4 ms·

Wouldn't the obvious answer seem to be to provide certainty factor estimates where varying degrees of evidence are available? Anyways, I've rarely ever seen hi

Wouldn't the obvious answer seem to be to provide certainty factor estimates where varying degrees of evidence are available? Anyways, I've rarely ever seen historians qualify their data points. Unfortunately humility is only for the scientists, and because they can't escape the same consequences of uncertainty when drawing conclusions from data (i.e. statistical significance). Confidence is for the rest of us, not because it is helpful, but because our human nature commands it.

nerdponx 9y ago
Instead of trying to conduct some kind of comprehensive probability analysis, admirable as the idea might be, you could just be clear and say "this is roughly the point in history I'm representing."
- wallace_f 9y ago
  I was thinking more along the lines of a metric indicating the probability the conclusion is correct, based on the strength of evidence.
stevenwoo 9y ago
You might be interested in Lies My Teacher Told Me by James W. Loewen. He's a sociologist but he brings up this point when talking about the different possible discovery dates fo the Americas by the Japanese, Africans, Polynesians and others at one point - history books have a tendency to wrongly paint many things as objective truth when there should be more subtlety. https://en.wikipedia.org/wiki/Lies_My_Teacher_Told_Me https://en.wikipedia.org/wiki/Lies_My_Teacher_Told_Me
jcranmer 9y ago
> Wouldn't the obvious answer seem to be to provide certainty factor estimates where varying degrees of evidence are available? The uncertainty involved in history is rarely quantifiable. How would you go about attaching a number to, say, the probability that Jesus existed? Where such quantifiability is possible--dates in radiocarbon and demographic estimates come to mind--you in fact do see estimation ranges from historians. > Anyways, I've rarely ever seen historians qualify their data points. Having read a few archaeological papers out of curiosity, yes, these data points are often qualified where prudent. This is much less the case in popular history books, but it's worth noting that popular science books are equally egregious in omitting such details. > Unfortunately humility is only for the scientists, and because they can't escape the same consequences of uncertainty when drawing conclusions from data (i.e. statistical significance). Let's not put scientists on some holy pedestal here. Having read science papers as well, I can tell you that scientists are quite bad at being humble and qualifying their results with limits of uncertainty.

nerdponx 9y ago

Instead of trying to conduct some kind of comprehensive probability analysis, admirable as the idea might be, you could just be clear and say "this is roughly the point in history I'm representing."

wallace_f 9y ago
I was thinking more along the lines of a metric indicating the probability the conclusion is correct, based on the strength of evidence.

stevenwoo 9y ago

You might be interested in Lies My Teacher Told Me by James W. Loewen. He's a sociologist but he brings up this point when talking about the different possible discovery dates fo the Americas by the Japanese, Africans, Polynesians and others at one point - history books have a tendency to wrongly paint many things as objective truth when there should be more subtlety. https://en.wikipedia.org/wiki/Lies_My_Teacher_Told_Me https://en.wikipedia.org/wiki/Lies_My_Teacher_Told_Me

jcranmer 9y ago

> Wouldn't the obvious answer seem to be to provide certainty factor estimates where varying degrees of evidence are available? The uncertainty involved in history is rarely quantifiable. How would you go about attaching a number to, say, the probability that Jesus existed? Where such quantifiability is possible--dates in radiocarbon and demographic estimates come to mind--you in fact do see estimation ranges from historians. > Anyways, I've rarely ever seen historians qualify their data points. Having read a few archaeological papers out of curiosity, yes, these data points are often qualified where prudent. This is much less the case in popular history books, but it's worth noting that popular science books are equally egregious in omitting such details. > Unfortunately humility is only for the scientists, and because they can't escape the same consequences of uncertainty when drawing conclusions from data (i.e. statistical significance). Let's not put scientists on some holy pedestal here. Having read science papers as well, I can tell you that scientists are quite bad at being humble and qualifying their results with limits of uncertainty.