Confusion of Semantics with Linguistic Pragmatics in the Defense of the TNIV

A Response to Blomberg, Wallace, and Bock on the Meaning of Aner (ανηρ)

by Michael Marlowe, May 2005

Soon after the appearance of the TNIV New Testament in 2002 some New Testament scholars published some articles in which they defended the TNIV’s use of gender-neutral language in the translation of the Greek word ανηρ. The arguments in these articles are very weak. After examining them, I conclude that these arguments would never have been advanced or even taken seriously by scholars if it were not for the desire to defend the “inclusive language” adopted in English versions like the TNIV. But I will give some attention to the arguments, because interaction with them may help to illustrate principles of lexical semantics which are of more permanent interest.


In 2002 an article by Craig Blomberg of Denver Seminary entitled Today’s New International Version: The Untold Story of a Good Translation was posted in various places on the internet, including the official website of the TNIV. On page 15 of the copy now posted on the TNIV site (accessed April 2005) Blomberg writes:

Third, a brief comment needs to be made about the TNIV’s renderings of aner. The CSG [Colorado Springs Guidelines] object to rendering this term that often means male (vs. female) or husband (vs. wife) with gender-inclusive language. But in fact, one well-attested meaning of the word is as a synonym for anthropos. 48 In James, probably every use of aner falls into this category. James 1:8, the first such usage in the epistle, clearly employs aner as parallel to the generic ‘man’ described as anthropos in v. 7, and a quick glance at all of the other uses of aner in this letter demonstrates that almost all clearly refer to men and women alike (1:12, 20, 23; 3:2; the possible exception is 2:2). This is what linguists call an idiolect, when a particular speaker uses a less common meaning of a term fairly consistently as part of his or her distinctive style. 49 Luke’s use of aner in translating introductory addresses to crowds of mixed gender in Acts reflects a similar idiolect (e.g., Acts 1:16; 2:14, 22, 29; 3:12, etc.). In each case the TNIV offers an improvement over the NIV. In short, each usage of aner must be evaluated on a case-by-case basis, in context, even if it is true that the sizable majority of New Testament uses do wind up referring to males as over against females.

48. H. Vorländer (“ανηρ,” in New International Dictionary of New Testament Theology, ed. Colin C. Brown, vol. 2 [Grand Rapids: Zondervan, 1976], 562-63) notes that the term in both classical Greek and the New Testament can mean simply ‘adult’ or be used ‘without emphasis’ on maleness, or refer simply to man ‘as genus’ and thus equivalent to anthropos. J. B. Bauer (“ανηρ, ανδρος, ο,” in Exegetical Dictionary of the New Testament, ed. Horst Balz and Gerhard Schneider, vol. 1 [Grand Rapids: Eerdmans, 1990], 98-99) notes that the term can refer to ‘human beings in general,’ supplying fourteen examples from the New Testament and two from the papyri. Walter Bauer (“ανηρ, ανδρος, ο,” in A Greek-English Lexicon of the New Testament in Other Early Christian Literature, rev. and ed. Frederick W. Danker [Chicago and London: University of Chicago Press, 2000], 79) lists as the second of three main definitions, “equiv. to τις someone, a person,” with nine New Testament and nineteen non-New Testament references.

49 So also Carson, The Inclusive Language Debate, 162.

When Blomberg says “one well-attested meaning of the word is as a synonym for anthropos” he is saying too much. It is true that lexicographers and other Greek scholars have noted places in which the word aner seems to be rather colorless, because the reason for using a word meaning “man” or “men” as distinct from women or children is not clear in the context. And so we see a kind of catch-all category in which lexicographers indicate the apparently colorless usage, which resembles the usage of the pronoun tis or the noun anthropos. But if we look at the lexicons we find that these cases are relatively few, despite the fact that the word aner is a very common word. Almost always the context indicates why aner is used instead of the less colorful anthropos, and the lexicons differ greatly in their short lists of the places where the “adult male” sense seems to be unimportant in the context. Thayer’s lexicon lists these under the sense “univ. any male person, a man; so where tis might have been used,” and there he puts Luke 8:41, 9:38, Acts 6:11, 10:5, etc. In all of these examples the referent is a man, and in some of them it turns out that the context does indicate why aner is used instead of anthropos when it is examined carefully enough. For example, in Acts 6:11 the reference is to witnesses giving testimony, and in ancient times women and children were not considered to be credible witnesses. Likewise the Abbott-Smith lexicon gives a sense “In general, a man, a male person: = tis,” and lists there Luke 8:41 and Acts 6:11. The third English edition of Bauer’s lexicon lists such occurrences under the heading “equiv. to tis, someone,” and gives nine examples: Luke 9:38 and 19.2; John 1:30; Romans 4:8; Luke 5:18; Acts 6:11; Luke 8:27; Acts 10.1; and James 1:12. Here again we see Acts 6:11 listed, despite the fact that the context does indicate a reason for the sense “adult male,” and the same is true of some other occurrences listed here. Romans 4:8 is really a quotation from Psalm 32 according to the Septuagint rendering, and so Paul’s reason for using the word aner here must be understood in light of his source and his characteristic manner of using biblical quotations. The immediate context does not tell the whole story, and I doubt that the occurrence here belongs under the weakened sense “someone.” But in any case, the Bauer lexicon is here simply noting that aner is sometimes used in a relatively colorless way, where the word tis might be substituted without any obvious change in meaning. These lexicon articles do not assert that the word has an entirely gender-neutral sense in these places. Blomberg also cites Vorländer’s article in the New International Dictionary of New Testament Theology, but I note that this article sermonizes on the theme of the “equality of the sexes,” which indicates that sober lexicography is mixed with something else there. One piece of information which must be borne in mind is the fact that aner is never used in reference to a woman. This is not mentioned in the lexicons because lexicons do not ordinarily indicate what a word doesn’t mean, but it must be kept in mind when we are discussing the gender-neutral Bible translations, which sometimes completely neutralize the word aner along with several other masculine terms. It seems to me that Blomberg’s argument is depending upon the reader’s ignorance of this fact.

Implicit in Blomberg’s argument is the assumption that the reader thinks anthropos is a gender-neutral word. His statement that aner may be used as “a synonym for anthropos” is really pointless without this assumption. But in fact anthropos is not entirely gender-neutral. It is a masculine noun that was not used in reference to an individual woman, though in the plural it may be used to encompass both men and women, much in the same way that the word ‘men’ has been used inclusively in English. If we were to define the word merely on the basis of its referential usage, we would have to say that the word anthropos in the singular means ‘a male human,’ and nothing more or less than that. But this is not a complete account of the meaning of the word, because it leaves out the connotative meaning, which involves a semantic foregrounding of the genus ‘human’ rather than an emphasis on the masculine meaning which attaches to the word through denotative usage. It is only on the basis of this connotative element of meaning that the word anthropos can be said to mean primarily ‘human,’ without emphasis on masculinity. Words are often laden with meanings that go far beyond the merely denotative meanings, and one cannot neglect the connotative aspects of meaning in giving a full account of the meaning of words. Now, despite the fact that he depends upon this principle with regard to anthropos, in his discussion of aner Blomberg completely ignores the connotative meaning, and he tries to reduce the meaning of aner to a denotational meaning indicated by the context. This is illegitimate enough in itself, but he goes even further astray when he claims that in the Epistle of James the uses of aner “clearly refer to men and women.” He ought to have said that here aner is used in statements which apply to both men and women. But application is not the same thing as reference. The legitimate and intended application of a statement commonly runs well beyond the bare referential meaning of its individual words.

For example, in the earlier portions of the Hebrew Bible the word pesel means “carved image,” and so a literal translation of Exodus 20:4 is, “You shall not make for yourself a carved image, whether in the form of anything that is in heaven above, or that is on the earth beneath, or that is in the water under the earth. You shall not bow down to them or worship them ...” Now, this commandment ought to be understood and applied in such a way that it prohibits any sort of idolatrous worship, for it is obviously not the intention of the commandment to limit the prohibition to carved images only. It applies also to images that are cast from metal (specifically prohibited in Exodus 34:17), molded, painted on a canvas, or projected on a screen with modern equipment. It should even be applied to such things as a pattern of accidental stains on a wall that some deluded person may be worshipping because he thinks it is a miraculous image of Jesus. Nevertheless, in Exodus 20:4 the word pesel in itself does not refer to any such images, it refers to the cultic images carved from wood or stone, which were so familiar to people in the ancient Near East. (1) If we were doing a word study we could not conclude that because the statement in which the word occurs applies to all idols, the word pesel must actually refer to all of the things which could be used as idols, or in certain contexts mean all of these other things, while ceasing to mean “a carved image.”

Perhaps an example from English will make this even clearer. Imagine a little boy throwing stones against the wall of his house outdoors. His father comes out and says, “you must not throw stones against the house, because you are chipping the paint.” Now, it’s perfectly obvious that here the father means to include any sort of hard object in that rule, not only stones, and even a child would know that the application goes beyond “stones.” He would not, for instance, think that he may continue to throw other objects such as marbles or little toy cars against the house, and he would not suppose that because his father used the word “throw” and not “shoot” he could go and get his slingshot and launch objects against the house with that. The application of the father’s statement goes beyond the meaning of the individual words, and the son understands this intuitively. But the English word “stone” does not, even in this context, have the generic sense “hard object,” nor does it refer to such things as glass marbles and metal toys; and the English verb “throw” is not used with the sense “shoot.” One cannot use the intended application of the father’s statement as evidence for such meanings for the English words. That is not how language works, and that is not how we learn languages. The stone-throwing boy in our example will not find the full meaning of his father’s instruction in the dictionary, nor does he need to find it there, and it would be wrong for anyone to think that “hard object” should be listed as a recognized sense for the word “stone” on the basis of examples like this.

Linguists commonly make a distinction between semantics, which has to do with the meaning of individual words and phrases, and linguistic pragmatics, which has to do with various aspects of intention, interpretation and application that need to be taken into account to fully understand how language is used and understood. Essentially this distinction is meant to express the rather obvious fact that people commonly intend more than they actually say, and that our understanding of the full intention of an utterance depends very much on common-sense inferences and knowledge about the situations in which things are said. The kind of inferences that the stone-throwing boy makes in order to fully understand his father’s intention when he says “you must not throw stones against the house” belong to linguistic pragmatics, not semantics proper. And that is why his common-sense interpretation of his father’s instruction, or any other similar interpretations of the intention of sentences, does not belong in a dictionary. The dictionary is a collection of semantic information, not a complete account of language which encompasses pragmatics. It would be impossible and inappropriate to try to include information like this in a lexicon, and it will only cause confusion to try to talk about these pragmatic aspects of language as if they belonged to lexical semantics.

In the oldest manuscripts of the Epistle of James, he addresses his readers in 4:4 as, “you adulteresses” (moichalides), not “you adulterers and adulteresses” (moichoi kai moichalides). Does this mean that he is adressing only the women at this point, or that moichalides by itself is a gender-neutral word? Certainly not. He addresses them as “adulteresses” because he is comparing them to unfaithful wives, in the same manner as the Old Testament prophets, who often compared Israel to an unfaithful wife. (i.e., Israel does not honor her husband but goes “whoring after other gods.”) And so he goes on to speak of God’s “jealousy” in the following verse, continuing the metaphor. The original Jewish readers of this epistle would have understood this language easily. Apparently, however, some later Christian scribes did not understand it, and so they made the language “inclusive” by adding the words “adulterers and,” as if James were merely accusing his readers of sexual immorality. These scribes knew the meaning of Greek words well enough, but they completely missed the point of James’ expression because they lacked the pragmatic competence of the Jewish readers. They did not understand this insulting metaphorical usage, in which the feminine word “adulteresses” is used to address male readers. This is an “inclusive” use of a feminine term, but it does not by any means lose its feminine connotation; rather, the feminine connotation is an important part of the intended meaning. The specific gender connotation persists regardless of who is being addressed with this word in the immediate context. The point is lost if we try to see this as a gender-neutral expression in keeping with the mixed gender of the audience.

Returning to Blomberg and his discussion of aner — we can see that he is basing his argument on a very inadequate assumption about how words get and retain their meanings, and how people perceive the full meaning of statements. He is either ignoring the whole pragmatic dimension of language-comprehension, or he is confusing semantics with pragmatics. He is arguing that if a statement containing the word aner applies to both men and women, then the word aner must in that context refer to and mean simply “a person,” without any masculine component of meaning. This will not do at all. Someone might care to argue that Blomberg has implicitly adopted a pragmatic theory of meaning, and that his statements must be understood in that context; but that does not rescue him from confusion, because he is not confining himself to statements about the intent of the author, he is trying to establish “the meaning of the word” aner and the things to which it “refers,” not the wider intention of the author. Or someone might argue that Blomberg is right because it is the job of a translator to make the proper application of biblical statements plain to readers, hence pesel should not be rendered “carved image” but “idol” (as in the NIV) and aner should not be translated “man” but “person” (TNIV). But in that case we are no longer talking about the actual meaning of the words pesel and aner in the original languages, we are instead talking about what needs to be done for modern readers of English versions who might not see the full application. It will only cause exegetical confusion to present such pragmatic interpretive measures as if they were a matter of Greek semantics, as Blomberg and others have been doing.

Blomberg tries to sweep away the ordinary meaning of the word aner by maintaining that its usage in the Epistle of James reflects a peculiar ‘idiolect’ in which the word, contrary to its normal usage in Koine Greek, had a gender-neutral sense. This effectively eliminates all the lexical evidence available to us outside of the Epistle for a determination of the meaning of the word within it, and I do not think such a procedure can be justified. His assertion that “Luke’s use of aner in translating introductory addresses to crowds of mixed gender in Acts reflects a similar idiolect” is implausible because the use of aner to address an assembly is a normal feature of Greek oratory. The phrase Ανδρες Αθηναιοι “Men of Athens” is quite frequently used by Demosthenes and the other Greek orators, as the normal way of addressing their Athenian audience. The aner used as a form of address in the speeches of Acts reflects this feature of the oratorical genre. It is a formulaic usage. So even if some women happened to be present in the audience for the particular speeches recorded here, this has no bearing on the meaning of the word aner.

We might compare this feature of ancient Greek and Roman oratory to the corresponding use of “brothers” or “brethren” in sermons to English-speaking audiences of the past few centuries. In published sermons of the eighteenth century we find that John Wesley and George Whitefield, for example, frequently address their hearers as “my brethren.” These sermons were delivered to crowds in the open air, and there can be little doubt that women were present. In the middle of the nineteenth century, Henry Ward Beecher was perhaps the most famous preacher in America. Many of the sermons he preached at Plymouth Church in Brooklyn were published soon afterwards. Looking at one volume of these at Google Books (Sermons by Henry Ward Beecher [London: J. Heaton & Son, 1864]), we find that he habitually addresses his hearers as “Christian brethren.” The word “sisters” does not occur in the volume. This is despite the fact that Beecher was one of the most prominent advocates of women’s suffrage in his day. Likewise, the most celebrated British preacher of those days, Charles Haddon Spurgeon, often addressed his huge audiences as “my brothers,” “brethren,” and “sirs,” along with an occasional “brothers and sisters.” The record states that these sermons were delivered on Sunday mornings, in places like the Metropolitan Tabernacle in London, where hundreds of women were certainly in attendance. (2) Will anyone care to argue that this shows that in Colonial America and in Victorian England the words “brothers” and “sirs” were gender-neutral words? Perhaps Blomberg is not very familiar with the rhetoric of older times, and finds it hard to understand why a speaker would habitually address the men in this way while women are also present. Obviously it does not accord with modern customs of polite discourse. But it would be fatuous to argue that preachers of the eighteenth and nineteenth centuries would not have addressed a mixed audience with “brothers” unless it was a gender-neutral word at the time. This manner of addressing an assembly was quite normal for at least four thousand years. Blomberg’s assumption that gender-neutral forms of address must have been normal for “addresses to crowds of mixed gender” in ancient times is anachronistic. The expectation that a public speaker should address his remarks to women as well as men is quite modern; it developed subsequent to the enfranchisement of women in the twentieth century. (3)

Another fallacy here is the idea that the meaning of a word can be safely inferred from the circumstances of its use in rhetorical parallelism. Blomberg points out that in James 1:8 aner is “parallel to the generic ‘man’ described as anthropos in v. 7,” as if this parallelism could actually restrict the meaning of the second word to the meaning of the first word. We might point out that if this were a legitimate principle of analysis it would work both ways, so that the meaning of the first term could be defined according to the usual meaning of the second. But the use of parallelism in discerning the meaning of words is legitimate only within certain limits. The most that can be said with confidence on the basis of such parallelism is that the words in question share a component of their meanings. There is no good reason to suppose that the words are identical in meaning. Often in rhetorical parallelism the second term is sharper and more definite in meaning than the first, and they need not be synonymous — in fact, the use of antonyms in parallelism is quite common.

Regarding James 1:7-8, I will venture to say that aner is probably used with the meaning “a grown man” because James wants to contrast the connotation of manliness in aner with the unseemly weakness of the grown man who doesn’t know his own mind, being “of two minds” and “unstable” like an adolescent. Arthur Carr in his commentary on James (Cambridge, 1896) points out that the word akatastatos (unstable) in 1:8 is ‘used of youthful fickleness’ by Polybius (p. 15; he gives the example of Polybius’ description of a young king having an “unstable character” in his Histories, 7.4.6). And if the saying “an aner of two minds is unstable in all his ways” was proverbial, we can easily see how James would use the word without any intention of being referentially ‘inclusive’ or ‘neutral.’ As for James 1:12, the use of aner there is appropriate to the subject because James is talking about a man of tested character, a man of experience who has endured under the hardships and trials of life. It is suitable to use the word for an adult man in this context, although obviously the statement in general applies to women also. Under Blomberg’s treatment of the diction here none of the ‘manly’ connotations of aner are supposed to be noticed, because he wants to reduce the meaning of the word as far as possible, even to the point where it means nothing more than ‘someone.’ But in general I think a better way of interpreting and profiting from Scripture is to draw out the meaning of the words. This is hard to do with a translation like the TNIV, which tries to hide the meaning for the sake of political correctness.


Now I will move on to a consideration of the arguments in another article posted on the internet. This short article, found here, is signed by Daniel B. Wallace of Dallas Seminary. It seems to be a piece he quickly dashed off in response to questions about the TNIV’s treatment of aner, in 2002.

Can Greek aner (“man”) sometimes mean “person”?

BDAG gives three basic definitions to aner, each backed up with numerous texts: (1) “an adult male, man husband”; (2) “equiv. to tis someone, person”; (3) “a transcendent figure.” Under def. 2, they give the following texts: Luke 9.38; 19.2; John 1.30; Rom 4.8; Luke 5.18; Acts 6.11; Luke 8.27; Acts 10.1; Jas 1.12. As well, several non-NT texts are listed, including: Sir 27.7; Ps 32.2; 1 Macc 12.1; 13.34; 1 Macc 7.7; Ignatius: Eph 4.2; 20.2; Trallians 13.2; Smyrnaens 5.1; Polycarp 1.3. The entry in the lexicon, in fact, has very little in terms of definitions and glosses here. Instead, most of this point is filled with data.

Let’s examine some of those texts to see if they have an adequate basis for this judgment. Luke 9.38 has aner while the parallel in Matt 17.14 has anthropos. John 1.30 has aner while 1.15 uses simply the masculine singular participle for the same statement from the Baptist. But in none of these texts is aner clearly used for simply person in such a way that a woman could fill the slot. Perhaps by this BDAG is simply noting that aner is sometimes used when the maleness of the individual is not emphasized. Still, I would have dropped these passages from the lexicon. Luke 5.18 is more promising, since it speaks of four men (aner) carrying a lame man. If Luke used Mark as his sole source here, then Luke is either interpreting the four who carried the lame man as four men (since Mark does not specify), or else aner can sometimes be used more generically for person. I would reject Acts 6.11, since in Jewish courts of the day the testimony of men was regarded as significantly greater than the testimony of women. I think the use of aner here is simply in keeping with Jewish custom (even the NRSV has ‘men’ here). Romans 4.8 and Jas 1.12 are the best instances in the NT, I believe, for there the generic principle can hardly apply only to adult males! Aren’t women whose sin is not imputed to them also blessed by the Lord? And don’t women who persevere under trial also receive a crown of life? I think that BDAG would have made out a better case if they had put the other NT texts in a dubious category, while retaining the principle clearly for Rom 4.8 and Jas 1.12. As for the non-NT passages, in Ignatius’ letter to the Ephesians 4.2, as well as his letter to the Trallians 13.2, he speaks of “man by man” (kat’ andra). But in Eph 4.2 the usage refers to those who sing praises to Jesus Christ—it can hardly refer just to the men. In Trall 13.2, it refers to those in the church who are to be loved: “love one another, man by man, with an undivided heart.” Again, this can hardly refer to adult males only. The same idiom is used in Smyr 5.1 and Poly. 1.3. These references are all from Ignatius, a bishop who died no later than AD 117. He is thus nearly contemporaneous with the NT writers. It is highly unlikely that he started to use the word aner in a way that was significantly broader than its usage a few decades earlier in the NT writings.

But more can be said: the LXX also uses the word more broadly than to mean merely adult male. Sirach 27.7 has a synthetic parallel between the two lines: in the first is aner and in the second is anthropos, both in the plural (“Praise no man before he speaks, for it is in this manner that men are tested”). This seems to be an excellent illustration of the point that BDAG want to make. Ps 32.2 is what is quoted in Rom 4.8. In one or two of the passages in 1 Maccabees use the idiom aner hos (‘a man who’) which, in BDAG’s estimation, is equivalent of simply “who.”

All in all, the evidence is not great that aner sometimes refers to a person rather than specifically to an adult male. But it is compelling enough, in some of these examples, to warrant its force as a legitimate meaning for the term in Koine Greek. Preference in most passages should be given to adult male as the meaning, but certainly not in all.

First of all, I find it remarkable that although Dr. Wallace correctly surmises that the third English edition of Bauer’s lexicon (BDAG) is “simply noting that aner is sometimes used when the maleness of the individual is not emphasized,” he goes on to critique the lexicon as if he thought it might be trying to establish a completely gender-neutral sense. But this is obviously not the case. If it were, the lexicon would not have cited under this heading the occurrences in which the word aner clearly refers to a man. So his critique of this lexicon article is really pointless, because he misunderstands the intention of the lexicographers. But apparently Wallace feels that the lexicon ought to have tried to establish a gender-neutral sense by presenting a different set of citations which are more suitable for that purpose. And so he goes on to attempt it himself, but without success.

Wallace relies upon some of the same fallacies of semantic analysis that I noted above in Blomberg’s article. Here again is the confusion of lexical semantics with linguistic pragmatics, the idea that every appropriate application of a statement must somehow be inherent in the meanings of its individual words. Hence he argues that because the statements in Romans 4:8 and James 1:12 apply to women, the word aner in those statements must be regarded as gender-neutral in some way. He does not commit such a blatant fallacy as does Blomberg, who explicitly claims that a word refers to both men and women if the statement in which it occurs applies to both; but by another path we are ushered to the same conclusion, through Wallace’s use of a suitably foggy phrase — “the generic principle.” But where did this shadowy “generic principle” come from, and what are its limits for the analysis of Greek words? Would Wallace care to argue that moichalides in James 4:4 is gender-neutral? Or is ishah (your neighbor’s wife) in Exodus 20:17 a gender-neutral word with the meaning “spouse”? Must we invent a gender-neutral sense for the statement “your ishah shall be as a fruitful vine” in Psalm 128 because the words are addressed to “everyone who fears the LORD”? Surely there are women who fear the LORD. What about the phrase “eyes full of an adulteress” in 2 Peter 2:14? Will Wallace argue that this, together with James 4:4, is “compelling enough” to establish some gender-neutral sense of the word moichalis? We could easily collect many statements from the Bible which might just as well have been written in gender-neutral or gender-inclusive style, though the expressions actually used by the authors exhibit their masculine orientation. The attempt to make the application of these statements into lexicographical evidence for gender-neutral senses for the words is clearly misguided.

Also like Blomberg, he gives an invalid argument from the circumstances of rhetorical parallelism, citing Sirach (Ecclesiasticus) 27:7. Strangely enough, despite his argument he translates the verse as, “Praise no man [aner] before he speaks, for it is in this manner that men [anthropoi] are tested,” which seems to indicate a lack of conviction that aner really means “person” here. In any case, the parallel to anthropoi is no sufficient reason to conclude that aner here means only “person.”

I do not understand why Wallace says both aner and anthropos are “in the plural” in this verse, because the form of aner here is andra, the singular accusative, not the plural andras. The verse reads as follows in the edition of Rahlfs:

πρὸ λογισμοῦ μὴ ἐπαινέσῃς ἄνδρα·
οὗτος γὰρ πειρασμὸς ἀνθρώπων.

It is possible that Wallace is looking at another edition of the LXX that has andras here; but if so, he cannot well ignore the fact that the standard manual edition has the singular if he is going to call attention to a plural reading to support his argument. And there is no plural variant noted in Rahlfs’ apparatus. I suppose that Wallace mentions the number of the noun because he thinks that if both aner and anthropos are in the plural, this would tend to strengthen the semantic parallelism. But it makes no great difference to his argument or to my response. (4)

We do not know for sure what the Hebrew vorlage of Sirach had at this point (the verse is not found in the Hebrew fragments), but probably the Greek translator used aner because the Hebrew had ’ish here (as usual in the LXX). If that was the case, then aner is probably meant to convey the same kind of honorific connotations that attach to the word ’ish, and the context also seems to indicate this connotation. The gist of the saying seems to be, “do not praise a man just because of his importance or his reputation, but wait and judge him by his mental characteristics as revealed in his speaking, as you would any ordinary man.” The phrase pro logismou, which Wallace translates “before he speaks,” would be more adequately translated “before [hearing] his reasoning,” as in the RSV, or at least “before his discourse,” because the primary meaning of the word logismos is “reasoning.” When used of speech, it connotes something more than small talk. So this saying looks like a piece of advice for sizing up men of repute which probably does not have women and their domestic speech so much in view. The ancient setting must be borne in mind. And it should also be pointed out that the entire book of Sirach has a strongly masculine orientation, as may be seen clearly enough in the sayings about women to be found in chapters 25 and 26. For example, we read that “a silent wife is a gift from the Lord” (Sirach 26:14), and many other sayings like it. The author did not ignore women at all, in fact he takes a special interest in them; but in referring to them he definitely speaks as a man, to men. This is normal in books written by male authors in ancient Mediterranean cultures, as David deSilva explains:

Women are discussed only in relation to men ... women were not viewed as independent entities but as in some sense ‘embedded’ in a male. Men were the ‘actors,’ and women had a role and place only insofar as they were embedded in such an agent, whether a father or a husband. That Ben Sira speaks of women (good or bad) only as they relate to men reflects this cultural situation. (5)

In 42:12 the author (or his Greek interpreter) warns his readers not to “sit [sunedreue] in the midst of women,” and in verse 14 he explains that the ordinary faults of men are less treacherous than a woman’s kindness: “the evil of man [andros] is better than the goodness of woman.” This is the wider context of culture and discourse in which the word aner is being used. Obviously the author had little reason to use the kind of ‘inclusive’ discourse that Dr. Wallace would like to find in his work.

There is one argument used here which Blomberg does not use, concerning the meaning of the idiom kat’ andra in the epistles of Ignatius. Wallace apparently thinks that the usage of this idiom in a second century source can tell us something about the meaning of the word aner when it stands alone in the New Testament. This involves another fallacy of analysis — the idea that the meaning of an idiom (a stock combination of words which functions as a single unit of meaning) can be attributed to the individual words of the idiom when they occur apart from the idiom. For example, in English we have a colloquial idiom in which a person may be called a “poor thing.” It is a way of referring to a person—usually a woman—who is an object of pity: e.g., “No one asked her to the dance, poor thing.” But this does not mean that the English word “thing” has a sense of “woman,” so that we would have to decide whether or not the word “things” in the sentence “the man brought his things from Memphis to Rome” includes his wife. The word “thing” does not mean “woman,” and the word “things” does not include people in any ordinary usage of the word. So even if a man did bring his wife, we would not be thinking about that when we say, “he brought his things from Memphis to Rome.” And this is despite the fact that many occurrences of the phrase “poor thing” clearly do refer to women. Another example: In English we have an expression, “the rank and file,” which refers to “the ordinary members” of an organization, especially of labor unions. This is a kind of military metaphor, which conveys the connotation that these members are like soldiers standing in a formation. But the word “rank” does not mean “ordinary member,” and neither does the word “file.” It is the combination of words which has this meaning, and not the individual words. Now, the Greek phrase kat’ andra, which means “to a man” or “each and every one,” was probably also a military expression at first, and this would explain why the word aner is used in it; but again, in idioms the meanings of the individual words tend to lose their original force, and so if this idiom were used in reference to a group including women it should not surprise us. We are no longer using the word aner alone, we are using an idiom which has its own proper meaning quite apart from the meaning of the individual words. For this reason, Wallace’s citations of this idiom can prove nothing about the meaning of aner in Koine Greek. It is as if some foreigner were trying to prove that our word “thing” has the meaning “woman” by citing passages in which the phrase “poor thing” refers to women. Wallace also neglects to consider the possibility that the phrase kat’ andra did have such a masculine connotation that the idiom was not used in reference to a group composed only of women (are there any examples of such a usage?), but that Ignatius used it in reference to the congregation because he habitually addressed his remarks to the men.

Wallace should have ended his article with “the evidence is not great that aner sometimes refers to a person rather than specifically to an adult male,” because his attempt to establish this gender-neutral sense in a few places is by no means compelling. Rather, after some apparent skepticism on his part, it becomes evident that he is merely assuming the thing to be proven. He just takes it for granted that anything which applies to both men and women must have been gender-neutral in the original. (“Aren’t women whose sin is not imputed to them also blessed by the Lord? And don’t women who persevere under trial also receive a crown of life?”) I doubt that Dr. Wallace really has such a naive understanding of Greek semantics.


I will respond to yet another article now, this one by Darrell L. Bock, also at Dallas Seminary. His article, “Do Gender-Sensitive Translations Distort Scripture? Not Necessarily,” first appeared on the website in the Spring of 2002, and was later published in the Journal of the Evangelical Theological Society 45/4 (December 2002), pp. 651-669. Bock writes:

One final argument remains. It is that the Greek noun for “man” (aner) should never be rendered “human” or in a way that includes women (either “men and women” or a generic “those”). The claim for this limitation is that there is no clear example of such usage. However, candidates for such generic reference to humans or to “ladies and gentlemen” to audiences that are mixed do exist in both the Old (if one is thinking about the force of the Greek in the LXX) and New Testaments. Here one could consider Psalm 84:5, 112:1, 5. Renderings of a generic “those” (84:5) or “person” (112:1, 5) make the point that the psalm is not just praising males, but the righteous person, who in the text is called man as part of a generic class. Romans 4:8 has already been mentioned as an example. More to the point are several texts in Acts, where a broad crowd is addressed (Acts 3:11-12; 13:38; 17:22; 19:35; 21:28; 22:1; note in many of these examples how the nearest antecedent for the audience is expressed in broad terms like “the crowd” or “the people”). In each of these contexts the address appears to be to a mixed audience. So the application of the desired response is for anyone in the audience who would respond. A gender specific rendering of “men” might suggest that only the men were being addressed. Some might wish to argue for this limitation in this ancient patriarchal culture. However, it really is a point that would need specific defending, given that the application in terms of the response called for by the speaker is for the benefit of anyone who is hearing the speech. Here is a case for which our modern “you guys” being used as a shorthand form of initial address may well be the better parallel. The biblical examples in this paragraph all use aner. The options here again show the nature of what is being discussed by each side with reference to how to render these texts. Gender sensitive renderings in some of these cases are likely to be adequate renderings for these texts. They are neither necessarily wrong nor are they doctrinally offensive.

Bock has coined a new term, “gender sensitive,” which needs an explanation. This seems to be his attempt to avoid all the quibbling about whether the controversial renderings in the TNIV should be called “gender neutral” (as most people would describe them) or “gender accurate” (as some defenders of the version have absurdly called them). Earlier in the article he explains that the “gender sensitive” versions generally use the “dynamic equivalence” method of translation, and he distinguishes “two basic types of gender sensitive approaches to translation: ideological gender sensitive renderings and translational gender sensitive renderings.” Throughout the article he implies that the TNIV is of the latter type, which he carefully defines thus:

Translational Gender Sensitivity: This approach renders terms to make clear the gender scope of passages, especially when they use an all-encompassing reference to man or mankind to address both men and women. So, for example, the rendering of a term that is translatable as “men” is made into “men and women” when the meaning intention [sic] or application of a passage is broad and not gender specific.

Here, as in the other articles quoted above, we see the important role that application plays in the discussion of the translational issues. Bock is saying that the TNIV translates so as “to make clear the gender scope of passages ... when the meaning, intention or application of a passage is broad and not gender specific.” I think this is an incomplete description of what is going on in the TNIV version, because the goal of avoiding offense at “sexist language” is probably more important to the translators than the goal of making the application clear. The proper application is not unclear to anyone with common sense, and the gender-neutral renderings are not necessary for that purpose. But at least Bock admits that here we are dealing with a version that aims to work the supposed application of a passage right into the translation, as often happens in the “dynamic equivalence” versions. It does not strictly confine itself to the proper meanings of the words in the original languages, as a “formal equivalence” translation does, it goes beyond these to indicate the “scope” or broader “application of a passage.” So far so good.

But when we come to his discussion of aner we find that this principle of explanatory paraphrase has become a principle of semantic analysis. The key statement in his argument is: “In each of these contexts the address appears to be to a mixed audience. So the application of the desired response is for anyone in the audience who would respond.” And again he says, “the application in terms of the response called for by the speaker is for the benefit of anyone who is hearing the speech” (my emphasis). Considerations like this clearly belong to linguistic pragmatics, not semantics. He is wrongly assuming that the meaning of a word can be established from the “application of a passage” in which it occurs. It is the same fallacy of semantic analysis that I have discussed above in connection with Blomberg and Wallace. But here in Bock’s article I think we can see where Wallace’s “generic principle” has come from. A principle belonging to the “dynamic equivalence” method of translation has been illegitimately carried over into the field of semantic analysis. The “tail” of application is “wagging the dog” of the original text.

Bock, unlike the others, acknowledges that “the ancient patriarchal culture” may be remembered when we see such phrases as “Men of Israel, hear these words.” (Acts 2:22). But it is rather silly for him to say that the masculine connotation for aner needs “specific defending” here — as if there ought to be some presumption in favor of his idea that the phrase differs in no way from a modern “ladies and gentlemen.” Explanations in line with the ancient patriarchal culture do not need specific defending. Rather, it is the presumption in favor of modern customs of discourse that need specific defending, when we are considering the meaning of ancient words. Again, the emptiness of this argument becomes plain when we try to transfer it to comparable features of sermons delivered in our own language to mixed audiences. Clearly the sermons of Wesley, Whitefield, and Spurgeon were given “for the benefit of anyone who is hearing the speech,” but we know very well that “brothers” and “sirs” have never been gender-neutral words in our language. No one will deny that Spurgeon wanted women and children to respond to his preaching as well as men, but this does not create any presumption in favor of the notion that in his days the word “sir” was used to address women. Rather, we understand that this element of his pulpit style conforms to a long-established convention of rhetoric, in which the speaker addresses men while expecting any women and children in his audience to make the appropriate applications to themselves. It is so obvious that it hardly requires an explanation, let alone a defense.

I do believe that Bock recognizes how vulnerable his arguments are here, because although he seems to be very interested in defending the TNIV throughout the article, his conclusion is rather modest. He only claims that gender-neutral renderings of aner in “some” places are “likely to be adequate,” and he does not use the word “accurate” in connection with them. Nor does he claim that these—or any other “gender sensitive” renderings—are necessary for English readers. Towards the end of his article he writes, “I do not subscribe to the view, as is often claimed by gender-sensitive translations, that English has changed so much that gender-sensitive renderings are necessary.”


So much for the arguments for gender-neutral renderings of aner. Those who wish to read more on the subject may read a very full discussion of the issue by Poythress and Grudem in appendix 2 of their book, The Gender-Neutral Bible Controversy.

The most common and most important element in all of these arguments is the idea that a word gets its meaning on-the-fly by some mental process in which the most general application of the statement in which it occurs completely determines the meaning of the word. We all know that language does not work like this. Words do have persistent meanings and connotations, and they are not just blanks which acquire all their meaning through some quick inferences about the application. In any case, I want to point out how dangerous this may be as a precedent for future discussions about weightier matters of Biblical interpretation. Once the idea has come in that Greek and Hebrew words can actually be defined according to the broad and indefinite notion of “application” used in the recent discussions of aner, the distinction between semantics and pragmatics has broken down, and we are in danger of losing our ability to do philological work in a rigorous manner. Habits of true and accurate exegesis depend upon our determination to resist fallacious arguments like this, even when they provide a way to justify renderings which are inoffensive to modern ears.

We should observe, in closing, that the old-fashioned style of preaching which directly addresses the men of a mixed audience carries with it a certain ideological freight—namely, an implication that the men and their response to the message have some priority in the mind of the speaker. This is no meaningless accident of discourse. The response of the men counted for more because they were the heads of their families. And when modern translators of the Bible wish to erase this feature, that is no accident either.

Michael Marlowe
May 2005

1. The Brown-Driver-Briggs Lexicon is not sufficiently precise on this point. David J.A. Clines writes, “the earliest uses of pesel are apparently exclusively of images that are hewn or cut, and it is at a later date (the time of Jeremiah, Deutero-Isaiah), that pesel can mean ‘an idol made of metal’.” [Clines, “The Etymology of Hebrew Selem,” in On the Way to the Postmodern: Old Testament Essays 1967-1998, Volume 2 (JSOTSup, 292; Sheffield: Sheffield Academic Press, 1998), pp. 577-84]. The same opinion is given in the Koehler-Baumgartner Hebrew and Aramaic Lexicon of the Old Testament (Leiden: E.J. Brill, 2001), vol. II, p. 949: “originally a divine image carved from wood or sculpted from stone, but later cast in metal.” Clines explains in a note, “in the only passages in which the composition of pesilim is mentioned (Deut. 7.5, 25; Isa. 30.22), it is clear that they are made of wood, and overlaid with gold or silver, and other references to pesilim are intelligible on this view. There is perhaps a case for understanding pesilim as metal images in the late passages 2 Chron. 34.4 (Josiah ‘broke in pieces [ibbar] the Asherim, the pesilim and the massekot, and he made dust of them’), and 34.7 (‘he beat the Asherim and the pesilim into powder’). But ibbar may be used of wooden images as well as of metal ones (cf. Isa. 21.9), and perhaps it was the silver and gold coating that was reduced to ‘powder’ and ‘dust.’”

2. Spurgeon’s sermons were published in the sixty-three volumes of the Metropolitan Tabernacle Pulpit, between 1855 and 1917. Many sermons from this series are now available online in the “Spurgeon Archive” at In a typical sermon, Justification by Faith Illustrated by Abram’s Righteousness (6 Dec. 1868), he addresses the congregation as “Friends” twice, “Beloved” six times, and “Brethren” seven times. I also note that the subjects treated in these sermons, and the manner in which they are treated, are often not very well adapted to a female perspective. In Spurgeon’s “Justification by Faith” sermon he exhorts his hearers to cultivate a “manly” faith in God. This tendency to focus on the men in the audience was not limited to Christian preachers, but also customary in Jewish preaching. For example, in a sermon preached in 1861 by Rabbi Bernard Illowy of Baltimore, we find that he addresses the audience in the synagogue with “my friends” four times and “my brethren” seven times.

3. See the many biblical examples of this tendency to address only the men of a mixed assembly cited in my article The Gender-Neutral Language Controversy, under the heading “The Patriarchal Bible Problem.” The speeches of Moses presented in Exodus and Deuteronomy provide some especially clear examples. The same thing is regularly seen in speeches attributed to Greek and Roman speakers in secular sources. Among other reasons, political orators would have addressed their speeches to men because only free adult male citizens were entitled to vote. Properly speaking, an “assembly” in ancient Greece and Rome consisted of the male citizens. Women (along with minor children and slaves) were excluded from all decision-making bodies. This was generally the case in all civilized countries up until the middle of the nineteenth century. In 1816 the American statesman Thomas Jefferson could still write: “Were our State a pure democracy, in which all its inhabitants should meet together to transact all their business, there would yet be excluded from their deliberations, 1. infants, until arrived at years of discretion. 2. Women, who, to prevent depravation of morals and ambiguity of issue, could not mix promiscuously in the public meetings of men. 3. Slaves, from whom the unfortunate state of things with us takes away the right of will and of property.” (The Writings of Thomas Jefferson, ed. Paul L. Ford, vol. 10 [1899], pp. 45-46.)   In connection with this it may also be noted that authors who write on weighty subjects have often assumed that their readers are men. Thus, for example, Adam Smith writes in his essay on The Theory of Moral Sentiments (1759): “To talk to a woman as we should to a man is improper: it is expected that their company should inspire us with more gaiety, more pleasantry, and more attention; and an entire insensibility to the fair sex, renders a man contemptible in some measure even to the men.” (Section II, capter 1.)

4. I notice that another article on the web makes this same mistake. Ann Nyland, an Australian feminist author, wrote a rather shrill article entitled “Against Grudem: Aner and Masculinist Misprisions of New Testament Meaning” that was published in the online journal Sea Changes: The Journal of Women Scholars of Religion and Theology vol. 3 (December 2003). In it she writes, “Among the Biblical references cited [in the BDAG lexicon] is Sirach 27.7, which has a parallel between the two lines. In the first is aner and in the second is anthropos, both in the plural: ‘Praise no people before they speak, as it is in this way that people are tested.’” Like Wallace, she does not mention what edition of the LXX she is using, and she does not mention a variant reading. This peculiar agreement with Wallace in what seems to be an error, together with other verbal resemblances between Nyland’s article and Wallace’s, suggests to me that the two articles are related in some way, although neither mentions the other. I will not interact extensively with Nyland’s article because it contains (in addition to the fallacious arguments dealt with here) so many false and misleading assertions that it does not seem to be worthy of serious attention. After her brief discussion of the BDAG citations (which she misunderstands) she gives a long list of other citations in ancient literature as “further instances of the inclusive meaning of aner,” in which first of all she declares that “the goddess Athena refers to herself as aner in Aeschylus, Eumenides, 911.” But when we consult the text of Eumenides we find that in line 911 Athena says, “for I cherish, like a man who is a gardener (ἀνδρὸς φιτυποίμενος δίκην) the race ...” Here Athena is comparing herself to a man who works as a gardener, by occupation; but it is not true that she “refers to herself as aner,” as Nyland claims. Her other citations are no more convincing than this one.

5. David A. deSilva, Introducing the Apocrypha (Grand Rapids: Baker, 2002), pp. 181-2.