Hacker News Reader

Comment by gruez | Hacker News Reader

>but is there anyone that actually defends these DEI statements with a coherent argument, or can you point me to one online

There was a recent debate on this[1] and even the debater for the pro DEI statement side (that they could find) admitted that DEI statements that were just ideological pledges were wrong, and he was only in favor of statements about concrete things you did to advance the DEI agenda (eg. "I did a, b, and c in my previous position to enhance DEI in my department"). He argued that was justified because DEI (at least the principles, not necessarily the specific policies like affirmative action or whatever) were ostensibly things that the university cares about, and therefore they were fair game to ask for.

[1] https://opentodebate.org/debate/are-dei-mandates-for-univers... (it's a podcast but there's a transcript tab on the page)

View on HN

I’ve heard this same argument for DEI statements every time the topic has been debated: They sidestep arguments about DEI statements and instead retreat to safer arguments about how advancing DEI is a good thing.

On one hand, I get it. Arguing for DEI in an abstract sense is much easier than arguing for specific interventions.

On the other hand, this is the textbook example of a Motte-and-bailey fallacy, where the debater conflates two topics and then retreats to the safer position while hoping that the audience will accept it as an argument for the more difficult position.

DEI statements have been quite unpopular as specific interventions, as noted in the article by the way that the majority of staff disagreed with them when polled privately. However, speaking out against them publicly was viewed as a very risky move and serious career mistake, so they slowly slipped into mainstream acceptance.

It’s interesting to see how they’re now quietly being removed from processes with as little attention as possible. Nobody wants to be known as the person who campaigned against them publicly, but I suspect there are a lot of people feeling relief in this case as they’re being quietly removed from the process.

Feels a little bit like making people take anti-theft pledges. I don't think anyone could reasonably expect that theft would be prevented by someone pledging not to engage in it yet it might be tricky to publicly remove because "what are you, pro-theft!?!"

Similar to DEI training in general which doesn't appear to do too much. It should be argued that "we shouldn't do these trainings" yet you also don't want to be the one saying that.

What does appear to work is a company actually hiring and promoting skilled diverse individuals and not just their buds.

> “feels a little bit like people making anti-theft pledges”

Slate Star Codex covered this so well on https://slatestarcodex.com/2017/06/21/against-murderism/

Imo it's very different bc the ideology pledge itself is mostly distracting from the real purpose of the dei statement (again imo), which is to provide a way to select employees based on race, etc, without it being the explicit reason.

A better comparison would be if instead of an anti-theft pledge, there were a ten-commandments plus beatitudes test/ statement, without referencing the source religion. Maybe call it a morality pledge.

It's essentially a religious test hiding behind another name and terms, and even if it doesn't explicitly discriminate based on religion, it's obviously designed so that people of certain religions do better.

And filters people not from those religions by boosting those who essentially share the same values, or can fake it.

In the same way, dei doesn't explicity select for diversity applicants, but for people who are explicitly pro-diversity-applicant.

Which doesn't sound quite as pure of you replace "diversity" with any of the approved / included groups...

'I'm not picking applicant based on them being black, but based on them having an ideology that is pro-black-applicant.'

Although sadly we've normalized it to crossing that line as well, with the last SCOTUS nominee.

> diversity applicants

Friendly nudge that this isn’t a meaningful term. Once you start noticing people saying things like “our hiring pool is 60% diversity” it’s hard to unsee.

I don't agree. "Diversity" has a commonly accepted meaning which in the US at least, means everything except white men. Female or non-white are "diversity." A great example are the statements and articles from 2020 when Biden announced the most diverse ever White House Communications Team which was 100% female. Biden talked about how critical diversity was and bragged about this 100% female communications team. All the articles I read about it had things like "most diverse White House Communications Team in history" to describe it. At least from a gender perspective, I don't see how they could make it more clear that diversity == women

Asians aren't counted as "diversity" either. This is why they're referred to as "an inconvenient minority" in the context of DEI.

For that, does "asians" mean people from the middle of asia (ie middle east), or people from south east asia (ie oriental)?

Asking because the term has different meanings in say the UK (asian -> from middle east) vs Australia (asian -> from south east asia).

In the US almost no-one would say "asian" for any country west of Nepal. Sometimes they say "south-asian" for India or the surrounding countries, but even that term is only sometimes extended out to pakistan.

>people from south east asia (ie oriental)

it is actually more common from the European and Middle Eastern context to call the (Turkish and Levantine and Arabian) Middle East "oriental"; in Israel, "oriental food" is hummus and felafel; the Orient Express train went to Istanbul. East Asian is the term for ... east Asians and of course the SE Asians you mentioned, and South Asian is the term for "India+Pakistan+" people. Central Asian is the "the -stans" and Mongolia and parts of Russia.

Wow, it's even more complicated than I realised. :)

An Asian-owned company in US can easily get 80%+ Asian. Is that "diversity"?

This depends on how you frame. It is not diverse for the company, but it could be in the larger social context. Putting the same thing in, say, San Francisco can be different from doing the same in Utah.

> "Diversity" has a commonly accepted meaning which in the US at least, means everything except white men.

Unfortunately this is common, because they do bad things with "diversity", even though diversity itself isn't bad. This isn't a very good defintion of what "diversity" should be, of course. There are many kind of diversity, and which are more important depends on the situation. But, regardless of it, diversity includes white men (and everyone else, too).

Visible diversity in skin colours, height, etc can be relevant for some things (e.g. movies that will have a lot of different people, or when doing research for a computer program that works on pictures of people (to do compression, colour correction for lighting conditions, etc)). Diversity in experience (even if all of them happen to be white men) can be relevant for many things (and is very helpful).

Of course, none of this should mean that you should deny application of other diversity because of their skin colours, height, gender, etc; they should not deny an application for such reasons. Having women in the White House Communications Team is not a bad thing, but that doesn't mean that having men is a bad thing!!! Having men is not a bad thing. One thing being good does not make the other one bad.

> none of this should mean that you should deny application

Is there any evidence that anybody was _ever_ doing this after about 1950 or so?

It may not be a category (imo it is), but it's at least a hierarchy.

It does go beyond simply making a pledge, though. Applicants are required to list all the ways they've furthered the cause of this very specific issue, regardless of whether the job itself would entail or even offer opportunities to exercise such experience or not.

Like requiring an applicant for a fullstack python/js position to list in detail all of their Erlang experience, projects, talks and initiatives, and gatekeeping on that.

Now Erlang itself is great, and understanding the principles that guide it will almost certainly help you in your career (even if you never write a single line professionally). But to make this a central focus of acceptance for the job is just plain silly.

Perhaps that is a good analogy, because I agree, and that is true whether it is Erlang or if it is DEI statements or if it is something else, even if they can be good things it is certainly no good to make this a central focus of acceptance for the jub.

Or like the pledges that university students have to fill out before each exam, promising not to cheat in any way. As if "you broke the pledge you signed" carries any more weight than "you cheated".

But the problem is they can’t even get “diverse” right. Do they mean diverse in ideas? In thought? In opinion? No.

My LARGE company is on a tirade to ensure we use diverse supplies. We’re collecting metrics on this so we can proudly state how many diverse vendors we we use. When asked what we define as diverse they stated (among other things) that they were members of the LGBQ community. So I asked “so our company is very much concerned if the owners of the companies we work with are bisexual?”

Im married to a woman, but that doesn’t preclude the possibility that I’m bi-sexual. So if, in addition to being attracted to my wife, I’m also attracted to men, then all of a sudden I’m a priority for the company.

It boggles my mind.

>On the other hand, this is the textbook example of a Motte-and-bailey fallacy, where the debater conflates two topics and then retreats to the safer position while hoping that the audience will accept it as an argument for the more difficult position.

Motte and bailey accusations only work if the statemnts in question are being made by the same person. For the debater in the post above at least, is there any evidence he personally is engaging in this? Or anyone else? Otherwise it's a bit fallacious to lump everyone who's pro DEI as one entity.

> Motte and bailey accusations only work if the statemnts in question are being made by the same person.

I think it's perfectly justified in cases where debater self-identifies as part of the "bailey" community -- that is, if in other contexts the debater wouldn't oppose people arguing for the bailey position.

> On the other hand, this is the textbook example of a Motte-and-bailey fallacy

In a textbook, it would be called "equivocation".

> where the debater conflates two topics

Yep.

I continue to be somewhat disturbed by the fact that giving a new name to a problem we've recognized for thousands of years has apparently increased awareness. Did people not realize equivocation was bad before Scott Alexander?

Because it's not just equivocation. Read more of the actual original definition/usage of motte-and-bailey.

Motte-and-bailey isn't just an argumentative form, it's also a political strategy. It's people pushing/doing controversial thing A but every time they get called on it they bring up uncontroversial thing B. Or, they even retreat to thing B, but then when the pressure is off they come back out and start pushing/doing thing A again.

"Equivocation" covers a wide variety of situations, but motte-and-bailey is more specific and includes the notions of a strategy executed over time which includes action.

You haven't actually described any differences between the terms. (Well, you did claim one, that "motte-and-bailey is more specific", but combined with your appeal to "actual usage", that is clearly untrue.)

Motte-and-Bailey speaks to equivocating about positions in argument. There are many other ways to equivocate, e.g. over the meaning of a single word, used in multiple ways in the same argument, as opposed to a retreat to an easier-to-defend position over the initial one presented/being discussed - as in the thread above.

The relationship between the two words is akin to the one between “rectangle” and “square”.

Alternatively, you're reversing the causality. As such fallacies have become borderline ubiquitous in many relevant aspects of life, greater appreciation and understanding of what's happening led to the emergence of language that's not only more figurative and visual, but also more precise. Because while all motte and bailey fallacies will be false equivocations, not all false equivocations will be motte and bailey fallacies.

Engaging in a genocidal action and then claiming it's self defense when scrutinized is a textbook motte and bailey fallacy. Claiming that people who oppose said genocide are supporters of fringe radical elements within the genocided, is a false equivocation, but not a motte and bailey fallacy. Indeed, you could make your exact argument about association fallacies, or a wide array of other fallacies for that matter - as most tend to involve false equivocation at some point.

The original SSC article about Motte-and-Bailey used "strategic equivocation" as an alternate name, not just plain old equivocation. I.e. intentionally equivocating whenever is convenient for some political/social/whatever goal

I suspect you could get away with speaking out publicly on this if you said something like "DEI statements aren't actually advancing DEI" first. You need to break the link between the statements and the goal first, so that you don't look like you're attacking the goal.

You have a much more optimistic view of the current state of public discourse than I do.

There are plenty of organizations and people ready to tear down anything seen as against the party line for their own benefit. It's so much easier to criticize than contribute, and it only takes a few criticisms that ignore nuance and present things in a different context for critical mass to gather behind those misinterpretations.

The same things that make certain public forums resistant to this are the same things limit their growth, or that limit the spread of the message (content bubbles), so I'm not sure how to get around that without a major societal and cultural shift to how information is consumed.

Thanks very much! I haven't watched this yet, but this was exactly the kind of honest discussion I was looking for that I didn't previously find, so much appreciated.

Would it make more sense to have a generic statement, where one could do more dei side or it could be reaching out to undergraduates.

Basically, a "aside from your teaching and research ability, what have you done to improve your campus that would make us want to hire you"

It's much more ambiguous, on what is valuable.

(I have no idea what the hiring process is like, just a random thoughts)

The exact question doesn't even matter that much. They could just as well give you a blank page.

What's more important is how they grade and rank you and make decisions. Eg imagine they only hired you if filled that blank page with a picture of the Pope, even if there are no explicit instructions or questions asking for it.

> (eg. "I did a, b, and c in my previous position to enhance DEI in my department")

What if they worked for a dept. that didn't promote these things and may have had policy against them? Why should a candidate need to please yet another committee or reviewer of their "body of work" to now include and require certain achievements in an area that isn't even understood by many and all but squashed now by the supreme court as a tool to recruit students?

> What if they worked for a dept. that didn't promote these things and may have had policy against them?

Then they lack experience relevant to the job, right? That's like asking "what if the applicant has no experience teaching computer science because he previously worked in the history department", except they can still make up something about feeling oppressed or fighting back against a culture they disagreed with.

This seems like the least problematic thing about mandatory DEI statements: if you value DEI, you should value candidates with some previous experience of working in the DEI industry.

Reply on news.ycombinator.com