The lesser evil

Danielle Navarro — Mon, 19 Oct 2026 11:00:00 +1100

Blood in the cut

Danielle Navarro — Sun, 18 Oct 2026 11:00:00 +1100

And so comes to pass that she starts another blog. It is not a choice she makes lightly, having written and discarded oh so many of these things in the past. But sometimes we do as needs must, and of late I've noticed that my personal and political thoughts have started creeping into my data science blog in ways I do not enjoy. That blog is special to me. I write about art, and code, and science, and statistics. And oh my do I take joy in being a crass transsexual bitch in the footnotes. From time to time the darker undercurrents of living as a trans woman slide into that blog, but that is simply the nature of things. The writer is not distinct from the words.

Nevertheless, there are some limits to the extent in which I would like my worlds to collide. The Notes from a data witch blog is not meant to be a place for writing about politics per se, or queerness per se, or any of those ancillary things that do sometimes creep into the narrative. They are the background, not the foreground. Were that state of affairs ever to change on that blog, I would likely lose all joy in writing it. I'd prefer not to allow that to happen, but when I found myself writing a post on that blog about Dining With Wolves, a post that had no art, no science, no statistics, and no code, I'd inadvertently opened the door and invited the vampires into my home. I'd exposed my delightful home on the internet to something darker. That state of affairs was simply intolerable to me.

Fortunately, having recently rewatched True Blood and taken great joy at having seven whole seasons of Alexander Skarsgård on my screen, I remembered that in many versions of the vampire myth you can simply rescind the invitation, and the undead are obligated to leave. Accordingly, I deleted the Dining With Wolves post from my data-centric quarto blog, created this entirely new eleventy blog, and moved that post over to here. Should I ever feel the need to write more posts of that nature, they now have a home here.

I do not expect that this blog will be as nice as that one. Any time that I am moved to write about the reality of living as a transsexual woman in the year of our lord 2026, I am never in a good mood. This is for the very obvious reason that people are consistently and exhaustingly cruel to us. Even were I a saint my patience for this behaviour would have been completely eroded many years ago. And as I am not at all saintly in my character, the charity I once extended is all but gone now.

That part of my life needs somewhere it can breathe. This is not a blog about art. Or code, or science, or statistics. This one is about the pain, the blood, and the terror. Because those stories also need to be told.

Dining with wolves

Danielle Navarro — Mon, 05 Oct 2026 11:00:00 +1100

Earlier today on Bluesky I had the misfortune to encounter this rather strange post from Justin Wolfers, an academic economist of no small renown.

The best cure for ideological bias is find someone on the other side and go to lunch with them.⁣ Sometimes we disagree about facts. Sometimes we disagree about values. Figuring out which is which is fun part.

This was his summary of a slightly longer set of remarks he made in a blog post, the gist of which is that he has evidently had some productive discussions with his libertarian friends, and comes to the deeply anodyne conclusion that concentration of political power can sometimes be bad.^[1]

Setting aside the blandness of the political claim (does anyone dispute this version of the thesis?), the thing that struck me as frustrating is that when -- inevitably -- hundreds of people started getting very, very angry at his "go have lunch with your ideological opponents" strategy, Professor Wolfers felt he was being misunderstood and tried to tell the audience that they were arguing against the wrong version of his argument. After all, the slightly longer version of his claim is this...

Find people you like and respect on the other side. Go to lunch with them. Try to understand how they see the world. Once you think through the issues from the perspective of someone you like and admire, you quickly stop straw-manning. Often you'll learn another perspective. You may still disagree, but if you're doing it right, it's respectful disagreement.

...and though his responses on Bluesky didn't say so explicitly, his pushback against his critics hinged on the "someone you like and admire" part of the preceding passage. He leans quite heavily on the idea that the ideological opponent with whom you should sit down and share a nice bottle of merlot over lunch is somebody likeable, admirable, and disagrees with you only on some abstract intellectual principle that can be the topic of a pleasant chat.

This is, to put it mildly, very foolish. It rests on a deeply flawed assumption about the world, namely that ideological disagreements stem from intellectual disagreements. In general they do not. There are of course exceptions, and academia in particular is the home to many of those exceptions. For example, statistics has been somewhat divided by an ideological split between Bayesian and frequentist ideas.^[2] The dispute between these camps is not a matter of factual disagreement; it is an intellectual and ideological one. To oversimplify a rather complicated story, the Bayesian position stems from the idea that probabilities are subjective degrees of belief; whereas the frequentist position stems from the idea that probabilities are properties of sequences of observed outcomes.^[3]

This is a kind of disagreement wherein a Bayesian such as myself can sit down amicably with an orthodox statistician and talk about our differences. I can (and do) accept pragmatic compromise positions in which I acknowledge the strengths of orthodox methods.^[4] These are ideological disagreements over which academics are well-qualified to offer commentary. Unfortunately for Justin, the vast majority of ideological disagreements as they apply to real life are very, very different in nature.

The people who are responding angrily to Professor Wolfers on Bluesky are not misguided; rather, they have correctly understood the nature of what an ideological dispute looks like in the wild. They are not misunderstanding his point; he is misunderstanding theirs.

Dining with wolves

In the lead up to the 2016 US presidential election, during which almost everyone I know found themselves aghast and horrified at the direction political conversations were headed in that country, someone made what I still consider to be the defining comment on the current political era:

This election is like when you're all trying to make dinner plans, and the options are either to get pizza, or to have your friends kill and eat you. Even if pizza wins you have a problem.

It is with this rather memorable comment in mind, and having spent the last decade living in abject terror as cisgender people of different political persuasions argue with one another over whether they should indeed kill and eat the transgender population -- only slightly metaphorically speaking, sadly -- that I have some remarks to add about the complete and utter worthlessness of trying to have a polite chat with one's ideological opponents.

Let me start with an anecdote. A couple of years ago I was having a quiet beer at my local pub in Newtown, Sydney watching somewhat sadly as the results of the 2024 US presidential elections were being reported. I figured this would be a pleasant place to distract myself, sit in the sun and maybe write some code. It was not to be. Although my local is usually a nice, politics free zone, the tranquility was rather rudely interrupted by two women in Trump t-shirts who sat down at the table next to me and started loudly gloating about the triumph of their preferred candidate.^[5] They went on at some lengths about how they believed that RFK Jr was going to be a fabulous force for good health in America, about Trump was going to build a wall to keep the illegal immigrants out, and -- with a certain amount of inevitability -- what the new regime was going to do to punish the transsexuals.

And oh my did they have a lot to say about trans women: according to their account we are all rapists, perverts, mentally ill degenerates, and predators. No exceptions. Our mere existence in society was prima facie evidence that society had lost its moral compass. So passionate they were about this topic that as they grew more animated I found it difficult not to glance in their direction, terrified at what might happen if they realised that the woman sitting right next to them was in fact one of these degenerate perverts whose extermination they were openly endorsing.^[6]

But they did not clock me as trans. Instead, they mistook my "freezing in terror" response as a form of acquiescence, and concluded that I must in fact be a cisgender woman who they could recruit to their cause. They spent the next hour or so recounting the long list of sins of which I am guilty -- by my very nature as a trans woman -- and mark me as a moral abomination whose presence cannot be accepted in society. In my desperation to maintain my "oh dear lord please don't let them notice what I am" posture, I spoke with them politely the whole time. I did all the things you're "supposed" to do in this sitaution. I was not confrontational, I listened to them non-judgementally, and only made very gentle counterarguments to suggest to them that perhaps they might be mistaken.

It should surprise nobody that my milquetoast manner did not persuade them to change their beliefs, nor did their venomous attacks persuade me to change mine. But strangely, by the time they decided to leave me in peace and move onto the next pub, they were completely certain that they had converted me to their anti-trans stance. They mistook fearful politeness for agreement. They joyfully welcomed me to "Team TERF" as they left.^[7]

In real life, this is what an ideological disagreement looks like. It almost never takes the form of a polite conversation between people who share mutual respect but merely disagree on an issue of substance that they can calmly analyse together. Ideology is dark, it is violent, and it is not built atop reasoned beliefs. You cannot have a polite conversation at the pub with someone who openly wants to kill you.^[8]

The academic fallacy

The story I've just recounted is merely one of many experiences I've had when encountering the anti-trans folks in real life, and in terms of how bad this one was it's probably middling? The least awful of those encounters are the ones where I don't feel any sense of physical threat, but spend the entire time being forced into a degrading conversation about my life. It is humilating to have to try to persuade people that I deserve to be treated with a minimal level of kindness, and for them to want to "debate" whether I have the right to use the bathroom in peace. It is exhausting. And that's the best case scenario for Justin's "engage with your ideological opponents" strategy: I have to subject myself to repeated degradation and humiliation for the sake of someone else's intellectual curiosity. But then there's the other end of the spectrum, where it actually gets much, much worse than the transphobic-women-at-the-pub story. There's the time I was physically assaulted on the 370 bus to work. There's the time an academic colleague threatened to punch me in the face if I used the women's bathroom. And then of course there are the several occasions in which a cisgender man raped me with total impunity because he knows perfectly well that trans women can't report sexual assaults to the police. All of these things -- yes, even the rapes -- have an ideological component. Political ideologies tell us who deserves to have rights and who doesn't. If you fall on the wrong side of someone else's ideology, a polite conversation isn't going to solve any problems whatsoever and has a good chance of putting your safety at risk.

Most people understand this. It is a very basic fact of life for most of us. Ideological disagreement is almost never over questions like "are probabilities objective or subjective things?" or "what should the marginal tax rate look like?" Contrary to what one might experience as a tenured academic of a certain social strata, most ideological disagreement turns on questions like "should immigrants be tossed into concentration camps?" and "do trannies deserve to live?" In my experience, academics -- perhaps including Mr Wolfers himself -- make the grave error of thinking that ideology revolves around intellectual disagreement. It almost never does; it is visceral, it is personal, it is emotional, and most of the time you cannot argue with people about it in a useful way.

Dining across the ideological divide is not fun. It is not productive. It is terrifying. It is a game that is safe only for the most privileged players. You do not get points for being a wolf who chats politely with the other wolves, thinking that this is the way disagreements between wolves and hens get resolved in the real world. That is naivete of the most tedious kind.

As an aside, from the transgender perspective I find it very difficult to believe that libertarians as a group are particularly committed to the idea of decentralisation of power: if they were even slightly honest about that, they would not be in favour of anti-trans legislation that has become popular in a great many jurisdictions. I am more cynical about them: I think their only real committment is that other people shouldn't be allowed to tell them what to do. Most libertarians I have encountered in real life believe in their own political freedoms, but not in the political freedoms for people who belong to different social and demographic classes to them. For the most part libertarianism is a fig leaf political identity that people claim to disguise the naked selfishness of their views. ↩︎
Arguably, the strength of this disagreement has dissipated a lot in the last 20 years, but the ideological split between these two camps was very intense during the 20th century. ↩︎
Please do not reply to me with any kind of "but but acshualllllly it's more complicated..." response: yes, of course it is more complicated than that, but statistical ideologies are not the primary topic of this post, and the overly crude "Bayes = subjectivist, frequentist = objectivist" summary is about as close as you can get in a single sentence. ↩︎
For example, one area where classical statistics is very strong is in basic inferences on a consideration of counterfactual possibilities. There are merits to this when doing inferences that should not be discarded on the basis of ideology. ↩︎
Honestly, why Australians continue to care about US politics continues to escape me, but as a nation we seem more interested in their politics than with our own. ↩︎
This, incidentally, was the day when it viscerally sunk in that I genuinely do pass for a cisgender woman in my everyday life. Because their venom was overpowering; there is absolutely no way these two women could have held back and spoken to me politely if they thought I was one of the transsexuals they despise so deeply. The hatred dripped off them. ↩︎
I then had a quiet conversation with the bar staff about the fact that the pub has a "no transphobia" policy prominently placed on the front door that they had neglected to enforce. But of course by then the damage was done. ↩︎
I mean, technically you can: that's precisely what I did. But you should not mistake that conversation for a "political discussion". It was verbal assault: two cisgender women publicly terrorising a trans woman who was too frightened even to disclose her status for fear of what would happen next. ↩︎

Notes from a transsexual bitch

The lesser evil

Blood in the cut

Dining with wolves

Dining with wolves

The academic fallacy