I Am Guilty of posting “Hate Speech” on social media …according to Facebook’s algorithms and whoever (or whatever) reviewed that declaration when I appealed.
The terrible, horrible, hateful idea I had posted in response to a comment: that women shouldn’t have to hide their intelligence in order to be appealing to men.
The gall. I mean, how could I be so insensitive, and insulting, and probably several other “in” words!?
Although I didn’t even call that type of insipid man a “pig” (or even insensitive and insecure) — after all, that would have — or at least should have! — triggered their algorithm banning hate speech.
But for the record, inscrutable, inscient men who assume all women are less intelligent than they are simply because of their gender are not only wrong (insensible, even!), but they are, in the long and storied vernacular of the women’s liberation movement, “sexist pigs” who almost certainly not only have a small intellect, but probably also vanishingly small (aka insignificant) penises.
The Insightful (aka Inciting) Post
It all started as part of my “Think Series” — short, stark memes encouraging more thinking in the world, in part because True’s Mission is to …Encourage More Thinking in the World. Number 6 (of 52 — one a week on Thursday evenings), posted May 13, was “Think. Because it’s sexy.”
“Sexy”? Uh oh.
Melba Jane — a woman — posted the first comment, four minutes later: “In today’s world it appears fewer and fewer are using their thought process. 🙁 ”. Dan O’Brien posted the second: “But it’s SOOOOOO hard.. Well for a lot of people it is..” And Marion Wyse posted the third: “We must have grown up in different generations, Randy!” — to which I clicked “LOL” and responded, “Shh! I’m trying to effectuate a mindset shift!”
Marion clicked “Care” (the smiley face hugging a heart) to that and responded, “To be clear, several men of my age group found me sexy … until, as one engineer put it, i opened my mouth. 😉 YOU are already indicative of the shift!” She appreciated that there are now men who encourage smart women like her to show their intellect, rather than tell them to shut up because it shows they (the men) don’t have such intellect.
I clicked “Love” on her comment, and replied, “Yeah, I’m not particularly attracted to dumb women, and definitely not attracted to smart women who play dumb.”
OK, did you catch the “hate speech” in that thread? Go back and look!
Apparently it’s “hateful” to not be “particularly attracted to dumb women” or, at the very least, “definitely not attracted to smart women who play dumb” — women who actively hide their best, most humanizing attribute in order to appease inferior men. Men who suddenly find a woman is intelligent and say they are sexy “until [they] opened [their] mouth” — they dared to say something intelligent.
I do notice something about Marion’s typing: it’s pretty typical for her to spell the word “I” in lower case. It makes me wonder if she still subconsciously tries to diminish herself a little by being “smaller” and not standing out. Couldn’t blame her if that’s so, but I find it sad that many women do that — I’ve noticed it before.
I’ll note that Facebook’s action was to delete my comment and to warn (and label) me. It did not suspend my account nor take other punitive action, as many of my real-life friends have noted has happened to them for ridiculously insignificant transgressions. This is the first time I have received such a warning in my many years of publishing sometimes very provocative things there.
“Algorithms are always unambiguous and are used as specifications for performing calculations, data processing, automated reasoning, and other tasks,” Wikipedia says (emphasis added). “The word algorithm itself is derived from the name of the 9th-century mathematician Muhammad ibn Mūsā al-Khwārizmī [a Persian polymath who produced vastly influential works in mathematics, astronomy, and geography. Around 820 CE he was appointed as the astronomer and head of the library of the House of Wisdom in Baghdad.], whose nisba (identifying him as from Khwarazm) was Latinized as Algoritmi.”
You know, just the sort of exceptionally intelligent foreigner (and an Arab at that!) who a certain class of inferior “supremacists” would find “inferior.”Anyway, algorithms work on huge data sets to try to sift out things that would normally need a human to make a judgement upon — such as “Is that comment trying to put someone down?”
Yes, I can certainly understand why Facebook’s algorithms might trigger on my comment: it’s nuanced, and subtle. But surely anyone with (hah!) intelligence can see that nuance and subtlety and “get” that I’m praising female intellect and women who display it, not dismissing the very idea.
I appealed, and four minutes later, that appeal was denied.
But let’s establish what Facebook’s “Community Standards” are regarding “Hate Speech” — since that really is a significant problem on social media.
Facebook’s ‘Community Standards’ regarding Hate Speech
The following is lengthy and is thus collapsed unless you wish to read it (click the + to expand), and don’t want to be tracked by Facebook as having sought out their policy on this matter. (Though if you do wish to, the following is copied from: Facebook Community Standards, Hate Speech.)
Note that the text of their policy would absolutely trigger their algorithms: it contains examples of the words and phrases that the algorithms are designed to detect and exclude, which means they cannot even be discussed on Facebook itself — yet another reason why I have chosen to discuss the shortcomings of their algorithms here instead of there. Interestingly, the classic “N-word” is not among the many disparaging terms explicitly called out.
In summary, “We define hate speech as a direct attack against people on the basis of what we call protected characteristics [including gender]. We define attacks as violent or dehumanizing speech, harmful stereotypes, statements of inferiority, expressions of contempt, disgust or dismissal, cursing, and calls for exclusion or segregation.” (Emphasis added.)
My comment does not meet that definition, so I appealed the automated determination of “hate speech” as applied to that comment. In other words, I hoped that a human, who has much more capacity to understand subtlety than a computer program, would review the context of the remark and understand the absolute encouragement I was expressing.
No such luck. Not that this is unusual: they screw it up all the time — a link supplied by Marion. One telling passage from that article: “Algorithmic systems lack an ability to capture nuances and contextual particularities, which may not be understood by human moderators who test data used to train these algorithms either.”
Either the same algorithm scanned it again after my appeal, or an overworked human who lacked appreciation for nuance “didn’t get it” (or just went by inflexible guidelines — you know, the sort of thing promoted by unthinking zero tolerance), and affirmed the computer’s “judgement” …just four minutes later.
So I appealed to the Facebook Oversight Board, which was in the news recently as having affirmed Facebook’s decision to ban Donald Trump from Facebook and Instagram “for violating its Community Standard on Dangerous Individuals and Organizations,” but slamming the company for not following its own policies in doing so. (Note: this page will not go into whether that was a good or bad decision on the part of Facebook or its Oversight Board, nor will I approve comments about that.)
Board Appeals: Not for the Faint of Heart
The problems about my making such an appeal are that 1) they warn that “very few” appeals are “selected” for review (aka, don’t expect them to even read it) — which is what prompted me to write this page so that the public could see the sort of procedures Facebook has, and 2) they threaten (my word) that to submit such an appeal, I must agree to allow the Board “to share details that could easily identify you” in its published decision, including sharing “data with special protections about you.”
Wouldn’t that very requirement on all Board appeals severely intimidate the very sort of people their policy is trying to protect?! Even I — a white man of privilege who has a large, well-established platform where I openly publish my opinions to an audience large enough that this has been my primary source of income for more than a quarter century — took pause before agreeing to such a thing.
They provided links to define these aspects:
What are “details that could easily identify me”?
These details include:
- your first and last name
- the city you live in
- your profile picture (or a description of it)
- your cover photo (or a description of it)
- your follower and friend counts
- the date your Facebook or Instagram account was created
- the locations that you have chosen to indicate on your account
What is “data with special protections”?
“Data with special protections” is information about your racial or ethnic origin, political opinions, religious or philosophical beliefs, trade union membership, health, sex life or sexual orientation, criminal conviction or offenses. This information is subject to special protections, including under EU law. [emphasis added*]
If you submit an appeal or provide the board with further context about an appeal on your content, you will be asked whether you give permission to let Facebook process data with special protections that you choose to provide about yourself. If you later wish to withdraw this consent, you can go to Case Status and choose “delete my information.”
You will also be asked whether you give permission for the board to share data with special protections about you in the public, written explanation of its final decision. If you later wish to withdraw this permission, you can go to Case Status and do so. If the board has already published a decision on your case, the board will remove the explanation from this website and take out your data with special protections. The board will then re-post the new, edited explanation. Please note that the board cannot change any commentary about this case that has already been published by other people or organizations.
* Do you think Facebook has correctly divined your “political opinions, religious or philosophical beliefs, trade union membership, health, sex life or sexual orientation, criminal conviction or offenses.”? I sure as hell don’t, yet in order to be heard, I have to agree to let them publish, without prior review, their opinion as to such personal characteristics, which I consciously keep out of the public eye as a professional journalist; True is, by design, nonpartisan.
I submitted the formal appeal anyway.
Who Cares What I Think?
Indeed, who does care (other than some percentage of that large paying audience who makes it possible for me to make a living publishing social commentary online)?
Anyone with an open mind.
But maybe I got it wrong. I am wrong sometimes, and perhaps I have a blind spot that others (especially women — and especially Marion, who I was replying to), might see.
First, in a private place on Facebook, I announced to some colleagues that “Today, I was accused by FB of ‘hate speech.’ I appealed, was denied. I appealed to the ‘Oversight Board’ …and was warned that few appeals to them are reviewed. Well, I’ll be blogging about it in the next few days. The accusation: defending a woman’s right to assert her intelligence rather than ‘play dumb’ is ‘hate speech.’”
Reactions to that included “Like”, “Care”, “Wow”, and “Angry”.
But one comment was really chilling, made by a friend who is a young widow and lives alone: “If you write about this, feel free to include that I recently received an honest to God no foolin’ death threat, and FB said that that wasn’t against their rules.”
Got that? Saying women should be encouraged to not hide their intelligence is “hate speech” against theoretical victims, but “honest to God no foolin’ death threats” against real, identifiable women are just fine and dandy.
If that doesn’t tell you about Facebook’s selective “zero tolerance” mentality and, frankly, grossly immoral position, I don’t know what does.
So what does the supposed “target” of my supposed “hate speech” think? After all, shouldn’t the “victim” of a “direct attack against people on the basis of what we call protected characteristics” (where “attack” means “violent or dehumanizing speech” against them) have a say?
First, I used some of my small amount of space to encourage the Oversight Board to contact her directly and ask.
Second, I contacted her directly and asked.
I sent Marion the first graphic above that had my comment, and asked if she had seen it before they removed it. I also solicited “your honest reaction to my comment: did you consider it dismissive of your intelligence, or supportive of you (and by extension all women) being able to express intelligence rather than hide it to please men? I don’t want you to say the latter unless you really believe it: if you think I was out of line in ANY way I’d like to hear about it with no holds barred.”
I made it clear I would publish her reply without changing anything.
Well, she really let me have it: “Thanks for including me in your conversation with Facebook. No, i didn’t see this comment and yes, i support your position 100%. I would have clicked the WOW icon for reasons given below. You are welcome to use anything that i say here.”
“Too often in the first third of my life,” she continued, “i had to play dumb in order to survive a male with socially-sanctioned economic and/or emotional power over me. I’m divorced twice; the common reason was that each ex stopped displaying respect for me as soon as the marriage vows were said. Through these earlier traumas i learned deeply that most men of my age group (mid-wave baby boomers) talk the talk without walking the walk, and i concluded that such lip service is simply a way to ensure acceptance within their social group. It was a courtship show, in other words.
“The rare men i’ve met of any age who walk the walk have become mutually supportive life friends. If I’d had the smarts when young to hold out for a mate who truly respected my abilities as I would his, my life trajectory would have worked out better emotionally and economically. As my dad used to say, ‘Too soon old, too late smart.’ Our caring surmounted earlier mistakes and he [my father] learned to give to me the respect he’d always vouchsafed to able women outside the family. Old scars have adhesions, as I learned to say about those like me who are survivors of such attitudes, and I steer clear of any situations or persons who consistently and intentionally poke at those scars while being careful not to inflict wounds on others myself.”
As to the post itself, she continued, it “did not poke. It did evoke memories and awareness of how untrue such a statement has been for me, though I fervently would wish it otherwise. And that kind of response by one of your readers is okay. I can see that the world evolves in a good way due to such mentoring as you offer. Your This is True site promotes respect for others and holds up a mirror to the stupidity of those who do not act out of respect.
“This has always told me that you live in that respect: your work, whether volunteer or site-based, has demonstrated this for decades, which is why i’ve been a subscriber even during years of low income ~ you refresh my faith in human potential for decent behaviour. You’re a living example of the business ethic attitude which i taught and occasionally had to fight for.”
But, I’ll guess, Facebook won’t care whatever about her opinion — which indifference would absolutely be a patronizing pat on Marion’s head.
Interestingly, Marion wrote back later to add, “I was shocked that your comment was labeled ‘hate speech’ by an algorithm, which was then defended when challenged. What I’ve always wondered is why anyone would trust algorithms to make moral decisions.”
Some will read this and cry “Censorship!” or “Unconstitutional denial of free speech!” I’m not. Facebook, while seen by many as a public utility, is a private playground. They get to set the rules. The problem here is, I didn’t actually violate those rules, which I’ve reproduced above.
My Appeal in Detail
The full text of my appeal is again in a “click to open” format in case you’d just as soon skip it. The original appeal is limited to clicking a limited number of possible selections to provide a basis for the appeal, which is obviously very much slanted against the appellant.
The Oversight Board appeal form allows for free text entry, but the appellant is forced to be very concise: there’s a “countdown” under each space showing how many more characters can be used. That lends itself to a somewhat stilted approach, but that’s the way it is.
There is also no ability to use italics or bold, which limits emphasis to ALL UPPER CASE.
Submitted to the Court of Public Opinion
I’m not ashamed to say that it’s quite daunting to be accused of a vile thing by perhaps the most socially powerful corporation on the planet. “Hate speech” — against women! My wife has so much faith in me she chuckled at that pronouncement even before I told her what it was about. (Thanks, sweetie!)
Again, I do know that hate speech is absolutely a massive problem in the world, and especially in the semi-anonymous world of social media, and agree Facebook (and other such companies) should do their utmost to stop its spread.
Yet to misclassify actual praise in that category both “proves” to those who actually spew hate that the social media companies get it completely wrong and therefore whatever they said was misinterpreted, and it trivializes real hate speech.
Both of those facts are sad commentary on social media.
I look forward to your comments.
No, I haven’t heard anything back from Facebook (or its “Oversight Board”), but I did hear about an even more outrageous case on the same lines. Or, more specifically, in the same lines as “Saying women should be encouraged to not hide their intelligence is ‘hate speech’ against theoretical victims, but ‘honest to God no foolin’ death threats’ against real, identifiable women are just fine and dandy.” (Above)
On June 16 political commentator Jim Wright, who is retired from the U.S. Navy, reported (on Facebook where he has more than 200,000 followers) that he was suspended from Facebook “for a few days.”
He had reported to Facebook about a honest to God no foolin’ death threat he received via a Facebook message, and sure enough they blew it off as “not a violation” of their rules.
Yet when Wright took a screencap of the threat, blocked the person’s name so he wouldn’t be identifiable, and posted that openly? Wright was suspended for “bullying” that unidentifiable person.
Reminds me of high school, when someone finally pushes a bully back: a teacher sees that and the victim is marched to the principal and suspended. The bully? Well they didn’t see him do anything wrong!
Yet “The message that Facebook specifically told me wasn’t a violation of the [Terms of Service] when some goon sent it to me somehow became a violation of the TOS when I posted it.”
Which takes us right back to:
Any company who does this without rational backup from humans who can demonstrate that they can make moral decisions has abdicated their moral responsibility; they are evil.
– – –
Bad link? Broken image? Other problem on this page? Let Me Know, and thanks.
This page is an example of Randy Cassingham’s style of “Thought-Provoking Entertainment”. His This is True is an email newsletter that uses “weird news” as a vehicle to explore the human condition in an entertaining way. If that sounds good, click here to open a subscribe form.
To really support This is True, you’re invited to sign up for a subscription to the much-expanded “Premium” edition:
Q: Why would I want to pay more than the minimum rate?
A: To support the publication to help it thrive and stay online: this kind of support means less future need for price increases (and smaller increases when they do happen), which enables more people to upgrade. This option was requested by existing Premium subscribers.