Pretty much anyone who was alive in 1620 will by now either have no descendants or a ridiculously large number of descendants.
If his descendants continued to breed at the rate of 2.5 children each at the age of 25 then by now he'd have approximately 1.6 million descendants in the latest generation (plus another million or so from their parents and grandparents' generations, still alive). Bump it up to an average of three children and we're looking at 30 million descendants. Bring the breeding age down to 22 and we're talking 287 million descendants, which just happens to be the population of the US. Of course you'd need to adjust that downwards since there'd be plenty of (mostly rather remote) inbreeding along the way.
If you could get a complete list of your (seven hundred thousand or so) ancestors dating back to the 17th century you'd probably be totally thrilled by the number of famous names on it. But of course, everyone else would have a similar number of famous names on their own.
Without analysis of the counter-factual, the list seems impressive. But had Howland died there is a high probability that there still would have been a 32nd, 41st, and 43rd US president; prominent 19th century poets; Alaskan governors; pediatricians and Hollywood leading men. And listing both Bush's seems gratuitous - why not list all the Baldwin brothers?
Probably the only two individuals on the list whose impact on US history might be considered irreproducible are Smith and to perhaps a lesser extent, Young because creating enduring spiritual systems [Mormonism] and political states [Utah] seems more dependent on individual characteristics than the roles filled by other individuals.
There's a famous Twilight Zone episode about a time-traveling assassin who kills baby Adolf Hitler. The episode concludes that nothing really changes, as another child growing up in that environment becomes just as bad as an adult.
I'm not sure if I believe it, but it's an interesting argument. There almost certainly would have been people in most of those positions, though perhaps not with the same quirks.
In other words, it is likely that Dole [had he won the Republican nomination] would have defeated Dukakis as readily as Bush did in 1988, or that Carry Grant could have starred in The African Queen.
On the other hand, the specifics of Smith's personality and circumstances are hard to separate from the movement he spawned.
If you're a man and you have children, anything you did differently before you had them would have resulted in different children because your sperm would have shifted.
This may or may not be possible, depending on certain factors.
First, assume that this would cause a paradox, basically, an exception of type CausalityFailException.
If the universe has runtime checking, you go back in time, eliminate Howard, all those people, and then they never existed, your motivation never existed, and you can't have gone back in time. Basically, the stack of history is now in an unknown state and calculating the state of the universe causes the temporal stack to overflow.
IF, however, the universe had compile-time checking, you'd get a simple, much cleaner
13/12/2010 09:43:23 ERROR: Cannot compile module "yummyfajitas-timetravel" - Method "EliminateTheFuckwits" has uncaught Temporal Side-Effects.
This leads us to believe that any language which can do compile-time checking is necessarily better for debugging, quality, and universe stability.
If you're wondering whether the story is true or not, it appears to be. All of the relevant facts are present in John Howland's (heavily cited) Wikipedia article.
If his descendants continued to breed at the rate of 2.5 children each at the age of 25 then by now he'd have approximately 1.6 million descendants in the latest generation (plus another million or so from their parents and grandparents' generations, still alive). Bump it up to an average of three children and we're looking at 30 million descendants. Bring the breeding age down to 22 and we're talking 287 million descendants, which just happens to be the population of the US. Of course you'd need to adjust that downwards since there'd be plenty of (mostly rather remote) inbreeding along the way.
If you could get a complete list of your (seven hundred thousand or so) ancestors dating back to the 17th century you'd probably be totally thrilled by the number of famous names on it. But of course, everyone else would have a similar number of famous names on their own.