Consequences of Conscience

How reciprocal altruism produced the voices of the gods

I’d like to add a third postulate to Galton’s Lexical Hypothesis:

1. Those personality characteristics that are important to a group of people will eventually become a part of that group's language.

2. More important personality characteristics are more likely to be encoded into language as a single word.

3. The primary latent factor represents the direction of social selection that made us human.

In the previous post, I argued that the Primary Factor of Personality (PFP) can be summed up as the tendency to follow the Golden Rule: do you treat others as you would like to be treated. That it should play a role in our evolution is actually a reframing of an observation made by Galton’s cousin, Darwin.

After the power of language had been acquired, and the wishes of the community could be expressed, the common opinion of how each member ought to act for the public good, would naturally become in a paramount degree the guide to action. ~The Descent of Man

Inside you there are two apes

The PFP is a description of what creatures dependent on a complex society are encouraged to become. This dependence is baked into our bodies. Consider our weak jaws and digestive tract that require cooked food, or our neotenous faces. We live by the good graces of others.

It’s also baked into our outsized brains. Our long development period and unparalleled ability to communicate. Our need to join a group. And our defining feature, our conscience, the voice of society in our heads. As Darwin says:

I fully subscribe to the judgement of those writers who maintain that of all the differences between man and the lower animals, the moral sense or conscience is by far the most important.

This post explores the idea that humans evolved inner speech as a type of proto-conscience—“an inward monitor [that] would tell the animal it would have been better to have followed the one impulse rather than the other”

¹

As an epistemic note, the origin of inner language is more speculative than the origin of outer language. According to Wikipedia, debate of the latter experienced a century-long moratorium as the dearth of evidence and abundance of theory is crack cocaine for scholars (particularly physicists

²

). In other words, perfect material for this blog.

A variety of voices

Before we hypothesize about the beginning, let’s look to the many modern forms of inner voice.

There are many types of internal voice. They share in common an ability to model different points of view

There is significant variation in how much time people spend with an active inner voice. For example, see responses to these popular tweets. Or, as one person without an inner voice describes:

Rivera said not having an inner monologue has been good for her in some ways, because she can block out negative memories or thoughts relatively easily. 

It also brought some challenges. She said that when she was growing up, her mother often told her to think before she spoke, but she couldn't. 

"I can be blunt and I can have no filter. Sometimes I say things I shouldn't say," she said. "People often know what I'm thinking because I will say exactly what I'm thinking."

Basically what you’d expect, less rumination but also limited social grace. The little niceties that require deception—white lies—just aren’t there.

Our inner voice does not always represent us; we can model a dialogue where we act out another perspective. However, it’s surprisingly common to also ‘hear’ other perspectives that spring up from outside our agentic control. Depending on the definition, a majority of people may experience occasional auditory hallucinations in their lifetime. The earliest such survey—the 1894 Census of Hallucinations —interviewed 17,000 (!) healthy adults. That study reports an 8% lifetime incidence for men, and 12% for women. More recent surveys of college students indicate much higher rates. Posey and Losch find that 71% of students experience at least brief, occasional hallucinated voices, and 39% report hearing their thoughts spoken out loud. Another study of college students found that 30-40% report hearing voices, and of these, almost half reported an occasion happening in the month preceding the survey. This is a huge range that depends on what is asked. Sounds (footsteps, for example) and one’s own name are most common. Entire sentences are comparatively rare.

A study of 1800 children between 5-12 found that 46% reported having an imaginary friend, which the authors argue have similar topographies to that of normal adult hallucinatory experiences.

Worldwide there is significant cultural variation in how we interpret voices, even those that may be classified as schizophrenic. For example Nga Whakawhitinga (standing at the crossroads): How Maori understand what Western psychiatry calls ‘‘schizophrenia’’ includes some interviews.

For me hearing voices is like saying hello to your whanau [family] in the morning it is nothing unusual. (CSW)

My understanding of that is that I absolutely accept that if someone tells me that they see someone standing in the room that I can’t see that there actually is. They actually can see it. I understand that. (KAU/MAN).

They come to me when things are about to get bad... they sometimes tell me what to do and if I do it then I get through. I used to think them coming meant I was going crazy again but now I realize that when times were tough, they were there to help me through. (TW) Yep a lot of them have something that needs to be done. You will know when you’re supposed to do it, they’re not subtle, they will show you what you need to do and they won’t stop until you do it. (TW)

It’s fascinating that such a fundamental aspect of our mindscape can vary so much between different people and societies. The experiences represent a sort of spectrum, but also one between kinds; there are stark phenomenological differences between auditory hallucinations, which include sense of losing agency, vs a ruminative internal dialogue with oneself.

Function of inner voice

Geva and Fernyhough offer an overview of the competing theories for the origin of inner speech in A Penny for Your Thoughts: Children’s Inner Speech and Its Neuro-Development

There are contentious debates on whether language evolved as mechanism for symbolic thought (using inner speech) (Everaert et al., 2015, 2017) or as means of communication (Pinker and Jackendoff, 2005; Corballis, 2017). Jackendoff (1996) and others (Rijntjes et al., 2012) have discussed the importance of inner speech in human evolution, suggesting that the development of inner speech supported more complex and abstract thought. However, Pinker and Jackendoff (2005) emphasize that, in their view, language evolved initially as means of communication, and that inner speech is a “by-product”: a later evolutionary development which is a result of internalizing external speech, which in turn supports more complex thinking.

That is, inner speech evolved either to make us better at abstract thought, or it was a result of outer speech that happened to make us better at abstract thought as well as outer speech. Within this general framework a recent paper argued specifically that inner speech was an adaptation for deception—the ability to think words without revealing them to those around us. I suggest another option: that what we call inner speech is downstream auditory hallucinations of societal demands. That is, the first inner voices would have been agents of domestication encouraging hearers to consider the will of the tribe. Don’t hit, share food, agree with superiors. Be good.

While abstract reasoning

³

, deception, and better outer speech are indeed useful, securing reciprocal altruism is essential. As Darwin said, once we have outer language, the main selection pressure is on finding favor with the group. The other abilities associated with inner speech could have come later.

The content of the demands are intimately familiar to any reader who lives with an angel on their shoulder (more on the devil later). However, the way the proto-conscience was experienced is likely foreign. There is no reason to think that identifying with the internal voice came at the same time the voice started talking. Indeed, there is considerable evidence that such identification came within the bounds of oral history.

Julian Jaynes and the bicameral mind

This is the argument made by Julian Jaynes in The Origin of Consciousness in the Breakdown of the Bicameral Mind. He contends that until about 3,500 years ago humans hallucinated the demands of society on one side of the brain, then heard and executed them on the other. There was no introspection, rumination, or consciousness. As summed up in a recent article:

According to Jaynes, prior to the emergence of consciousness, the human mind was bicameral i.e., it was split into two parts: a decision-making part and a follower part. Importantly, neither one of these separate parts was conscious. For simple actions, bicameral people were creatures of habit, following well-established routines and patterns of behaviour. Every so often, however, a situation would arise for which routines and habits were not sufficient. In these situations the decision-making part of the mind was recruited. This would direct behaviour by issuing an auditory command. Crucially, these commands were not regarded as self-generated. Instead, bicameral people experienced them as being issued by an external agent. For Jaynes, this property of the bicameral mind explains the origin of gods in human societies—humans regarded these auditory hallucinations as the words of their god(s).

For this post, don’t get hung up on the totality or recency of his claims. As the quote above points out, for Jaynes it was volition that was essential for consciousness. Jaynes was a man built to joust windmills, and spent the first chapter of his book making the case that consciousness could be reduced to volition rather than reason or feeling. For this post, it is sufficient to believe that thinking using language was a Big Change.

So, who was Jaynes? A colleague described him as “an old-fashioned amateur scholar of considerable depth and tremendous ambition, who followed where his curiosity led him”. A heavy drinker and son of a preacher man, he enjoyed spotty employment as an untenured professor and playwright. He spent time in jail by refusing the draft, including administrative duties for the military. He resigned with a note to the U.S. Attorney General: “Can we work within the logic of an evil system for its destruction? Jesus did not think so … Nor do I.” Describing the Eureka moment that led to his book:

In my late twenties, living alone on Beacon Hill in Boston, I had for about a week been studying and autistically pondering some of the problems in this book, particularly the question of what knowledge is and how we can know anything at all. My convictions and misgivings had been circling about through the sometimes precious fogs of epistemologies, finding nowhere to land. One afternoon I lay down in intellectual despair on a couch. Suddenly, out of an absolute quiet, there came a firm, distinct loud voice from my upper right which said, “Include the knower in the known!” It lugged me to my feet absurdly exclaiming, “Hello?” looking for whoever was in the room. The voice had had an exact location. No one was there! Not even behind the wall where I sheepishly looked. I do not take this nebulous profundity as divinely inspired, but I do think that it is similar to what was heard by those who have in the past claimed such special selection.

Without interruption, he goes on to explain the nature of Egyptian deities.

Osiris, to go directly to the important part of this, was not a "dying god," not "life caught in the spell of death," or "a dead god," as modern interpreters have said. He was the hallucinated voice of a dead king whose admonitions could still carry weight. And since he could still be heard, there is no paradox in the fact that the body from which the voice once came should be mummified, with all the equipment of the tomb providing life's necessities: food, drink, slaves, women, the lot. There was no mysterious power that emanated from him; simply his remembered voice which appeared in hallucination to those who had known him and which could admonish or suggest even as it had before he stopped moving and breathing.

God-king from the afterlife: “moar harem.” (Somehow dalle does not mark harem as haram)

According to Jaynes, the pyramids are monuments to bicameral minds. The subjects remembered the voice of the god-king, and labored for his benefit after his death. At this time there was no space for us to reflect.

In the bicameral era, the bicameral mind was the social control, not fear or repression or even law. There were no private ambitions, no private grudges, no private frustrations, no private anything, since bicameral men had no internal ‘space’ in which to be private, and no analog to be private with.

Truthfully, I can’t believe that much of the world was populated by automatons (his word!) before the age of discovery. It’s too recent, a psychological bridge too far. I can believe that humans at some point were bicameral before identifying with the inner voice, and it would be remiss not to mention the man who so artfully described this. Those wanting a more complete account can read the many reviews of his work: Matt McClendon, Kevin Simler, Scott Alexander and Nautilis. There is also a somewhat active forum at Julian Jaynes Society.

The Paradox of Schizophrenia

Living the Golden Rule requires theory of mind (ToM). “If I were in their situation, how would I like to be treated?” Effective communication more generally also requires ToM. (“How can I get them to understand how to make this arrowhead? What are their current misconceptions?”) The development of language must have unlocked an enormous selective pressure for towards minds that were increasingly capable of simulating other minds. Molecular biology provides some clues about knock-on consequences of this ability in Interrogating the Evolutionary Paradox of Schizophrenia (2019):

Schizophrenia is a psychiatric disorder with a worldwide prevalence of ∼1%. The high heritability and reduced fertility among schizophrenia patients have raised an evolutionary paradox: why has negative selection not eliminated schizophrenia associated alleles during evolution?

…

In Figure 4, we offer a simple preliminary framework that integrates our results within an evolutionary context. Our framework adopts the by-product hypothesis’ notion that the number of schizophrenia risk alleles increased with the development of the social brain, language, and high-order cognitive functions (Crow, 2000; Burns, 2004). Aligned with this notion, we speculate that around 100,000 – 150,000 years ago (Burns, 2004), before the migration of modern humans out-of-Africa (Stringer and Andrews, 1988), there was a “turning point” at which time the number of schizophrenia risk alleles plateaued. Thereafter, risk alleles for schizophrenia have been progressively but slowly eliminated from the modern human genome while undergoing negative selection pressure.

GWAS is still young and the authors offer many caveats. But this timeline both explains the population genetics of schizophrenia and aligns with a plausible commencement of inner speech at peak schizophrenia.

At a certain threshold, genes that aide theory of mind also produce auditory hallucinations. When there was no inner voice, ToM and schizophrenia alleles were under positive selection. Once auditory hallucinations became a phenotype—sometime after the development of language—there was selection against hallucinations (but still for ToM).

What were the voices like?

Imagine the first human to hear voices gathering berries when the forest goes eerily quiet. A voice shouts, “Run! There’s a bear!” It’s not clear this person would have had the ability to think “what was that voice?”, which may require modern inner speech. Even if abstract deliberation were possible, a likely interpretation of the voice would be some spirit agent. These beliefs are present in every culture to this day.

Millenia later, all we can do is speculate on what the voices would have said. If they were produced by our over-active theory of mind designed to live the Golden Rule, then they would have been the demands of society.

Jaynes argues this produced god-kings and spandrels as large as the pyramids at Giza. Society, especially once it is hierarchical, can ask all sorts of things. But in general it was individually fit. Those with functional inner voices produced more offspring. (Could this have been our edge over the Neanderthals? The madman theory of inter-human politics.)

Fitness enhancing prosocial voices might have instructed their hosts to conform and be a good teammate. “Patience! Share! Protect the sacred cow!”

Friends and relatives, living and dead, would have been reoccuring cast members in the bicameral mind. Parallels the Internal Family Systems model of psychotherapy.

One can also imagine voices urging “lie! cheat! steal!”. This is indeed sometimes fit, but these behaviors are more complex when interacting with the same 50 players your entire life. As Darwin said, for humans, reciprocal altruism is the name of the game. I think we developed the devil on our shoulder—the ability to defect for personal gain—later.

Of course, this was a rudimentary system with many failure modes. What to do when society makes competing demands? Or demands too much? What memetic god calls upon the fanatic to self-immolate, for example? Descent of Man includes an interesting case where duty to familial specters misfires.

Dr. Landor acted as a magistrate in West Australia, and relates, that a native on his farm, after losing one of his wives from disease, came and said that, "he was going to a distant tribe to spear a woman, to satisfy his sense of duty to his wife. I told him that if he did so, I would send him to prison for life. He remained about the farm for some months, but got exceedingly thin, and complained that he could not rest or eat, that his wife's spirit was haunting him, because he had not taken a life for hers. I was inexorable, and assured him that nothing should save him if he did." Nevertheless the man disappeared for more than a year, and then returned in high condition; and his other wife told Dr. Landor that her husband had taken the life of a woman belonging to a distant tribe; but it was impossible to obtain legal evidence of the act.

Perhaps, when we first heard voices, we were specialists at fitting in the same way squirrels are specialists at hiding nuts. They have a sophisticated strategy that takes into account the availability of other nuts, whether others are watching, and the time of the year. According to one squirrel expert, "Animals are as smart as they need to be, including humans. They have evolved to solve a particular type of problem, and for squirrels that problem is storing food and finding it later. They are really good at that problem." Why should solving social problems be any different at the start? Squirrels can’t introspect on their nut strategy. Likewise, in the bicameral theory, perhaps humans could not introspect on their social strategy or tell when it was leading them astray. We were ‘narrow’ self-domestication experts: Homo Schizo.

Pictured: Cro-Magnon finds God

Altruism and group selection

There is an ongoing debate as to whether altruism requires group selection. The calculus of selfish genes dictates I should lay down my life for two of my brothers or eight of my cousins. But we see people sacrifice for strangers all the time. It is one of our defining features, yet there are still papers written claiming it was an evolutionary mistake. One explanation for altruism is group selection, which suggests more altruistic groups tend to displace selfish ones. But group selection theory suffers from the free-rider problem: the most fit strategy is to be a selfish member of an altruistic tribe. Most evolutionary biologists reject group selection. How then did humans become so altruistic?

A 2020 paper argues that, in fact, the strong form of group selection is not needed. It is only requisite that a group come up with altruistic norms and punish defectors. The bicameral mind is such a mechanism this selection could have produced.

To give a sense of other solutions, consider the 2012 possible genetic and epigenetic links between human inner speech, schizophrenia and altruism. It uses the framework of exaptation and mis-exaptation to square these three traits. Exapatation, or co-opted adaptation, is when a trait is produced for one reason, then is selected for another use. Feathers are the prototypical example. Original selected as thermal regulators, they then were co-opted for flight. A mis-exaptation is when one adaptation produces new, harmful traits. The paper argues that both schizophrenia and altruism are mis-aptations of inner speech.

Reports of inner speech as an exaptation employed for ethical purposes are exceedingly numerous: Socrates is one of the best known examples, as reported by Plato, especially in the Apology, where his inner speech demonstrates its relevance for ethical behaviour (see, e.g., Reale, 2001). Another interesting example, but one that is border-line between psychic homeostasis, physiology and pathology, is the dialogue between Tasso and his familiar, Genius. It is described in a passage by Leopardi (1834) in which Genius (i.e., Tasso's inner speech; see Appendix 1 Panel a) consoles the poet during his captivity. It should be noted that Tasso, in a schizophrenic attitude, accepts his Genius's voice as an external, real entity. Even in later literature (see Appendix 1 Panel b), the conversations between the fanatic, Naphta, and the humanist, Settembrini, in Thomas Mann's ‘The Magic Moun- tain’ (Mann, 1924), and the dialogue between The Devil and the composer, Adrian Leverkuhn, in Mann's ‘Doctor Faustus’, can be regarded as the externalisation (in the form of Mann's manuscript) of the author's own inner speaking, touching on the nature of ethical conflicts which torment a creative genius.

The importance of inner speech should again be pointed out since it allows us to widen the range of empathy and caring from what is available to other animals. In particular, the capacity for drawing analogies allows the detection of similarities between one human group and another, unfamiliar group, and which can promote, by means of such ‘external considering’, empathy towards those people (Eisler and Levine, 2002). If the connections between rational and emotional processes in our brains (in which the orbitomedial prefrontal cortex is especially vital) are working at their best, analogy – the sensing of another's predicament – leads to empathy (see Barnes and Thagard, 1997; Eslinger, 1998) and, hence, to cooperation and mutual caring between different individuals and groups being a much more likely outcome.

Based on the bicameral theory, perhaps we should consider the flipped evolutionary path instead: that reciprocal altruism selected for pseudo-schizophrenic minds that were susceptible to pro-social auditory hallucinations. Later we experienced a breakdown of the bicameral mind that resulted in our modern inner speech. This accounts for all three traits without treating our defining trait as an evolutionary accident. Additionally, it is in line with the understanding of Tasso, Socrates, and the Maori.

Is consciousness a matter of degrees?

Considering the inner lives of man and beast, Jaynes writes:

The chasm is awesome. The emotional lives of men and of other mammals are indeed marvelously similar. But to focus upon the similarity unduly is to forget that such a chasm exists at all. The intellectual life of man, his culture and history and religion and science, is different from anything else we know of in the universe. That is fact. It is as if all life evolved to a certain point, and then in ourselves turned at a right angle and simply exploded in a different direction.

This is in line with comments Darwin made allowing for human exceptionalism:

If it could be proved that certain high mental powers, such as the formation of general concepts, self-consciousness, etc., were absolutely peculiar to man, which seems extremely doubtful, it is not improbable that these qualities are merely the incidental results of other highly-advanced intellectual faculties; and these again mainly the result of the continued use of a perfect language.

As promised this post does not hinge on the magnitude of the change wrought by the bicameral breakdown. It could be a matter of degrees and the proposed mechanism remains interesting. However, I believe that "I think therefore I am” would have been a category error in the bicameral era. There would have been no self with which to think or identify. “I” came later.

For our needs it suffices that inner speech is “very important for cognition”. Even if the effect is a matter of degrees, these are the degrees most unique to our species.

Eye thinks

Conclusion

How did language first get inside our head? This is the question that sparked my journey to the bicameral mind. The transition to modern cognition must have followed natural selection. In a world of only outer speech, the most fit subsequent adaptation would capture reciprocal altruism. Verbal hallucinations of societal demands could be such a mechanism. Even today, schizophrenia is widespread despite negative selection pressure, and most people have experienced hallucinations.

Jaynes finds echoes of the bicameral psyche in our bronze age mindset. Why did stone-age humans from Easter Island to Egypt litter the globe with monuments to the gods? Why is the concept of self in the Iliad so foreign? As he found evidence in text, Jaynes dates the end of the bicameral culture on this side of the written record—1,500 BC. This is hard to square that with accounts such as Cabeza de Vaca, a shipwrecked conquistador who lived with Native Americans for eight years. Or the complex philosophical system of the Aztecs. In Dawkins’ view:

It is one of those books that is either complete rubbish or a work of consummate genius, nothing in between! Probably the former, but I'm hedging my bets.

In my view bicameral framework eventually languished in obscurity because of the unbelievable dating and reliance on memetic evolution. After the Breakdown there would be a stark fitness landscape for a brain that readily accepts the new ToM. We are not just culturally removed from bicameral man.

We should remember there must have been some cognitive revolution in the last 100k years. Language is a new invention running on old hardware, and has rapidly entered center stage of mind. Was the selection pressure towards the development of inner speech strongest from abstract thinking? Language ability? Or altruism? If not the latter, why are humans so altruistic? Why are hallucinations so common? The evolutionary path must have left cultural and genetic marks.

What we need is a paleontology of consciousness, in which we can discern stratum by stratum how this metaphored world we call subjective consciousness was built up and under what particular social pressures. ~Julian Jaynes

The next post will look at what selection pressure may have caused the bicameral mind to break down, and when that may have happened. The result is our modern mind-space, all too capable of rumination. One of the most remarkable realizations for me has been that our inner voice may be downstream our ancestors trying to follow the Golden Rule. “I” may have emerged in an attempt to model the mind of others. Or in the words of Hemingway:

No man is an island entire of itself; every man is a piece of the continent, a part of the main; if a clod be washed away by the sea, Europe is the less, as well as if a promontory were, as well as any manner of thy friends or of thine own were; any man’s death diminishes me, because I am involved in mankind. And therefore never send to know for whom the bell tolls; it tolls for thee.

Polls

I’d like to experiment with polls to see which points my (very bright) readers find compelling. Please no lizardman constant!

1. Descent of Man, again.

2. Keeping physicists out of your discipline is an important survival instinct. Never let a competitor with better technology in the gates. Psychometrics, to it’s credit, brought in LL Thurstone, an engineer. Or rather joined forces to help him found the Psychometric Society and the journal Psychometrika.

3. It seems as though strong selection for abstract thinking works best with group selection. Improvements to arrowheads occur over the timeframe of centuries or millenia; the style is a good way to date specimens. The inventor captures only a fraction of the fitness boost from any discovery as humans are quite good at copying superior technology. As such, is it the entire tribe who benefits? Are they the unit that evolution is selecting on?