What Doesn’t Kill Me Makes Me Stronger

Yeah... what about bacteriocidal antibiotics that cause mitochondrial DNA damage? The people affected by that stuff don't seem to just like recover and get stronger.

A very interesting concept, will definitely need to read more on this!

But what about the kind of research that talks about the dangers of living in cities... like "living in a polluted city is like smoking X cigarettes a day"?

Mitteldorf has an interesting take on programmed aging, which is to say the collection of theories suggesting aging is a evolved program that acts to shorten life because there is some very global selection advantage in it. It is worth reading as a counterpoint to the more common viewpoints of that school (such as the hyperfunction theory of aging that is a modern take on antagonistic pleiotropy in the context of programmed aging) espoused by some of the Russian gerontology community. There's a link somewhere in the article linked above.

Programmed aging is, however, a minority view in the aging research community as a whole. The consensus view is that aging is caused by an accumulation of unrepaired damage, though there are many factions and a lot of debate within that tent. Programmed aging seems to be gaining some ground, but it's rather hard to tell from the sidelines as some of the advocates (e.g. Blagosklonny and his views on mTOR) are very prolific in their publications.

Hormesis as a phenomenon to be measured and evaluated can stand apart from either of these views on aging for the purposes of evaluation and investigation of molecular mechanisms. It is a robustly demonstrated thing in animal models, though as for all these things translating those findings into human health is ever a challenge. For things like calorie restriction, exercise, and intermittent fasting, where hormesis is thought to play an important role, the human and rodent responses in the short term are very similar. There is a small mountain of papers on this topic - just go look at PubMed and search for hormesis and longevity.

Hormesis works because some forms of damage - such as mild oxidative stress - trigger repair responses that last long enough and are proficient enough to produce a net benefit in cell health throughout tissues. There is a dose-response curve to all of this of course. This is may be how you get a variety longevity mutants in nematode worms wherein they live longer if you either reduce or increase the flux of reactive oxygen species emitted from the mitochondria. Less means less damage and more means less damage because it produces more aggressive repair.

There are plenty of ways to damage tissue that will cause incremental damage over time, but are not hormetic, and will not produce benefits. It all depends on how the repair mechanisms handle the specific case in question.

Also:

"Walking around with a 40-pound backpack has the opposite effect of carrying an extra 40 pounds of belly fat."

Actually walking around with a 40-pound backpack is a bad idea, as compared to carrying the weight at the level of your belly (all other things being equal) because the risk of back damage is much higher.

I'm very skeptical.

Aging isn't there because our genes "need" it. It's there because our genes don't need long lifespans. The gain in reproductive viability that we'd get with a >60 year natural lifespan just isn't enough to justify the constraint that a much longer (or indefinite) lifespan would impose on our genetic "search space". We can live forever and be simple, or we can be complex and get a job done and die.

There probably is some anti-fragility in us. I don't buy the LNT threshold of radiation, for example. All that said, I'm not sure that hormesis is useful as a general concept. There are mild stressors (such as cold, when prepared and for short duration) that turn out to have positive effects, but there are a great number of stressors that seem to have no positive effects, whether you're talking about biological agents (e.g. cadmium, lead) or psychological experiences (e.g. rape, war).