, all textual variants are inherently equal. After all, why should one form of the New Testament text be considered genuine and not another? Who is to say which text is right? Different Christians in different regions experienced different textual variants (and to them these variants were the word of God). It seems, then, that Ehrman is being pulled back and forth between these two competing positions—historical textual criticism that privileges one reading over another and the Bauer thesis, which suggests no reading can really be regarded as superior. The latter position seems to be prevailing when Ehrman declares, "It is by no means self-evident that [reconstructing the original text] ought to be the ultimate goal of the discipline . . . there may indeed be scant reason to privilege the ‘original' text over forms of the text that developed subsequently."59 

Thus, Ehrman's Bauer-driven approach to textual criticism is more radical than one might first realize. His claim is not simply that the battles over heresy and orthodoxy altered the original text, but he goes one step further to say that the battles over heresy and orthodoxy imply that there is no original text. Put differently, the Bauer hypothesis does not just explain the cause of textual variants, but it determines what our attitude should be towards textual variants. They are all equal. Once again, it is clear that Ehrman's conclusions are driven less by the discipline of textual criticism and more by his prior commitment to the Bauer thesis and the pluralistic nature of early Christianity. 

Thesis 4: The Impact of Unresolved Variants: the remaining number of truly unresolved variants is very few and not material to the story/ teaching of the New Testament 

The prior section has argued that even "significant" variants do not present a problem for the integrity of the New Testament because our text-critical methodology allows us to determine, with a reasonable degree of certainty, which is the original text. However, a very small number of significant variants remain where our methodology is not always able to reach a certain conclusion in either direction. In such a case, we may have two (or more) different readings and not know for sure which one is the original. Although these "unresolved" variants are quite rare, they are the only legitimate places where the New Testament text is genuinely in question, and therefore they need to be addressed. 

Examples of Unresolved Variants 

Needless to say, the question of what constitutes an "unresolved" variant is not always easy to answer (and cannot be fully resolved here). Certainly we cannot regard a variant as "unresolved" simply because there is some disagreement about its originality amongst scholars—after all, it seems that some sort of argument could be made for almost any variant reading if someone really wanted to try. Instead, we are talking here about a situation where there are two (or more) possible readings and the evidence for each reading (whether external or internal) is relatively equal, or at least close enough that it is reasonable to think that either reading could have been original. Again, a few examples may help. 

In Mark 3:32, the crowd sitting around Jesus said to him, "Your mother and your brothers are outside, seeking you." However, evidence from some other early Greek manuscripts (A D) and Old Latin, Old Syriac, and Gothic witnesses (combined with some strong internal considerations) suggest that the original may have been "Your mother and your brothers and your sisters are outside, seeking you." Even the editorial committee of the UBS Greek New Testament was divided on the question, which has prompted a number of English translations to include a footnote in this verse with the variant reading.60 Whichever way one decides, very little is at stake here. We know from other passages that Jesus had sisters (Matt. 13:56), and no doubt they would have been concerned about him along with the rest of the family. Another example, Mark 7:9, reads, "And he said to them, ‘You have a fine way of rejecting the commandment of God in order to establish (stesete) your tradition!'" But, a number of majuscules (א A K L Δ Π ), some of which are quite early, substitute "keep" (teresete) for the word "establish" (stesete). Given the similar spelling and similar meaning of these words, it is quite di"cult to determine which gave rise to which. However, either way, it leaves the meaning of the passage virtually unchanged. 

Both of the above examples are typical "unresolved" variants—not only are they very rare, but most of the time they affect the meaning of the text very little (and thus are relatively boring). But Ehrman has suggested that there are some other hard-to-solve variants that do impact the meaning of the text in a substantive manner. For example, Luke 22:43-44 describes the anguish of Jesus in the garden: "And there appeared to him an angel from heaven, strengthening him. And being in an agony he prayed more earnestly; and his sweat became like great drops of blood falling down to the ground." These verses are attested by a number of important witnesses including Justin Martyr, Irenaeus, Hippolytus, Eusebius, and other church fathers. However, these verses are also omitted by a number of important witnesses as well as Clement of Alexandria and Origen. Consequently, it is difficult to be sure whether the verses are original to Luke.61 The question, then, is whether either option raises a substantial problem or changes any biblical doctrine (Christological or otherwise). We know from other passages that Jesus felt great anguish in the garden of Gethsemane (Matt. 26:37-38; Mark 14:34), and that he was a real human being that could suffer temptation and sorrow (Heb. 2:17-18). Moreover, we have other accounts where angels attended Jesus in times of great need (Mark 1:13). These realities remain unchanged whether we include or omit this reading. Thus, either option seems to be consistent and compatible with what we know about Jesus and his ministry. 

Ehrman offers another example from Mark 1:41 ('()) where Jesus sees a leper and was "filled with compassion" (splagchnisthei). Though this reading has superior external support in its favor (א A B C K L W Δ Θ Π f1 f13), Codex Bezae (D) and a number of Old Latin witnesses declare that when Jesus saw the leper he was "filled with anger" (orgistheis). Although the external evidence is in favor of "filled with compassion," a number of internal considerations (e.g., which reading would the scribe have likely changed?) suggest that the original may have been "filled with anger." In short, it is difficult to know which reading is original.62 So, again, we ask whether either option raises a substantial problem or issue related to the teaching of the New Testament. Although "filled with anger" certainly changes our understanding of the passage—Jesus was likely expressing "righteous indignation at the ravages of sin"63 on the world, particularly the leper—this perspective on Jesus fits quite well with the rest of the book of Mark, where he shows his anger in 3:5 in a confrontation with the Pharisees and in 10:14 as he is indignant with his disciples. But it is also consistent with the Jesus of the other Gospels. Particularly noteworthy is John 11:33 where Jesus is faced with the plight of Lazarus, and the text tells us that he was "deeply moved" (enebrimesato), a term that can better be understood to mean Jesus felt "anger, outrage or indignation."64 Was Jesus angry at Lazarus? No, the context suggests that he was angered over the ravages of sin on the world, particularly as it affected Lazarus. In John 11:33, then, we have a vivid parallel to what might be happening in Mark 1:41—both are examples of Jesus showing anger toward the effects of sin in the midst of performing a miracle of healing and restoration. In the end, whichever reading in Mark 1:41 is original, neither is out of step with the Jesus of he New Testament. 

Unresolved Variants and Biblical Authority 

It is here that we come to the crux of the issue regarding biblical authority. Do we need to have absolute 100 percent certainty about every single textual variant for God to speak authoritatively in the Scriptures? Not at all. When we recognize not only how few unresolved variants exist but also how little they impact the overall story of the New Testament, then we can have confidence that the message of the New Testament has been sufficiently preserved for the church. All the teaching of the New Testament—whether regarding the person of Jesus (divinity and humanity), the work of Jesus (his life, death, and resurrection), the application of his work to the believer (justification, sanctification, glorification), or other doctrines—are left unaffected by the remaining unresolved textual variations.65 Belief in the inspiration of the original autographs does not require that every individual copy of the autographs be error-free. The question is simply whether the manuscript tradition as a whole is reliable enough to transmit the essential message of the New Testament. As we have seen above, the manuscript tradition is more than adequate. It is so very close to the originals that there is no material difference between what, say, Paul or John wrote and what we possess today. 

Of course, as we have seen above, Ehrman has taken a very different approach. For him, the quest for the original text is somewhat of an "all or nothing" endeavor. Either we know the wording of the original text with absolute certainty (meaning we have the autographs, or perfect copies of the autographs), or we can have no confidence at all in the wording of the original text.66  Unfortunately, this requirement of absolute certainty sets up a false dichotomy that is foreign to the study of history. As historians, we are not forced to choose between knowing everything or knowing nothing—there are degrees of assurance that can be attained even though some things are still unknown. This false dichotomy allows Ehrman to draw conclusions that are vastly out of proportion with the actual historical evidence. Although his overall historical claim is relatively indisputable (that the New Testament manuscripts are not perfect but contain a variety of scribal variations), his sweeping conclusions simply do not follow (that the text of the New Testament is unreliable and unknowable). We can have reliable manuscripts without having perfect manuscripts. But it is precisely this distinction that Ehrman's "all or nothing" methodology does not allow him to make. 

As a result, addressing the historical evidence (the nature and extent of textual variants) will not ultimately change Ehrman's conclusions about the New Testament. It will not change his conclusions because it is not the historical evidence that led to his conclusions in the first place. What, then, is driving Ehrman's conclusions? Ironically, they are being driven not by any historical consideration but by a theological one. At the end of Misquoting Jesus, Ehrman reveals the core theological premise behind his thinking: "If [God] really wanted people to have his actual words, surely he would have miraculously preserved those words, just as he miraculously inspired them in the first place."67 In other words, if God really inspired the New Testament there would be no scribal variations at all. It is his commitment to this belief—a theological belief—that is driving his entire approach to textual variants. Of course, this belief has manifold problems associated with it. Most fundamentally, one might ask, where does Ehrman get this theological conviction about what inspiration requires or does not require? How does he know what God would "surely" do if he inspired the New Testament? His approach certainly does not reflect the historical Christian positions on inspiration (except perhaps those in the King-James-Only camp).68 Instead, Ehrman seems to be working with an arbitrary and self-appointed definition of inspiration which, not surprisingly, just happens to set up a standard that could never really be met. Does inspiration really require that once the books of the Bible were written that God would miraculously guarantee that no one would ever write it down incorrectly? Are we really to believe that inspiration demands that no adult, no child, no scribe, no scholar— not anyone—would ever write down a passage of Scripture where a word was left out for the entire course of human history? Or is God prohibited by Ehrman from giving revelation until Gutenberg and the printing press? (But there are errors there, too.) 

It seems clear that Ehrman has investigated the New Testament documents with an a priori conviction that inspiration requires zero scribal variations—a standard that could never be met in the real historical world of the first century. Ironically, as much as Ehrman claims to be about real history, his private view of inspiration, by definition, prevents there from ever being a New Testament from God that would have anything to do with real history. Not surprisingly, therefore, Ehrman "concludes" that the New Testament could not be inspired. One wonders whether any other conclusion was even possible. 


Did the battles over heresy and orthodoxy in earliest Christianity affect the transmission of the New Testament text? Yes. No doubt a variety of scribal changes are due to these early theological disputes. But do these changes affect the text in such a way that we cannot be sure what it originally said? Not at all. Since the New Testament is a historical book that has been passed down to us through normal historical means (copying manuscripts by hand), then it inevitably contains the normal kinds of scribal variations that we would expect from any document of antiquity. No doubt some of these scribal variations were intentional and motivated by the theological debates of the day. However, the New Testament is different from most other ancient texts in a fundamental way: the wealth of manuscript evidence at our disposal (both in quantity and date) gives us good reasons to think that the original text has not been lost but has been preserved in the manuscript tradition as a whole. Given the fact that the vast number of textual variants is "insignificant," and given that our text-critical methodology can tell which "significant" readings are original and which are secondary, we can have confidence that the text we possess is, in essence, the text that was written in the first century. 

For footnotes from "Tampering with the Text," click here

EDITOR'S NOTE: Excerpt taken from The Heresy of Orthodoxy by Andreas J. Kostenberger and Michael J. Kruger (Crossway).

The Heresy of Orthodoxy
Copyright ©2010 Andreas J. Kostenberger and Michael J. Kruger
Published by Crossway Books 
A publishing ministry of Good News Publishers
1300 Crescent Street
Wheaton, Illinois 60187



Originally published July 27, 2010.