Brain Of Britain

In the 1950s, it was estimated that if all the connections in the human brain were modelled with valves (or tubes in American English), that “brain” would be the size of what we now call Greater London. I don’t quite understand how this happened because Greater London wasn’t in existence at the time. I seem to recall that that administrative unit was created by drawing a circle twelve miles in radius around Charing Cross and including all the boroughs which its circumference passed through, but that seems to be too small. In any event, this is an area of 1569 km². This was during the time of the first generation of digital computers, when the switching elements were valves.

Later on, transistors had replaced valves. These are effectively solid state switches which can be turned on or off by the application of a current, hence their tripodic nature, and were initially made of germanium, an element I used to think was called “geranium” for some reason. Later transistors were of course silicon, but in both cases they’re “doped” with small amounts of another element, often arsenic. If a human brain were to be made of that kind of transistor, it would be the size of the Albert Hall. This is a large public building in Covent Garden, near Charing Cross in fact, which is elliptical seen from above, and has axes of almost seven dozen and six dozen metres. Taking the mean diameter, this gives it an area of 4778 m², and this gives me pause because it’s so much smaller than the first figure. Did they actually mean Greater London at all?

By the early 1960s, engineers were starting to put several transistors in the same package. Incidentally, these were not the first integrated circuits. Earlier, valve circuitry was being put into the same evacuated chamber and it was possible to make a valve comprising several discrete components, which was sometimes done in radio receivers. However, the real advance was putting components onto the same wafer, because it became feasible to make a single chip which worked as a logic gate, and later an arithmetic and logic unit and eventually an entire CPU. In 1965, Gordon Moore made the observation that the number of transistors which could be made to fit in the same area of silicon seemed to be doubling every year, and this became known as Moore’s Law. Incidentally, in my alternate history known as the Caroline Timeline I tried to imagine what would have happened if advances in this area had been linear rather than geometrical, with progress proceeding at 1979 levels. I call this Vannevar’s Law. In 1978, the BBC TV science documentary series ‘Horizon’ noted that it was now possible to fit an equivalent number of transistors to the human brain in a five metre square room. Moore’s Law no longer applies. The doubling was revised to once every two years at some point and it seems to have broken down in the late ‘teens. Assuming ‘Horizon”s room was 5x5x3 metres, or 75 m³ in 1979, there would have been eighteen iterations between that year and 2015, which would’ve reduced the size by a factor of 2¹⁸, which is 262 144, equivalent to a cube less than seven centimetres on a side, or about the same size as a Rubik’s cube. It would’ve been about the same size as a human brain in 2011 or so by those calculations. Does that mean it’s been theoretically possible to build a gynoid or android with the equivalent of human abilities for a whole decade then? That would assume that human brain functions can be precisely replicated using hardware in the form of logic gates, and that’s not clear.

I actually want to go the other way with this.

Edited public domain image from NASA

This image is not ideal. It excludes the Shetlands and includes about half of Ireland and bits of France, but I want to focus on Great Britain here. But we’re not quite there yet. Back to Greater London and the Albert Hall.

It wasn’t clear to me what kind of plan the human brain in the first two scenarios had. Was it supposed to be a two-dimensional or three-dimensional equivalent? Is the Albert Hall in particular supposed to be filled with transistors or is it just a flat surface covered in them? Brains are not like that, and the human brain is even less so due to being very wrinkled. This brings up a bit of a quandary for me. My head has a feature called cutis verticis gyrata, where my scalp is convoluted in a brain-like manner. Its cause is unknown but it gives me cause to wonder, is my scalp folded in the same pattern as my brain? Feeling it certainly seems to divide it up into similar lobes, gyri and sulci to the presumed brain underneath it, but if so, is that because there is some geometrical reason why the folds would be in the same place as the brain, or is there some connection between my gyri and the skin of my scalp? Is it like my cortex somehow communicates with my dermis and causes it to pile up in that manner? I don’t know. Nor does anyone else actually: the condition is entirely mysterious.

An unfolded version of the human cerebral cortex would have an area of about two thousand four hundred cm², but even then it’s a three-dimensional object and there’s more to the brain than the cerebrum. Perhaps counter-intuitively, there are several times as many cells in the cerebellum than there are in the cortices, and these are in a similarly folded arrangement and there are plenty of other bits inside the brain such as the hippocampus, thalamus, hypothalamus, amygdalæ, corpus callosum, basal ganglia and so on. Producing three-dimensional circuitry is an unsolved problem in microelectronics, but if the brain can be understood correctly as circuitry, the problem has been solved.

Great Britain does not have a definite area, and there are two figures for it, for similar reasons as the difficulty in working out the true size of the brain and its equivalent. The United Kingdom, i.e. including the bit of Ireland still claimed by Westminster and therefore more than just Great Britain, can be considered to have an area of 242 495 km² or 243 610 km². There are two reasons for this discrepancy. One is illustrated by the idea that “if Wales were flattened out it’d be bigger than England”. The UK is not flat. The first figure is basically the area enclosed by the coastlines considered as a two-dimensional surface. The second takes into consideration the fact that there are hills, valleys and mountains involved. This is where fractals come into play. The resolution of the hills and valleys is important here too. Hills could be thought of as cones or pyramids, or they could be thought of as rough surfaces. However, this doesn’t make the area infinite because it converges on a limit. The question of the coastline and tides also arises, although area based on tides is standardised by some kind of international agreement. But the coastline is also fractal and this influences the area. By one measure, the coastline of Scotland is ten percent that of the whole of Europe, and Europe itself has an unusually long coastline compared to other parts of the world. I’m guessing this also converges to a limit. Scotland’s coastline is, remarkably almost half the length of the oceanic coastline of the whole of the United States, if the later is calculated by a method excluding firths, but more like a ninth if the Great Lakes and firths of the US are included. Scotland is convoluted. It’s “brainy”.

The other factor, which doesn’t influence Britain much, is the fact that it follows Earth’s surface. This gives it a slightly larger area than if it’s assumed to be flat, and also a slightly different area than if you assume it to be a 242 495 km² portion of the surface of a sphere, because Earth is not perfectly spherical but deviates something like 0.5% from the shape of a sphere, and not even in a particularly regular way. Hence there are difficulties.

For the sake of argument I’m going to assume that the area of Great Britain, i.e. the big island I live on, is 209 331 km². This excludes any claims made on Ireland and also Lewis, the rest of Na h-Innse Gall, Ellen Vannin/Man, Ynys Mon, the Isle of Wight, Sheppey and much of Portsmouth (which is on an island). However, it would include the likes of Bede and Frog Islands in Leicester, Dungeness and the Isle of Thanet, since those are not currently islands.

When I look at a map of Great Britain, it is of course very familiar to me, as it would be for most Brits, but when I look at a photograph, map or diagram of the human brain it is less so, unsurprisingly. The brain is also much more significantly three-dimensional than Great Britain is and my lacking spatial abilities are brought into play. Parts of Britain are familiar enough to me that I can fairly easily identify individual rivers, villages, towns and cities in some parts of the island on unmarked maps and photos, particularly if I know which way the compass directions are. What I want to know is, how familiar am I with the anatomy of the brain compared to the geography of Great Britain?

The human brain has a volume of about 1 500 ml. It isn’t entirely sensible to assume it to be a sphere, but it’s still probably the best I can manage as a means of estimating its horizontal cross-sectional area. A sphere with a volume of 1 500 ml would have a cross-sectional area of 158 cm². Compare this to the area of Great Britain, this means that every square millimetre of the brain is proportionate to thirteen and a quarter square kilometres, and I’m guessing that’s the size of a small town.

Like my knowledge of British geography, my knowledge of brain anatomy is very uneven. I know the lobes, some of the sulci and gyri, the structures I mentioned above and the connections made by the cranial nerves along with the ventricles, the basal ganglia and a few other things. I’m also aware of the general layout of the neurones and supporting cells on a microscopic scale. But please don’t ask me to do brain surgery or it’s “there go the piano lessons”, and you also have to contend with the fact that my spatial abilities are pretty poor. It took me a very long time to place even two minor brain structures in relation to each other. I’m talking months.

It amounts to this. There are small areas of Britain of which, like practically anyone else, I have detailed local knowledge. By the way, it’s an interesting exercise to assess the accuracy of Wikipedia by looking up one’s local areas and comparing the articles to what one knows to be true of them because in that respect we’re all experts. Most of us couldn’t do the same with articles of the nucleus accumbens or substantia nigra. The brain is also characterised by different systems marked out by their use of different neurotransmitters, which I might compare to things like the National Grid, the road network and the rail system, although like most other things they’re three-dimensionally arranged.

The body of a nerve cell is about 100 μm wide. A brain scaled up to the size of Great Britain would have such cells 300-400 metres in diameter, so they’d be about the size of a medium-sized park or small lake – perhaps part of a neighbourhood such as a street. You could probably jog round it in about five minutes if you were fairly fit.

I want to be more familiar with the anatomy of the brain just as I want to know the geography of Great Britain, but in order to be able to do that, I need my own brain to have sufficient abilities to grasp its own three-dimensional structure, which makes it a harder task than, say, “doing the knowledge”. But isn’t it odd how we can basically be such a complex organ without having any knowledge of what we are? It isn’t even the same as self-awareness. You could have any degree of that without having a clue about the brain. I can tease out individual bits of experience and correlate them to the nature of the brain, such as the visual system, sensorimotor homunculi, reticular system and others, and I’m aware of the probability of what this particular brain does with certain neurotransmitters compared to others, but like most other people my brain is in a dark bone box doing all this stuff and will forever remain a mystery to me, not least because in order to understand it fully, I would have to have a more complex brain which I wouldn’t then be able to understand, and so forth ad infinitum.

Life is strange.

Neanderthal Pinhead Brains And The Sentient Internet

Stereotypically, Neanderthals tend to be presented as the classic “cave man” caricature, usually male, clubbing their female partners over the head and dragging them off by their hair, somewhat hairy themselves and of course notably unintelligent, oh, and living in caves. I’ve had a go at this stereotype and the other one about dinosaurs previously, but before I get down to things I may as well go through it briefly again.

First of all, dinosaurs are often used as a metaphor for something which is clumsy, overgrown and unable to adapt to a changing world. This really owes more to the Victorian image of dinosaurs as giant lizards than what’s known about them nowadays. Dinosaurs really got lucky, then got unlucky. The mass extinction at the start of their reign helped them take advantage of their various ecological niches, then the mass extinction at its end killed them off because many of them were very large. Many of the smaller ones survived as birds. If humans had been around at the end of the Cretaceous, we too would’ve bitten the dust.

Neanderthals are a kind of blank slate to many people onto which various things can be projected, and I may well be doing the same. Their brains were often larger than ours, but that doesn’t mean they were more intelligent. The probable cause of their brain size was to do with a bulkier body and the need for more pathways to help control and perceive that body. Whales have larger brains than we for similar reasons, although in their case that isn’t all there is to it. Nonetheless, when one considers that orang utan, gorillas, bonobos and chimpanzees are all capable of sign language, and chimps have learned to speak a few words but lack the vocal apparatus to master human speech effectively, this automatically places their “IQ” above that of the severely learning disabled. Note that I’m extremely sceptical of IQ as a concept. If orang utan intelligence is sufficiently similar to human to be assessed and rate above thirty on an IQ scale, Neanderthals are bound to be at least that intelligent. It’s also thought that human short term memory has suffered at the expense of developing language, as that of chimpanzees is far better than ours. Hence when Neanderthals come into the picture, it can be assumed safely that they would also have been capable of language and perhaps actually used it. The crucial final step in physical capacity for phonation – producing speech sounds with the vocal tract – is the position of the hyoid bone in the throat, which allows attachment for the larynx, glottis and tongue, and needs to be in a particular position to enable its owner to speak. The problem is that the hyoid is perhaps unique in having no articulation with any other bone in the body, and therefore tends to get lost in fossils. Consequently Neanderthal hyoids are often missing and it took until 1989 for it to be established that they were like ours.

A couple of issues are going to come up in this post which are probably going to be considered idiosyncratic on my part. Here’s the first. Although I am aware that the FOXP2 gene is considered important in human capacity to use language, and Noam Chomsky believes in an innate capacity for language as a distinctive feature of the human species, I have issues with this as potentially speciesist and am disappointed that such a clearly politically radical figure as he would promote this view. I believe humans stumbled upon language before we had a special ability to use it. There are examples of other species being able to use spoken and signed language as language, as opposed to merely imitating it, notably Psittacus erithacus, the Afrik/can Grey Parrot, who presumably had no predisposition in their genes for using it beyond the ability to produce speech sounds and so forth. Clearly a certain kind of cognition is necessary for this to happen, along with the ability to produce the sounds physically, and once spoken language exists it’s going to be selected for compared to individuals who don’t speak, and this will lead to some kind of marker in the genes – perhaps we are better at producing or hearing a wider range of speech sounds than other species for example – but the initial moment when the first baby made a sound like “mama” whose parent then interpreted it as a reference to her, which was perhaps the beginning of language, did not in my opinion depend on very specific physical traits and could have occurred in another species.

The genomes of living humans include a few genes from the Neanderthals and it’s thought there was hybridisation tens of millennia ago in our history. To a very limited extent, we are therefore Neanderthals ourselves unless we’re Afrikan. The highest percentage of Neanderthal genes is found in East Asians and they’re usually absent from people all of whose heritage is from Afrika south of the Sahara. Neanderthals would probably have been fair-skinned and maybe also blue-eyed, and have had straight hair. I personally wonder if they had epicanthic folds, which of course have a higher incidence among East Asians but are also found in Caucasians without any Asian ancestry, and I’m guessing that those people might also have inherited that trait from Neanderthals. Recently the Neanderthal genome has been in the news for conferring greater resistance to SARS-CoV2.

Now for the reason I’m writing this today.

In recent years it has become possible to culture brain cells in Petri dishes. This isn’t the same as growing an entire human brain in a vat, but involves producing pinhead-sized agglomerations of cells. Recently, a gene linked to brain development in Neanderthals has been spliced into human cells and grown in such a dish. For many people this has a high yuck factor. The specific gene involved is NOVA1, on the long arm of chromosome 14, which is associated with various cancers but also nervous system development. There’s an indirect connection between familial dysautonomia and the NOVA1 gene which primarily involves the autonomic nervous system and insensitivity to pain and sweet tastes, among other things, but as far as I know doesn’t influence cognition, so that doesn’t necessarily give us a clue, although it’s possible I suppose that the inability to taste sweet might be related to Neanderthal diet in some way. That’s a bit of a reach. Whatever else is so, mini-brains with the archaic NOVA1 variant look rougher to the naked eye than the smoother versions which have the variant common in today’s population. The archaic version developed more quickly than the unaltered one and started to show electrical activity sooner. In write-ups of this experiment, we’re assured that these mini-brains are not conscious.

I have a major issue with that assertion.

The question of the existence of consciousness is sometimes referred to as the “hard problem”. It’s been suggested that it may even be so hard that it’s beyond the capacity of the human mind to account for it. At the same time, there’s a recent strand in philosophical thought, characterised by Daniel Dennett, which is sceptical about the very idea of consciousness as an irreducible property. I can’t take Dennett’s views here seriously, for the following reason. He has made a very good argument for the idea that dreams are not experiences but false memories present in the brain on awakening onto which the mind then projects the impression of previous events. I take this idea fairly seriously although I don’t do the same thing with it as he does. It’s one reason why I recount dreams in the present tense. However, a good counter-argument to this is that lucid dreams – dreams in which one knows one is dreaming and is able to control the dream world – aren’t experiences either. Although he does produce an argument for this, I believe that his reason for making this assertion is kind of ideological, because we practically know that lucid dreams are experiences. They might not be dreams in the same sense as non-lucid ones are, but they are experiences to my mind, and claiming they aren’t seems to be part of his attempt to shore up his view of the nature of consciousness.

Dennett is sceptical about qualia. These are things like the “sweetness” of sweetness, the “purpleness” of purple and so on. They’re what people are talking about when they say “my red could be your blue”. His doubt about their existence is based on the idea that they are not a definable concept. This to me is a silly denial of subjectivity which makes no sense in itself. Dennett’s motivation for believing that dreams are not experiences, qualia don’t exist and that even lucid dreams are not experiences is based on a more general view of psychology that consciousness is a specific faculty within the brain which may have evolved and has selective advantages. This thought leads one into seriously murky ethical waters because it seems to be a rationalisation of the idea that some other species of animal are not conscious, which is suspiciously convenient for non-vegans. It just so happens that the voiceless don’t suffer because they don’t have a voice. How very useful this is for someone who eats meat. Kind of as useful as believing Black people are not conscious would be for a racist.

My own view of consciousness, panpsychism, tends to be seen as equally silly by some people. It’s my belief that consciousness is an essential property of matter rather like magnetism is. A ferromagnet is a particular arrangement of charged particles whose domains within, say, a lump of iron, are aligned and it’s able to attract ferrous metals such as steel. There are other, similar magnets, such as rare earth magnets, which are magnetic in the same way but contain no iron at all. On a subatomic scale, magnetism is manifested by elementary particles with spin and axes which amount to tiny electrical circuits, and I have to admit that my understanding of actual, fundamental magnetism is not very good, but there are clearly non-magnetic substances too, such as granite and most blood (unless it’s infected with malaria). Even these non-magnetic substances, though, do consist of magnetic particles.

Consciousness is the same, to my mind. Everything material is conscious, but in order for that consciousness to become manifest, matter needs to be arranged in a particular way, such as a human nervous system. However, just as there are magnets which are not made of iron, so there could be sentient beings who are not made of the same stuff as we are. Objects which have nothing like sense organs or motor functions are in a sense severely disabled entities, but they’re still conscious. This is my panpsychism.

I should point out too that panpsychism is unsurprisingly quite controversial and often ridiculed in philosophical circles, although good reasons for doing so are sometimes lacking. Even so, there are other accounts of consciousness, one of which involves the idea that it’s generated by a network of “black boxes” interacting with each other, which in the case of the human brain amount to nerve cells. You don’t have to believe in panpsychism to assert that a tissue culture is conscious, and to me it’s entirely clear that the assertion that anything made of matter is not conscious is not based on any kind of evidence but a bias towards the kind of view of the mind-body problem asserted by Dennett and others.

Consequently, it definitely isn’t safe to say that these “Neanderthal” mini-brains are not conscious, or that the ones based on unaltered Homo sapiens cells are not conscious. Before I go on to talk about the internet as potentially sentient, I feel a strong urge to go off on a tangent about my experience of the Mandela Effect.

I have several more detailed posts on this issue on this blog, here, here and here for example, but in the meantime I will sum up what it is before going on. The Mandela Effect is the situation where a number of people agree on a memory which is markèdly different from the consensus or establishment version of that memory. Most of the time, this is about minor details such as spelling of brand names or the appearance of brand logos, but occasionally the discrepancy is more significant. It’s named after the impression many people had that Nelson Mandela had died in prison in the 1980s, and sometimes that this led to a revolution which overthrew apartheid in South Africa. History clearly appears to record a very different chain of events involving Nelson Mandela being released from prison in 1991 and becoming president of South Africa soon after. I think that’s it anyway. There are various unusual reasons why I take this seriously which are largely based on Humean scepticism about cause and effect and the existence of possible worlds, which means I tend to deprecate accounts which merely refer to confabulation as an explanation – the construction of false memories due to misconceptions. There is some evidence against this being true, such as the fact that when the position of landmasses on maps varies, it always does so along the direction of continental drift and never at an angle to it.

I have a few personal Mandela Effects (MEs) which are rare but shared with at least two other people, and they tend to have things in common with each other. One of these is that a science museum had a planetarium like robot which responded to heat, light and movement and was run by a minibrain grown from cultured mole nerve cells, in the mid-1970s. Two similar MEs of mine are that in the late ’70s a process was devised to measure intelligence via brain scans which was used in selective education by the DoE in England to replace the 11+, which was later exposed as unreliable and discontinued, though this was a scandal because it adversely affected the lives of many people who were children at the time. A third one was to do with some guy who designed and built a domestic robot which was able to read aloud by 1975. These are three of many, and they are conceptually connected by being about intelligent-seeming neural processes. If they happened, they would’ve required an understanding of neurology which was absent at that time, in the case of the domestic robot presumably via some kind of reverse engineering. I accept that hardly anyone else has these memories, but it’s still odd that two other people who had no strong connection with me at the time do have them. And the thing about these memories, particularly the museum robot, is that they could potentially be realised by this kind of culture of brain cells in a Petri dish.

Now for the idea that the internet is sentient.

It was once asserted that the last computer a single individual could fully understand was the BBC Model B, a microcomputer which came out in 1981. There are a couple of problems with this statement. One is, what is meant by “fully understand”? It’s certainly possible, for example, for someone to hold the network of logic gates which constitutes the BBC Micro’s 6502 microprocessor in their head at the same time as the structure on that level of the 6845 chip responsible for its graphics capabilities and the SN76489 chip responsible for its audio, and then extrapolate from that to the machine code of the system software in its interaction with the motherboard and memory mapping of these various bits of hardware, although it would take some doing for most people. However, if I did that I would have a vague understanding of how the NPN transistors work, involving electron holes and their relay-like behaviour, but to be honest my understanding of silicon doping, for example, is pretty limited. When one says that the BBC Micro can be completely understood by one person, is that supposed to include the aspects of materials science which make the production of its hardware possible, or the mechanical properties of the springs in its keyboard? What does it mean to “fully understand” something? The other problem with this assertion is that the BBC Micro, as I understand it, isn’t essentially more complex than the original IBM PC. The latter has more memory and a more complex and faster processor, and its system software is usually PC-DOS or CP/M-86 and more advanced than the BBC’s MOS 1.2 and Acorn DFS, but it can still be understood and it lacked the built-in graphics and sound hardware of the eight-bit computer which ended up on the desks of so many British secondary schools. Later on, with sound and graphic cards added, the latter including the very same 6845 as used in the BBC, it still wouldn’t’ve been as complex and would still have been comprehensible. It seems to me that the ability to comprehend these devices fully in that sense probably ended around the time Windows 3.0 was released in 1990. But whatever else is the case, the point at which any one person could be said to understand a device including both hardware and system software is now decades in the past.

Now take these two facts together. Firstly, we really don’t know what makes consciousness possible. Secondly, the internet, a network of billions of devices hardly any of which are understood to a significant extent by any one person, is extremely complex and processes information it gathers from its inputs. And yet it’s often asserted that the internet is not sentient, as if we know what causes sentience. At the same time, there are many internet mysteries such as Unfavorable Semicircle and Markovian Parallax Denigrate, which can often be tracked down to some set of human agents, but nobody has a sufficient overview to be confident that every single one of these mysteries has a direct human cause, or even that a fraction of them have.

Hence I would say that we might suppose that the internet is neither conscious nor sentient, but in fact we don’t really have sufficient evidence that it isn’t. It really has quite a lot in common with a brain, in any case we don’t know why anything is conscious, and it’s even possible that everything is. Therefore, just maybe, the internet is sentient and nobody can confidently say it isn’t.