I keep finding myself looking at questions involving the concept of gender neutrality of words or phrases. (Check out the RELATED panel down to the right there.) "Craftmanship" for example - is it gender neutral or not? I'm frustrated because the discussions often seem aimless, anecdotal, or like a kind of culture-skirmish. In the answers to these questions I see various kinds of evidence offered:

  • anecdotal usage ("It sounds fine to me")
  • subjective opinion ("I don't like to say..." or "I'm looking for an alternative...")
  • etymology ("The root word is 'craftsman'")
  • diachronic evidence ("It originally meant...")
  • armchair linguistics - "Imagine you're at a talk and the speaker says..."

For acceptability, we can cite dictionaries, or if a usage is new, cite contemporary writing or a Google search. What would constitute objective or at least 'citable' evidence for or against the gender-neutrality of a word? I'm not asking for specific cites or websites (but that'd be fine), but even kinds or classes of evidence ("published sociological studies" or "brain scans"!) Otherwise it's just a bunch of people clashing their assumptions together!

  • I think you need to ask a more specific question, or this is likely to get closed as "Too broad".
    – Barmar
    Oct 31, 2014 at 18:43
  • 1
    Well, these type of questions are already abuzz with opinion, or opinion hiding behind uncompelling evidence. I'm looking for some way to reduce that.
    – Spike0xff
    Oct 31, 2014 at 19:12
  • 1
    I agreed with everything in your question until your final statement. I don't think "word geeks" argue about this so much; these arguments are more for people with a stake in the political correctness game. :^)
    – J.R.
    Oct 31, 2014 at 19:52
  • 1
    Gender neutrality or the lack of it is a provable fact about a word. Bride and Groom (in a wedding context) are not g.n. Ballplayer is still g.n even if 95 percent of them happen to be male. Man in the context of the race is g.n. too, since more than half the race is female.
    – Oldcat
    Oct 31, 2014 at 22:14
  • 2
    There's been a request to move this question to Meta. I support this idea, but I'm ambivalent enough to leave it to the community to decide. I think there is something to be said for having clearer standards or at least an agreement about how to make these kind of judgments.
    – Kit Z. Fox
    Nov 2, 2014 at 20:37

3 Answers 3


I originally had this as a comment, but I think I'll expand it out.

Google's N-Gram viewer is an excellent opportunity to see not just how a word was used throughout history and how it is used now, but it can give an insight into the current trend to allow us to see how the gender association for a word might be down the road.

I'd say there's four general trends that we might see:

  1. Single-gender consensus. For example, the word actress is undoubtedly femenine (he is/was/'s an actress registers exactly 0 hits). Adam's apple is going to be intensely associated as masculine in spite of being used occasionally for women.

  2. Cross-gender consensus. For example, calling someone sexy is quite evenly distributed and has been since its introduction. Notice that I used several different forms of the phrase. When I first searched with just he/she is sexy, it swung in favor of women but still registering a quite sizable (33%ish) of masculine uses. Being completely unscientific and arbitrary, I'd sayif after including as many variations as feasible, if a word is in the 15-85% range for one gender's use (hence 85-15% for the other), we can conclude that it can be safely used for both genders.

  3. Converging use. This would be a term that used to be heavily gendered, and is now seeing increased use for the other gender and, based on the trend lines, showing an increasing amount of parity in recent times. For example, with nurse or actor. While it may strike some or many as odd the use with either gender, that is the direction these words are going towards.

  4. Divergent use. These are words that may have been less gendered in the past, but now are trending toward gender-exclusivity. While I'm sure there are others, the best one I can come up with is handsome. Since 2000 the female use has a marked uptick after a steady trend to obscurity, though the gains on the male side are larger. These are still usable for both genders, but maybe start to sound strange to some when used with a particular gender (I know most of my students find it quite odd to hear a woman called handsome).

Sometimes the graphs aren't perfect. For example, we can all agree (I hope) that wrestler can be appropriately used with each gender. That said, none of the N-Grams I tried could register a use with women — only men — until I tried male/female wrestler. Bingo, huge surge for female. Since clearly it was a male dominated term, we can see its close gender association from in the past, such that people felt the need to prefix female in front of it. The same happen with nurse where we can also see after the idea of male nurses caught on, there was a reduction in usage of the term with a continual increase in the use of female nurse.

Context of course is lost, so jocular vs serious uses might not be easily distinguishable (calling a guy vs gal "pretty" for me, at least, has vastly different connotations). But it at least gives some data to back up (or not) what might be found in dictionaries whose lexicographers presumably have already done much of the work to determine the gender associated with words in actual use.

  • The problem with Ngrams is that they are unreliable reference for everyday speech. Street slang, expressions, phrases etc can be around for years before anyone gets round to writing them down, and it could run into decades before a "dictionary" actually includes that term in its pages. What is accepted as current speech is not necessarily reflected in books and journalism.
    – Mari-Lou A
    Nov 1, 2014 at 8:57
  • @Mari-LouA this is absolutely true. But if a word has never registered in a book under a male or female context, it's unlikely that the majority of speakers today would find it readily acceptable. Most of the questions have revolved around formal terminology which would be more likely written down first (for example, I've seen sportswomanship written once or twice, but never in my life heard it spoken, despite being an athlete my whole life in a 50/50ish coed sport with 75/25 female/male coaches). Nov 1, 2014 at 12:52
  • 1
    I'm not saying NGrams definitely prove anything, but they can provide a good base of hard evidence that, like anything else, must be contextualized and evaluated given historical and social backgrounds. Otherwise, using the nurse or wrestler example, we might presume that there were more male nurses for a few decades or more female wrestlers. The context can explain why that is, while the NGrams demonstrate evidence of actual accepted use. Nov 1, 2014 at 12:54

One man's meat is another man's (or woman's) poison.

In one sense, it comes down to what you can get by with without your audience being distracted by the metalinguistics. (Unless the subject IS metalinguistics.) If the audience is easily distracted by your use of polarizing language, you have no choice but to submit or be damned.

You've listed all of the cases where correctness is driven by subjective opinion or hidebound dependence on authoritative resources and reasoning. This can be, for all intents and purposes, ignored by the ones you want to communicate with.

One group will castigate you for not being more forward thinking about gender neutral language and another group will criticize you for going too far. Either way, if they want to, they will see red meat or poison if you make the wrong calculation. And then you have lost them.


I would say that the best evidence is anecdotal evidence. Why? because that's how words and phrases are actually used now. Current usage always trumps any dictionary because dictionaries are out of date by the time they are published because a language will have changed during the publishing cycle. Dictionaries, like grammar, are descriptive rather than prescriptive. They describe a language, but they don't define it.

So you can use dictionaries, but the best information will come from Google.

  • 4
    W e l l . . . An anecdote may be evidence of misuse, too, and in many cases the worst information will come from Google as well. Meanings must be decoded as well as encoded. Oct 31, 2014 at 19:11
  • I'm feeling foolish already, which I wish didn't happen so often on SX. For 'craftmanship' I should have started with a survey of dictionary definitions! If they don't refer to men, then that's one kind of objective evidence that the word is gender-neutral.
    – Spike0xff
    Oct 31, 2014 at 19:15
  • 1
    You look for a concensus among the anecdotes. Currently nucular is considered an incorrect pronunciation of nuclear, but if everyone starts saying it, it becomes correct (I'll bet there was a time when people pronounced February and Wednesday the way they're spelled).
    – Barmar
    Oct 31, 2014 at 19:15
  • 1
    I'm understanding this answer to be: Survey dictionaries first, and then to make sure current usage hasn't superseded them, look for consensus in a suitably broad sample of actual current usage. And Google's your friend.Yeah, I'm afraid someday nucular will be the preferred pronunciation ;-)
    – Spike0xff
    Oct 31, 2014 at 20:54
  • @Spike0xff I think even something as straightforward as dictionary survey will leave you with trouble if you do it on a level applicable to human thought. To wit: In the case of “craftsmanship” I don’t know of a definition that names a gender directly, but there is at least one dictionary that refers to a “craftsman” in the definition (following the transparent derivation) which is then defined as “a male artisan”. Nov 5, 2014 at 18:57

Not the answer you're looking for? Browse other questions tagged or ask your own question.