Pinker gave many reasons, but mostly from high above, using general statistical patterns to make predictions on a very specific issue.
Rees's argument was short. It was one plausible scenario.
I often find the latter more compelling than the former.
Rees's argument was short. It was one plausible scenario.
I often find the latter more compelling than the former.
The bias itself is probably a cousin of availability heuristic - https://en.wikipedia.org/wiki/Availability_heuristic.
You can't really say "Rees must be having a cognitive bias, and his estimate of the bioterror/bioerror has to be grossly overstated". Rees is not a population. the Law of large doesn't apply. He might have been too optimistic. Even if he is overly pessimistic, maybe the inflation of risk isn't that significant. Or it is significant, but the risk is so great, that even after subtracting the inflation, it is still dangerous.
The point is, applying a statistical heuristic without supplying the actual parameters is meaningless. The heuristic itself is not bullshit. The application in this case certainly is.
Now, this may be a situation where the sides think they're talking about different things: Pinker may be arguing that the probability is <50%, while Rees may be arguing that the probability, while low, is non-negligible. In which case, in my opinion, they'd both be right: But Rees' bet (again in isolation as a bet) isn't rational (since the expected payoff is negative), while Pinker's is.
Let's take a minute to remember that our brains are shaped, by millions of years of evolution, to model existential threats and act accordingly. I would say that what we deem "imaginable" is a pretty good proxy to what is "probable", especially in the long run.
There's also a dishonest sleight of hand here - when determining a course of action, it's not the probability alone you look at, but the probability multiplied by the expected harm. The probability of getting in a car accident is low, but the expected harm is high enough that the hassle of wearing a seatbelt each time you go for a drive is still worth it. If he were being consistent, Pinker would say: "why are you putting on a seatbelt? Just because you can imagine a car accident, doesn't mean it will happen today".
The purpose of the 'long bet' exercise isn't really about the money - it's about our ability to reason about the future. So talking about these things as abstract probabilities as if they are the outcomes of sporting events, not the lives of a million people, is dishonest bordering on morally repugnant.
It is not a good proxy at all. The environment in which we lived for millions of years does not correspond well to our modern civilization. Being killed by a member of another tribe was a likely event for most of human and animal history. Today, being killed by a toaster is many orders of magnitude more likely than being killed by a terrorist. Yet many people fear terrorists, while no one fears toasters.
Who would have thought that a brain evolved by taking in and balancing things based on interacting with a couple hundred other members of their species and maybe a 100 miles in any given direction wouldn't be very good when taking in information from billions of other people and millions of miles of territory.
I don't think this is true. Googling it, it seems to originate from the claim "you should be more scared of toasters than sharks" based on the number of toaster deaths (source of this data is unclear - Fermi estimate?)
The only data I could find cites 21 deaths in the US in 2000 from "countertop cooking." By comparison, there were 211 mass shooting deaths in the US in 2019 per Northeastern University. https://www.cpsc.gov/s3fs-public/hazard_housewares.pdf
Why or why not by the end of 2020? Why not by the end of 2021 or 2025 or 2050?
They're both mute on this point.
Yeah, I get that. But why or why not should it happen in 4 years? Why not 6 or 8 or 3?
These numbers seem just pulled out of a hat, and their arguments would make as much or as little sense with a different date.
Let's say you're trying to argue why we'll run out of helium in, say, 10 years.
You could say: we have 10X quantity of helium, we use 1X per year, therefore we'll run out in 10 years.
Or you could say: we use a lot of helium. We're bound to run out of helium sometime, therefore we'll run out in 10 years.
The arguments presented in this bet are of the second sort. They do not say anything about why what they're betting on should happen in 10 years or in any other timeframe.
Sure, the bet is for the end of 2020, but in order to be persuasive you actually have to make an argument for why what you predict does or does not happen in that timeframe.
No such arguments were even attempted here. That's what's so frustrating.
Simply put, they guessed; it's a gut feeling.
And whether an argument is persuasive depends on what the argument is and on the people evaluating it.
Arguments with using hard data and actual reasons for picking a certain time frame are not automatically persuasive.
Take a look at the Peak Oil arguments that were made. They had pretty hard numbers and reasons for their predictions, but plenty of people were not persuaded.
Another example is Ray Kurzweil's predictions about the singularity, which also had their reasons, statistics, etc, but plenty of people disputed them and made counterarguments.
But least you could tell why they believed their predictions would happen within that timeframe.
No such evaluation is possible with the "bioerror" predictions, however, since they didn't make arguments for why it should happen by the end of 2020, but instead just expressed their "gut feeling", as you point out.
That's absolutely worthless.
A gut feeling is absolutely an argument. It may not be a well-reasoned, evidence-supported, or persuasive argument but it absolutely 100% is an argument.
Hence Pinker's remark about 15 years of hindsight, and the fact that some of the comments below the line were posted 19 years ago.
With this in mind, I'd suggest that 2020 was picked as an arbitrary "sci-fi future" date rather than for an arbitrary term of 4 years. Still arbitrary yes, but more understandable in my view.
We might find one compelling, but what matters is which is more accurate (or maybe neither of them are).
Rees basically said, I bet this is going to happen before 2020 ends, because my gut said so.
Pinker multiplied a bunch of statistical principals without a number. I really don't understand how to reach a conclusion based on his line of reasoning.
At least the gut feeling of an expert is somewhat a reasonable point.
Often intelligent academic people can justify almost anything using their toolset of theories from different fields when they intuitively feel something is right. That intuitive feel is what really matters most anyway