Re: Neutral language in W3C specifications

I recently was working through some use cases with a group where instead of using names or gendered pronouns, I used “Person A” or roles as identifiers (i.e. “Participant” or “Engineer”). One thing it provided was a way to keep the scenarios neutral, which was key to the discussion we were having, but there was another element that I think comes in handy, we were able to question bias.

Singular “they” has been in use in the English language for a lot longer than most people give it credit (records on this go back to the 1300s!) [1], where people today trip on it is where it’s used as a pronoun. But for speakers of English (native and otherwise), it’s long been a comfortable alternative to gendered pronouns where you don’t know the gender of the person you’re speaking about. I know that this is a more significant struggle for more gendered languages, I’ve been researching French and German approaches to this problem, and it’s something every language is trying to figure out, even when the grammar is more challenging.

I understand wanting to make use cases and scenarios more personal with names and genders, to give the reader or user a lens through which to hopefully develop empathy for the case. I think, based on the discussion here, we have a few options, with varying degrees of difficulty to implement.


  1.  We continue the practice of using names, gendered pronouns or other identifiers, and in the spirit of inclusivity, we also include examples using neutral pronouns (they/them and other variants). To ensure this practice is equitable, groups would need to confirm they’ve analyzed all of the use cases for equal use of the identifiers, and they’ve considered bias in their selection of scenarios. (NB: IDCG is looking at creating an Equity Review process, which could help groups with something like a checklist or similar self-check system).
  2.  We move to a practice of gender-neutral use cases and language, stripping out any mentions of gender or gender-identifying language. This would use the singular they or identifiers like “Person” or “Individual”. A good case for this is made by the APA [2].

I lean towards the second for the purposes of equity, but also because as I mentioned before, it opens up an opportunity to have some hard discussions about bias. When we ascribe gender to a scenario, even if it’s offhandedly, unconscious bias can creep in. By removing that factor from scenarios, it opens up the chance for groups to ask themselves, “how did you visualize this when you read it? Did you picture someone in particular? What is the first thing that comes to mind?”. From those conversations we can build stronger, more equitable use cases and specifications.

Language is ever-evolving, new words get added yearly, definitions change [3], and common usage evolves. We might have to have this discussion again in 5-10 years as the world becomes more conscious of these issues, and that’s totally ok! Part of learning to be better allies is identifying and acknowledging our own biases and working to overcome them.

I’ve also linked to a resource I found that had some more resources on gender-inclusivity I thought were useful so I wanted to share [4].

[1] https://www.merriam-webster.com/words-at-play/singular-nonbinary-they

[2] https://www.apaonline.org/page/nonsexist

[3] https://www.bbc.com/news/world-us-canada-52993306

[4] https://consciousstyleguide.com/gender-sex-sexuality/




From: "Mccool, Michael" <michael.mccool@intel.com>
Date: Thursday, July 30, 2020 at 11:06 AM
To: "Levantovsky, Vladimir" <Vladimir.Levantovsky@monotype.com>, Janina Sajka <janina@rednote.net>, Ivan Herman <ivan@w3.org>
Cc: "lwatson@tetralogical.com" <lwatson@tetralogical.com>, Gregg Kellogg <gregg@greggkellogg.net>, Denis Ah-Kang <denis@w3.org>, Spec-prod <spec-prod@w3.org>, "chairs@w3.org" <chairs@w3.org>
Subject: Re: Neutral language in W3C specifications
Resent-From: <chairs@w3.org>
Resent-Date: Thursday, July 30, 2020 at 11:05 AM

Conversely, the use of “they” as a third-person gender-neutral pronoun is becoming fairly acceptable in common usage.
Traditional prescriptive grammar – including the grammar taught to second-language learners – says it is unacceptable.   But this is increasingly no longer the case – language changes.

One place it does get odd though is in subject-number agreement, which unfortunately English has to worry about (not all languages do).   So it works ok for possessives, etc. but is harder for subjects.  The simple answer though is to replace the pronoun with the referent in this case (e.g. use “Alex” instead of “They” to avoid weird constructions like “They is*”).

Michael McCool

From: "Levantovsky, Vladimir" <Vladimir.Levantovsky@monotype.com>
Date: Thursday, July 30, 2020 at 11:54 AM
To: Janina Sajka <janina@rednote.net>, Ivan Herman <ivan@w3.org>
Cc: "lwatson@tetralogical.com" <lwatson@tetralogical.com>, Gregg Kellogg <gregg@greggkellogg.net>, Denis Ah-Kang <denis@w3.org>, Spec-prod <spec-prod@w3.org>, "chairs@w3.org" <chairs@w3.org>
Subject: RE: Neutral language in W3C specifications
Resent-From: <chairs@w3.org>
Resent-Date: Thursday, July 30, 2020 at 11:54 AM

Quoting from both references Janina provided:
1) “A person who dictates how people should write or speak is called a prescriptivist or a prescriptive grammarian.”
2) “prescriptive grammar, is the attempt to establish rules defining preferred or "correct" use<https://en.wikipedia.org/wiki/Usage> of language. … <snip, read the whole 1st paragraph> … They may also include judgments on socially proper and politically correct language use.”

Does the effort to neutralize the W3C specification language make us all “prescriptive grammarians”?
I have no doubt that the diversity of the participation in creating the W3C specifications is very valuable, but the specs themselves, as the final product of these efforts, should first and foremost serve the target audience (i.e. the implementers) and should be clear, easy to read, and as unambiguous as possible.

Vladimir


From: Janina Sajka <janina@rednote.net>
Sent: Wednesday, July 29, 2020 5:54 PM
To: Ivan Herman <ivan@w3.org>
Cc: lwatson@tetralogical.com; Gregg Kellogg <gregg@greggkellogg.net>; Denis Ah-Kang <denis@w3.org>; Spec-prod <spec-prod@w3.org>; chairs@w3.org
Subject: Re: Neutral language in W3C specifications

Ivan Herman writes:
>
>
> >
> > Doesn't lose anything if it becomes:
> >
> > "Alex is a developer who uses SVG as part of their job..."
> >
> > There are times when the gender of the person in a user story is important, but in the context of W3C I can't think of one that applies to technical standards!
> >
>
> i must admit that for a non-native English speaker the usage of the plural form as a gender-neutral pronoun sounds extremely strange in this case. I know it is coming to the fore but I am worried it would create lots of confusion. (It is certainly a usage that goes against my own English training.)
>
It goes against all our training, native or acquired English speakers.
The rule against using "they/their//them" in the singular is the legacy
of a group of academics known as the "Prescriptive Grammarians." It's
completely academic and artificial. Native English speakers have always
tended toward this usage, unless the schoolhouse knocked it out of them.

I offer two web pages with relevant explanations:

1.) https://www.thoughtco.com/prescriptive-grammar-1691668<https://protect-us.mimecast.com/s/VMkwCM82gJsq8y6EuNzvpZ>

2.) https://en.wikipedia.org/wiki/Prescriptive_grammar<https://protect-us.mimecast.com/s/x-x-CNk2jYt0xqMvslWF0b>


These are the same grammarians, by the by, who created the rule that the
English infinitive verb is never to be split. Why? Because the Latin
infinitive is never split.

Best,

Janina

> Ivan
>
>
>
> >
> > Léonie.
> >
> >> [1] https://www.w3.org/TR/vc-use-cases/<https://protect-us.mimecast.com/s/7T8xCOY2kPUpXL2qiR72Eh>
> > [2] https://github.com/w3c/idcg/issues/17<https://protect-us.mimecast.com/s/sGUZCPN9lzhKm15yu9SJ8L>
> >
> >>> Denis
> >>>
> >>>
> >>> On 7/22/20 1:21 PM, Léonie Watson wrote:
> >>>> Denis, this is a really positive step, thank you to you and the team.
> >>>> One suggestion - is it possible for PubRules to check for gender specific pronouns (he/she, him/her etc.) too?
> >>>> They are rare in specifications, but do sometimes feature as part of use cases or examples.
> >>>> Léonie.
> >>>> On 22/07/2020 09:40, Denis Ah-Kang wrote:
> >>>>> Dear editors and chairs,
> >>>>>
> >>>>> In order to offer the best environment possible to its
> >>>>> community, W3C is supporting the push for a more inclusive and
> >>>>> neutral language, especially in our specifications.
> >>>>>
> >>>>> In the upcoming weeks, pubrules [1] will show a warning if
> >>>>> terms like "master", "slave", "grandfather", "sanity" or
> >>>>> "dummy" are detected in a specification and this will also
> >>>>> be reflected in the Manual of style [2] with a list of
> >>>>> alternatives.
> >>>>> Note, since it may take time for the editors to change the
> >>>>> branch name "master" to something else, we will not flag the
> >>>>> URLs containing that word in the first place.
> >>>>>
> >>>>> Going forward, we will audit all the specification repositories
> >>>>> and open issues if they contain problematic terms.
> >>>>>
> >>>>> Let me know if you have any comments/suggestions.
> >>>>>
> >>>>> Denis
> >>>>> W3C Systems team
> >>>>>
> >>>>> [1] https://www.w3.org/pubrules/<https://protect-us.mimecast.com/s/Vj7RCQW2m6ikq1NzsVWO4p>
> >>>>> [2] https://w3c.github.io/manual-of-style/<https://protect-us.mimecast.com/s/BcQfCR60nPhvlDg7spaLQO>
> >>>>>
> >>>
> >
> > --
> > Director @TetraLogical
> > https://tetralogical.com<https://protect-us.mimecast.com/s/DmBMCVOkr7TxKnPOi82rS3>
> >
>

--

Janina Sajka
https://linkedin.com/in/jsajka<https://protect-us.mimecast.com/s/kXVKCW6lvQh5YLylsw5UzZ>

Linux Foundation Fellow
Executive Chair, Accessibility Workgroup: http://a11y.org<https://protect-us.mimecast.com/s/Okb6CXD0wrhXNPMpIgl254>

The World Wide Web Consortium (W3C), Web Accessibility Initiative (WAI)
Co-Chair, Accessible Platform Architectures http://www.w3.org/wai/apa<https://protect-us.mimecast.com/s/P7gbCYEnxytLZW6GsDv7h->