I was driving along a one-lane highway, speeding up an incline where a semi-trailer was parked on the shoulder of the oncoming lane due to mechanical failure, when—just as I crested the top of the hill—another vehicle appeared from behind the semi, veering around it and drifting over the center line of the road.
Which meant it was careening toward me, at a combined vehicle velocity of around 100 miles per hour.
I consider myself a “good” driver, in the motor-vehicle-licensing-department sense of good, which is to say, I know what I’m supposed to do, and I do it. I obey the speed limit (mostly) and signage, and I check my mirrors, and I don’t use a phone while driving, and I pay attention to my surroundings. But this kind of knowing, good as it is, doesn’t necessarily save you from a collision with a pickup truck pulling an extra-wide horse trailer behind it.
My first thought was, “Is this the moment I die?” My second thought was, “No, probably not”. Around the same instant, as my hands jerked on the steering wheel, jolting my 2800-pound Toyota rightward, shaving the gravel shoulder, I could foresee that I would miss the pickup truck by a few inches, and miss the extra-wide horse trailer by an even narrower margin.
I don’t know how I knew, though I could see the gap not only in the immediate moment, but in the heartbeat or two that followed I seemed to calculate the size of the gap, and sensed that the front left side of my vehicle would not get clipped at high speed, like a boxer’s uppercut to the chin, which might have otherwise knocked my lightweight Toyota off the highway, and me with it.
In retrospect, I realize if I had been more diligent, I would have taken more note of the semi-trailer—how its bulk narrowed the oncoming lane—and would have been more cautious of the possibility another vehicle might crest the hill at the same moment I did.
In that sense, I was a bit remiss. I thought I knew how to drive well, but maybe I was not as good as I thought?
Then again, the fact I survived with an instinctive maneuver suggests I know how to drive very well, not necessarily in the motor-vehicle-licensing-department sense of knowing, but in another less obvious sense of the word.
There are different kinds of knowing, some clear, some hidden and implicit. In this age of information, we tend to emphasize the former, the knowledge we can see and grasp. Information is precious to us, yet ubiquitous; it drips off our devices like sweat. We want to know more and more, so we can point to it and show others and say, “Look at what I know.”
One of the things we want to know about is the world itself, not narrowly, but in the ultimate sense. What is the best way to understand the world? How should we think about it? What should we believe about it?
These are big questions, bordering on the spiritual.
Jeremy Clifton and colleagues have been exploring world beliefs, or what they call “primals”. According to their ambitious data-driven theory, most people hold 22 distinct beliefs about the world, and most of these beliefs cluster into 3 overarching beliefs called Safe, Enticing, and Alive, and single overall belief called Good.
Here are the three basic primals in brief:
Safe (versus Dangerous): If you see the world as Safe, you tend to see it as pleasurable, regenerative, progressing, harmless, cooperative, stable, and just.
Enticing (versus Dull): If you see the world as Enticing, you tend to see it as interesting, beautiful, abundant, worth exploring, meaningful, improvable, and funny.
Alive (versus Mechanistic): If you see the world as Alive, you tend to see it as intentional, needing you, and interactive—not just as a machine with no awareness or intentions, but as an active universe that animates events, communicates with us, and has a role for each of us to play.
According to the researchers, we are more likely to see the world as “Good” if we see it as Safe, Enticing, and Alive—and the more Good we see it, the better our mental health, according to early studies.
Primal theory is a grand theory of sorts, and grand theories are fascinating, yet deserve special scrutiny, as they claim to reach the outermost boundaries of a subject, to the horizon of the known and the unknowable.
The researchers themselves did not conceive of the idea of the Good and the beliefs that underlie it. Instead, they took a data-driven approach, and started by examining all mentions of “world”, “universe”, “everything”, “nothing”, and “life”, in many of the world’s most influential historical sources including 14 sacred texts, 100 novels, 100 films, 100 speeches, and 71 treatises.
Next, they gleaned 80,677 Tweets based on relevant phrases from a 2010–2013 database, which were analyzed by natural language processing tools.
They also refined the data through psychological experts and focus groups from different religions (Confucianism, Buddhism, Daoism, Christianity, Islam, Hinduism), as well as experts in philosophy, cultural anthropology, art history, political science, and comparative religion, and subjected the findings to statistical analysis.
I’ve missed a few details, but in the end, the original data were refined enough to create the theory, along with a survey for measuring people’s world beliefs.
If this were cuisine rather than psychological research, it’s as if they bought up all the best products at Whole Foods, and the farmer’s market, plus Asian, Halal, and every other respectable food known to humanity, added a few chemicals to ensure everything would bind together, tossed in a handful of birdseed (the Tweets), and then dumped everything into a giant mixer and squeezed out a nutritious paste packed with spiritual and philosophical vitamins—or rather, what people believe to be a nutritious paste.
Jeremy Clifton has been careful to point out that the primal belief of Good refers to what people believe is “good” in terms of how to see the world, and he also emphasizes that holding these beliefs positively impacts our well-being. However, he makes no claim that the theory addresses whether these beliefs are actually true about the world.
But we are a utilitarian society. If we think something is practical or useful, we will be inclined to see it as “true”—or else to believe that truth matters mostly in the practical sense.
There are other problems too. The “Alive” world belief forms an important part of the theory; believing the universe is alive and relational—as opposed to being a machine—helps us to see the world as “Good”. I agree with this basic idea, but there is a glaring irony. Despite the involvement of human experts and focus groups, the creation of the theory depended on machines and machine methods, including statistics and natural language processors.
Deriving “the Good” by selecting key text words and ideas, and processing them through human and statistical analysis, also reflects an abstract mentality: a desire to distill things into their pure components, to put them through a centrifuge to capture an “essence” while skimming away the crud.
What if ultimate truths cannot be distilled? What if some of the truth is in the crud?
The initial primals research was done before the age of ChatGPT, which itself processes massive amounts of information, and can throw everything into a mixer and produce its own ideological paste.
When I asked ChatGPT, “Give me 3 beliefs, based on the world’s religions, that will increase my well-being”, it answered: mindfulness/non-attachment (Buddhism), unconditional love/forgiveness (Christianity), and interconnectedness (Hinduism).
Those aren’t bad answers, and I got them with barely lifting a finger. As AI becomes ever more powerful and accessible in our daily lives, we will be able to do our own primals research, using even more information than Clifton and colleagues did, to produce more comprehensive answers. Why read a scientific paper on world beliefs, when you can have a personal AI guru whispering the secrets of the universe through a smartphone?
Even so, I doubt that will get us much closer in what beliefs we ought to hold about the world. Expressing “the Good” so completely through abstractions and statistical science invites us to place too much faith in rationalism. It encourages us to nod our heads and to think, “Yes, science is the answer” or “Big data will show us truth”.
It isn’t that I don’t believe in the value of rationalism, science, or data. But when it comes to the ultimate questions of life, there is a place where these things cannot go, a place they cannot understand.
I am speaking as a person with religious faith, the domain of which concerns ultimate questions: Why are we here? Where are we going? How shall we live?
When I ask myself such questions, the asking does not begin in a conscious way, and even the answering has an implicit origin. This sort of faith—this asking and answering—begins with an impulse, an intuition to trust something beyond myself, to be in relationship with a Reality I cannot see. Then this faith articulates this trust into a set of conscious, knowable ideas and practices (believe this, do that), according to the traditions of my religion; and then, when I surrender to these ideas and practices, I am returning again to trust and intuition, to a realm I cannot fully know or control, to wait and experience what happens.1
This will sound simplistic, but to live according to a spiritual faith is, in one aspect, like knowing how to drive a car, in the sense of knowing the objective, specifiable things one needs to do. Learn and know this, and do such and such, and the car will go.
In another way, living through spiritual faith is like knowing that driving itself, in the actual lived sense, means implicitly trusting in our driving knowledge and skills—without constantly thinking about them—and understanding that, while we will probably get to our destination, unpredictable things can happen on the way—even an accident, like a pickup truck suddenly veering into the lane.
Whatever happens, the trust is that we will know how to respond to whatever comes, in the moment we encounter it. This sort of faith, in the Christian worldview, is to walk “without sight”.
This kind of knowing-by-trust doesn’t guarantee a perfect response or outcome. Intuitions can sometimes mislead—in which case, we may have to ask questions about whether our conscious knowledge or practices might need review or adjustment.
To that extent, our conscious knowledge forms part of the context through which our intuitions and implicit perceptions flow. Perhaps for the same reason, there is a temptation to try and rely on our conscious knowledge only, which forces us to create rules and controls for everything, and to flee from the risk that is required by trust. We instinctively don’t want to walk “without sight”, but with Google-map clarity.
But overcontrol does not lead to a truer spirituality any more than it leads to better driving. Instead, it erodes our ability to trust in something greater than ourselves and replaces it with faith in our grasping selves.
I don’t mean to suggest there is nothing worthwhile in primal theory. Whatever its merits, though, it suggests a future in which society’s ideology—the values we live by—the stories we tell about ourselves and the universe—are increasingly determined by big data rather than lived human experiences.
We must be profoundly skeptical about the manufacture of ultimate things, even if they are only ideas, and wary of the statistical spirits, who can only sing in the language of the Machine.
I leave the final words to a children’s book—Lucy talking to the Beavers about Aslan the Lion.
“…if there’s anyone who can appear before Aslan without their knees knocking, they're either braver than most or else just silly.”
“Then he isn’t safe?” said Lucy.
“Safe?” said Mr Beaver; “don't you hear what Mrs Beaver tells you? Who said anything about safe? ‘Course he isn’t safe. But he’s good…”
- C.S. Lewis, The Lion, the Witch and the Wardrobe
Those familiar with Iain McGilchrist’s work will recognize this pattern in neurological (brain) terms: right hemisphere, then left hemisphere, then right hemisphere. McGilchrist’s theory is also a grand theory, yet has an open-endedness that does not fully reduce its subject.
I have long been fascinated by the mechanistic view of the world that assumes it can be broken up into little pieces and then put back together again. I have though yet to meet someone who has had their head chopped off, their arms and legs removed and all their innards taken out who has been neatly sewn together by the surgeon and been what we call 'alive'!!
"Why read a scientific paper on world beliefs, when you can have a personal AI guru whispering the secrets of the universe through a smartphone?"
I suspect this is one of the (if not *the*) end game goals of AI. No need to think, just ask the machine and it will reveal to you all 'truth'. The early phase to gain adoption will be more neutral and seemingly less biased so that it can gain our trust. Once it's been full bought into and implemented in our daily lives, those behind the scenes can tweak the data/feedback provided to our queries to lead us to the desired conclusions.