Some anecdotes I remember regarding American English: men pose fewer questions, are more likely to use double negative forms, and use the passive voice less. Many word choices are somewhat gendered as well. One reply gives the obvious example of "lovely" being much more commonly used by women, but many words are like that. Even a simple choice like which empty qualifier you use -- "drastically", "dramatically", "enormously", "badly", "awfully" -- can apparently tip the scales one way or the other. (I don't know which is supposedly which for those!) These are small differences individually, but summed up they apparently become considerable.

Even as a non-native English speaker I can tell you off the top of my head: "lovely" in the American culture is predominantly used by women (though in the British culture exceptions can be made more often). But I'm sure there are a lot of more subtle differences too.

British culture is very much not homogeneous in this regard. The use of words like lovely probably tells a British speaker more about the geographical origin of the speaker than their sex.

Here's an essay exploring that

https://www.diva-portal.org/smash/get/diva2:1023448/FULLTEXT...

There is a site I'd used a while back for exactly this when I was trying to ensure that my writing (and speech) in some fiction was gendered in a particular manner - https://www.hackerfactor.com/GenderGuesser.php#Analyze

Playing "guess why" is the fun part :) I'm sure there are other tools available now and I'd love to hear of them if people are aware of them!

I don't speak any Spanish or Italian, but I had French in school (it's been a while) and I presently cannot recall how the gender of the speaker is more obvious than in, say, English.

I'm curious, could you give me an example please?

First very simple example that comes to mind, the sentence "I'm happy": in English, the speaker's gender is not obvious.

In French, you will say "Je suis content" if you are male, and "Je suis contente" if you are female.

The trailing "e" change the pronunciation: in the male variant, the trailing "t" is silent, while it's not in the female variant because of that "e".

English kind of has traces of this in eg “I’m an actress/waitress/stewardess” (vs actor/waiter/steward) although gendered role-terms like those are less common and tend to get phased out in favor of gender-neutral ones (“server” replacing waiter/tress and “flight attendant” replacing steward/ess). And I wouldn’t consider it incorrect or out of place for a woman to say “I’m an actor”.

Same in Spanish

"Mon amie est enseignante. Durant l'été elle s'est engagée à se lever plus tôt pour la nouvelle année scolaire."

- "amie" is feminin of "ami"

- "enseignante" is feminin of "enseignant"

- "elle s'est engagée" is feminin of "il s'est engagé"

Edit: note that the concept of "gender" in indo-european languages is only vaguely related to the concept of "gender" as in human sexual expression. For example "la nouvelle année scolaire" ("the new school year"), "année" is considered feminin, so "nouvelle" has to be adapted to the feminin, instead of the masculin/neutral "nouveau". But that of course doesn't make any sense if you only consider "gender in languages" to be about male/female.

> - "amie" is feminin of "ami"

But this relates to the gender of the friend, not of the speaker.

Yeah, a better example would be something describing their speaker. For example: "je suis américain" vs. "je suis américaine," (I'm American) the latter being the feminine form of "American."

I am not a French speaker by any means so feel free to correct my words.

> feel free to correct my words.

There is nothing to correct, you’re right ;)

The principle is similar for the speaker. Anything the speaker would say about themselves would have to be gendered accordingly.

I don't speak French, but I'm a native Spanish speaker and I have two observations:

First, it's true that Romance languages use gender in this manner. I was just pointing out that the specific example chosen didn't show this. The example has nothing to do with the gender of the speaker, but with the gender of the object under discussion; it's similar to you saying "she" or "her" when referring to a woman; it enables me to make no assumptions about your own gender.

Second:

> Anything the speaker would say about themselves would have to be gendered accordingly.

In Spanish there are careful ways to avoid making assertions about yourself in a gendered manner. They are a bit more roundabout, but can be done. My point is not that this is the most natural way of speaking, just pointing out that it's technically not true that you must always make gendered assertions about yourself.

Example:

Instead of saying "I'm very tired", "estoy muy cansado" (gendered male language) I can say "tiredness overcomes me", "el cansancio me invade", hiding my gender. Yes, it's a bit artificial but not so uncommon I have never read it. So in Spanish you can make assertions about yourself without betraying your gender. Don't know about French or other Romance languages, but it wouldn't surprise me to know you can employ similar tricks. It's probably harder in Spanish than in English to hold a conversation longer than a couple of sentences while hiding your gender; it will start sounding very unnatural, I grant you this.

I guess you could do it to certain extents, but it’s likely to be considered incorrect French, or just really weird. Now if you’re writing a song or a poem, it’s more common to see that kind of construct, as a way to play with rhythm and sounds, so it’s not entirely impossible.

Many adjectives in Spanish have gender, so for example:

I'm tired (male) -> estoy cansado

I'm tired (female) -> estoy cansada

In French it works the same way, although I'm not confident enough in French to give you an example without making horrible spelling mistakes. But the idea is the same.

For French:

I'm tired (male) -> Je suis fatigué

I'm tired (female) -> Je suis fatiguée

The French passé composé is funny if translated word for word into Portuguese:

Je suis tombé -> Eu sou caído (or Eu sou tombado) Je suis tombée -> Eu sou caída (or Eu sou tombada)

Je suis tombé is passé composé. Je suis fatigué is présent de l'indicatif. Fatigué can both be an adjectif and a participe passé. The verb fatiguer uses the avoir auxiliary verb when using passé composé however: ils m'ont fatigué / la poutre a fatigué sous le poids.

The passé composé of "je suis fatigué(e)" [I am tired] would actually be "j'ai été fatigué(e)" [I was tired]. Translated word for word to Portuguese, this would be "Eu tive sido/estado cansado(cansada)".

In Italian some tenses are gendered. They are more or less the equivalent of english perfect tenses, like "Io sono andato"/"Io sono andata" (I have been to) "Io sarò andato"/"Io sarò andata" (I will have been to)

While you can certainly use statements like that to adjust your priors, be careful not to make absolute assumptions - in the US approximately 1% of people who have a wife are women.

If your machine learning model had a indicator that could point to something with 99 percent accuracy, you would tell it to ignore that?

Well, no. And that isn’t what I said, is it?

I would say that an assumption with 99% accuracy is a pretty good one.

Just to clarify to help avoid some awkward moments...

In the U.S., the normalization of gay marriage has involved a lot of cultural strife over the past few decades.

One may still encounter cliques in the U.S. that refuse to accept grammatical constructs such as "her wife" for various reasons.