Gender Representation in Ancient Hebrew

    David E. S. Stein is a careful researcher of gender representation in ancient Hebrew. He has published a number of ground-breaking articles in the field. He is also well-known for his  gender-sensitive adaptation of the NJPS translation of the Torah , which I discuss here, here, and note this as well. He just presented at a NAPH meeting on the subject: I link to a video slide version of the presentation below. In a series of posts, I will reproduce and interact with David's responses to a few pointed questions I threw into his lap. 

Q #1: If I boarded the elevator in the lobby of the Empire State Building, and rode to the 80th floor with a fellow Hebrew scholar who asked me to state the rules of usage that apply to the use of grammatical gender in ancient Hebrew, I would say the following in the 45 seconds at my disposal. (1) If the subject or subjects of a verb are exclusively masculine, the gender of the verb must also be masculine. (2) If the subject or subjects are feminine, the gender of the verb must also be feminine. (3) If the subject or subjects of a verb comprise masculine and feminine of a given species, the gender of the verb will be masculine, unless the verb has an explicit compound subject in which one of these subjects is to be spotlighted, in which case the gender and number of the verb will agree with the subject to be spotlighted, not the gender and number of the compound subject. (4) If the grammatical gender of a noun is feminine, but the social gender of the referent subject is masculine, the gender of the verb will be masculine. (5) I can't think of any examples offhand of the opposite, in which the gender of the noun is masculine, but the social gender of the referent subject is feminine. 

You will notice that I haven't used up my entire 45 seconds. What would you add to the above? Where do the above statements stand in need of correction? 

    Here are David’s first seven responses to this question (more to follow):

1.  At this point in my research, I avoid elevators whenever possible and take the stairs instead! I am so aware of complexity that I nearly despair of being able to succinctly articulate what I observe. But I’ll try. . . .

2.  Let’s explicitly limit ourselves to speaking only about references to persons. (References to animals and to inanimate objects do not necessarily follow the same rules. They are not burdened by our interest in the referent’s social gender.)

3.  Your starting point appears to be the extralinguistic reality that the text designates by its wording. You are asking how the speaker or writer of biblical Hebrew would make reference to a given person with a known social gender. What I try to do instead is formulate the rules from the audience’s perspective -- to start with the text (linguistic expression) as the given, and then ask what it is (and is not) saying about the referent’s social gender. That’s what I mean by the term “referential gender”: what the expression is saying about the referent’s social gender.

4.  Because the terms “feminine/masculine” are so ambiguous, lately I prefer using the terms “marked/zero-marked” to describe syntactic gender, versus “womanly/manly” to describe referential gender (as distinct from grammatical gender).

5.  The key factor in referential gender is often overlooked: the reference’s specificity. We must distinguish between two basic kinds of reference: categorizing versus identifying. With any linguistic reference, the audience must disambiguate whether it is referring to a category of persons (“anyone who fits the description” -- what I call a categorizing reference) or to a particular person (what I call an identifying reference). Some expressions can be employed to make either type of reference; it depends on the situation and shared knowledge.

6.  So here is the crucial addition that I would make to your elevator rules: If the reference’s wording (e.g, the verb) is syntactically zero-marked (masculine) and the reference is categorizing, then the referent’s social gender is not specified. The reference per se is gender-neutral or gender-inclusive.  

7.  For that reason, the mental leap from linguistic realm to extralinguistic (personal characteristics) needs to be explicit. Yet in some of your assertions, it’s not clear. For example, "If the subject is exclusively masculine...” Better: “If the social gender of the syntactic subject’s referent is exclusively manly.” Your way of stating it is simpler, but unfortunately such imprecision can easily lead us astray.  

    Since it’s “Summertime, and the livin' is easy, fish are jumpin' and the cotton is high” (Gershwin), it may not be easy to distract fellow Hebraists into joining a conversation on this topic at this time. On my part, I would respond to David’s points along the following lines.

    First of all, I would say that gender representation in language, like gender construction in culture, is both extremely routinized and nonetheless subject to subversion and inversion in specific cases. I remain interested in stating rules that apply to gender representation, not only in reference to the social gender of persons, but to that of animals and abstract concepts such as wisdom and foolishness. It is a linguistic universal that so-called inanimate objects are endowed with social gender by those who speak of them. As much as we might want to draw a sharp distinction between grammatical gender and social gender – and we need to as linguists - the tendency of language speakers, individually and collectively, is to conflate the two for a variety of ends. Even in a language like English, which is impoverished from the point of view of marking gender, overt marking of gender, albeit “unnecessary” for the “normal” needs of communication, is still common. That is why in English a car becomes a she in a variety of contexts or genres. In ancient Hebrew, attention to the metaphoricity of grammatical gender pays dividends, both from the point of view of a performer of ancient Hebrew, and a spectator. 

    It is true that I stipulate rules from the point of view of a performer of ancient Hebrew, not a spectator, and that I have social gender in mind, in reference to persons and to members of a species, in my rules (1)-(3), and social gender-by-convention in mind in reference to “inanimate” objects. In my view, it is of great heuristic value to do so. I realize that it is contrived to do this, but no more and no less contrived than to stipulate rules from the point of view of a spectator of ancient Hebrew. I see David’s point about distinguishing between social and grammatical gender. What is of interest is the interplay of social and grammatical gender, not only at the level of syntax, but at the level of realities and categories as varied as ant, god, and wisdom.        

    Finally, I wonder about David’s “crucial addition” to my elevator rules. My problem is that I am not sure it is practically true that speakers or listeners think in terms of the binary David proposes, categorizing versus identifying. I imagine that speakers and listeners think more or less exclusively in terms of identifying references. Still, I agree with David that the social gender of a referent cannot be read off from grammatical gender, because of what I refer to as the “coed” goes to “masculine” rule. One might pose the question in this way: If I am speaking about a category of individuals that are all male or who will be assumed to be all male under normal circumstances, do I use “masculine” verbs in concord with that fact or do I use “zero-marked” verb forms for lack of an alternative? Put another way, is the “zero-marked/marked” binary an etic distinction, or an emic distinction?

    For David Stein's narrated slide show, based on his NAPH presentation, entitled “Meaningful Manipulations of Grammatical Gender: Explaining a Set of Exceptions to So-Called Masculine Precedence in Biblical Hebrew,” go here.

Comments

[David E. S. Stein]

John, you wrote: "I imagine that speakers and listeners think more or less exclusively in terms of identifying references." Rather, I submit that listeners constantly evaluate the specificity of the speaker's references, but usually we do so quite automatically, without reflecting on the possibilities. Consider this scenario:

I'm in an outdoor supply store with a puzzled look on my face. A salesperson comes up and says, "May I help you?"
I reply, "I'm looking for a tent...."
"Sure, what kind of tent? Three-season or all-season? To sleep how many persons?"
"No, I'm looking for the Marmot Sanctum. It was right here last week but now I don't see it. Are you out of stock?"
"Actually, that model has been discontinued. But I can show you another one like it."

She quite reasonably assumed that my statement "I'm looking for a tent" was a categorizing statement, because people say such things all the time. As soon as I clarified that I had a particular model in mind, she shifted her thinking and responded accordingly.

[John Hobbins]

David,

Now I get you. Excellent point.

I'm still not sure how that works for gender at the intersection of grammatical and social gender. I think it depends on context. For example, if I am supervising a day camp for children in Italian, a camp that includes both boys and girls, and I want to get the boys together apart from the girls, I can say, "Portami i ragazzi prima di tutto" (i and ragazzi are masculine plural) "Bring me the boys first of all" and I am making an *identifying* reference, just as I would be if I said instead, "Portami le ragazze prima di tutto" (le and ragazze are feminine plural) "Bring me the girls first of all."

Do you see my point now?

[Steve Pable]

For my part, I'd probably be looking for a different elevator...

;)

[Abram K-J]

Indeed, it is summertime, and I've just come back from a great cookout with friends. Reading this post has made for a great nightcap.

I appreciate (in point #4) the desire to clarify beyond the ambiguous "masculine"/"feminine"... but might there be a risk with “marked/zero-marked” of perpetuating the idea of masculine as self/subject/standard and feminine as other/object/deviant? (even though this is referring to syntactic gender specifically)

In other words, couldn't one view calling masculine syntactic gender "zero-marked" as potentially hegemonic? Surely there is social hegemony; can there be syntactic hegemony? And could the latter (even if unwittingly) produce the former?

I realize this comment could easily come across as satire, could be easily mocked if one were so inclined, etc. But I do ask the question sincerely.

[Mitchell Powell]

In Hebrew, as in Spanish, it is generally true to say that the grammatical masculine is "zero-marked" and that the grammatical feminine is "marked."

I could entertain the notion that, just maybe, the way grammatical gender is used normal spoken Hebrew or Spanish speech does something of some kind to help enforce certain notions of gender. (Whether this is for better, for worse, or for both is yet another question.)

But the use of the terms "zero-marked" and "marked" as they are used in this article are simply descriptions of what is going on from a linguistic standpoint. If we must tiptoe around describing how language actually works even in in-house linguistic discussions, then we risk seriously compromising the integrity of the social sciences.

[David E. S. Stein]

Abram, I think you've raised a fine question. My answer is: Let's understand the term "masculine" as referring merely to the prototype of the various ways that we use the zero-marked form in Hebrew. It's only typical — not comprehensive! For we use that same form also to refer to someone whose social gender is unknown, or whose sex is ambiguous (hermaphrodite) or indeterminate, or when the collective is of mixed gender, or when gender is irrelevant. So it's more an all-purpose or generic way of speaking/writing about persons than it is a "male" way.

In short, to call it "masculine" is a bit misleading because that obscures the asymmetry of how gender references work in Hebrew.

By the way, of the many languages that have two grammatical genders, there are a few that reserve the "masculine" gender solely for references to manly persons. Any of the other referents that I mention above are expressed via the "feminine" gender. But if you think about it, I bet you'll see that this arrangement is not per se any more or less sexist than the arrangement in Hebrew or Spanish. (For those languages, one could say: "Men are so special that they get their own gender form all to themselves!")

Finally, it's worth noting that the concern for "masculine" hegemony in the Hebrew language has been asserted (and decried) with extreme seriousness by a few feminist scholars, such as Athalya Brenner. Their critique is independent of whether the form is called "zero-marked" or not. Personally, I find their claim to be based on midrashic reasoning, for the reasons stated above.

[David E. S. Stein]

John, can you come up with a sentence in Italian that employs i and ragazzi in a categorizing reference?

Back to Hebrew Bible: In Alison Grant's 1977 analysis of how the words אדם and איש are used to make reference, she pointed out something remarkable: The word אדם is almost always used in categorizing references (more than 70% of the time). The only exceptions are the references to the same special individual: the mythical progenitor of the human species. And those come pretty close to being categorizing references, given that they're used to identify the prototype human being, who represents us all (from a social perspective, not a biological one). Arguably it's the non-specific way that it's most often used to make reference that gives the word אדם its generic feel.

Much the same can be said of איש. Grant found that in only 20% of its instances is it employed to make an identifying reference. (My terminology, not hers.) Unfortunately she didn't publish her data set, but the results seem pretty robust in their lopsidedness.

My point is that the Bible makes categorizing references much more often than most of us realize.

[John Hobbins]

Hi David,

There are many instances in which "i" and "ragazzi" in Italian are categorizing as opposed to identifying references. The same is true no doubt for masculine gender plural nouns in Hebrew comparable to ragazzi "boys" [plural]. Context is the only safe guide to knowing what is going on, on a case by case basis.

Conversely, in Hebrew and other inflected languages I know, masculine gender singular nouns like ragazzo "boy" are almost always deployed in order to, among other things, identify the gender of the referent.

Different again is a word like uomo, "man." Even more so than masculine gender plural nouns, whether or not it serves to serves to categorize or to identify gender depends on context. It would be wrong to say that there is a default sense that uomo "man" carries. I feel the same way about איש and אדם in Hebrew.

The complexity of the issue in a specific case may be illustrated by Psalm 1. Does the use of איש serve to identify gender or to categorize according to species? One might reply that it serves to do both in context, in the sense that איש in the Psalm is never an איש in the abstract but an איש who does and does not do certain things.

The things the איש is said to do and not do in Psalm 1 apply readily to the kind of things a literate "loya jirga attending" Israelite freeman might do and not do, and only by extension to the kind of things a female Israelite, a slave, or a non-Israelite might be expected to do.

In a case like this, I wonder how helpful it is to say that איש serves to categorize, not to identify. The same thing applies to awilum "man" in Hammurapi's law code; awilum serves to categorize in shumma clauses. But that seems to capture only a fragment of what is going on. It serves, at the same time, to identify.

[David E. S. Stein]

John, it seems that we're using some words differently, so let my try to clarify what I mean.

• Some identifying references with איש (pointing to a particular individual):
‏וַיִּגְּשׁוּ אֶל־הָאִישׁ אֲשֶׁר עַל־בֵּית יוֹסֵף (Gen. 43:19)
‏זֶה מֹשֶׁה הָאִישׁ אֲשֶׁר הֶעֱלָנוּ מֵאֶרֶץ מִצְרַיִם (Exod. 32:1)
‏לֵךְ וְאַרְאֶךָּ אֶת־הָאִישׁ אֲשֶׁר־אַתָּה מְבַקֵּשׁ (Jud. 4:22)
‏אָרוּר הָאִישׁ אֲשֶׁר בִּשַּׂר אֶת־אָבִי (Jer. 20:5)

• Some categorizing references with איש (applying to whoever fits the description):
הָאִישׁ אֲשֶׁר לֹא־יִשְׁמַע אֶל־דְּבָרַי (Deut. 18:19)
הָאִישׁ אֲשֶׁר אַתָּה נֹשֶׁה בוֹ (Deut. 24:11)
‏מִי־הָאִישׁ אֲשֶׁר בָּנָה בַיִת־חָדָשׁ וְלֹא חֲנָכוֹ (Deut. 20:5)
‏הָאִישׁ אֲשֶׁר לֹא הָלַךְ בַּעֲצַת רְשָׁעִים (Ps. 1:1)

• I did not claim that איש itself categorizes, rather that in the Bible it is employed mostly in order to categorize. Referential specificity is not a property possessed by words or expressions. Rather, it depends upon how those expressions are used, which the audience then interprets based upon the discourse topic, the situational context, and social convention.

• A speaker could potentially refer to the same referent by any of several designations. The speaker chooses one such designation in order to highlight some particular aspect of the referent. Regardless of the type of reference, איש designates its referent in terms of a relationship. So do the nouns בן and עבד. In some settings, all 3 nouns can more specifically indicate subordination. Those are their semantic components, which is distinct from the nature of the reference in which they are employed.

• All references in law codes are categorizing, by definition. The apodictic laws refer to "you, if you are in this situation." The casuistic laws refer to "anyone who is in this situation." The designations used may be in terms of a particular social status, such as עבד or גר. But those designations don't point to a particular individual, which is how I'm using the term "identify."

[Abram K-J]

David, thanks for your reply. Makes perfect sense.

[Mitchell Powell]

Now, what would really be interesting here would be to see you, John, or anyone else who wants to take a stab at it, give a comparable "elevator speech" on the verbal "tense" system of biblical Hebrew.