5 ms·

I’m sceptical, it’s too easy to cherry-pick data to make it look supportive of one’s political agenda. Not saying it’s the case here, I just wouldn’t trust too

I’m sceptical, it’s too easy to cherry-pick data to make it look supportive of one’s political agenda. Not saying it’s the case here, I just wouldn’t trust too much "obvious" patterns and easy to understand maps from a twitter thread. More often than not, such threads are omitting data that doesn’t support the view of the author. Reality is rarely obvious and simple.

alephxyz 4y ago
It's more of a weak indirect correlation between historical map -> economic prosperity -> political preferences. The modern political maps for France, Germany, Portugal and Romania at least are essentially proxies for GDP per capita (not sure about the others).
- hammock 4y ago
  One might add something to the front of your analysis, as maps don't form out of thin air. In fact, those lines on a map are formally called "political borders." historical governance -> historical map & economic prosperity -> political preferences
  - lrem 4y ago
    Political borders also don't just happen to be drawn on a piece of paper. They tend to snap to physical barriers, mostly rivers and mountain ranges.
LeanderK 4y ago
I totally assumed the examples to be cherry picked but it's still fun to look at. Not every past border is visible on modern maps, but still some are clearly the consequence of those, for example the german maps where the old GDR is the reason for the high rate of atheism or popularity of more extreme parties. I think that often past borders are the consequence of different demographics, which clearly translate into modern differences.
- hammock 4y ago
  By definition the examples are cherry-picked. I noticed he didn't pull up old Visigoth ranges, nomadic farmers or Babylonian borders. That doesn't take away from the insight that certain political histories still resonate today. Just as you can find examples of geologic features (like mountain ranges or rivers) influencing political divides, so too can you find examples of geologic features not influencing them. Doesn't mean the former is "misleading" or something.
- derbOac 4y ago
  When I was reading through the examples I sort of wondered why sometimes these geopolitical phenomena persist, and other times not. For example, at one time long ago in the US, politics were dominated by "frontier-interior" versus "urban-coastal" dynamics, and then as the US grew, and the civil war came about, it established a lot of the geopolitical patterns evident today (although I'd argue the urban-rural distinction is maybe reemerged in a more distributed way today). To explain cherry-picking versus something else, you'd want a theory for why patterns sometimes change and other times remain the same.
orbital-decay 4y ago
Pretty much this. Without deep historical analysis, this doesn't mean anything. It's like amateur etymology that is based on how similar certain words are, which typically gets things laughably wrong.
not2b 4y ago
Oh, come on: he had numerous examples from all over the world where the political implications are quite different. The author's view was simply that old political boundaries have lingering effects, and he demonstrated that very well.
- qsdf38100 4y ago
  As I said, I’m not saying it’s the case here. It’s just collateral damage of the disinformation age we live in. I’m not trusting "trends" in data unless I spend 1 or more hours looking it up myself and hoping I can avoid being mislead. Here I guess it didn’t help that the first map was trying to convey a link between Macron voters and British inhabitants. Anytime I see suggested links between voters, culture, immigration, etc. I’m prudent. It’s surely interesting in its own, but I can’t resist asking myself "why is the author mentioning this? Did he came across this data and then found interesting correlations? Or was he looking for trends that would bring support to some views he has? It’d be ok-ish if such approach was open, like, "here’s my thesis, and here’s data supporting this thesis." But when there’s no thesis, just "interesting trends", pretending it’s up to the reader to make his own conclusions… More often than not, it’s a dishonest cherry-picking with some more or less forced untold conclusions waiting to be made. So, I just stay away from such articles, even if I know some of them aren’t dishonest. It’s collateral damage of the disinformation war.
ssnistfajen 4y ago
Absolutely right to be skeptical. The regional demographics in many of these overlapped regions have gone through comprehensive changes over centuries under drastically different regimes. The map about English posessions in 1154 vs. first round of French presidential election is just an absolute straw-grasping joke. ~900 year's worth of cultural, demographic, and economic shifts reduced to basic visual pattern recognition AKA the thing that causes people to see Jesus on a piece of toast. The map about Charlemagne's empire vs. 6 original members of the future EU is a joke too. The latter was a product of the reality of post-WWII European politics (Iron Curtain, Allied occupation of Germany and Austria, and Spain being a neutral country ruled by de facto dictatorship). It's fun to look at patterns and think they are neat. It is absolutely not fine to pretend these resemblances somehow have a direct & causal relationship without properly navigating the long and complex multi-faceted history behind them.

alephxyz 4y ago

It's more of a weak indirect correlation between historical map -> economic prosperity -> political preferences. The modern political maps for France, Germany, Portugal and Romania at least are essentially proxies for GDP per capita (not sure about the others).

hammock 4y ago
One might add something to the front of your analysis, as maps don't form out of thin air. In fact, those lines on a map are formally called "political borders." historical governance -> historical map & economic prosperity -> political preferences
- lrem 4y ago
  Political borders also don't just happen to be drawn on a piece of paper. They tend to snap to physical barriers, mostly rivers and mountain ranges.

LeanderK 4y ago

I totally assumed the examples to be cherry picked but it's still fun to look at. Not every past border is visible on modern maps, but still some are clearly the consequence of those, for example the german maps where the old GDR is the reason for the high rate of atheism or popularity of more extreme parties. I think that often past borders are the consequence of different demographics, which clearly translate into modern differences.

hammock 4y ago
By definition the examples are cherry-picked. I noticed he didn't pull up old Visigoth ranges, nomadic farmers or Babylonian borders. That doesn't take away from the insight that certain political histories still resonate today. Just as you can find examples of geologic features (like mountain ranges or rivers) influencing political divides, so too can you find examples of geologic features not influencing them. Doesn't mean the former is "misleading" or something.
derbOac 4y ago
When I was reading through the examples I sort of wondered why sometimes these geopolitical phenomena persist, and other times not. For example, at one time long ago in the US, politics were dominated by "frontier-interior" versus "urban-coastal" dynamics, and then as the US grew, and the civil war came about, it established a lot of the geopolitical patterns evident today (although I'd argue the urban-rural distinction is maybe reemerged in a more distributed way today). To explain cherry-picking versus something else, you'd want a theory for why patterns sometimes change and other times remain the same.

orbital-decay 4y ago

Pretty much this. Without deep historical analysis, this doesn't mean anything. It's like amateur etymology that is based on how similar certain words are, which typically gets things laughably wrong.

not2b 4y ago

Oh, come on: he had numerous examples from all over the world where the political implications are quite different. The author's view was simply that old political boundaries have lingering effects, and he demonstrated that very well.

qsdf38100 4y ago
As I said, I’m not saying it’s the case here. It’s just collateral damage of the disinformation age we live in. I’m not trusting "trends" in data unless I spend 1 or more hours looking it up myself and hoping I can avoid being mislead. Here I guess it didn’t help that the first map was trying to convey a link between Macron voters and British inhabitants. Anytime I see suggested links between voters, culture, immigration, etc. I’m prudent. It’s surely interesting in its own, but I can’t resist asking myself "why is the author mentioning this? Did he came across this data and then found interesting correlations? Or was he looking for trends that would bring support to some views he has? It’d be ok-ish if such approach was open, like, "here’s my thesis, and here’s data supporting this thesis." But when there’s no thesis, just "interesting trends", pretending it’s up to the reader to make his own conclusions… More often than not, it’s a dishonest cherry-picking with some more or less forced untold conclusions waiting to be made. So, I just stay away from such articles, even if I know some of them aren’t dishonest. It’s collateral damage of the disinformation war.

ssnistfajen 4y ago

Absolutely right to be skeptical. The regional demographics in many of these overlapped regions have gone through comprehensive changes over centuries under drastically different regimes. The map about English posessions in 1154 vs. first round of French presidential election is just an absolute straw-grasping joke. ~900 year's worth of cultural, demographic, and economic shifts reduced to basic visual pattern recognition AKA the thing that causes people to see Jesus on a piece of toast. The map about Charlemagne's empire vs. 6 original members of the future EU is a joke too. The latter was a product of the reality of post-WWII European politics (Iron Curtain, Allied occupation of Germany and Austria, and Spain being a neutral country ruled by de facto dictatorship). It's fun to look at patterns and think they are neat. It is absolutely not fine to pretend these resemblances somehow have a direct & causal relationship without properly navigating the long and complex multi-faceted history behind them.