BeDuhn, Jason. Truth
in Translation: Accuracy and Bias in English Translations of the New Testament.
For an exchange between the reviewer and Jason BeDuhn indirectly related to this book review, click here .
Professor Jason BeDuhn, chair of
the religion department at the
Now, this claim is almost automatically offensive to the sense and sensibilities of nearly all biblical scholars, conservative and otherwise, who view the New World Translation [NWT] as a decidedly unscholarly production tendentiously prepared to support the doctrines of the Jehovah’s Witnesses, so that the NWT is itself a rather egregious example of a biased translation. BeDuhn includes quite a bit of helpful analysis and information throughout the course of his treatment, but does he in fact prove his overall thesis? I hope to demonstrate in this review that he does not, and that his treatment of the evidence itself, despite his claim to be a “neutral investigator” (Preface, p. ix, cf. his comments on p. 167-68) is itself heavily biased. In so contradicting nearly the entire scholarly consensus, BeDuhn has a heavy burden of proof, and he has failed to meet that burden. Since this work, like that of Rolf Furuli and (to a lesser extent) Bart Erhman, will be used extensively by the Jehovah’s Witnesses and perhaps other Arian type cults to support their theological agenda, this review will be somewhat longer than the average in order effectively to engage a good sample of his various points.
BeDuhn begins his work in the introduction
by summarizing the linguistic and textual history of the NT documents,
sufficient to demonstrate why the modern reader needs a translation in the
first place (although I don’t remember him specifically so stating, his work
appears to be pitched to the educated laymen, not biblical scholars
specifically). He gives a fair summary
of what requirements translator must meet in order successfully to engage in
the work of translation. In chapter 1,
“Origins of Modern English Bibles” (a very cursory treatment, but adequate to
his purpose), he makes the claim that “Bible translation is usually undertaken
by people with theological training who
also happen to be reasonably
competent in biblical languages” (p. 8, emphasis mine) and “Although
biblical scholars have been the key players in identifying the more accurate
Greek text of the New Testament, most have never been involved in a Bible
translation project” (p. 9). Now, I
hasten to point out that BeDuhn provides absolutely no evidence to support
these rather naked ad hominem claims.
Most modern major Bible publishers provide a list of the translators. It would have been a simple matter for BeDuhn
to go down the lists and check qualifications.
Had he done so, he would have found that many, in fact the majority, of
the primary translators had advanced degrees qualifying them for their
work. To wax anecdotal, I personally
know several translators who have worked on both the NIV and the more recent
ESV: all of them have Ph.D.’s from institutions such as Harvard or
The ad hominem can cut both ways. I did some research on BeDuhn, and I note that while he is certainly an accredited scholar, his Vitae is not what I expected to support the book he has written. He is essentially a history of religions scholar with particular emphasis in Manichaeism, and I was able to uncover one article particularly related to NT exegesis. While it is not impossible for a scholar to produce good work outside his normal field of expertise, this Vitae does not automatically inspire confidence that BeDuhn will have the breadth of experience in NT studies that such a work invariably needs. The rest of his book tends to support my initial impression.
BeDuhn then goes on to give an
adequate summary of the difference between formal correspondence and dynamic
equivalence theory, pointing out the difference in translations as one of
degree rather than kind (he includes the NIV as a formal correspondence
translation with various dynamic equivalency moments, p. 32). I appreciate his comments on the Living Bible
and the TEV: though all translations involve interpretation, these two often
cross the line from translation into theological commentary, and truly are
biased in the accepted sense of the term.
However, his discussion of the various translations he intends to use
for comparison are not always sound. I
was quite surprised at his comments on the NKJV. He seems unaware that the choice of the
textual basis was a deliberate philosophical choice on the part of the
translators. He fails to comment on the
“total equivalency” theory advanced by the translators in their introduction,
nor does he mention that they several times mark important textual variants,
such as at 1 John 5:7, with complete awareness of the text critical issues. While he may be correct not to include the
NKJV in his comparisons, seeing it as simply a cosmetic makeover for the KJV
(though I think it is more than that), he simply does not accurately report the
philosophical approach employed in the translation.
His comments on the NWT reveal something of the bias of the author himself. Amazingly, he discusses far less of the history and controversy surrounding the translation than he does his other sample translations. Edmund C. Gruss, in his We Left the Jehovah’s Witnesses, presented eyewitness and documentary evidence that the alleged translation committee of the NWT had very little in the way of the qualifications for translation work which Dr. BeDuhn has helpfully outlined earlier in his work.[[1]] While these claims could conceivably be disputed, they should not be ignored. BeDuhn points out that the translation committee was kept anonymous, though he does not mention why, and compares this to the NKJV committee and the Lockman foundation, apparently unaware that the list of translators for the NKJV is available from Thomas Nelson upon request, and that the Lockman Foundation has made their list of translators public.[[2]] He dismisses claims of bias concerning the NWT with the observation that all Bible translations are biased (quite a non sequitur) but then implies that the NWT is less biased because “[it] is free of the shadow of the King James” (this in the context of explaining that although the translation philosophy is similar to the NRSV, it reads quite differently from the KJV-dependent NRSV). In other words, the purported theological independence of the NWT makes it a better translation. I mention this last point particularly because several times in his work he simply asserts that the various translations are following the KJV example, as though the translators were unable or unwilling to form independent judgments on the meaning of the texts. He does not prove this, but simply assumes it, and in effect, uses the differences of the NWT to support his point, and therefore assumes as evidence what needs to be proved. Of course, apart from real evidence, it is possible that the resemblance to the KJV may mean that the translators have simply gotten it right.
In all fairness to Professor
BeDuhn, he does attempt to present a good amount of evidence to support his
claims, using the Greek as his reference point, but the evidence that he
presents is often inadequate or poorly handled, creating at times the
appearance of misrepresentation. Allow
me to present an analysis of two major discussions in his work, his chapter (11)
on John 1:1, especially his justification of the “a god” rendering of the
anarthrous θεος
in 1:1c, and his comments on the personhood of the Spirit in chapter 12.
Since John 1:1c is translated “and the Word was God” in nearly all major translations, and since a tremendous amount of exegetical effort has been expended on this passage, Professor BeDuhn has a great deal of work cut out for him to defend the rendering “a god” found in the NWT. Unfortunately, BeDuhn does not provide nearly enough support properly to engage the subject. I was extremely surprised that he did not interact significantly with the extensive secondary literature on the subject. Admittedly, a popular work is not going to have the same level of interaction that we would expect from a truly scholarly monograph, but BeDuhn certainly could have demonstrated more familiarity with the major works on the subject, and would also have avoided making certain mistakes in his evidence and argumentation.
Instead, as is typical of his methodology throughout, he begins by discussing the grammatical principle that he feels to be most pertinent, in this case the nature of the article. The gist of his description on p. 114-116 is that the presence of the article means that the noun is definite; the absence of the article means that it is indefinite, and normally should be translated with English indefinite article “a/an.” This is quite an oversimplification, and ignores the well known fact that the article and its lack does not always, in all contextual circumstances, equate to the definite article in English or the indefinite. On page 115-116 he states:
If John had wanted to say “the word was God,” as so many English translations have it, he could easily have done so by simply adding the definite article “the” (ho) to the word god (theos), making it “the god” and therefore “God.”
Many exegetes have argued, however, that the presence of the article before θεος in 1:1c would actually identify the term with τον θεον in 1:1b, hence stating that the λογος is actually God the Father. This would mean that John was advancing a form of modalism, something which the context clearly contraindicates. As Harner (whom BeDuhn cites extensively, see below) argued in his essay on the subject (interacting with Colwell’s Rule), the lack of the article is irrelevant to the definite/indefinite nature of the θεος in 1:1c, since it is intended to emphasize the qualitative nature of the noun, and is an assertion that the λογος shares in the same quality of godhood that is possessed by God the Father. This is accentuated, in my thinking, by the simple but effective chiasm that we see between 1:1b and c:
και ο
λογος ην προς
τον θεον,
και θεος
ην ο λογος.
In a careful author such as John, such a construction can hardly be accidental, and emphasizes in a particular way the key nouns in the chiastic arrangement, strengthening the impression that it is the divine nature of the Logos that is being highlighted without identifying the Logos directly with God the Father. Now, this interpretation is itself subject to debate, but my point is that while these arguments are well known in the secondary literature, BeDuhn doesn’t even mention them. In fact, he does very little real exegesis, and seems unaware of how more complicated syntactical structure and contextual nuance might overturn his rather simplistic use of grammatical principles.
BeDuhn finishes the quote given above:
He could have simply written ho logos en ho theos…or ho logos ho theos en…. But he didn’t. If John didn’t why do the translators?
He then goes on to blame the KJV translators, arguing that since they normally used the Latin Vulgate, and since Latin has no article either definite or indefinite, they were therefore used to seeing the noun without the article, and that this carried over into their English translation. To say that I find this claim incredible would be an understatement. In the first place, whatever else one might find to say about the KJV, the NT translators show a very high level of competence in correctly translating the article throughout. They knew Greek as well as any other scholar of their period. Why would they only get it wrong here? BeDuhn could have made the argument that the KJV translators were following the earlier English versions, but he doesn’t even mention that (nor would the claim be particularly true, as the KJV translators felt quite free to depart from the earlier English versions if they felt justified in doing so). Secondly, BeDuhn has just accused the translators of the various modern versions of slavishly following the KJV, as if they have no ability independently to read the Greek text for themselves. If good exegetical support existed for rendering 1:1c as “the word was a god” surely at least some of the major modern versions would have so rendered it, but in fact, they have good exegetical evidence to avoid that translation.
In order to prove that the Bible talks about “a god” in the sense that the NWT intends here (that Jesus is a mighty being but not God Almighty), BeDuhn supports his case by citing various verses where he thinks the anarthrous usage of θεος indicates an indefinite usage. However, the verses selected are highly problematic. BeDuhn has neglected the fact that θεος, when referring to God, is practically a name or proper title. While it often happens that the article might be used in such contexts, the use of the article to mark a proper name or title is optional, and is often varied for stylistic or other reasons. In every case that he cites (Luke 20:38; Mark 12:27; 2 Cor 1:3; Rev 21:7; Phil 2:13; 2 Thess 2:4) the context clearly indicates that it is God who is in mind, and substituting “God” or even “the God” does not at all violate the sense. Even in 2 Cor 1:3, the translation “a God” works, because the idea is descriptive of what type of God he in fact is, without necessarily suggesting that other gods have equal validity.
BeDuhn does discuss Colwell’s rule, and correctly points out its limitations (in my opinion, Colwell’s rule has long focused interpreters too narrowly on one grammatical aspect of the passage, at the expense of broader structural and contextual exegesis). When he comes to Harner, however, he selectively cites him in order to support his own spin, but seems to ignore Harner’s actual conclusions. On page 129:
I am in basic agreement with Harner that theos in John 1:1[c] is used qualitatively. I think the best translation would be: “And the Word was divine….” John is trying to stress that the word has a divine character, or belongs to the class of divine things, however that is to be worked out technically.
However, this is how Harner saw his rule of the anarthrous qualitative predicate being applied here:
Perhaps the clause could be translated, “the Word had the same nature as God.” This would be one way of representing John’s thought, which is, as I understand it, that ho logos no less than ho theos had the nature of theos.[[3]]
Finally in this section, it is not difficult to quibble with BeDuhn’s idea that the Word is included in a class of “divine things” as though he were one among many. John consistently uses the term qeoj of the one true God throughout his prologue. To suggest that a 1st century Jewish-Christian monotheist would begin to speak of the existence of other gods as real stretches credulity to the breaking point. For John, θεος is in a class by himself, and it is in that unique class, so to speak, that the Logos shares.
In chapter 12, “The Spirit Writ Large,” BeDuhn continues this discussion and applies it to the Holy Spirit. He argues on p. 136-7 that the readers of the NT would draw their understanding of “spirit” (πνευμα) from the larger cultural context, which had a much wider range of usage for the word. He contends that because “we” have a much narrower range of usage, “we tend to run together in our mind the distinct things called “spirit” in the New Testament.” Due to centuries of Christian theological reflection, “modern readers and translators…think of the Holy Spirit as a “who,” or even a “he,” rather than as an “it” that transcends human measures of personhood.”
First of all, I think it is a foundational hermeneutical mistake to look first to the broader cultural context to determine the usage of NT Greek words. While the authors did use normal language of the period to communicate their messages, they did so with a keen and predominant awareness of previous canonical communication, and we should look primarily to this context. Certainly we do not ignore the usage of vocabulary in the non-canonical and general literary context, but this consideration needs to be made secondary, or we may easily misread the text. In more popular terms, asking what a particular vocabulary item may have meant to the general pagan reader would have quite possibly given a distorted picture of the meaning, much as giving an out of context Bible verse today to someone who has never read the Bible and asking her to comment would very likely not result in capturing the author’s intended meaning. What the word meant in context to someone familiar with the total canonical literary-community context, i.e., the “biblically educated” reader, might be quite different than what the word meant to someone familiar with its usage in quite a different literary-community context.
BeDuhn then proceeds to support
his contentions based once again on certain principles of Greek grammar and an
examination of the passages which he thinks proves his arguments. Once again, his support fails to satisfy the
burden of proof. He notes that there are
87 usages of the term “holy spirit,” about half of which are anarthrous. He correctly reports that the lack of the
article in all contexts does not mean that the term is indefinite, particularly
in prepositional phrases (citing Smyth 1128).
He fails to note that “Holy Spirit” is tantamount to a technical term, name
or title, and hence the usage of the article is going to be optional even when
not employed in a prepositional phrase.
As it is, he uncovers 7 verses where he thinks it is fine to render “a
holy spirit” (Acts
An examination of each of the remaining citations will yield similar results. This does raise the question of why these particular usages are anarthrous. Since 6 of the 7 usages are Lucan, this might be attributed to some peculiarity of Luke’s style. Since, however, each of these is in the context of the giving of the Holy Spirit or the empowering thereof, I think it is more likely that the lack of the article subtly emphasizes the Holy Spirit as power of God which in some way qualifies the recipients in their relationship with God (this would also fit the one Johannine citation). This emphasis, however, certainly does not mean the recipients are being given a different holy spirit, nor does it suggest that the Holy Spirit is only a principle of power, since other Scriptures make plain the personal nature of the Holy Spirit. Regardless of the precise construction we wish to derive from these examples, it still clear that BeDuhn has not adequately dealt with all the exegetical and contextual details.
We now move on to BeDuhn’s discussion about the appropriate pronouns to use in reference to the Spirit, and what it is that might motivate translators to use the personal (masculine) pronouns in English, when πνευμα is clearly a neuter word.
BeDuhn correctly notes that Greek words have grammatical gender, masculine, feminine, and neuter. He further observes that English nouns do not really have the same quality, but are either personal or impersonal. Personal nouns tend to take pronouns which are gender qualified, since persons are either male or female. Impersonal objects take the pronoun “it” which emphasizes the non-personal nature of the object. BeDuhn further clarifies that in Greek many nouns which are masculine or feminine in grammatical gender are still impersonal. It is the meaning of the noun which determines what pronoun it takes in English translation, not the grammatical gender. I have more than once been amused at the result when a beginning Greek (or Latin, since the same rule applies) student waxes a bit too literal in rendering the pronouns.
However, BeDuhn makes a grievous error when he asserts “But “neuter” nouns are used only for impersonal things…” (p. 140). There is no soft way to say this: the claim is simply erroneous. Even in the NT itself, the words τεκνον and παιδιον are personal neuter nouns. In such cases, it is optional whether the neuter pronoun or the masculine/feminine is used to refer to the noun, although at least in the NT the preponderance seems to be for the grammatically neuter (cf. Smyth 1013). Is it correct, then, to refer to a child with the pronoun “it” in English? Normally not, since we tend to think of even small children as persons, and therefore gender qualified (especially so in modern English: the pronoun “it” in reference to an infant is really an obsolete usage).
Now, in a sense, BeDuhn is correct: using the personal pronoun in English to refer to the Spirit is a theologically biased translation, but it is a correct theologically biased translation, one that is in accord with the various descriptions in the NT concerning the Spirit. Throughout the NT, we see numerous references which indicate the personhood of the Holy Spirit. If we translate ad sensum rather than according to literal grammatical gender, then we must use a personal pronoun in English. Interestingly enough, the KJV several times uses “which” and “it” in reference to the “Holy Ghost,” of which BeDuhn approves, but he fails to note that this weakens his usual argument that the modern versions are heavily KJV influenced.
BeDuhn goes on to discuss other usages of the word “spirit” which do not involve the Holy Spirit. In some cases, there is real ambiguity whether or not the Holy Spirit is in view or whether the term is being used in another sense, but the majority of his examples in that section are valid. With some variation, most of the modern versions translate these passages correctly.
In his various chapters, I found many other claims and arguments which could be contested in similar fashion to the examples above. Throughout his book, BeDuhn’s writing is a mixture of some helpful information combined with fallacious arguments and inadequate treatments of various texts. These are not easy to untangle, and especially so for the one who has not been trained in biblical studies and the original languages. In this reviewer’s opinion, the weaknesses of the monograph sufficiently outweigh the strengths so as to render the work essentially worthless for the purpose intended. There are much better texts available on the subject of biblical translation, and interested readers would find it more profitable to consult them instead.
N.E. Barry Hofstetter
The Center for Urban Theological Studies
[1]
[2] Listed
on the Web at the Lockman Foundation’s website, http://www.gospelcom.net/lockman/nasb/nasbtrans.php
[3] Philip
B. Harner, “Qualitative Anarthrous Predicate Nouns: Mark