Saturday, June 17, 2017

Nonsense in NatGeo's “Year Million”

The National Geographic channel has a new TV series entitled “Year Million.” A recent episode in the series was entitled “Never Say Die,” and was about the prospects of human immortality. The episode was long on slick visuals, and short on intelligence. The episode presented the ideas that we will greatly extend human longevity by using nanobots and genetic engineering, and that later people will upload themselves into computers.

The episode presented a simplistic view of biological life, one that is purely mechanistic, epitomized by a person who said, "We are still biologically hindered by the squishy computer in our heads.” The assumptions seemed to be that biological life is merely some mechanical phenomenon that humans can completely master through technologies such as nanobots and genetic engineering. These assumptions are quite dubious because of all the profound mysteries involving biological life.

It is generally admitted that we do not understand the origin of life. It is generally admitted that we do not understand the mystery of protein-folding (why proteins conveniently form into three-dimensional shapes needed for our existence). It is generally admitted that we do not understand the mysteries of morphogenesis, how it is that a fertilized egg is able to progress to become a baby who then grows into a full-grown human. We also don't really understand the origin of species, as the theory of evolution by natural selection is merely a theory of accumulation rather than what we actually need to explain the mountainous degrees of coordination, structural fine-tuning, and functional coherence in our bodies: a theory of organization explaining the arrival of the fittest, not just the survival of the fittest. As discussed here, we also don't really understand where it is that the body plans of organisms are stored, because DNA is a minimal “bare bones” language unsuitable for describing three-dimensional body plan blueprints; so while DNA might be storing a list of chemical ingredients, it does not actually seem to have a specification of our body plans.

We therefore have every reason to suspect that biological life involves some mysterious forces or factors very far beyond our understanding. If that is the case, then the prospect for vastly increasing the human lifespan through injecting humans with tiny robots called nanobots and tinkering with DNA may be much less bright than many think.

But there is some reasonable hope that genetic engineering or nanobots might extend the human lifespan, so this part of the TV show wasn't nonsensical. The show did become nonsensical when it discussed the very dubious concept of mind uploading in a completely uncritical manner. Among the reasons why mind uploading is such a dubious concept is that we have no understanding of how a brain could even be storing memories that last for decades, given all the rapid molecular turnover and protein turnover that occurs in brains; and we have no idea of how brains can generate abstract ideas. Mind uploading cannot possibly work unless some reductionist model of consciousness and memory is correct; and there are very good reasons (discussed here and in this series of posts) for doubting that any such model is correct.

The show quoted neuroscientist Michael Graziano, who made this very dogmatic proclamation:

A lot has to be done before we can figure out whether we can take a pattern of connectivity from a brain and upload it or copy it. It's going to happen; I'm positive of that. Everything's moving in that direction.

Graziano's certainty on this matter is laughable. There is no basis for believing that minds can ever be uploaded. There is no progress at all in mind uploading. Instead of “everything's moving in that direction,” the situation is actually, “nothing's moving in that direction.” And why is Graziano saying he's sure it's going to happen, just after saying, “A lot has to be done before we can figure out whether we can take a pattern of connectivity from a brain and upload it or copy it”? Those two thoughts contradict each other.

The show then quotes string physicist Michio Kaku, who nowadays seems to have his head popping up on every science or futurism documentary put on television. Kaku states, "Believe it or not, we can actually upload memories, and record memories in mice.” This statement is false. The actual claims that have been made by certain mice researchers (researchers in optogenetics) is that memories in mice can be blocked or activated. Such claims are not well founded, and are based on a few doubtful studies using a dubious methodology. The studies typically report weak levels of statistical significance. See here for a discussion of the flaws in some of these studies.

Making claims about mind uploading, the show's narrator says this:

Once we shed our physical bodies, we will exist only as a computer copy of our brains. It's hard to imagine exactly, but basically you'd be digital information stored on a server.

The TV show assures us that mind uploading is the key to eternal life. The narrator says this:

But when we start talking about immortality, that's a whole other can of worms. If we really are serious about getting to forever-land, we'll have to replace our carbon based human bodies and upload ourselves to a super-computer.

The show then tells us this existence will be like some afterlife heaven. We are told:

We're talking about everlasting life in a digital paradise....All of society will live digitally in a virtual world called the metaverse that's a thousand times more intense than the one you live in now.

There are quite a few reasons why such a thing is very unlikely to happen. First, the brain does not store digital information, so the mind is very likely not something that can be uploaded into a computer. Second, there is not the slightest evidence that the brain uses any type of readable code to store information. DNA uses the genetic code, something we can understand and read. But we have zero knowledge of anything like a brain code that we can use to read information from the brain. Nor can we plausibly imagine how such a code could have originated, as it would have to be something almost infinitely harder to explain than the genetic code (the origin of which is very hard to explain).

Third, you could never electronically capture the exact state of a particular person's brain, even if you tried to use microscopic nanobots to do such a thing. Imagine trying to exactly map every synapse and neuron in a brain with nanobots. Each of the brain-mapping nanobots would have to somehow be aware of its own exact position in the brain, so it can record the exact position of each neuron and brain connection it encounters. So if a nanobot comes to a neuron that is 1.334526 centimeters from the left edge of the skull, and 2.734538 centimeters from the right edge of the skull, and 5.292343 centimeters from the back of the skull, then those exact coordinates must be recorded. But how can a microscopic nanobot do that? You can't supply a microscopic nanobot with a little GPS system allowing it to tell its position. So it would seem that nanobots are completely unsuitable for any such job as mapping the exact physical microscopic structure of an organ with billions of cells and synapses packed together in a small space.

Then there's the duplication problem. Imagine if there was a machine that could scan your brain, and then upload your mind to a computer. Even if that process was done perfectly, the computer would not have your mind. It would instead have a copy of your mind. You may realize this just by considering that if this uploading process didn't kill you, and the computer with your “mind upload” existed at the same time as you, there wouldn't be two you's. There would be one you, and a copy of you living in the computer. If you then died after this upload, it would not at all be true that you had survived death by the fact that a copy of your mind was in the computer.

None of these difficulties are discussed on the “Year Million” TV show. The show presents mind uploading as a sure thing, ignoring all the reasons for thinking that it can't ever happen. Not only did the TV show present mind uploading as something that will likely happen; the TV show assured us that mind uploading is the key to true immortality – on the grounds that once you are living in a computer server, you are guaranteed to live forever.

This is nonsensical, because the programs running on computer servers sometimes crash and stop working; the programs running on computer servers are sometimes deliberately halted; and the computer servers themselves sometimes crash, lose power, or are turned off. If you were living in a computer server, there is no reason why you should be confident that you will be around for more than a century or two. 

mind upload
 After the father and daughter mind-uploaded

The next episode of the “Year Million” TV series continued to be largely about mind uploading, speaking as if this extremely doubtful idea was on solid ground. There was also a pitch for the idea that we are already living inside computers. The “Year Million” TV series is on track to be the silliest documentary series ever put on about the human future.

Given our very low state of knowledge about the most basic riddles of biology, memory and consciousness, fantasies of mind uploading are rather like the fantasies of some boy who knows almost nothing about the contents of the sun, but who fantasizes that he will one day reorganize the sun into a shape and color more pleasing to him.

No comments:

Post a Comment