Invasion of the Meme Snatchers

From: Mark Walker (
Date: Fri Feb 15 2002 - 03:48:10 MST

Suppose a nanocivilization sent out nanobots aboard a von Neumann probe (or
otherwise) to our "lovely" planet. Their mission is to destroy any evidence
gathered by humans of the galactic civilization that in fact exists out
there. Is this scenario physically possible? No doubt the nanobots would
have no problem destroying various forms of physical evidence, presumably
the nanobots could munch-up photos, computer files, etc in quick order. For
example, SETI instruments might detect hundreds or thousands of signals
every night but the nanobots destroy the evidence before it is processed.
Suppose this process is not perfect and occasionally humans become cognizant
of some evidence of the existence of ETs. The question I am wondering is
whether there is any principled reason why the nanobots could not also
selectively destroy the associated human memories. It seems likely that a
task of this complexity would require a network of nanobots, perhaps
connected to a CPU. In fact, if the nanobots can perform this sort of
selective memory purge (and perhaps replacement) then it seems quite
possible that ETs could be visiting us here on junkets from other stars.
Imagine that they walk down the main street of your hometown. Humans might
realize for a few seconds or minutes what they have seen (i.e., ETs) but
then the nanobots set to work doing damage control, namely, memory purging
and replacing, film eating, file destroying etc. So I ask (again) is there
any principled reason you can think of that would stop selective memory
destroying by nano in humans?
    Is such a scenario plausible? Well, we can say at minimum that it is
logically compatible with all the evidence we now possess, since by
any countervailing evidence is destroyed. This conjecture is both refutable
and confirmable. The sort of evidence that would refute this conjecture is
if we discovered the existence of extraterrestrial intelligence, if SETI
succeeds for example. We can also imagine the conjecture being confirmed, if
the ETs decided one day to confess to their deception.
    Can we rule out this conjecture with occam's razor? Maybe, maybe not. In
terms of the Fermi paradox this conjecture is a variant on what is called
the 'zoo hypothesis': we are purposely being isolated from the galactic
civilization. So the question of whether this violates occam's razor is not
so straightforward as it might seem. The alternative solutions to the Fermi
paradox are not particularly inviting or necessarily simpler in themselves,
e.g., that we are the first technological civilization out of the gate, that
ETs exist but we just have not seen them or detected them yet. It is not
always clear how to head the advice, pluralitas non est ponenda sine
     Another related place the meme snatcher conjecture is relevant is in
connection with Nick Bostrom's simulation argument. I pointed out in the initial round of
the discussion of this argument that it is simply a special case of the more
general principle that we might be created by others. Thus, one way to make
our world is to do it in a computer simulation, another would be to create a
physical world and populate it with created humans. Both logical
possibilities seem easily implemented by a technologically advanced species.
As I pointed out at the time, obviously the physical alternative does not
look very cost-effective in terms of matter/energy utilization. But who
knows? Perhaps they choose physical implementation because of its retro
appeal, or to heed some ethical imperative, or because creating
matter/energy is no sweat for them. In any event, at the time Robin Hanson
pointed out that this probably would require building some sort of screen
around our planet or solar system. Well, the meme snatcher conjecture offers
an alternate and perhaps more cost-effective means to hide the galactic

Cheers, Mark.

Dr. Mark Walker
Research Associate, Trinity College, University of Toronto
Editor-in-Chief, Journal of Evolution and Technology

This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 13:37:39 MST