3 ms·

> We should be extremely wary of highly educated technocrats who attempt to impose a "scientific" viewpoint on political events, always fail, and yet remain utt

> We should be extremely wary of highly educated technocrats who attempt to impose a "scientific" viewpoint on political events, always fail, and yet remain utterly convinced by their oversimplified models. Could you detail that one? Is it arguing for more emotional driven decisions over data based ones?

vharuck 7y ago
It sounds more like a politician who predicts a situation and backs it up with good-sounding logic. But then, when the situation doesn't happen as predicted, the politician finds an excuse that doesn't invalidate his justifications.
- deogeo 7y ago
  Or a scientist that claims all objects are pulled with equal acceleration by gravity, yet when he drops a pillow and a brick from a balcony, the brick hits the ground first.
  - baked_ziti 7y ago
    Pardon me if you're being sardonic and I'm not detecting it, but you're pointing out the flaw is this form of criticism yes?
    - deogeo 7y ago
      Yes.
guelo 7y ago
One example is the Laffer curve economics theory with proponents claiming repeteadly over decades no matter how many times it fails that tax cuts will increase government revenue.
jasonlotito 7y ago
No, though I understand how you could interpret it that way. The key part is this: "and yet remain utterly convinced by their oversimplified models." That's emotion talking. A data driven approach would accept the data of the failure, and look to improve. It's warning of people who use science or data to enact a plan, and when that plan fails, falls back on emotion and belief rather than listening to the evidence.
repolfx 7y ago
No. What it's saying is that data based decisions are not always right and may in fact often be more wrong than an emotional decision. The argument can be found in much greater detail in the book "Wrong" by David Freedman: https://www.amazon.com/Wrong-us-Scientists-relationship-consultants/dp/0316023787 https://www.amazon.com/Wrong-us-Scientists-relationship-cons... and is explored from a slightly different angle in this essay by Scott Alexander: https://slatestarcodex.com/2019/06/03/repost-epistemic-learned-helplessness/ https://slatestarcodex.com/2019/06/03/repost-epistemic-learn... Here's a brief summary of the book. The world is full of people, presented to us by the media as experts, who constantly make predictions. These predictions are not sometimes wrong, but it turns out are mostly/almost always wrong. Some data is provided to back this up. These people usually share the same characteristics. They are wrong because they radically over-estimate their understanding of complex systems, and apply simple models as a consequence. Sometimes it may be impossible to construct a model sophisticated enough to do what they're trying to do, but this doesn't stop them. Sometimes the media garbles or exaggerates what's being predicted, or doesn't communicate uncertainty. And then when their predictions turn out to be wrong, they are isolated from any kind of accountability or reality check. Freedman focuses quite a bit on 'experts' in health, dieting etc (e.g. does red wine cause cancer or not?) but examines other areas too, and arguably they're more important. For instance economists are always wrong unless they're predicting a continuation of the current trend line, which is pretty useless, but no economist is ever fired because their predictions turned out badly. And he argues that one reason there's no accountability is that many people refuse to downgrade their belief in these experts, even when presented with repeated evidence of their wrong beliefs. The second link, the Alexander essay, explores why people may often be right to make an apparently emotion-driven decision. His argument boils down to evaluating arguments being really hard, and it being easy to be convinced by arguments that are wrong, and the world being full of wrong arguments that would have radically devastating outcomes on your life if you really took them seriously and tried to live by their conclusions, so it's often smarter to just ignore (apparently) rational arguments and reject anything that sounds "weird".

vharuck 7y ago

It sounds more like a politician who predicts a situation and backs it up with good-sounding logic. But then, when the situation doesn't happen as predicted, the politician finds an excuse that doesn't invalidate his justifications.

deogeo 7y ago
Or a scientist that claims all objects are pulled with equal acceleration by gravity, yet when he drops a pillow and a brick from a balcony, the brick hits the ground first.
- baked_ziti 7y ago
  Pardon me if you're being sardonic and I'm not detecting it, but you're pointing out the flaw is this form of criticism yes?
  - deogeo 7y ago
    Yes.

guelo 7y ago

One example is the Laffer curve economics theory with proponents claiming repeteadly over decades no matter how many times it fails that tax cuts will increase government revenue.

jasonlotito 7y ago

No, though I understand how you could interpret it that way. The key part is this: "and yet remain utterly convinced by their oversimplified models." That's emotion talking. A data driven approach would accept the data of the failure, and look to improve. It's warning of people who use science or data to enact a plan, and when that plan fails, falls back on emotion and belief rather than listening to the evidence.

repolfx 7y ago

No. What it's saying is that data based decisions are not always right and may in fact often be more wrong than an emotional decision. The argument can be found in much greater detail in the book "Wrong" by David Freedman: https://www.amazon.com/Wrong-us-Scientists-relationship-consultants/dp/0316023787 https://www.amazon.com/Wrong-us-Scientists-relationship-cons... and is explored from a slightly different angle in this essay by Scott Alexander: https://slatestarcodex.com/2019/06/03/repost-epistemic-learned-helplessness/ https://slatestarcodex.com/2019/06/03/repost-epistemic-learn... Here's a brief summary of the book. The world is full of people, presented to us by the media as experts, who constantly make predictions. These predictions are not sometimes wrong, but it turns out are mostly/almost always wrong. Some data is provided to back this up. These people usually share the same characteristics. They are wrong because they radically over-estimate their understanding of complex systems, and apply simple models as a consequence. Sometimes it may be impossible to construct a model sophisticated enough to do what they're trying to do, but this doesn't stop them. Sometimes the media garbles or exaggerates what's being predicted, or doesn't communicate uncertainty. And then when their predictions turn out to be wrong, they are isolated from any kind of accountability or reality check. Freedman focuses quite a bit on 'experts' in health, dieting etc (e.g. does red wine cause cancer or not?) but examines other areas too, and arguably they're more important. For instance economists are always wrong unless they're predicting a continuation of the current trend line, which is pretty useless, but no economist is ever fired because their predictions turned out badly. And he argues that one reason there's no accountability is that many people refuse to downgrade their belief in these experts, even when presented with repeated evidence of their wrong beliefs. The second link, the Alexander essay, explores why people may often be right to make an apparently emotion-driven decision. His argument boils down to evaluating arguments being really hard, and it being easy to be convinced by arguments that are wrong, and the world being full of wrong arguments that would have radically devastating outcomes on your life if you really took them seriously and tried to live by their conclusions, so it's often smarter to just ignore (apparently) rational arguments and reject anything that sounds "weird".