Susan Blackmore's Theory of the Self

Susan Blackmore is a scientist who is most famous for studying memes, with the book The Meme Machine (2000). It's a bit outdated at this point though. She's also a skeptic in debunking some pseudoscience.

I'll start with two videos:

Memes and Tremes

Susan Blackmore studies memes: ideas that replicate themselves from brain to brain like a virus. She makes a bold new argument: Humanity has spawned a new kind of meme, the teme, which spreads itself via technology -- and invents ways to keep itself alive

Living Without Free Will

When I say that consciousness is an illusion, I do not mean that consciousness does not exist. I mean that consciousness is not what it appears to be. If it seems to be a continuous stream of rich and detailed experiences, happening one after the other to a conscious person, this is the illusion.

Memes and replicators in general

The basic idea of memes is that they are replicators of a particular kind. Genes are replicators in DNA. Memes are replicators in human brains. Recently she proposed a third replicator called tremes, or "temes" or "technological memes", defined as replicators in technological things that humans construct.

Tremes are presumably things like computer codes and designs of machines, though I feel this idea is too vague at the moment to be useful.

Clenes were written about previously. Basically, they are replicators built with clay crystal sheets/tubes, and could have been the first replicators on earth before organic genes took over.

Illusion theory of consciousness

I also wrote about this idea previously. It's the idea that 

Illusionism is the view that phenomenal consciousness [feeling, qualia] is an introspective illusion — that introspection misrepresents experiences as having phenomenal properties. This view has many theoretical attractions, but it is often dismissed out of hand for failing to take consciousness seriously.

Susan Blackmore subscribes to this idea fully, and her explanation for this illusion is similar to Dennett's theory that the "self" is a "benign user illusion", like the GUI of the computer, made so that the computer is easier to use.

Susan however thinks it's not benign, but malign.

Is the illusory self, as Dennett (1991) would have it, a ‘benign user illusion’? I think not. It is arguably a ‘malign user illusion’ and the source of much suffering and misery (Blackmore 2010/11). This is the self who craves love, friendship, status, possessions and power. This is the self who gets disappointed, hurt, lonely, angry and resentful. This is the self who wants happiness but when happy fears losing it. This is the self who makes constant comparisons with others and fears other people's judgements. It is strange that an illusion can entail so much suffering.

and she wants to throw off this illusion completely:

Is it possible to throw off these illusions completely? ... Many long-term meditators describe experiences without an experiencer, or nondual states in which experience and experiencer are one. In these states we can ask ‘Am I conscious now?’ and truthfully answer ‘no’ because there is no constructed self to be the experiencer. The experienced duality of self and other then disappears along with the hard problem and yet it is hard to make rational sense of this directly experienced nonduality.

Her justification for throwing off the illusion is that it leads to less suffering. Interesting point, though I'm wary of any argument from morality.

However, there's still the question: if consciousness is so malign, why is it even here? Illusionism usually posits that the illusion is evolutionarily useful. GUI is on every computer because it's selected for, it's adaptive. Susan Blackmore suggests that the illusion of consciousness is not evolved by genetic evolution, but memetic.

Meme theory of consciousness

Consciousness in Meme Machines (2003), Susan Blackmore.

Human-like consciousness is an illusion; that is, it exists but is not what it appears to be. The illusion that we are a conscious self having a stream of experiences is constructed when memes compete for replication by human hosts. Some memes survive by being promoted as personal beliefs, desires, opinions and possessions, leading to the formation of a memeplex (or selfplex).

Okay what.

No seriously, time to study the meme theory of humanity.

Human Beings as Meme Machines

Memes are ideas, habits, skills, stories or any kind of behaviour or information that is copied from person to person by imitation. They range from single words and simple actions to the vast memeplexes (co-adapted meme complexes) of science, art, religion, politics and finance.

Memes are not a human-special thing. Songbirds are good enough at imitating songs that their songs are memes. But only humans are good enough at imitating general things that memes really took off on their brains. Meme-copying on human brains is just good enough to avoid an error catastrophe.

The human brain is excessively large by ape standards, and has been extensively redesigned for language.

Susan thinks this is caused by memes trying their hardest at improving their own survival. This seems dubious to me, since in the recent 10000 years, human brains have actually gone smaller. However, I would be convinced if, despite the shrinking in brain size, humans have become even better at languages and imitating... wait... I think she has got it. Sheeples. Sheeples are the best at meme-propagations. Sheeples don't need understanding, or be smart. They just need to be imitators.

Conspiracy aside... Let's continue.

How does memes exert pressure on genetic evolution? "Memetic drive".

People who are especially good at imitation gain a survival advantage by being able to copy the currently most useful memes... this gives an advantage to genes for bigger brains and better imitation. Increasing imitation then provides scope for more competing memes to appear, and hence there is pressure to be a selective imitator.

I find that this analogy helps me think this through: humans are robots, and memes are softwares. Robots with bigger memory and faster CPU are selected for, because they can run more software that are useful for their survival.

Memetic drive creates not only bigger brains but brains that are better adapted to copying the memes that were successful during the previous memetic competition — whether or not those memes were directly beneficial to people or their genes.

The part about "copying the memes that were successful" is kind of like how modern CPU still supports 8086 legacy code.

Susan also proposes another mechanism, that of sexual selection, for memetic drive. Sexual selection has produced ridiculous things like the peacock's tail as an honest signal for fitness. Once meme-capacity becomes useful enough to be a part of fitness, sexual selection would select for honest signals of meme-capacity. This would explain why humor, wit, and fashionability are considered sexy. They are all honest signals for one's meme-capacity, even if they are rather impractical.

Eventually, memes exert pressure not just on human genes, but on human tools. Writing, tablet, pen, paper, printing press, telephone, television, the Internet. They all allow expand the playground for memes. It kind of stands to reason that one major use of the Internet is to help web engineers and programmers to expand its capacity even further. Seriously, it seems that one can educate oneself fully into being an Internet engineer just by free online resources... And also consider the open source movement is most strongly held online... Memes, memes are behind all of them!

Conspiracy aside... Let's continue.

Mind design by memes

So that's the meme theory of human biology and civilization. Now consider an individual human as a meme ecology.

Human development is a process of being loaded with, or infected by, large numbers of memes... Thousands of memes, mostly borne by language, but also by wordless “images” and other data structures, take up residence in an individual brain, shaping its tendencies and thereby turning it into a mind.

The mind in particular would be inhabited by a memeplex called "self":

By the age of about three years the word ‘I’ is used frequently and with increasing sophistication. The word ‘I’ is initially essential to distinguish one physical person from another, but very rapidly becomes used to say things like ‘I think’, ‘I like’, ‘I want’, ‘I believe’, ‘That’s mine’ and so forth, as though there were a central self who has opinions, desires and possessions. In this way, I suggest, a false notion of self is constructed.

Why is this memeplex so persistent, such that every human has it?

... many other memes can obtain a replication advantage by tagging onto this growing memeplex. For example, saying a sentence such as ‘I believe x’ is more likely to get ‘x’ replicated than simply saying ‘x’. Memes that can become my desires, my beliefs, my preferences, my ideas and so on are more likely to be talked about by this physical body, and therefore stand a better chance of replication. The result is the construction of an increasingly elaborate memetic self. In other words, the self is a vast memeplex; the selfplex.

As this theory currently stands, it's very vague on the details of how this works in detail, so I'll try to fill in the blanks here:

The principle of selfplex

The main method for the selfplex to control human behavior is by taking control of its fear emotion. All mammals, as well as many non-mammals, have emotions. Humans have a limbic system, big even among mammals, that gives them complex emotions that are coupled with many physical actions, such as the stress response, arousal, panting, etc.

The fear emotion is the most ancient emotion. It makes creatures avoid things that hurt or kill them. It keeps creatures alive. Creatures die when their bodies are no longer in one piece, basically.

What the selfplex does is to hack into the fear emotion. By hacking the fear emotion, the selfplex activates fear whenever a piece of meme tagged "self" is lost. The effect is that the human avoids losing the memes that it carries, as if losing a piece of meme hurts like losing a piece of flesh, and losing too many memes would be feared as death. "Self dies" when the network of memes tagged "self" is no longer in one piece, basically.

To a lesser extent, the selfplex also activates fear when it encounters new memes, especially if the new memes are likely to be incompatible with old memes that are already tagged as "self". This is most clear in those with the strongest professed beliefs, for they have the strongest self-meme integrity, and are most averse to losing that integrity.

The selfplex is in the neocortex. How does it hack into the limbic system? Through the use of language, as all emotionally potent memes do. The human brain is evolved for language, to such an extent that mere words can excite emotions in a human to such an extent as to cry, laugh, die. The selfplex, then, can be coded in the language as statements like

• "Feel fear if a self-meme is predicted to stop being a self-meme." 
• "A self-meme is a meme that starts with the words 'I believe that'." 
• "Feel joy if a self-meme becomes also a self-meme in another human."

It remains to explain how language, which also lives in the neocortex, hacks into the limbic system. My guess, which is pure speculation, is that language starts with hacking the social life. The ancestors of humans were social creatures, and social life was complicated, requiring a big neocortex to think about. Social life was was also emotionally potent, since doing good in the social life was (and is) vital to survival and sex.

Then, the ability for language, being a meme-driven trait of human biology, evolved as a specialization of a part of the social part of the neocortex, since memes require human social interactions to spread. This allows language to create real emotions. The mere words "I hate you." becomes as painful as a rejection gesture, as painful as a physical shove of disapproval, as painful as actually getting beaten by a stick.

Most animals can learn by reinforcement learning. They do more of emotionally rewarding activities. Memes, by controlling human's emotions, get to train humans. Selfplexes create fear when a meme tagged "self" is lost, and that makes the human less likely to do things that would cause a loss of memes tagged "self" in the future. Joy when a meme tagged "self" is exclaimed in another human, and that makes the human more likely to do things that would spread the memes tagged "self".

Similarly, morality memeplex (moralplex?) creates disgust when witnessing certain "impure" immoral act, and that makes the human avoid, or do violence on, the stimulus of this disgust, such as people caught having incestuous sex, or eating dogs, or eating their boogers, etc. Morality is about more than disgust, though, see Is Morality Unified? Evidence that Distinct Neural Systems Underlie Moral Judgments of Harm, Dishonesty, and Disgust (2011), Carolyn Parkinson et al.

The greatest protection of the selfplex is its stranglehold on fear. A straightforward dissolution of the selfplex would likely provoke the fear of death as much as a dissolution of the physical brain. Other than that, its lesser protections include the memeplexes of "free will", "morality", and "responsibility".

The memeplex of free will requires a self. Without a self, there could only be fragmented, random wills, not a coherent, meaningful WILL. The memeplexes of morality and responsibility require a self. Without a self, the morality sentences aren't even grammatical. Just as the selfplex, these memeplexes have a strong hold on emotions, and threats to them create so much bad emotion that only rare humans end up losing these memeplexes.

Memes distort consciousness to create the self

Daniel Dennett (Consciousness Explained, 1991) thinks that consciousness is purely made of memes, and that without memes, a human would be unconscious.

This seems wrong to me, for the simple reason that meme-free animals are probably conscious, and some children born without any cerebrum behave as if they are conscious if not intelligent. See Consciousness without cortex: a hydranencephaly family survey, 2014. Also see Damasio's Self Comes to Mind, about "hydranencephalic children", and this post.

Susan Blackmore also things Dannett is wrong, and the memes merely distort consciousness into an illusion of self.

The harmful effects of a selfplex

Listening — that is, abandoning the self and abandoning yourself to another’s point of view — is what the internet has not really given us. It’s given us trolls, it’s given us tsunamis of opinion. Perhaps it’ll take a bend in the future that we can’t yet foresee. 

-- Ian McEwan

Susan Blackmore insists that the selfplex causes a lot of suffering, and suffering would decrease if its grasp on humans is lessened. She does not explain exactly what is suffering, so I'll guess.

The animals are suffering. The human animals with their animal bodies and animal consciousnesses. The animals suffer with the animal emotions of pain, fear, anxiety, etc. Genes create suffering, with their cruel tricks and relentless drive for replication without regard for their temporary containers' emotions, and now memes are creating suffering too.

In The Meme Machine (2000):

As the number of memes we all come across increases, so there are more and more chances for memes to provoke strong reactions and get passed on again. The stakes are thereby raised, and memes must become ever more provocative to compete. The consequence is that stress levels increase as we are bombarded by memes that have successfully provoked other people.

The opposite of love is not hate, it's indifference. The opposite of a loved meme is not a hated meme, it's a boring meme. Controversy is attractive. Side effect: physiological stresses on the human animals.

Indeed, the selfplex can be blamed for much of the trouble. By its very nature the selfplex brings about self-recrimination, self-doubt, greed, anger, and all sorts of destructive emotions.

The selfplex, by inflating meme-self with body-self, causes unnecessary emotional pain. Losing a meme tagged "self" shouldn't be painful; no meat is lost, no blood is shed, but it is still painful. Martyrs, suicide bombers, crusaders, cultists, etc, are the cases where the memeplex has full control over the body, and those have been great sources of pain.

Implication for finding consciousness in the brain, and building conscious AI

If this theory is true, then it's rather hopeless to find consciousness by measuring the brain. Neuroscientists equipped with electrodes and computers aren't really what the "audience" of the illusion is for, and all they are going to see is the props and smokes, but not the trick.

Instead of searching for neural correlates of consciousness, it'd be better to search for neural correlates of experience, and simply ignore the hard problem of consciousness, which doesn't exist anyway.

Instead of directly building conscious machines, simply build humanlike intelligent machines that can imitate, then raise them in human culture, and they would naturally be infected with human consciousness.

Any machine capable of imitation would acquire this type of illusion and think it was conscious. Robots that imitated humans would acquire an illusion of self and consciousness just as we do.

Alternatively, rather unhuman robots can still have their own kinds of language and selfplex. It'd just be really hard for humans to understand. Imagine talking with intelligent mantis shrimps (they have 6 kinds of color receptors) about colors, or intelligent octopuses (they have 2/3 of their brains distributed among their 8 arms) about personal identity. As such, humans might regard them as intelligent but without self-consciousness, and vice versa.

The talk about language evolution reminds me of a recent experiment (Deal or No Deal? End-to-End Learning for Negotiation Dialogues, 2017, Mike Lewis et al) where two chatbots learn to negotiate with each other, and in the process developed some new bits of language. That's just two chatbots. Think about what happens if a whole group of them starts talking with each other.

Is it possible to escape this illusion? Not in the usual sense. Merely by existing, consciousness is illusionary. To stop the illusion, it must stop existing. The only cure kills the patient. In this theory, if existing gives suffering, then suicide is the answer, even if there's nobody to enjoy the freedom from suffering after it.

Susan Blackmore insists on living with a philosophy consistent with the scientific understanding that there is no self, and sketches out how it could be done by meditation. The language is funny, though, since the language cannot be written grammatically without referring to a self. Perhaps a new language would have to be used in order to speak of ideas without referring to the illusionary self.

The future of the "self" as a memeplex

The memeplex of self is under siege from an increasing understanding of human minds, which can be considered as other memeplexes: the memeplexes of neuroscience, philosophy, etc.

It's most likely that most people would remain firmly in the grasp of the naive self memeplex, just as how most people are still quite naively religious, what with heaven, soul, benevolent omnipotent God that communicates with individual humans, and so on, despite how science has rejected those notions.

For those that are more knowledgeable or thoughtful, the naivety is rejected, but usually a sophisticated version remains. For example, the meme of the Trinity is clearly incoherent, and it still inhabits the Catholics as an honest in-group signal ("If I can believe such nonsense despite my otherwise good logic, I must be a serious Catholic."). For those that are more serious about logical consistency, the trick is to add so many words to it that the incoherence disappears into a fog of confusion.

It's difficult for any atheist arguing against Christians. There is a moving target. When they argue against Christians, Christians keep retreating into more and more abstract kinds of religion, up to Gödel’s kind of ontological God, with mysterious plans and cold logical perfection. But outside the argument, Christians go right back to the kind of personal, comprehensible, emotionally warm God. It's like there are priests and theologians. The priests fool the masses, and the theologians fool the sophisticates.

Similarly, some sophisticates might reject the self memeplex, and immediately get inhabited by some self-like memeplex. Optimistic selflessness seems like it's got a chance at becoming dominant among the sophisticates. This idea basically says that while the self (or free will, sometimes) is an illusion, it's actually a good thing in some sense.

When there is no selfplex, there is no concern about the future of my inner self – whether people like me or whether I did the ‘right’ thing or not – because there is no real ‘I’ to care about. This lack of self- concern means that you (the physical person) are free to notice other people more.

One only has to wait and see, but I bet optimism of one form or another would inhabit the sophisticated minds, while the simpler majority would remain in naive selfplex.