by Vern S. Poythress
[Published in the Westminster Theological Journal 64/1 (2002) 185-92. Used with permission.]
The central problem with Today’s New International Version (TNIV)1 does not lie in this or that verse that has been translated in less than an ideal way. It lies in a pattern, a systematic policy, namely that it avoids using a male representative or example to communicate a general truth.2
We first consider four examples, then stand back to discuss the pattern.
Examples of meaning changes
First, consider 1 John 2:10.
NIV: Whoever loves his brother lives in the light, and there is nothing in him to make him stumble.
TNIV: Those who love their fellow believers live in the light, and there is nothing in them to make them stumble.
TNIV changes “his brother” to “their fellow believers,” avoiding the male-marking of both “his” and “brother.” “Fellow believers” loses some of the intimacy implied by “brother.” “Brother” indicates a family-like relation, implying family-like responsibilities to the other person. “Fellow believers” indicates only a common belief within a larger group. In addition, it replaces the family idea of brotherhood with the idea of belief, which does not occur in the original. “His brother” takes its starting point in a single case, which the reader is then to generalize, so that in the end the generalization includes both men and women. In this sense, either wording includes the same people within the scope of the principle, but the meanings by which one achieves the inclusion are distinct and different.3
TNIV also changes the singulars “whoever,” “his,” “him” to plurals (“those,” “their,” “them”). First John 2:10 sets forth a general, or “generic” truth, as the word “whoever” makes clear. But in the NIV subsequent references to “whoever” take the form of the masculine pronoun “his/him.” The masculine may suggest that a male sample case is illustrating the general principle, but it leaves intact the inclusive scope of “whoever.”4 (This kind of use of “he/his/him” is called “generic he.”)
By substituting plurals, TNIV avoids the male overtones of the masculine “his/him.” The general principle expressed may be similar in the two wordings, but the starting point for expressing it is different. If we use the singular, it is as if we start with a particular case, “whoever,” “he,” where the masculine pronoun hints that we think of the representative case as male.5 We then move from the representative case to observe that any one case stands for a principle applicable to a whole group of cases, including both men and women. On the other hand, if we use the plural (“those,” “they”), we start by designating all the members of the group, and we may move from there to conclusions about any one member of the group. The direction of inference goes either from one case to the generality, or from the generality of members to any one member. The difference between the two is subtle, yet real.6
In some cases, the change to plural also introduces an ambiguity between an individualizing and a corporate interpretation. If the singular occurs, we know unambiguously that the principle applies to each person, because the starting point is a single case. If the plural occurs, we are free to infer that the principle applies to each individual. But we might also infer that the verse is focusing more on the corporate life of believers together. Christian together show by their love for one another that they live in the light, and this corporate behavior serves as proof to the world, in a manner analogous to what Jesus says in John 13:35, “By this all people will know that you are my disciples, if you have love for one another.” The pluralizing of 1 John 2:10 slants the verse more towards thoughts of the members as a whole than does the singular. Again, this is a change in meaning.
Next, consider 1 Corinthians 14:28.7
NIV: If there is no interpreter, the speaker should keep quiet in the church
TNIV: If there is no interpreter, the speaker should keep quiet in the church
NIV: and speak to himself and to God.
TNIV: and speak to God when alone.
The NIV is correct in translating it “to himself.” In Greek the phrase in question is parallel in structure to the phrase translated “to God.” If there is no interpreter, Paul forbids addressing the church in tongues, because this would distract rather than edify (see 1 Cor 14:9-11, 16-17). Instead of addressing the church, one should speak “to himself and to God.” This instruction is clear enough.
But TNIV substitutes “when alone” for “to himself.” The 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, as it were, mumble under his breath while still in the church setting. Or he could pray out loud, not to address the church while everyone else is listening, but 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 their prayers out loud, the practice of simultaneous vocal prayer is common, even outside the context of tongues.) The operative issue for Paul does not concern whether one is literally alone, but whether one disrupts the church gathering by trying to address it in tongues.
But suppose, for the sake of argument, that we assume that the two expressions have similiar practical implications. They still differ in meaning. The NIV and the Greek achieve their meaning by specifying the addressee of the utterance. The TNIV does so by specifying the circumstances of the utterance, which is a different meaning.
Why should TNIV have undertaken this unfortunate alteration? The clue lies in the word “himself,” which is masculine. The general principle expressed in the verse undoubtedly applies to both men and women. And the phrase “the speaker” indicates this generality. It does not single out either a male or a female individual. The subsequent “himself” refers back to “the speaker,” and therefore retains the earlier generality of “the speaker.” At the same time, it may suggest to many readers that the individual case that we choose to represent the general principle is a male example. We have a male example representing a general truth.8 TNIV as a matter of systematic policy avoids such use of generic “he.”
Next, consider Hebrews 12:7b.
NIV: For what son is not disciplined by his father?
TNIV: For what children are not disciplined by their parents?
TNIV changes “son” to “children” and “father” to “parents,” removing male meaning components in both instances. It also changes the whole sentence to plurals, thereby avoiding the masculinity of “his.”
Are these moves are justified by the underlying Greek? The key Greek words for “son” (huios) and “father” (pater) are both singular. While the word translated “father” may sometimes have the meaning “parents” when it occurs in the plural (see Heb 11:23), it regularly has the meaning “father” (not “parent”) when it occurs in the singular in the context of ordinary nuclear family relations. Similarly, the word translated “son” regularly has the meaning “son” (not “child”) when it occurs in the singular in the context of ordinary family relations.9 The relation of discipline between a father and son illustrates the relation of God to his people. Doubtless the relation might be illustrated by merely talking generally about parents and children, rather than father and son, but Hebrews 12:7b (in Greek) has not chosen this way of doing it. TNIV expresses a general principle of discipline, but eliminates any suggestion that male examples are used to illustrate it.
Finally, consider Revelation 22:18.>
NIV: … I warn everyone who hears the words of the prophecy of this book:
TNIV: … I warn everyone who hears the words of the prophecy of this scroll:
NIV: If anyone adds anything to them, God will add to him the plagues described in this book.
TNIV: If any one of you adds anything to them, God will add to you the plagues described in this scroll.
The next verse, Revelation 22:19, exhibits an analogous change.
The NIV’s expression “anyone” becomes “any one of you” in TNIV, and the subsequent “him” becomes “you.” The changes introduce an ambiguity in meaning. What is the reference of “you” in “God will add to you …”? In informal writing, “you” could have the same referent as “any one of you,” which would mean that the plagues fall on whatever individual adds anything to the words. But “you” could also refer to the same people as the earlier occurrence of “you,” that is, “everyone who hears.” The implication would then be that if any one person among you adds something, all of you will receive the plagues. In more formal writing, one generally expects “you” to keep pace with the earlier occurrence of “you,” so that this second meaning actually becomes the preferred reading. Needless to say, this reading does not represent the force of the Greek original.
One might still hope that people will suspect that this spurious meaning is wrong. It seems “unfair” that a penalty should fall on everyone for the sin of only one person. But this kind of corporate punishment has a precedent in the case of Achan (Josh 7), so one cannot automatically exclude it here. To represent the meaning clearly and accurately, we definitely need another wording than what TNIV offers.
Responses from defenders of gender-neutral translations
Defenders of TNIV and other gender-neutral translations have developed a considerable number of arguments and explanations attempting to explain changes in meaning such as we see above.
The critics of TNIV are wrong in their interpretation of this verse.
There is room for scholarly disagreement about its exact force.
The main point is sufficiently expressed in both wordings.10
Translators may legitimately have different goals, and in this verse they allowed themselves more liberty in rephrasing the point.11
Look at how well they did in translating other verses.12
Or–the final fall-back–no translation is absolutely flawless.13
One of more of these defenses may or may not be relevant as an explanation for a particular verse. Arguments may fly back and forth, amassing ever greater quantities of information about possibilities for interpreting one particular verse, about theories of translation,14 about which aspects of meaning may or may not have undergone change, about whether the meaning changes are notable or negligible. These arguments all have their place, and Wayne Grudem and I have ourselves laid out a considerable quantity of arguments in our book The Gender-Neutral Bible Controversy.15
A pattern of deleting male representatives
But as words multiply we may miss a crucial point: the heart of the matter lies not in a problematic rendering in one or two verses, but with a systematic policy and a systematic pattern. TNIV along with all previous gender-neutral translations systematically eliminates hints of a male representative or of a male sample case used to illustrate a general truth.16 The above verses clearly illustrate the pattern. In 1 John 2:10 “brother” disappears, so that we have a statement that mentions “fellow believers” who may be equally male or female. One does not allow “brother” to stand as a male example of a principle that clearly does apply to both men and women. Similarly, one does not allow generic “he” to appear, because it suggests a male starting point for the general truth. In Hebrews one eliminates the maleness of “father” and “son” as sample cases of discipline in a family. In Revelation 22:18 and 1 Corinthians 14:28 one makes substitutes in order to eliminate generic “he.”
Anyone who doubts that this pattern exists needs first to observe the language of the prefatory section of TNIV, entitled “A Word to the Reader.” It says,
Among the more programmatic changes in the TNIV is the removal of nearly all vocative “O”s and the elimination of most instances of the generic use of masculine nouns and pronouns.17
The expression “generic use of masculine … pronouns” refers to the elimination of generic “he” (which occurs in 1 John 2:10, 1 Cor 14:28, and Rev 22:18 above). “Generic use of masculine nouns” probably includes instances like the change in 1 John 2:10 eliminating “brother,” as well as changes replacing “son(s)” with “child(ren)” and “father(s)” with “parents.”18
Anyone interested can now read through 1 John in the NIV, noting each time (there are quite a few) where generic “he” occurs. One can predict that each time TNIV will change the verse in such a way as to eliminate generic “he.” Sometimes TNIV converts the verse to plurals, as in 1 John 2:10 (above). Sometimes an initial “whoever” gets followed by “they” (1 John 2:5). Sometimes TNIV substitutes “you” (1 John 2:15) or “we” (1 John 4:20). Sometimes “his” drops out by other kinds of rewording (1 John 3:15).19 One way or another, the job gets done–consistently.
A similarly pattern occurs with the use of “brother” and “son” in generic contexts.20 To this one must add the use of “man” to designate the human race, such as in the statement, “God created man in his own image.” This use of “man,” always without any article “the” or “a,” and without any quantifier “every” or “all,” occurs in the secular press,21 but not at all in TNIV.
Freedom
I am advocating freedom to use the whole spectrum of English expressions currently in use. Generic “he” and “man” for the human race are in current use, so we may use them.22 Modern speakers and writers, if they wish, may also refrain from using them, and instead use only various other ways of speaking. It is up to them.
But translators are under unusual constraints. Modern writers control their own meanings, but translators do not. They are responsible to convey the meanings of the work that they are translating. Because of differences between languages, translation becomes complex and challenging, and it becomes excruciatingly difficult, even using all the resources of the English language, to capture every nuance.
In fact, the difficulties make it imperative that translators actually use the resources of the English language without artificial restrictions. They must have at least the same freedom that ordinary writers have. In particular, they must be able to use generic “he,” “man,” “brother,” “father,” and other words if that is the best way to express the meaning.
It is an instructive exercise to try to write a paragraph or an essay without using some part of English. One refrains from using linking verbs, or from using overworked words like “interesting,” or even from using the indefinite article “a/an.” It can be done. One changes one’s meanings around in whatever ways are necessary in order to avoid the forbidden words. Likewise, one can write in English without using generic “he,” or “man” for the race. But if one does that in the process of translation, one must find work-arounds every time. The work-arounds inevitably lead to alterations in meaning. Defenders may argue that after the resulting alterations, particular verses are still “accurate”–more or less. But the difficulties of translation imply that, as a matter of general principle, the overall result for the Bible as a whole cannot possibly be as accurate as it would be if one permitted oneself to use the extra resources of the English language from which one has abstained. The issue is particularly important when it comes to the use of generic “he,” because it is needed in a large number of verses that involve singular examples expressing general truths.23
The debate then focuses on the question, whether generic “he” and “man” (for the race) are still usable resources of the English language. They are being used in mainstream secular publications, so the immediate conclusion is that, yes, they are available.24 Systematically not to use them in translation sacrifices meaning. TNIV made the wrong decision at this principial level, and cannot regain standing merely by arguing about a small list of verses.
1 TNIV appeared in a “New Testament Preview Edition” on January 28, 2002. This preview edition was not offered for sale, but was distributed to key evangelical leaders and was available in full at the website <www.tniv.info>. The formally published version (New Testament only) became publicly available for sale on April 22, 2002 (according to information at <www.tniv.info>). My analysis is based primarily on the Preview Edition. But as of May 16, 2002, the website shows no changes in the four verses discussed below. I am therefore assuming that the published edition contains few if any changes from the preview edition.
2 This issue also lies behind the gender-neutral Bible controversy of 1997. See Vern S. Poythress and Wayne A. Grudem, The Gender-Neutral Bible Controversy: Muting the Masculinity of God’s Words (Nashville: Broadman and Holman, 2000), especially pp. 1-36; Vern S. Poythress, “Gender in Bible Translation: Exploring a Connection with Male Representatives,” Westminster Theological Journal 60/2 (1998): 225-53. In defense of gender-neutral translation (though with some minor reservations), see D. A. Carson, The Inclusive Language Debate: A Plea for Realism (Grand Rapids: Baker, 1998); and Mark L. Strauss, Distorting Scripture? The Challenge of Bible Translation & Gender Accuracy (Downers Grove, Ill.: InterVarsity Press, 1998).
3 See Poythress and Grudem, Gender-Neutral Bible, 267-76. This book discusses earlier gender-neutral translations, but its arguments are all relevant to TNIV. In fact, very little that is genuinely new has cropped up in TNIV, and the arguments presented in the book still hold with just as much force today.
4 See ibid. 111-61; a similar phenomenon occurs in Hebrew, Aramaic, and Greek (ibid. 142-46; 335-47). In 1 John 2:10 the Greek has two occurrences of the masculine singular pronoun (autou, auto). The third occurrence in English arises from an effort to make the English smooth.
5 “Thus he is not really a gender-neutral pronoun, rather, it refers to a male who is to be taken as the representative member of the group referred to by its antecedent” (The American Heritage Dictionary of the English Language [3rd ed.; Boston/New York: Houghton Mifflin, 1996], 831). See further discussion in Poythress and Grudem, Gender-Neutral Bible, 142-43. Hebrew and Greek exhibit a similar pattern of using a masculine pronoun to designate a sample male person who represents a general truth (ibid. 143-45, 335-47).
9 See ibid. 251-63. The immediate context of Hebrews 12:7b includes some occurrences of the plurals of pater and huios, but the context is fully consistent with the normal translation of the half verse 7b. Pater occurs in the plural once in Heb 12:9, but also in the same verse in the singular with reference to God. TNIV agrees that the singular in 12:9 should be translated “Father,” and nothing prevents the translation of the preceding plural as “fathers.” To get “parents” in 12:7b one must make no less than three unjustified steps. (1) Since the plural of pater can mean “parents” where the context clearly indicates this meaning (Heb 11:23), one concludes that the plural must mean “parents” even in a context (12:9) where “fathers” makes good sense and where there is an obvious comparison with God as Father. (2) If the plural means “parents,” one concludes that the singular can also be forced to mean “parent,” even though there are no examples elsewhere of such a meaning for the singular. (3) The singular in Greek can be replaced by a plural “parents” in English, even though in this verse it shifts the meaning from a focus on a sample case to the focus on the plurality of parents and children.
Huios occurs in the plural once in 12:5, once in 12:7a, and once in 12:8. It occurs in the singular in 12:5, 6, and 7b. Curiously, TNIV translates with “son” for the singular in 12:5, but elsewhere uses “child(ren).”
16 Vern S. Poythress, “Gender in Bible Translation: Exploring a Connection with Male Representatives,” Westminster Theological Journal 60/2 (1998): 225-53; Vern S. Poythress, “Searching Instead for an Agenda-Neutral Bible,” World 13/45 (November 21, 1998): 24-25; Poythress and Grudem, Gender-Neutral Bible, 142-48, 232.
18 See Poythress and Grudem, Gender-Neutral Bible, 285-89. We allow that there may be cases where using “children” or “parents” is legitimate. We object to the principial, systematic elimination of cases where a general principle is expressed using a male representative.
19 Here are the verses for comparison:
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 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.
1 John 3:15
NIV: Anyone who hates his brother is a murderer, ….
TNIV: Anyone who hates a fellow believer is a murderer, ….
20 See ibid. 255-76. Note again that with respect to some verses and some contexts the issue is complex. The appropriateness of using “brothers and sisters” or “children” varies with context.
22 On current use in the secular press, see ibid. 203-212, 239-42. On the principial issue, see ibid. 111-232, 233-42, 355-66.
24 The situation has not changed in a fundamental way since the publication of the book by Poythress and Grudem in 2000, where we cite evidence for the use of generic “he” and “man” in the secular press (ibid. 203-210, 239-42). In 1998 Mark L. Strauss, to his credit, attempted to present evidence to show that generic “he” and “man” (for the race) were unusable (Distorting, 140-46). But Poythress and Grudem show that the evidence in question can easily be seen as confirming their position that the underlying issue is really male representatives for general truths (Gender-Neutral Bible, 224-29). D. A. Carson, in his 1998 book Debate, scatters hints that he will present an argument regarding the usability of “he” and “man,” but in the end he gives none (Poythress and Grudem, Gender-Neutral Bible, 355-66). For a full discussion of issues of gender sensitivity and other kinds of objections to generic “he,” see ibid. 135-232.