by Vern S. Poythress
[Published in Journal for Biblical Manhood and Womanhood 7/2 (2002) 21-30. Used with permission.]
Like earlier gender-neutral translations,1 Today’s New International Version (TNIV) consistently eliminates generic “he,” and by doing so changes meanings.
The discussion of gender-neutral translations in the book The Gender-Neutral Bible Controversy (GNBC) continues to be relevant and applicable to this translation.2 The TNIV, a revision of the New International Version of 1984, is definitely gender-neutral in its treatment of terms like “father,” “brother,” “son,” and “man,” as well as in its avoidance of generic “he.”
In this article I consider only the policy of avoiding generic “he.” “Generic ‘he'” describes the use of the masculine singular “he,” “his,” “him,” “himself” in the context of a general statement, typically a statement starting with “anyone,” “no one,” “everyone,” “whoever,” or “each.”3 We exclude general statements beginning with “he who,” because “he who” can legitimately be re-expressed with other phrases such as “the one who,” “anyone who,” or “whoever”4 But in most other occurrences, generic “he” can only be eliminated by more extended rewording, rewording that inevitably introduces meaning changes. The book GNBC devotes considerable space to the topic, because many factors and several kinds of arguments and counterarguments need discussion.5 I do not propose to repeat or review all the arguments at this time, but rather to inspect the details of how TNIV avoids generic “he.”
We gather verses together into five categories according to the route that they use.
1. Pluralizing
First, TNIV converts many statements from third-person singular to third-person plural.6
Consider Revelation 2:26-28.
NIV: To him who overcomes and does my will to the end, I will give authority over the nations–‘He will rule them with an iron scepter; he will dash them to pieces like pottery’–just as I have received authority from my Father. I will also give himthe morning star.
TNIV: To those who are victorious and do my will to the end, I will give authority over the nations–they ‘will rule them with an iron scepter and will dash them to pieces like pottery’–just as I have received authority from my Father. I will also give them the morning star.
In these verses TNIV converts all the third-person singulars (“he/him”) to third-person plurals (“they/them/those”). The meanings are fairly similar; but they are not completely identical. TNIV opens up a potential ambiguity between an individualizing and a corporate interpretation. In the individualizing interpretation, each individual victor rules. In the corporate interpretation, they exercise a single joint rule, with one “iron scepter” (singular). In the corporate interpretation they also jointly receive “the morning star,” a single gift to all of them together. The retention of singulars for “scepter” and “star” may push readers in the direction of a more corporate understanding.7
Because we are all united to Christ, who is the chief ruler, we can deduce that doubtless there is a corporate dimension to the rule of the saints. But that is the implication from other passages, not the explicit teaching of this passage.
Note also a more subtle effect. Even if we ignore the ambiguity, the use of the singular invites us to use as a starting point a sample case, “him,” from which we infer a general principle applicable to every case. TNIV starts with the generality, “those,” from which we infer applicability to any particular case. The directions of inferences are subtly different, and this is already a difference in meaning.8
Now consider 1 John 4:16.
NIV: … Whoever lives in love lives in God, and God in him.
TNIV: … Those who live in love live in God, and God in them.
The NIV again has a principle applicable to each individual. But the TNIV is ambiguous. It allows a corporate interpretation in which “living in God” and “God in them” refer to the totality of Christians together—the church lives in God and God in the church. This thought is theologically true (see 1 Cor. 3:10-15; 1 Thess. 1:1), but it is not the assertion of this verse, and the TNIV alters meaning by allowing it there.
Consider 1 John 3:3.
NIV: Everyone who has this hope in him purifies himself, just as he is pure.
TNIV: All who have this hope in them purify themselves, just as he is pure.
Again TNIV opens up a potential ambiguity between an individualizing and a corporate interpretation. In the individualizing interpretation, the purification takes place when each individual purifies himself. In the corporate interpretation, the purification takes place as each individual purifies everyone else as well as himself, or as each individual purifies the whole body through actions that help the body corporately. Corporate growth of the body of Christ is taught elsewhere (Eph. 4:11-16; 1 Cor. 12-14), so this latter interpretation is reasonable. But the singulars in Greek and in the NIV indicate unambiguously an individualizing interpretation (which, in the light of teachings elsewhere, will doubtless have some indirect corporate effects).
Consider 1 John 3:9.
NIV: No one who is born of God will continue to sin, because God’s seed remains in him; he cannot go on sinning, because he has been born of God.
TNIV: Those who are born of God will not continue to sin, because God’s seed remains in them; they cannot go on sinning, because they have been born of God.
Once again, TNIV converts to plurals (“those”), and thereby allows a corporate interpretation, namely that the body of people whom God has brought together through spiritual birth do not continue to sin, as a group, when looked at as a whole. But there may be some few exceptions that do not ruin the observation with respect to the general whole. By contrast, the singular (“no one,” “him,” etc.) is more explicit about disallowing exceptions. Naturally, this is not an all-or-nothing issue. We must allow that, for certain verses, the context may indicate that there are exceptions even in a case that is formulated using the singular. The point is that the singular is stronger in pushing one away from allowing exceptions.
Revelation 3:12.
NIV: Him who overcomes I will make a pillar in the temple of my God. Never again will he leave it. I will write on him the name of my God and the name of the city of my God, the new Jerusalem, which is coming down out of heaven from my God; and I will also write on him my new name.
TNIV: Those who are victorious I will make pillars in the temple of my God. Never again will they leave it. I will write on them the name of my God and the name of the city of my God, the new Jerusalem, which is coming down out of heaven from my God; and I will also write on them my new name.
TNIV again pluralizes and introduces the possibility of a corporate interpretation. Is the name of my God written on each person individually or once on the group as a whole? If one pictures literal pillars, one might most naturally think of the name being written on each one, but it is also possible that the name would be spelled across a group. In any case, the language of pillars is metaphorical, so it remains possible that in the reality to which the metaphor points, the name would be inscribed once on all of “them” as a group.
TNIV has converted to plurals every one of the seven promises to the “overcomers” (2:7, 11, 17, 26-28, 3: 5, 12, 21). The cumulative shift from single case to plural members of a group, over a considerable number of verses, is a fairly pronounced change.
Matthew 16:24-25.
NIV: … If anyone would come after me, he must deny himself and take up his cross and follow me. For whoever wants to save his life will lose it, but whoever loses his life for me will find it.
TNIV: … Those who want to be my disciples must deny themselves and take up their cross and follow me. For those who want to save their life will lose it, but those who lose their life for me will find it.
The retention of singular “cross” and singular “life” makes this saying ambiguous between a corporate and an individual interpretation.9 (The parallel passage in Luke 9:23-24 shows a similar problem.)
John 14:12.
NIV: … anyone who has faith in me will do what I have been doing. He will do even greater things than these, because I am going to the Father.
TNIV: … all who have faith in me will do the works I have been doing, and they will do even greater things than these, because I am going to the Father.
John 15:20
NIV: No servant is greater than his master.
TNIV: Servants are not greater than their masters.
TNIV will not allow a male element even in this parable-like analogy to Christian living.
1 John 5:12.
NIV: He who has the Son has life; he who does not have the Son of God does not have life.
TNIV: Those who have the Son have life; those who do not have the Son of God do not have life.
The NIV could have been changed by using “one who” or “anyone who” instead of “he who.” It is unclear why the TNIV pluralized the whole. Once again, the change opens up the possibility of a corporate interpretation, where all together have a life together. This is doubtless theologically true, but not the focus on the verse.
One could produce many more examples. As a sample of the frequency of this usage, I have found the following additional examples from the Gospels: Matt 10:10, 24, 25, 38, 39; 12:35; 13:12, 19, 21, 23, 57; 23:12; 25:29; Mark 2:22; 4:25; 6:4; 8:34; 13:13; Luke 4:24; 5:37; 6:40, 45, 47-48; 8:18; 10:7; 11:10; 12:21; 14:11, 27; 16:16; 17:33; 18:14; 19:26; John 3:20, 21; 4:14, 36, 44; 7:18; 11:9-10; 12:25, 35, 45, 47, 48; 13:10, 16; 15:15.
2. Change from third person (“he”) to second person (“you”)
On occasion TNIV substitutes “you” for generic “he.” The general principle may still be the same, or similar, but the starting point for illustration is no longer someone “out there,” but “you.”10
Luke 16:13.
NIV: No servant can serve two masters. Either he will hate the one and love the other, or he will be devoted to the one and despise the other. You cannot serve both God and Money.
TNIV: No one can be a slave to two masters. Either you will hate the one and love the other, or you will be devoted to the one and despise the other. You cannot be a slave to both God and Money.
The parallel in Matthew 6:24 has an analogous change.
The final sentence, “You cannot serve both God and Money,” is second person is Greek and in the NIV. So far so good. But TNIV shifts into second person in the second sentence, one sentence before the Greek changes to second person. This change might not seem to be too bad, until one realizes that the first two sentences are presented as generalities about life. They picture for the listener a slave “out there.” Only in the third does one hit the application, when one shifts from the generality “out there” to “you.” TNIV ruins the surprise punch of the third sentence by prematurely making the second already “you.”
TNIV also makes the second sentence awkward in another way. “Hate” and “love” are obviously hyperbolical when applied to a typical master-servant relation. The hyperbole can remain effective when applied to a servant “out there.” But it is more likely to seem extreme when it is directly describing “you.” The immediate reaction might be, “No, not me. That doesn’t describe my actual experience.” Again, the effectiveness of the whole saying is subtly damaged.
TNIV’s change is all the less justified because this verse is a kind of mini-parable. In Greek “no one” and “slave” are both masculine in gender. There is no reason why one should not think of a male slave as an example of a principle. When, in the third sentence, one leaves the realm of the parable and goes to the application to “you,” the “you” is obviously inclusive of both men and women. There is no need to tamper with it.
Luke 9:26.
NIV: If anyone is ashamed of me and my words, the Son of Man will be ashamed of him when he comes in his glory ….
TNIV: If any of you are ashamed of me and my words, the Son of Man will be ashamed of you when he comes in his glory ….
TNIV’s change from “anyone” to “any of you” runs the danger of restricting the range of “anyone.” Now it is no longer “anyone at all, throughout all ages,” but “anyone of you present to hear.” Perhaps one can still infer a broader application, but the broad sweep is not as unambiguous nor as directly and emphatically stated. (Mark 8:38 is similar.)
Matthew 16:26.
NIV: What good will it be for a man if he gains the whole world, yet forfeits his soul? Or what can a man give in exchange for his soul?
TNIV: What good will it be for you to gain the whole world, yet forfeit your soul? Or what can you give in exchange for your soul?
Mark 8:36 and Luke 9:25 are similar. (It would be allowable to use “a person” here instead of “a man” [Greek anthropos].11 The problematic change lies in the shift to second person [“you”].)
1 John 2:15.
NIV: … If anyone loves the world, the love of the Father is not in him.
TNIV: … If you love the world, love for the Father is not in you.
1 John 3:17.
NIV: If anyone has material possessions and sees his brother in need but has no pity on him, how can the love of God be in him?
TNIV: If any one of you has material possessions and sees a brother or sister in need but has no pity on them, how can the love of God be in you?
1 John 5:16.
NIV: If anyone sees his brother commit a sin that does not lead to death, he should pray and God will give him life. …
TNIV: If you see any brother or sister commit a sin that does not lead to death, you should pray and God will give them life. …
1 John 5:16b.
NIV: I am not saying that he should pray about that.
TNIV: I am not saying that you should pray about that.
In all the cases from 1 John “you” can refer to the immediate recipients of John’s letter. It is not as clear as before that the principle holds in general, not just for the recipients.
3. Change from third person (“he”) to first person (“we”)
1 John 4:20.
NIV: If anyone says, “I love God,” yet hates his brother, he is a liar. For anyone who does not love his brother, whom he has seen, cannot love God, whom he has not seen.
TNIV: If we say we love God yet hate a fellow believer, we are liars. For if we do not love a brother or sister whom we have seen, we cannot love God, whom we have not seen.
The third person, “anyone,” leaves open the possibility that, in the historical context of the letter, John is thinking mostly of people who belonged to a dissident group and had already separated themselves from the church (1 John 2:19). Changing to “we” in the TNIV suggests instead that the issue at hand is primarily one of hypocrisy among those whom John is directly addressing. There is a difference of meaning here, affecting how we see the situation that John is addressing.
4. Dropping generic “he”
In some passages TNIV tries simply to drop generic “he.” But this too can produce changes in meaning.
1 Corinthians 14:28.
NIV: If there is no interpreter, the speaker should keep quiet in the church and speak to himself and to God.
TNIV: If there is no interpreter, the speaker should keep quiet in the church and speak to God when alone.
The NIV is correct in translating it “to himself.” The phrase is parallel in structure to the phrase translated “to God.”12 TNIV’s expression “when alone” not only leaves out completely the idea of speaking to himself, but adds the idea of being alone, which is not there explicitly in the Greek. And it is not clearly implied either, since the person in question could speak in tongues quietly, mumbling under his breath while still in the church setting. Or he could speak in tongues out loud in a context of a small number of other Christians who were each praying out loud to God, and with none disturbing another. (I understand that in some cultures, more given to expressing all prayers out loud, the practice of simultaneous vocal prayer by many is common, even outside the context of tongues.) The operative concern for Paul seems to be in not disrupting the church gathering by trying to address it in tongues, not in a literal restriction to being off by oneself.13
1 John 3:15.
NIV: Anyone who hates his brother is a murderer, ….
TNIV: Anyone who hates a fellow believer is a murderer, ….
5. “They” with singular antecedent
Finally, TNIV uses “they” with a singular antecedent in order to avoid generic “he.” Let us consider an example, John 14:23:
NIV: … If anyone loves me, he will obey my teaching. My Father will love him, and we will come to him and make our home with him.
TNIV: … Anyone who loves me will obey my teaching. My Father will love them, and we will come to them and make our home with them.
In the second sentence, TNIV substitutes “them” for “him” three times.14 To whom does “them” refer? The context is set by the preceding sentence, which uses “anyone.” “Anyone” is grammatically singular. It invites us to start with a particular case (one person), but that case is an example of a general principle applying to a whole group, namely all human beings, and then that large group is narrowed down to “anyone who loves me.” The group is composed of a plurality of members, and “they” is sometimes used in contemporary English, as it has been for centuries, as the follow-up pronoun in such contexts.15 We shall call this usage “‘they’ with singular antecedent.”
How do we evaluate this kind of use of “they,” which is fairly frequent in TNIV? The question is complex, partly because different people may react differently to the same verse. A portion of the English-speaking public quite regularly uses “they” with singular antecedent, sometimes without realizing it. On the other hand, some people have heard from school grammarians that this usage is “wrong,” and consciously try to avoid it. A portion perceives “they” with singular antecedent as improper, perhaps because of the influence of school pronouncements. A portion would see it as out of place in formal written English, but be more tolerant of its appearance in informal conversation.16
The potential for misunderstanding rears its head, because some people may look for a plural antecedent to “they.” Others may interpret it as a fully plural usage, and conclude that the Father and the Son will make a single corporate home, “our home,” with “them,” that is, with all who love the Son. In that case the sentence is interpreted in a corporate sense, as having to do with “them” as members of the group to which God comes and which he loves, rather than as individuals.17
We can illustrate this possibility by imagining that Jesus had said something like this: “Anyone who loves me will obey my teaching. My Father will love them, and we will come to them and make our home with them, and they will share a new life together.” In this new statement, the wording “share … together” indicates unambiguously a corporate experience. A single person cannot “share” by himself. The word “they” must refer to Christians together, more than one at a time. It has plural reference, in spite of the earlier word “anyone,” which refers to a single sample person within the group. But this means that, even before a reader comes to the key extra expression “share … together,” he must allow for the possibility that the word “they” is referring to a plurality of members. And then one must also allow that this plurality of members might function together. And so the expression “our home with them” can mean a single “home” with them together, a corporate dwelling of the Father and the Son with “them”together, a dwelling of God in the church.
The use of “they” with singular antecedent does have one distinct advantage over all the other routes for avoiding generic “he.” In the usual case, at least, it makes it possible for a sophisticated reader to reconstruct accurately what the actual meaning is. Substitute generic “he/his/him” for each occurrence of “they/their/them,” and you have it! But of course that also raises a question. Why introduce the ambiguity of a corporate interpretation, when you can just use generic “he” and achieve your purpose immediately?! I know, I know, there are all kinds of concerns generated from ideological sources, and for those I must once again refer readers to the book GNBC.
Consider one more case, Revelation 3:20:
NIV: If anyone hears my voice and opens the door, I will come in and eat with him, and he with me.
TNIV: If anyone hears my voice and opens the door, I will come in and eat with them, and they with me.
In this case, there is a potential plural antecedent, namely “those whom I love” in 3:19. If “they” refers back to this plural antecedent, then Jesus is saying that if anyone–even one person–opens the door, Jesus will come in and eat with “those whom I love,” with the whole group of people in the Laodicean church, and by extension with any other church with similar problems. Such, of course, is not the meaning of the original.
One can hope that many people, because they remember this famous verse from other translations, will realize right away that this is not its meaning.18 Others will quickly realize that the most obvious antecedent is the more immediate one, namely “anyone.”
But even so, the potential for a corporate meaning does not disappear. Consider the following sentence: “If anyone hears my voice and opens the door, I will come in and eat with them, and they with me, and they will share a new life together.” The final occurrence of “they” has an unambiguously plural reference, and the other extra words speak unambiguously about a corporate experience. This result illustrates the fact that even the earlier words, “eat with them and they with me,” allow a corporate interpretation. Under this interpretation, all the people who fit into the class indicated by the preceding occurrence of “anyone” may together have the experience of a communal meal with the Lord. And indeed, the Lord’s Supper is just such a communal meal. As long as we do not add extra words like “share … together,” the meaning is not unambiguously corporate. But neither is it unambiguously individual. Readers are automatically open to the possibility of a corporate function for “they,” as the expression “they share” illustrates.19
Moreover, the thought of communal fellowship is a reasonable theological inference from teachings elsewhere about the Lord’s Supper and about Jesus’ fellowship with the church. But it is not directly the meaning of this verse, and opening up a corporate interpretation to this verse changes its meaning.
Evaluation
This collection of verses confirms what the book GBNC said two years ago. The techniques for avoiding generic “he” are roughly the same as in earlier gender-neutral translations, though with greater frequency for the use of “they” with singular antecedent. With respect to this last use, reactions may vary. For the other uses, changes in meaning nuances are regularly visible, though the translators tried to keep them small.
My judgment remains what it was in GNBC: the translators should have discarded the underlying policy of avoiding generic “he.” In translation, generic “he” is needed for maximal accuracy. The firm commitment to avoid it leads to unacceptable degradations of meaning. We lose a valuable resource in the English language, and with it a whole host of nuances in verse after verse.
The defenders of gender-neutral translation have a raft of replies, for which I must refer readers again to GNBC.20 The appearance of the TNIV has led to a number of new papers, but for the most part they take up themes already discussed in GNBC (see the response below).21 For now, let me be brief: generic “he” continues to be used in English in the secular press.22 For Bible translation, we need it. Then let us use it. I continue to believe what I wrote in 2000 in The Gender-Neutral Bible Controversy:
In fact, the “problem” with generic “he” is not with a single occurrence but with the pattern of thought in the Bible, a pattern that more often than not uses male examples as a starting point to express or illustrate truths that apply to both men and women. This pattern of thought a translator is not free to change or tone down in translation.23
An Addition: Response to Craig Blomberg and D. A. Carson
As we have seen, the TNIV changes person and number in order to avoid generic “he” (sections 1-3 above). In defense of the TNIV Craig Blomberg cites a number of cases where parallel passages in the Gospels differ in person or number (Matt. 5:3//Luke 6:20; Matt. 9:17//Mark 2:22; John 3:5//3:7), and passages where New Testament authors shift person or number in stating general truths (John 15:15; Rom. 4:7-8; 13:2; etc.).24 Much could be said about these and other examples. For the sake of brevity I note only the following points:
First, in each of the passages that Blomberg cites, he perceives no meaning difference. But I find the usual subtle meaning differences due to person and number, just as GNBC did (pp. 112-124). The passages therefore prove nothing.
Second, Gospel parallels are not Bible translations. If they did give us a model for what is permissible in translation, then a modern translation could freely interchange the wordings and meanings found in any of the parallels, producing a “gospel harmony” without differences between the Gospels. Every New Testament scholar would find this completely unacceptable. Blomberg, who is a New Testament scholar himself and is familiar with Gospel parallels, should see the fallacy in using these examples as if they provided a principle for translation.
Third, passages for which New Testament writers are the original authors are not translations. An original author is free to vary person and number. With the third person he makes a statement that directly focuses on the general case, while with the second person he makes a statement that directly focuses on the addressee(s). These complementary foci reinforce one another, rather than proving absolute identity of meaning. They therefore offer no principle by which to justify meaning alterations in translation.
Fourth, Blomberg needs not merely a few examples where (he alleges) person and number make no difference, but a general principle that will justify all the places where the TNIV makes changes. Such a general principle is clearly contradicted by any of the passages above that show subtle but demonstrable meaning changes.
In sum, Blomberg fallaciously uses Scriptural examples that are not about Bible translation in order to justify flawed translation.
Carson’s article
Now let us consider D. A. Carson’s recent article, “The Limits of Functional Equivalence.”25 Most of the article discusses the general subject of dynamic equivalent translation, and makes points many of which are compatible with GNBC 57-81. But when Carson comes to discuss gender, he misrepresents our position. We can only touch on the main problems.
First, Carson criticizes us for permitting some changes but not others: “They are making such changes … all the time, …. But when others make similar changes with respect to the pronoun ‘he,’ Poythress and Grudem condemn them for distorting the Word of God” (p. 22; see also p. 26 and p. 24n54). Carson’s word “similar” makes it sound as if we have no standards or are making arbitrary judgments in permitting some changes but not others. But this is not true. We already explain this issue in GNBC, chapter 5, where we talk about “Permissible Changes”; and the Colorado Springs Guidelines mapped out areas of permissible change.26 Roughly speaking, changes toward generic English are permissible when we are not losing a male meaning component in the original. And changes with respect to generic “he” are permissible when they do not produce significant meaning loss. (For instance, GNBC 111 permits “he who” being replaced by “anyone who” or “whoever,” because there is no significant meaning loss from the original.) Carson is of course free to disagree with where we draw the line. But instead he describes us as if we give no reasons.
Carson also paints a harsh picture of us by saying that we “condemn them,” that is, condemn the translators.27 That is not true. We most pointedly do not do so. GNBC repeatedly makes a distinction between the translators and the resulting translation:
We are not criticizing the personal motives of the translators. (GNBC 7)
We must be careful not to jump to conclusions about individuals. For convenience we have spoken of what translators do, but all we actually have is the product, the resulting translation. We know neither what was going on in translators’ minds nor the motives that underlay their thinking. ….
Thus, it is inappropriate to make this issue an occasion for personal attacks. We must beware of overreacting and firing ourselves with a zeal that “is not based on knowledge” (Rom. 10:2). (GNBC 293).
Carson overlooks these explicit statements in suggesting that we are attacking the translators.
Carson more than once complains about overheated rhetoric coming from people on our side. But this distortion of our position on his part is not only unfair–it is overheated, and may unwittingly encourage those on our side, if they believe Carson, to imitate the harshness that they mistakenly think we advocate, thereby further heating up the situation. Similarly, on p. 26, Carson says, “… their wrath knows few bounds when the TNIV deploys a plural instead of a singular.” Colorful, no doubt, but also unfair and dangerous.
Second, Carson has this to say:
I cannot help remarking, rather wryly, that in the light of the ESV, the argument of Poythress and Grudem sounds a bit like this: “The language is not changing, so we do not need to respond to the demands of inclusive language. But if it is changing, the changes are driven by a feminist agenda, so they are wrong and must be opposed if we are to be faithful to Scripture. Because of the changes, we will make some minor accommodations in our translations, but if others make any other changes, they are compromisers who introduce distortions and inaccuracies, and should be condemned, because changes aren’t necessary anyway!” (p. 24n54)
Carson here uses the unfair language “condemned” and “compromisers,” again falsely accusing us of attacking the translators personally. In addition the picture that he paints is totally off-base. Carson’s depiction repeats the errors of his earlier caricature of us in Debate, 183-184, which we pointedly refuted in GNBC 358- 360, and now he produces an even more distorted version, to which he adds our personal names!
Yes, it is intended to be humorous, but in the context of other misrepresentations the humor may manipulate readers. First it lowers the normal demands for fairness and evidence (“a bit like this”). Then by the effects of witty distortion it leaves a colorful, lasting impression that at bottom our position must be hypocritical and ridiculous. One wants to believe the picture because it is witty, not because it is true. And then, if objection is made, will we be told that it is “only a joke”?
Third, Carson says Poythress and Grudem “abuse their own theory by not admitting that basic translations really cannot frequently rise much beyond level 2.” But this statement misrepresents us. Far from “not admitting” translation limitations, in the very chapter 4 to which Carson refers, we explicitly discuss at some length the limitations of translations in conveying meaning (GNBC 58-81, especially 79-80). The four levels laid out in GNBC 82-90 are not levels for translations to achieve, as Carson’s wording here makes them out to be, but levels of analysis of meanings, whether the analysis is directed toward translations or toward other texts. (The whole section is entitled “Excursus: Analyzing linguistic complexity,” GNBC 82.) Our point is not that translators can achieve perfect representation of meaning, but that they should not be content with “basic meaning” in cases in which a fine-grained analysis shows that they can achieve more (see GNBC 189-190). All this is clear in GNBC, and Carson abuses our position by making it sound otherwise.
Fourth, Carson misunderstands my statement that Carson and Strauss “could not frankly discuss the ideological connotation of generic ‘he'” (Carson, p. 22). In context, I was obviously not saying that they would find it literally impossible to discuss (which would be a rather absurd claim), but that discussing it in any detail, and genuinely weighing the problems (as in GNBC 111-232, and especially 163-175), would weaken the case for gender-neutral translation, by removing some people’s impression that it is all a question of “neutral” stylistic preferences or of adjusting to “neutral” facts about the current state of English. Interestingly, in his latest response Carson still does not discuss the ideological connotation of generic “he”–in particular, the fact that ideology continuously maintains some people’s aversion to hearing it. Instead, he repeats generalities from his book about the influence of feminism, the reality of language change, and the lessening use of generic “he,” all points to which we have already responded in GNBC 355-366. He has shifted the issue instead of discussing it frankly as he claims (p. 22). Ironically, his continued avoidance of this one particular topic confirms rather than undermines the point in my review.
Fifth, Carson has missed the point of the quote from GNBC 202, which says, “The underlying assumption in this objection is that only what can easily be conveyed into all languages is worth conveying in English.” On the preceding page (GNBC 201) we introduce an objection that appears to us to have been made in Carson’s book and elsewhere: “Gender systems differ among languages. Therefore, you should not insist on mapping a masculine form in Hebrew onto a masculine in English.” We then explicitly indicate that we agree that gender systems do differ (201). We also indicate (202) that we are not talking about all masculine forms in Hebrew, but “a third-person-singular masculine pronoun used in a generic statement.” (But Carson in spite of this statement describes our position as perfectly general: “where we have the masculine pronoun in Hebrew,” p. 25.) With respect to this special kind of use, we offer considerable evidence that the meaning match between Hebrew, Greek, and English is generally very good (GNBC 335-47). Our argument does not rest on the mere assumption that one should always use formal equivalence, as Carson suggests (p. 25; see GNBC 190-91, 202n24, 61, etc.).
Carson’s 1998 discussion of gender in other languages in Debate, chapter 4, is relevant and helpful as a general illustration of the form-meaning contrast, which we ourselves recognize (GNBC 61, 85, 86n37, 190-91). But in and of itself it cannot settle the questions about how best to translation the meaning (not merely form) of masculine singular generics in Hebrew and Greek.
With respect to this narrow question, the discussion of other languages (outside Hebrew, Greek, Aramaic, and English) would have weight only if someone–not necessarily Carson personally–falsely assumed “that only what can easily be conveyed into all languages is worth conveying in English.” Without this assumption, the fact that a particular loss of meaning nuance is inevitable in translating into Polish does not lead to the conclusion that a similar loss in English is O.K., even though it is avoidable in English. (I do not think that we differ substantively with Carson on this point, only that Carson misunderstands us.)
Of course, with respect to generic “he” Carson might want to claim that there is a trade-off between different kinds of loss, because generic “he” is “offensive” on university campuses (see Carson, p. 21). But this is another topic, to which we respond in GNBC 163-175, 180-182.
Sixth, Carson says that he provides many examples pertaining to nonequivalence of gender systems, but “Poythress and Grudem tackle none of them” (p. 25). This is not true. We discuss Numbers 5:6 (from Carson, Debate 97) in GNBC 341-43, and Carson’s discussion on feminine subjects in Hebrew with masculine verbs (Debate 96) corresponds to GNBC 336 (but we mention singular subjects while he mentions plural). In addition, we explicitly indicate our agreement with the main point of Carson’s discussion of gender systems in GNBC 86n37. Carson’s word “tackle” suggests that we need to refute his examples. But this completely misses the point. The clear statement in GNBC p. 86n37 indicates that our view is completely compatible with his examples.
Carson repeatedly misunderstood and misrepresented us in his book Debate (for documented cases, see GNBC 77n22, 92n1, 94n3, 107-8, 130n30, etc.). He continues to do so in this latest article. I can only tell readers not to draw conclusions until they read what we say in GNBC.
2 Poythress and Grudem, GNBC.
3 GNBC 111-12.
4 GNBC 111.
5 GNBC, chapters 7-11.
6 See GNBC 112.
7 See GNBC 117-23.
8 GNBC 112-15. In fact, the ambiguities produced in this and other verses probably arise primarily from the initial difference in the starting perspective between the singular and the plural. If one starts with an individual case (“whoever, he”), the individualizing character of the application is plain. If one starts with a reference to a plurality of members of a group, it may remain unclear whether the statement applies to each member separately or to the members’ interaction with one another in a more corporate fashion. Even with a corporate interpretation, there is always some kind of application to each individual. But what kind? Does the individual receive the morning star himself, or is he part of the group that receives it as a group (the corporate interpretation)?
9 See GNBC 112-17.
10 See the discussion in GNBC 112-14, and the response to Craig Blomberg near the end of this article.
11 GNBC 354n1.
12 Mark Strauss (“Response to Vern Poythress,” Christianity Today [Oct. 7, 2002], 45) claims that “the Greek dative eauto in 1 Corinthians 14:28 probably means ‘by himself’ (= ‘when alone’) rather than to himself.” But this view must be rejected for several reasons: (1) the dative eauto (“to himself”) is obviously parallel to the dative to theo, “to God,” and the parallelism is reinforced by the word kai (“and”) linking them together. Thus both datives indicate the addressee, and are properly translated with the English word “to.” (2) The clause as a whole (verse 28b) does not fit together well under Strauss’s interpretation. One would have to translate, “by himself let him speak and to God,” or “by himself let him speak also to God,” or “by himself let him speak even to God,” all of which are awkward. The problem is to make sense of the word kai (“and”; sometimes “also” or “even”). In Greek it is clearly functioning to link the two datives (“to himself” and “to God”). But once Strauss reinterprets the first dative to mean “by himself,” it is hard to account for its presence. (3) “By himself” in a spatial sense is not a normal function of the dative in relation to a verb like laleito, “let him speak,” while the dative of addressee is a normal function (for example, “I have spoken to you” in John 6:63).
Strauss compounds his error by citing Robertson and Plummer’s and Fee’s commentaries as if they supported the meaning “by himself.” Actually, both commentaries explicitly contain the wording “to himself”! They then infer from this meaning that the speaker should wait until he is alone. The inference may or may not be correct, but in any case ought not to be pushed back into translation.
13 From Vern S. Poythress, “Systematic Pattern in TNIV,” Westminster Theological Journal 64/1 (2002) 187.
14 TNIV also changes the NIV by changing a conditional sentence, beginning with “if,” to a sentence with a restrictive relative clause, “who loves me.” The NIV is a more literal representation of the Greek. The change has the effect of eliminating the offending “he” in the second clause, “He will obey my teaching.” The meanings are very similar, but not identical. Using “anyone,” the NIV starts with a potential pool of examples as wide as humanity. It considers what may be true, if a sample person out of this wide pool is in fact found to love Christ. The TNIV starts with the pool already narrowed to consider only the people who do in fact already love Christ. The NIV is slightly more open-ended, in suggesting a look at people who are not yet Christians but might become so. The TNIV focuses more on those who are already committed. This kind of change occurs in a number of other verses as well, and could easily be made into a sixth category along side the five in the main text of this article.
15 See GNBC 216.
16 The convention distinguishing formal writing from informal conversation may not be as arbitrary as it sounds. Face-to-face communication normally decreases the potential for misunderstanding, because much collateral information is supplied from the situation. Written communication cannot rely on the collateral information, and must take greater care to head off misreadings of potentially ambiguous uses. In some contexts, where both a singular and a plural antecedent offer themselves, “they” is potentially ambiguous. The maxim to allow only plural antecedents helps to disambiguate, if both writer and reader abide by it. The use of generic singular “he” also helps head off the potential ambiguities of “corporate” interpretations.
17 See GNBC 117-118 for a further discussion of the problem of a corporate interpretation of plurals in John 14:23. The example in GNBC is from the NRSV, which begins with “those who …” rather than “anyone who …” (TNIV). “Anyone who …” is considerably more individualizing, but the danger of slipping into a corporate interpretation later on in the verse does not disappear (see the further discussion in this article).
18 It should go without saying that it is unwise for translators to rely on the clarity of other translations as a consolation and a protection, covering the lack of disambiguation in the one they are producing.
19 D. A. Carson thinks that the earlier word “anyone” guarantees individuality in this verse (“The Limits of Functional Equivalence in Bible Translation–And Other Limits, Too,” manuscript distributed by Zondervan to ETS and IBR members, [Sept. 4, 2002], p. 27). But he fails to realize that the later occurrences of “they/them” reintroduce the possibility of a corporately-oriented interpretation of the part in which they occur.
20 See GNBC, chapters 7-11 and appendix 3. Opposite to our viewpoint, for the most systematic defense of gender-neutral translation (though with some reservations here and there), see Carson, Debate, and Strauss, Distorting.
21 See also footnote 12 for a response to Mark Strauss.
22 GNBC 203-212.
23 GNBC 232. In some ways the single most significant datum confirming this conclusion is the oscillating use of generic “she” and generic “he” in some authors (GNBC 150, 170, 362). In these contexts, generic “he” is seen by feminists as acceptable, because it is accompanied by an equal weight of generic “she.” This oscillating use also confirms the fact that, with both generic “he” and generic “she,” readers easily understand that a general principle is being articulated that applies to both male and female.
On the question of whether the Greek and Hebrew generic masculine forms suggest a male example, see the discussion in GNBC 142-146, 335-347.
24 Craig Blomberg, “Today’s New International Version: The Untold Story of a Good Translation,” manuscript distributed by Zondervan to ETS and IBR members, [Sept. 4, 2002].
25 See note 18.
26 There is even more support. It is not widely realized that in 1989 the Lutheran Church–Missouri Synod requested one of its commissions to study inclusive language issues. Nine years later the Commission produced the report “Biblical Revelation and Inclusive Language,” A Report of the Commission on Theology and Church Relations of The Lutheran Church–Missouri Synod (February 1998). The report is compatible with the Colorado Springs Guidelines in all the areas on which both speak, but it appears to have been produced independently. The report does quote once from an article by Wayne Grudem, from October 27, 1997 (p. 31). But there are no direct signs of interaction with the Colorado Springs Guidelines. In any case, most of the eight or nine years’ work by the Commission was presumably done before the Guidelines appeared publicly on June 3, 1997. The Commission’s report thus represents an independent witness to the fact that our principles have logical coherence, and do not arise merely from arbitrary personal whims.
27 In the quote from p. 22, reproduced above, “them” might possibly refer to “changes,” but is most naturally taken as referring to “others,” that is, the translators.