Thursday, July 11, 2019

Princeton Astrophysicist's "Doomsday" Math Shows Humans Have Less Than 1,000 Years Left

Movie 'Threads'  which depicts the end of civilization from nuclear war,  starting with a U.S. attack on Iran. Dr.J. Richard Gott whose 'doomsday' math ought to have our collective attention. 

The article,  'The Unsettling Possibilities Of Doomsday Math' (WSJ, Review, July 6-7) by William Poundstone, makes compelling reading though I fear too many may be inclined to dismiss it. Or to put it into the same category as the idea that some fraction of UFOs may actually be advanced craft piloted by extraterrestrials.  

In this case the doomsday math concerns work done by astrophysicist J. Richard Gott III.  As noted by author Poundstone (ibid.):

"DrGott invoked math and population data to predict that the end of the Earth is likely to occur in under a thousand years."

Then elaborating:

"The core of Dr. Gott's reasoning is known to statisticians as 'the German tank problem'.  When the Allies were planning the Normandy landings, they wanted to know roughly how many tanks the Germans could be expected to have on D-Day. One clue was the meticulous consecutive numbering of German tanks and their components.

Army statisticians reasoned that captured tanks and parts could be regarded as randomly drawn from all those that Germany produced.  The serial numbers gave certain evidence of the minimum number (which couldn't be less than the highest serial number captured). All the numbers provided a statistical probability - though not a certainty -  of how many tanks existed - and based on that how many more would be produced by D-Day. 

After the war captured German records revealed that the statisticians' estimates had been almost exactly right."

How does this connect to any doomsday forecasts?  The author goes on:

"To apply this to the future of humanity, the doomsday argument begins by asking us to imagine a list of all past, present and future people, sorted in their order of birth. No one alive knows how long the list is or where their own names fall in it, so we can consider ourselves randomly drawn. That means that statistically there is an equal (50 % chance) that we are in the first half or the last half.

Humans don't come with serial numbers but Dr. Gott reasoned that demographers have already roughly estimated the number of humans so far. They put the total number of people who have ever lived - from the beginnings of Homo sapiens to the present day - at 100 billion. That means that my birth order serial number - like for anyone alive now - is somewhere around 100 billion. 

Since it is equally likely that those of us living today are in the first or second half of all past and future human births - let's say we are in the second half - that would mean there are 100 billion births yet to come.  There is a 50 % chance that is true, which at the current global birthrate (about 131 million a year) translates that we have at most 760 more years of births.  A changing birthrate would modify the estimate but the calculation is that simple."

Poundstone adds that:  "the purely circumstantial nature of Dr. Gott's prediction unsettles many who hear of it".

 And it did perturb one WSJ letter writer, a David E. Weisberg. ( 'Doomsday Math Neither Accurate of Helpful' , July 9) who wrote:

"The conclusion assumes that the list (of humans) is finite because it is meaningless to speak of the first half or second half of an infinite list.  And the assumption the list is finite contradicts the premise that no one alive knows how long the list is."

But Weisberg's error is twofold: first there is no contradiction of the premise given that the relevant assumption was not of any absolute enumeration, but for its relative length in time. I.e. after 780 years, 300 years etc., which length is determined by the birth rate.  The birth rate itself- no matter how low (i.e. 1 % per year) -also dictates the list cannot be infinite.

 Regarding the last, we have a brilliant mathematical computation by  Univ. of Colorado Emeritus physics professor Albert Bartlett-   that such a list would have to be finite. You cannot have an infinite number of humans occupying the planet.

Prof. Barltett's exposure of this nonsense - in another context to do with the exponential function and computing future population -  can be found in this lecture :

Go to time index 13:04 and continue watching as he computes the global population projections at an estimated rate of 1.3 percent per year.  Within a minute or so we see the graphic appear:

Which Prof. Bartlett specifies as the full stop to further population increase, i.e.  the absolute arrival of  "zero population growth".   (Though he goes on to show that at the same assumed birth rate the mass of humans would equal the mass of the whole planet in 2400 years.)  As he notes:

"It's absolutely clear people could not live at that density on Earth. So zero population growth is going to happen!"

In other words, there exists an absolute limit to human population growth - and hence the "list" of Prof. Gott. It cannot go on indefinitely!  Prof. Bartlett even includes a humorous cartoon to reinforce the point for any simpletons:

Why is this not possible?   Well because each human needs a defined living space to support him over a lifetime, not merely the minimal space his body may occupy. This is owing to the fact humans need resources to live which are extracted from a given land area.  This life support basis is a function of the rate of consumption of resources in this living space, and the birth rate.  The current average – when all humans are factored in – amounts to nearly 2.1 hectares per person .  This reckons in crops, lumber,  water,  meats-cattle etc.,  consumed over a life .  A square meter therefore cannot be a life support space. Only a certified moron would believe so. (Unless we are talking about immobilized humans put into a deep coma or deep freeze and remaining in stasis).

Thus, no high population density scheme that doesn’t factor in acreage impact  can be taken seriously.  Hence, Weisberg's argument concerning an "infinite list" of  potential humans , i.e.

"Unlike German tanks at a particular moment in time, human beings over the entire course of human history - past, present and future - might be conceivably infinite in number."

Is pure poppycock given the list terminates at the maximum density (more likely even well before that) which is certainly well short of an infinite list!   And as Prof. Bartlett shows, anyone spouting similar nonsense also needs his or her head examined.  Or, they need to go back to school to try to learn exponential math!

Yet more evidence that the "Flynn effect" is pure, unadulterated codswallop!

As for Dr. Gott, his use of Bayesian statistics also saw him (earlier) predicting closing dates for 44 Broadway and off-Broadway shows. When he checked back 4 years later, 36 that had closed had done so within the prediction window. As Poundstone observes:

"Like many of those involved in the doomsday debate, he views the discussion as a wake up call."

I'd say the first "wake up" needs to be for people to process exponential growth which is why the link to Dr. Bartlett's lecture is so critically important. I strongly urge all readers to at least watch the time segments indicated - if not the whole 1 hr 14m lecture.

No comments: