The meaningless computer worlds of cognitive science
The meaningless computer worlds of cognitive science
Jerome Bruner helped create cognitive science and there's a fascinating reason why he didn't care for the metaphor of the mind as a computer.
The study of the human mind is so difficult, so caught in the dilemma of being both the object and the agent of its own study, that it cannot limit its inquiries to ways of thinking that grew out of yesterday's physics.
Jerome Bruner, Acts of Meaning
One of my favorite stories is the novel Use of Weapons by the late Iain M. Banks.
No spoilers here, because you really need to read it yourself to experience what I'm about to describe.
The story is about a secret agent operating on the seedy margins of the Culture, a machine-administered "perfect" society that is anything but. The story itself is nothing unorthodox if you know your noir and spy-thrillers.
Then you get to the final scene, the last five pages in a book of 400+ pages, and watch in awe as one trivial detail, revealed in a single line, obliterates the story you just read and replaces it with another.
It's a masterful gut-punch that left me gasping for air. I have envied Banks's writing skill ever since.
Think about how weird that is. In the space of a single line, a whole sequence of events changes because of one new piece of information.
How can a piece of information hiding in the future affect events in the past?
The logic of narrative runs rough over the usual intuitions about time and causality. The effect isn't limited to suspense stories deliberately playing with your perceptions, either. Narrative permits the future to influence the past by revealing new information.
That could never happen with physical causes.1
And that's strange, because we talk of causality as the master principle of stories as well as the physical universe. The key idea is easy enough. Every event has a cause. This happens, and because of that, this other thing happens.
What's that word "because" mean? Easier said than explained.
It seems obvious to many good folk that causality is a physical thing. Throw a rock at the window, the window breaks. One thing slams into another thing. Everything from the cosmic structure of the universe to the action of sub-atomic particles obeys that law. It's the foundation of everything from physics to the plot of your favorite movie.
But then we say things like, "Jim's in the kitchen because he's making a sandwich." We're now talking about a person whose actions are caused by his intentions, which are mental attitudes.
Like the total reframing that happened in Use of Weapons, the causality involved in Jim's sandwich-making requires more than knowledge about the physical events.
We have to interpret his actions.
That's a little bit tougher to fit into the rock-breaks-window kind of causality.
Cognitive scientist Jerome Bruner, who was there with Noam Chomsky and George Miller at the beginning of what is now called cognitive science, understood the limits of the machine-model more than most.
Bruner tells of how the Cognitive Revolution was originally meant to restore the idea of meaning to scientific psychology. For decades before then, psychologists had insisted on treating mental activity as observable behaviors. The mind didn't matter.
Cognitive science was meant to restore the inner life of mind, but that project got hijacked by the concept of information. Bruner wrote over 30 years ago that
[E]mphasis began shifting from "meaning" to "information", from the construction of meaning to the processing of information. These are profoundly different matters. The key factor in the shift was the introduction of computation as the ruling metaphor and of computability as a necessary criterion of a good theoretical model. Information is indifferent with respect to meaning. In computational terms, information comprises an already precoded message in the system. Meaning is preassigned to messages. It is not an outcome of computation nor is it relevant to computation save in the arbitrary sense of assignment.
Acts of Meaning
Processing of information is all blind technique. There's no need to understand what it is processing, only how to do it by shuffling signs around. Computers show the way. Any stupid machine can do the "how" as long as it's got the rule book.
Today you know the rule book as The Algorithm. Today, it's still no better at understanding the content of messages, no matter how convincing ChatGPT can be when it isn't hallucinating and making death threats.
Whatever meaning it spits out has to be put there.
That detail didn't stop the computer lovers. It wasn't long before information processing absorbed the concept of the mind, while computability replaced meaning. Mental activity doesn't have anything to do with knowing what you're talking about. It could all be explained as dumb machines playing with symbols.
And that went for your own squishy brain, too. If you don't need meaning and you can do all this neat stuff with computing, you don't need to talk about different operating principles for human minds, either.
Mind is nothing but machinery. Cognitive science and Artificial Intelligence are close cousins, each founded on the same ruling idea.
This is called computationalism. Its defenders want machines to eat your soul.
If you pay attention to these writings around here, you might have noticed that I'm careful — some might say nitpicking — about the words used to describe human beings.
There is a reason that I refuse to describe real-live people as brains, dopamine, genes, glands, hormones, and other metaphors borrowed from inanimate nature and mechanical objects.
The names we give to things, even unconsciously and in passing, shape our experiences and understanding. Words, learned and used by custom and convention, are as great a part of our world as the physical objects around us. Language is magic.
Unless it isn't. Unless words and behavior are nothing more than by-products of mechanical cause and effect in our brains.
But then, that's a story told in narrative form.
Materialists make sense of reality with a story that there is no story. Story is mere opinion, neurological fabrication, psychological projection. Whatever it is, it isn't real like physical stuff is real.
Which is a story. Someone has to tell it, and you have to buy into it.
Bruner's point, and it is worth taking seriously, is that scientists of the mind are not going to get too far if they insist on using a simplified model of mental function borrowing the best physics of the 19th century.
The study of mind should be an interdisciplinary field drawing on anthropology, developmental psychology, and even history and literature. There's no need to limit ourselves to one preferred field with one preferred way of doing things.
Which is exactly what cognitive science has gotten us.
If the computationalism is right, then all the common-sense vocabularies we use to talk about our mental lives, our actions, and other people — known as "folk psychology" — are sophisticated illusions generated by a brain struggling to understand other brains.
Sure, there's that nagging mystery about consciousness that preoccupies a certain audience, but that's a trivial thing and one day we'll get around to showing how that phenomenon appears out of blind information processing, too.
But computationalism isn't right. I draw your attention to the fact that computing and information are metaphors. They are guiding insights that direct and organize experiment, theory, hypothesis creation, and interpretation of data.
Metaphors belong to the world of narrative.
But if this is only a story, then what isn't?
Is there no truth outside of narratives? That doesn't sound very helpful either. What's going on here?
The point I'm lurching toward is that interpretations of meaning are as essential to our own mental life and our understanding of reality as physical cause and effect.
Meaning is integral to life and mind as sure as physical causality.
The logic of narrative shapes a definitely-not-physical kind of causal order on events. This makes it no less respectable or less essential. Narrative organized around meaning is so key to the way we understand the world that we can't just say "computers work, therefore it's not real".
(Nobody worth listening to actually says that, but with the proliferation of unthinking nerds repeating slogans on the internet, this must be said.)
Antimentalistic fury about folk psychology simply misses the point. The idea of jettisoning it in the interest of getting rid of mental states in our everyday explanations of human behavior is tantamount to throwing away the very phenomena that psychology needs to explain.
So it is.
Bruner goes on to divide cognitive functions into two basic types of process. One is the paradigmatic mode of cognition which we expect in logic, mathematics, and natural science. The other is the narrative mode, which creates meaning by using concepts of intentions, actions, persons, beliefs, purposes, and all that neat stuff we find in stories.
We take the paradigmatic seriously, far too seriously, to the point that it absorbs all kinds of mental activity. The power of the computer metaphor in psychology, and now in the wider culture, testifies to this.
This leads us to forget how much we rely on the narrative mode. Forms of expression like poetry and novels and drama can reveal truths about ourselves and the human condition as sure as scientific theories.
Indeed, Bruner tells us that the two forms of mental process are not opposed, pitting humanities against science, but that they manifest as two distinct but equally necessary ways of coming to terms with ourselves. One is not more important than the other. They're two equally valid and useful ways of organizing and understanding experience.
But there is also a good point about limiting untamed subjectivity into scientific explanations. The threat of radical relativism also lurks here, tempting us with the cheap slogan that "all stories are equally valid".
Storytellers we may be, but how do we decide between stories to determine matters of truth?
If psychology isn't value-free and we can't study behavior and mental activity as objective physical processes, then what stops the slide into cultural relativism?
Are we going to need a unique psychology for every people, place, nation, and culture around? How are we going to pull that off if we don't have some hard universals to work with across cultural differences?
The divide in human evolution was crossed when culture became the major factor in giving form to the minds of those living under its sway. A product of history rather than of nature, culture now became the world to which we had to adapt and the tool kit for doing so.
There is no such thing as human "nature" with optional language and culture upgrades. Culture does not simply adjust and modify naive biological needs and drives. Paraphrasing anthropologist Clifford Geertz, we are incomplete animals who complete ourselves through culture.
It is man's participation in culture and the realization of his mental powers through culture that make it impossible to construct a human psychology on the basis of the individual alone.
Culture completes our nature. It is part of what we are as sure as our kidneys, eyes, and hands.
Bringing meaning into the scientific picture doesn't commit us to wild subjectivism or relativism. The meaning-making processes that connect individuals to culture are public and shared. They exist in language, not only the "dead" words, but the living acts of speech and writing that we use in communicating.
Our culturally adapted way of life depends upon shared meanings and shared concepts and depends as well upon shared modes of discourse for negotiating differences in meaning and interpretation.
Culture is our tool kit for completing our biological natures. And that belongs to no single individual. The situation is exactly the opposite — the existence of the single individual depends on shared culture.
Why? Because folk psychology itself is a cultural product and a cultural tool. We use it to anticipate, predict, judge, explain, and understand our own actions and those of other people. Folk psychology reflects shared ways of valuing as well as knowing.
Biology is at best a constraint or condition on culturally shaped action, not its cause. It's not so easy as sandblasting the rough edges to find the true objective causes of mental activity.
Narrative organized around meaning is essential to human self-understanding. At least as essential as objective descriptions and explanations of human biology.
I have to tell you, this is wild stuff.
Most of psychology, and this is certainly true today, has not appreciated these insights. Facile fears about relativism and subjectivity have kept the majority of psychology and cognitive science on the straight and narrow path grounded in mechanical causality.
That's unfortunate. It leaves us with rather ugly and pessimistic stories about ourselves. Freudian psychoanalysis was the first in that genre. Hard determinism, whether it comes in Freudian or behaviorist or computationalist packaging, leaves our present and future in the unalterable grip of the past.
If Bruner's getting at something right, then Freud's dark vision and other kinds of mechanical determinism cannot get off the ground in psychology.
The past cannot hold us hostage because the meaning of the past is re-created in each act of memory. Your present behaviors cannot be fully caused by past events because the past depends on present understanding and future intentions. A fully material determinism is impossible.
A cultural psychology appreciates the past as a condition while acknowledging the importance of the present and the future.
The time of physics is not the time of story. The future we intend to create can transform the meaning of the present and the past.
Your own life plays out the story in a book. The meaning isn't fixed by what's already happened, not when the next event can transform the meaning of the whole sequence.
The other week I wrote about an alternative to the orthodox reading of Plato's dialogues. Plato wasn't some austere moralist and rationalist, but a playful dramatist exploring how the soul can find salvation.
In another piece I wrote of how the self appears in a process of struggle and negotiation, which is not simply physical but shaped by meaning and culture.
These ideas weave into a single, larger picture. We humans are individuals struggle to find ourselves, yet we belong to a wider transcendental reality. That negotiation to be authentically ourselves cannot happen in a vacuum. We are always with and of others.
The more we retreat into individuality, the more we ground ourselves in shared culture. There is no impersonal or amoral self. Every story we tell about ourselves articulates a sense of the worth of things, for or against, which is already present for us.
Like reading a great story with a twist ending, the meaning depends on what happens next.
Thanks for reading.
-Matt
p.s. If you found this valuable, interesting, funny, or it made you upset that you had to use your mind for something besides infinite scrolling, I ask that you do me a favor and share it with just one person.
Here's a handy button for you ⤵
If you're new here, go ahead and subscribe. Do it.
You don't have to point out quantum physics and general relativity to me. Some of the conclusions of these theories are so perplexing exactly because they challenge the mechanical causality that is the most basic presupposition of physical science. It's quite the problem, but we'll have to tackle that another day.
Unless I'm completely misunderstanding you, it sounds like you're misunderstanding Freud. Freud didn't say that the past completely determines the present. He presented the idea that if you're completely in the thrall of the past, you're suffering from an illness. He was a doctor; he believed this illness is treatable, that it is possible to free oneself from the tyranny of the past. For Freud, the fact that, as you point out, each new experience of memory is an active re-creation of the past, is exactly why some people can get trapped in it; they keep re-creating past traumas again and again for themselves. His idea was that through analysis, you can become conscious of doing this, and then you'll be able to stop doing it.
On the other hand, what you're writing here is the perfect description of the current brand of psychiatry, which believes that emotional problems can be solved with drugs, and what you are writing certainly explains why this drug therapy model is a total failure.