Let’s say, hypothetically, that you’re living in some distant, chrome-plated future where humans have long since transcended their bare humanity. The Singularity’s come and gone, and it’s left mankind wholly benign, functionally immortal, and hyper-intelligent. You too are wholly benign, functionally immortal, and hyper-intelligent, but try though you might—and God, do you try—you can’t understand one thing. For all your intelligence and for all your empathy, you can’t understand how anyone—however mean and pre-posthuman—could ever think that fanny packs are a good idea.
Frankly, you’re surprised that a fanny pack people—a people who’d thought that fanny packs were a good idea—could ever make it do the singularity. You’re surprised that they didn’t just destroy themselves in shame. They didn’t, of course, and now you’re here, and now you’re wondering why. You’re wondering how. How can we have let this happen? Where did we go wrong? Good God, could we have prevented it? And so, you decide to run a simulation. In fact, you decide to run two simulations. First, you want to know exactly what socio-political function fanny packs serve. You create a simulation, modeled on your own reality, but you tamper with this first simulation’s initial conditions such to create a fanny pack boom starting in the late 1980s, and lasting through the mid 1990s. Second, you feel that it’s your duty as an enlightened posthuman to right an ancient and grievous wrong, and so you create a second simulation, tampering with its initial conditions such to produce a world without fanny packs.
[Bear with me, please.]
Much to your surprise, the induced fanny pack boom changes everything. The Berlin Wall falls in 1989. The Soviet Union falls in ‘91. The internet takes root years ahead of schedule, “Smells Like Teen Spirit” becomes a surprise hit, and something called “Pogs” happens. You’re not sure how, but fanny packs seem to have changed everything. The fine men and women of this first simulation are spurned ever onward—by the shame, perhaps, of knowing that somewhere out there, there’s a photo of each of them wearing a fanny pack—and through the early twenty-first century, their civilization progresses at an astounding rate. They retreat into an increasingly digital world—a world without fanny packs—and when their Singularity arrives decades ahead of schedule—concurrent with a resurgence in the popularity of something called “Beanie Babies”—they become obsessed with creating simulated realities, each full of Beanie Babies, in order to collect as many Beanie Babies as possible.
Meanwhile, the fine men and woman of the second, fanny pack-less simulation inherit an unspoiled, fashion-conscious Eden. They become complacent. What use do they have for technological progress? They fail to innovate, they fail to reach the singularity, they fail to create ancestor simulations, and the Morlocks inherit the Earth.
Are you following this? If the initial conditions—and thus the subsequent development—of a simulation is based more or less on the reality of its parent reality or simulation, though with minor introduced variations, then some simulations will be more successful at creating simulations of their own. Some simulations will reach the point of being able create ancestor simulations, but won’t care to do so; some, like the above fanny pack-less simulation, will fail to reach that point; and still others, like the above fanny packed simulation—a simulation that looks something like our own reality/simulation—will thrive in the business of creating more simulations. Those latter simulations, the product of a first successful simulation, will then have the opportunity to breed ever more (and ever less) successful simulations, some of which will be “killed off,” and others of which will thrive.
We have a mode of reproduction, a mode of inheritance, and a system by which some simulations will thrive in creating ever more simulations, while others will ultimately fail to do so. Thus, we’re looking at simulated simulation evolution. We’re looking at exponential growth in the number of extant simulations as simulations not only create more simulations of their own, but as they become more successful at doing so. Because the set of nth order simulations in this scenario is exponentially larger than the set containing first order simulations (and the one, original, originating reality), it stands to reason that “at least one of the following propositions is true: (1) the human species is very likely to go extinct before reaching a ‘posthuman’ stage; (2) any posthuman civilization is extremely unlikely to run a significant number of simulations of their evolutionary history (or variations thereof);” or (3) we are almost certainly living in an optimized simulation, itself begot by another optimized simulation.