2024i3, Wednesday: Unsure.

This isn't the first time I've mentioned these little heuristics, I realise.

But as we go into another year likely to be dominated by shouted appeals to the demons in our souls, rather than the angels of our better natures, something I read on New Year's Day - pointed out by game designer Dave Morris, who's lucky enough to count one of my favourite writers of thoughtful journalistic prose as a friend - made me think they were worth mentioning again.

As my daughter grew up (a blessedly smart and considered person that she is), I used to worry to the point of obsession about how she and her generation were enmired in so much more information than we had been, with much less clear provenance. And I wondered whether there were thinking tools that could help her (and me!) navigate the waters and avoid some of the rocks and rapids.

I came up with four. None are terribly original; some, indeed, are direct lifts from brighter people than me. Some do rather overlap. But together I think they work rather well.

First: Distrust certainty. I remember the first time this thought crossed my mind: I was in a cafe in Jaisalmer, in Rajasthan in western India, about 30 years ago, with some bloke from South London pontificating about conspiracy theories. He was so sure of everything he was saying, despite clearly having paid no real heed to the bases thereof. I remember seeing the mental red lights flashing. As a paid-up god-botherer, this always seems still more important a heuristic: if you believe in an omniscient, omnipotent single deity, by definition you can't possibly really understand everything they are saying to you. Everything, particularly scripture, is seen through a glass darkly. So when confronted by anyone who says they absolutely, certainly know: run away. Now.

(I once came across a lovely definition of a fundamentalist which I was sure was a Pratchettism except that I can't find it anywhere in his writing: "Someone who knows exactly what God would want if only God, like he, had all the facts." Quite.)

Second: Try to imagine the person in front of you saying: I could be wrong. And meaning it. This is the one whose recall was sparked by Dave Morris's piece. The post it linked to, by sci-fi author David Brin, included the following quote:

"[Here's a simple catechism] spoken not just by scientists, but by any person with an ounce of decency or maturity. 'I might be wrong. Let’s find out.'”

The post is worth reading in its entirety, as is its Part 2. But this little snippet is critical. It links rather heavily to the first heuristic, of course: people who are certain are unlikely to admit they're wrong. But I think it's worth something on its own. Because it goes to the heart of true intelligence and wisdom: the rueful realisation that you're always working on partial information, and the next thing you learn could change your world-view radically.

I realise this does rather require stepping into someone else's shoes, or brain. But that's a key part of being human as well.

Third: Evil always starts with treating people as things. (Or less than people.) Any people. OK. This is wholly unoriginal. And definitely a Pratchettism. One reason why people like me keep reading Pratchett is because his take on humanity just makes sense: humans can be small, and petty, and narrow-minded, and selfish, and just plain godawful. And the very same humans can be noble, and wise, and generous, and kind, and move mountains to help another. Most of us are both. People, in other words, are people. Don't expect them to be otherwise.

But more than that: ALL people are people. As we used to say in reporting: the only difference between us and the rich and powerful is that they've got more money and power than we have. They're not better. No-one is. Because the moment you start dividing the world into who's better and who's worse, as individuals or groups or nations, you're starting to treat people as things. And the moment that you do that, that you start seeing them not as full-fledged human beings in all their glorious, splendid contradictions, but as pieces on a board.

And that - as the Porter said - is the first step along the primrose path to the everlasting bonfire.

Now, obviously sometimes people need to be seen in the aggregate. Politicians, for instance, can't plan for each individual in turn. But there's a difference between aggregating and still knowing, and owning, the effect of what you do on individual people, and lumping them together into a mass that can be used, exploited or just overlooked or forgotten.

Fourth: Good ideas don't need lies told about them to win out. Again, entirely unoriginal. This is from Dan Davies's well-known (among a certain class of nerd) "One-Minute MBA". It was prompted by the invasion of Iraq, but has broad application. If the only way an idea can gain mass acceptance is by convincing people that something is true when it's not, then it's a crap idea. End of. Like so many heuristics (I'm looking at you, Dunning-Kruger), it's amazing how much explanatory power it has when you start looking around keeping it in mind. Although it's also astonishingly depressing to do so.

(The less-well-known "One-Minute MBA Part 2" is good as well. Both quick reads, with lots to think about.)

All of these, I think, are relevant to my life as an advocate, as well as my life as a human. Either practically (in the sense of what's persuasive) or ethically (in both the professional and can-I-look-at-myself-in-the-mirror senses). Or often both:

1. Certainty is strangely unpersuasive. If I read back draft submissions and they come across as suggesting that things are 100% black and white, I usually sigh and rewrite. As part of the reflective practice that we all need to engage in, I have to acknowledge that some of my worst failures stem from straying into stridency. Avoid.
2. As far as recognising that you could be on the wrong track is concerned, one of the most fabulous pieces of advocacy I ever saw was in the High Court in 2019 where Tom Cross, of 11KBW, argued a point for the government in an education case which appeared on its face to be slightly ludicrous. He accepted that, acknowledged that was an understandable view, and then spent 45 minutes oh-so-gently easing the Court round to deciding that in fact his was the proper approach. No tub-thumping, no assertion. Just subtle, honest argument. It was breathtaking.
3. This is more an ethical point. It's an easy point to take to make someone look stupid - especially when they're acting that way. And it's then a short step from implying (albeit almost never saying out loud) that this stupidity is baked-in. Don't. It's just wrong. And from the effectiveness perspective, a lot of tribunals will feel pretty uncomfortable when you give the impression you think someone's an idiot. Far better, almost always, to give them the opportunity to prove it themselves. And then just let it sit there. Don't jump on the bandwagon; let it roll straight on to oblivion all by itself.
4. This shouldn't really have anything to do with us, should it? After all, ethically we're not allowed to tell porkies. (And to the surprise perhaps of many outside the profession, most of us would be horrified by the idea of doing so.) But there are outright porkies... and then there's shades of truth, handwaving, gaslighting, exploitation of unvoiced assumptions, and so much more. All these are acknowledged advocacy techniques; but to my mind they're crap ones. Working on the assumption that your opposition is as smart if not smarter than you are, and the tribunal is smarter than either of you (an assumption that gains painful truth as one climbs the court hierarchy) leads to the conclusion that any of this stuff is more likely than not to get found out. So accept the facts. Highlight the useful ones. Contextualise the others. And get the job done right.

And what about the rest of life? Well, 2024 is going to be a race between dickish pols and black-box-trained LLMs to see how much straw-manning (the former) and hallucination (the latter) can be spilled into our collective mental atmosphere. Not all these rules are going to work on LLM output, of course. But they'll help when dealing with those who repurpose it without attribution. Godspeed to us all.