I can't understand why people who think about longevity miss the logical error built into it -- as biological causes of death are remedied, the prospect of non-

I can't understand why people who think about longevity miss the logical error built into it -- as biological causes of death are remedied, the prospect of non-biological causes of death go up until they become the cause of death instead. This graphic (PDF warning) -- http://www.nsc.org/news_resources/injury_and_death_statistics/Documents/2014-Injury-Facts-43.pdf http://www.nsc.org/news_resources/injury_and_death_statistic... -- shows that a person's lifetime (say, 75 years) probability of dying of any of the top four non-biological causes of death (suicide, car crash, toxic exposure, falling) is 0.033 (roughly 1 in 30). Let's say this is for a normal biologically limited lifetime of 75 years. Using this as a basis, the Binomial Theorem shows that we have a probability of .000448 per year of dying of any one of these causes, and, for 75 years, this probability produces the quoted chance of 0.033 for any of these four causes. For longer lifetimes, this outcome means that: Age Probability ---------------------- 75 0.0330 100 0.0438 200 0.0857 500 0.2007 1000 0.3612 2000 0.5919 5000 0.8936 This outcome, for four common non-biological causes of death, means that there is no "indefinite lifespan" -- all that happens is the way we die changes. I hate to rain on your parade, but this discussion of longevity fails to notice an elephant in the room. My Python code for this result: #!/usr/bin/env python # -*- coding: utf-8 -*- from scipy.stats import * k = 0 p = .000448 print(' %8s %12s' % ('Age','Probability')) print(' %s' % ('-' * 24)) for n in (75,100,200,500,1000,2000,5000): r = 1 - binom.pmf(k,n,p) print(' %8.0f %12.4f' % (n,r))