Some thoughts are harder to think than others.
"How hard" varies from person to person for specific thoughts, depending on the individual's nature, training, and experience. A physicist finds it harder to think about plants than a botanist, a calm person finds it harder to think catastrophic thoughts than an anxious person.
Some thoughts tend to be harder for all neurotypical people, to the extent that it might be possible to define different types of neurodiverse people in terms of the kind of thoughts they find it easier to think than the norm.
It might also be possible to define societies by what thoughts they make the hardest to think. Monarchies made thoughts of regicide hard. Autocracies made thoughts of democracy hard. Democracies make thoughts of autocracy hard. Not impossible, though.
Other thoughts, which violate the conditions of thinking, really are impossible to think. Those thoughts are beyond the scope of this essay, but we encounter them when we try to think about the universe as described by quantum mechanics.
This meditation is motivated in part by reading Sam Hughes' novel There is No Antimemetics Division, which is about thoughts where the act of thinking them erases the act of having thought them: once thought, they cannot be remembered, nor can the act of thinking them be remembered. They leave a blank spot where they once were, and the nature of that blank spot is such that to the victim it has always been true that that blank spot has been there. Anti-memes re-write the past by re-writing memories.
In a period of extreme stress in my life I actually had an experience that is very much like this, but it would go too far down a complex rabbit hole to get into it here. Suffice to say I found Hughes' description uncannily accurate.
The evolutionary biologist Richard Dawkins--who is apparently a terrible human being--coined the term "meme" to describe thoughts that propagate themselves easily, and Hughes--who writes under his Internet handle QNTM--pulls a Pratchettesque inversion on the idea: if some thoughts naturally push themselves into our consciousness, others must naturally resist being thought.
The novel itself is bizarre in the best possible way. It is one of the most remarkable works of hard science fiction I've read this century, taking an idea and romping through metaphorical space with it, right on the edge of comprehensibility.
As we find out in the first few pages, the most powerful anti-memes don't just erase themselves from the thinker's consciousness and memory, they erase their antecedents, so the chain of thought that leads to them vanishes as well. They become practically unknowable, which makes it hard for the Antimimetics Division to counter them. It's an idea that could be milked for humour, like Monty Python's killer joke, which no one could be aware of without dying of laughter, but Hughes focuses on the serious side.
The novel reminds me a little of Zinoviev's "The Yawning Heights", a Soviet-era Russian doll that describes an experimental project whose purpose is to discover which participants are opposed to it, so they can be liquidated.
"Roadside Picnic" by the Strugatsky brothers is another book that comes to mind, as does Algis Budrys' "Rogue Moon" and maybe some of Lem's works: "Solaris", "Chain of Chance", "His Master's Voice"... all dealing with aspects of reality that we can't think the right thoughts to comprehend.
China Miéville 's "The City and the City" has similarities as well, although in the case of "There is No Antimemetics Division" the ontology of the antimemes is unambiguously physical, which is not obviously the case with Miéville's divided metropolis.
But in these cases the incomprehensible is mostly alien in origin, or the product of natural phenomena we don't (and maybe can't) understand, and not reflexive, whereas antimemes are a consequence not of the nature of the universe but the nature of the human brain, which is not a general purpose information processing machine--although it is capable of emulating one--but a special-purpose embodied prediction engine that is much better at some information processing tasks than others.
Antimemetics is a fertile metaphor for human thought. Consider:
"Millions of people died, and we just forgot. Is that what you want to show me? You want me to write that down?"
Ever read a political history of the 14th century? How about anything covering the First World War and its immediate aftermath? Have you wondered how Islam burst forth upon the world stage virtually unopposed?
How about what happened five or six years ago, and is still happening today?
Millions of people died, and we just forgot.
And then we are mostly incapable of understanding why we forgot.
I saw someone proudly say that they were incapable of understanding how it is that people are against Gestapo thugs terrorizing US cities, but are OK with tens of thousands of US citizens dying of covid every year. They then made no attempt to answer the question: no analysis, no inquiry, no nothing. Just "I don't get this" and then... "the rest is silence".
That incurious silence is a good indicator of an anti-meme at work: the people affected are probably incapable of thinking the answer to their question. So, very likely, are you.
If not an anti-meme, this kind of question-and-silence is definitely a memetic defense: it's a practised snippet of thought designed to protect the thinker from having to think thoughts that might put them in a position of facing a reality they really really don't want to see.
The fundamental unthinkable thought in these cases is that neurotypical humans are almost completely incapable of reasoning about causes that are not other human beings.
That is, neurotypical humans are good at understanding and responding to human action, and bad at understanding and responding to everything else: pandemics, system effects, natural disasters, you-name-it.
After all, why would they be any good at such reasoning?
A pro-social being with limited attention and processing power should be extremely optimized by Darwinian selection to ignore almost everything except threats and opportunities from other people, because those will be by far the biggest, most important, and most common threats and likewise the biggest, most important, and most common opportunities. Expending precious, limited, processing power and attention on non-human causes is a waste, so our very special-purpose brains are just not very good at it.
This explains why performance on the "Watson Selection Task" depends on whether the logic puzzle it embodies is posed in human or non-human terms. The puzzle is that there is a rule that says "A is true if B is true", and the subject is presented with some cases they can use to test the rule. They are asked to figure out which cases they need to check to see if the rule holds.
For example, the non-human version is usually that there are cards with a number on one side and a colour on the other, and the rule is "A card is blue if the number is even." Given four cards, one of which is showing 3, one showing 8, one showing blue, and one showing red, which ones do you need to turn over to check the rule for those four cards? Only about 10% of people get it right (I didn't).
Conversely, if we make it a problem about humans the rule could be something like, "You must be over 18 to drink alcohol" and the people involved all have drinks. One you know is sixteen has a glass of something, as does one you know is twenty, and one is drinking a can of Coke and one is holding a beer. Who do you have to check to see if the rule is being conformed to? Easy, right? You check what the sixteen-year-old is drinking and how old the person drinking beer is.
Almost everyone gets the human version right, whereas with the card version we tend to say we need to flip the 8-card, the blue card, and the red card. But the blue card is equivalent to the person you know is 20: it doesn't matter what they're drinking. The number on the back of the blue card could be odd or even and it wouldn't be relevant to the rule, because the rule only says, "If the number is even, the card must be blue" not "If the card is blue, the number must be even", which would be equivalent to, "If you are over 18 you must drink alcohol."
Reasoning about human causes is extremely easy, reasoning about non-human causes is extremely hard. This is just a fact, and it is a fact that is consistent with the idea that our brain evolved to deal with human causes, because of course it did: they were by far the most important thing in the lives of our ancestors, who lived in relatively small groups that worked together to "fill the Earth and subdue it." Pretty successfully, too: travelling out of Africa to reach the Americas and everywhere else in less than seventy thousand years.
Furthermore, we know that human perception is tuned up to deal much better with faces than anything else. Why wouldn't human reasoning be tuned up to deal with human causes better than anything else? And just as the highly specialized "fusiform face area" of the brain can be repurposed to differentiate other similar objects from each other, so can the highly specialized cognitive processes of the brain be--with difficulty and training--repurposed to think about things other than human causes. Just not all that easily.
Why do people insist, contrary to evidence, that we are as good at dealing with non-human causes as human causes when there is any amount of data showing we routinely ignore non-human causes that completely dominate historical eras?
I propose that the answer to this is that the claim that humans are not the only important causal agents in the universe is an anti-meme.
You're probably rejecting the claim right now based on your belief that plagues couldn't possibly be that big a deal because if they were they would be written about in histories at greater length, instead of those histories focusing on the important things: human choices and actions.
Think about that argument for a moment.
When human agency kills or hurts people, it's a big deal. When non-human causes do the same, they are not processed, at least not without enormous effort, decades of training, or exceptional intelligence, because our brains are not evolved to be general information processing units. Again: why would they be? There as no selective pressure in the ancient veldt to be a better general information processor. Our brains evolved to allow us to be good members of tightly integrated gangs that work together like no other species on Earth, while at the same time getting us as much power, influence and mating success within that gang as we can.
So today, hundreds of thousands of years later, non-human causes are simply ignored if it's at all possible. When they're too big to be ignored--as in the acute phase of the ongoing pandemic--they are experienced as trauma: there is no way to integrate them into the heroic narrative everyone tries to build around their life. Although we do try. Health care workers were briefly repurposed to serve as communal heroes, until it required actual work, at which point we threw them under the bus and washed our hands--literally--of them.
Both as individuals and as societies we experience the effect of non-human causes as trauma.
Unprocessed trauma is what leads to PTSD: experiences that can't be remembered in a way that is consistent with the narrative we use to construct our identity rattle around loose inside our psyches, bashing into things and sprawling all over otherwise benign interactions. Because they aren't well-integrated into our narrative network of memories and beliefs, they are hard to recall. We don't have many ways of approaching those thoughts and memories compared to better-integrated memories, and when we do bring them close to the surface, they hurt. They are anti-memes: thoughts that make it hard to think them.
One way humans try to deal with this is to integrate non-human causes by attributing them to a human cause, which is why the "lab leak" thing gets so much attention. It has no practical import: we know lab leaks happen, because they have happened before. We know bureaucrats lie to cover their asses, because they have done so before. Knowing those things happened in this case--which is what the "lab leak" idea posits--changes nothing and contributes nothing to planning for future pandemics. But people obsess about it because they really really need a pandemic that killed millions of people to have a human cause. Anything else is trauma.
Now that the acute phase of the ongoing pandemic is over, interest in the lab leak idea has faded because ignoring the fact that covid killed millions of people has become easier. The dead are silent, as always. There will be no memorials to them.
Since humans don't believe in what they can't see, no one will ever accept this analysis, and everyone will instead continue to engage in the kind of performative anti-memetics that is the other thing that prompted me to write this. And to be clear: I'm writing this to clarify my own thoughts, not to convince anyone, because it won't: it can't, because neurotypical humans are not capable of entertaining the idea that non-human causes are enormously important in human life and human history.
Some thoughts are harder to think than others.
It is impossible for most people to think: "My life is dominated by events that are beyond the control or influence of any human being or organization."
This has extremely far-reaching consequences for individuals and society. Amongst other things, in times when non-human causes have the greatest impact there will be a collective desire to find human causes that distract from this fact. This comes out in many different ways. People invent human causes for diseases from cancer to covid. And they look to leaders who claim to protect them from imaginary human threats lurking within the populace. Anything to distract from or repurpose the generalized anxiety that being repeatedly brought face to face with a persistent anti-meme creates.
People will burn the world before they will admit that non-human causes dominate their lives. Just watch them.