So we already have evidence that something with long term negative impacts (and smoking has some pretty immediate easily noticeable negative impacts as well), can persist without this being obvious for a while.
In addition, population level data is messy. For example, we know there's been a massive increase in depression amongst children, and deaths of despair in the US over the past decade. What if those are related to cell phone use? That would mean the effect is obvious but connecting it to the primary cause is not easy, despite the fact that there is indeed correlations with cell phone use (the correlations people have pointed out is to the rise of social media, which can quite easily also be a correlation to cell phone use). Sounds far fetched? Possibly. But then again, the US saw a massive unexplained drop in crime over the last 3 decades, and arguably the best explanation we have so far is the phasing out of lead from gasoline about 2 decades before the drop actually happened. When the link was first suggested it sounded equally far fetched to most.
I don't think there's enough evidence to panic yet (although it would be prudent to follow suggestions around keeping distance from your cell phone), but it's far too early to say the conclusion is settled and that there is no health risk, especially considering the limited research on this topic yet.
I didn't say that we'd stop using cell phones if they posed significant dangers. I said that the abrupt increase in usage would make those dangers obvious if they existed.
To use your example, Tobacco was seen as bad from the git go because the impact was obvious. King James wrote a book against it in 1604:
>A custome lothsome to the eye, hatefull to the Nose, harmefull to the braine, dangerous to the Lungs, and in the blacke stinking fume thereof, neerest resembling the horrible Stigian smoke of the pit that is bottomelesse.
And the fact that the dangers were known to some, but yet society widely considered tobacco safe and even good for you, does not help the idea that the dangers would be obvious. If we continue the analogy, it's very possible people 50 years from now will be pointing to this paper to show that the dangers of cellphone radiation was known in the early 2020s, even though society and the scientific community broadly haven't accepted these dangers.
The US population was smaller at that time, so it was probably more. It also peaked at about 4500 in the 60s [1], so it was even higher then.
I do suspect the pareto effect applied here, and the heaviest 20% of smokers raised the mean for the entire population.
For example, about 70-80% of alcohol consumed in the US is consumed by the top 20% drinkers.
1: https://nap.nationalacademies.org/read/11795/chapter/4#42
Dear god, each US citizen smoked 5 cigs per day on average in 1930? It's mind-boggling.
