Sentience: The Invention of Consciousness

SENTIENCE

The author (left), with (clockwise) Marvin Minsky, Stephen Jay Gould, Daniel Dennett, John Brockman, Eastover Farm, Connecticut, 1995 Photograph: Katinka Matson

Also by Nicholas Humphrey:

Consciousness Regained

The Inner Eye

In a Dark Time (ed. with Robert Lifton)

A History of the Mind

Leaps of Faith

The Mind Made Flesh

Seeing Red

Soul Dust

SENTIENCE

The Invention of Consciousness

NICHOLAS HUMPHREY

The MIT Press

Cambridge, Massachusetts

London, England

© 2023 Nicholas Humphrey

All rights reserved. No part of this book may be reproduced in any form by any electronic or mechanical means (including photocopying, recording, or information storage and retrieval) without permission in writing from the publisher.

The MIT Press would like to thank the anonymous peer reviewers who provided comments on drafts of this book. The generous work of academic experts is essential for establishing the authority and quality of our publications. We acknowledge with gratitude the contributions of these otherwise uncredited readers.

This book was set in Albertina by Oxford University Press.

Library of Congress Cataloging-in-Publication Data is available.

ISBN: 978-0-262-04794-4

10 9 8 7 6 5 4 3 2 1

d_r0

CONTENTS

PROLOGUE

1: SENTIENCE AND CONSCIOUSNESS

2: FOOTHILLS

3: THE TOUCH OF LIGHT

4: BLYTHE SPIRITS

5: WHAT THE FROG’S EYE TELLS THE MONKEY’S BRAIN

6: BLINDSIGHT

7: SIGHT UNSEEN

8: RED SKY AT NIGHT

9: NATURE’S PSYCHOLOGISTS

10: ON THE TRACK OF SENSATIONS

11: EVOLV ING SENTIENCE

12: THE ROAD TAKEN

13: THE PHENOMENAL SELF

14: THEORETICAL MISPRISIONS

15: COMING TO BE: SENTIENCE AND BODY SENSE

16: SENTIENCE ALL THE WAY DOWN?

17: MAPPING THE LANDSCAPE

18: GETTING WAR MER

19: TESTING, TESTING

20: QUALIAPHILIA

21: THE SELF IN ACTION

22: TAKING STOCK

23: MACHINA EX DEO

24: ETHICAL IMPER ATIVES

ACKNOWLEDGEMENTS

REFERENCES AND NOTES

INDEX

PROLOGUE

Hello? (hello) (hello)

A place for big thoughts is a hot tub at night at the edge of the Mojave Desert. It's a couple of hours’ drive from San Diego, where I've been at a meeting about human evolution. The lodge where I'm staying is surrounded by cactus-like trees. The Mormon pioneers called them Joshua trees because they seem to stretch their arms heavenwards. Lying on my back in the gurgling water, I gaze at the stars and sink into the vastness of space.

Is there anybody there?

If there are extra-terrestrial intelligent beings somewhere in our galaxy, they may be looking at the very same stars I am. Are they conscious of visual sensations like mine? Do they experience this phenomenal blackness, pricked with points of light?

I lie back, arms out sideways holding the rim of the tub. I feel the warmth of the water on my skin, smell the scent of the desert grasses. I'm flesh and blood. I'm soul. How can that be?

Hello, E.T., if you can hear me. Do you have this dual nature too? Is the light on inside your head? Do your sensations have the same eery immaterial quality that mine do?

I want to think so. I want this to be shared.

I hear a coyote bark, then another. Where else on Earth does sentience reside? Do dogs feel pain like mine? Does an earthworm enjoy smells? Are machines ever going to have conscious feelings? Do they already? How could we know?

Barking again. Have the coyotes caught a rabbit? Poor rabbit. One minute she's comfortably scratching her ear, the next a coyote has her by the neck.

What to say about the downside of sentience? The philosopher Schopenhauer wrote: ‘If the reader wishes to see whether the pleasure in the world outweighs the pain, let him compare the respective feelings of two animals, one of whom is engaged in eating the other.’

There's a rock in the next valley, sculpted by the winds into the shape of a huge human skull. The hill in Jerusalem where Jesus was crucified was called Skull Rock—Golgotha in Aramaic, Calvary in Latin. Schopenhauer might have compared the feelings of two humans, one of whom is nailing another to a cross. During the crucifixion it's said that day turned to night. The stars came out.

But suppose conscious beings like us have not evolved anywhere else.

Suppose consciousness as it exists on Earth is a one-off accident of evolution. Astronaut Frank Borman, looking from the window of Apollo 8, remarked, ‘the Earth is the only thing in the universe that has any colour’. This can't be strictly true. But it could be true that the Earth is the only place where sensations of colour exist. Or sensations of anything: sweetness, warmth, bitterness, pain. Which would be better: a universe without either joy or tears or a universe with both? Philosopher Thomas Metzinger agrees with Schopenhauer: the net utility is negative. He says that if an all-powerful and all-knowing ‘superintelligence’ could look across the world of pleasure and pain, and do the sums, it would conclude that it had a moral obligation to eliminate conscious life.

I think he's wrong. We don't live by bread alone. Pain and pleasure can't be all that matters. But there's no question that they matter. When we have reason to think someone is suffering, we have a duty of care. Some people think we have an equal duty towards any sentient being—human, nonhuman, even robot. It's not self-evident. But it could still be a rule we choose to live by. In that case, we have a heavy obligation to get it right about what in the world is conscious and what isn't.

The Norwegian government permitted the drawn-out, ugly, killing of more than 1,000 whales, including breeding females. But the Swiss government has made it illegal to boil lobsters alive and the British government soon may do the same.

Descartes believed that it's only humans who have feelings. Non-human animals are all unconscious machines. That's hard to believe. But maybe some animals are unconscious. A moth lands in the water of the hot tub. I scoop it out and toss it aside. Descartes could be right about moths. I hope he is right about moths.

Could Descartes be right about extra-terrestrials? What if the life-forms that exist out there are simply jumped-up insects? They might be ever so clever and still not have conscious feelings. I don't believe there's any necessary connection between intelligence and sentience.

Many people—famous philosophers among them—still don't get this. They think that if an octopus can solve a picture-puzzle that a four-year-old child would have trouble with, it probably has sensations something like ours.

Frans de Waal asks, ‘Are we smart enough to know how smart animals are?’ He writes a beautiful book, Mama's Last Hug. He's sure animals have feelings on the same level that we do. But what he comes up with as evidence is no more than a laundry list of clever tricks.

What about the Argument from Continuity? People say evolution has been a gradual process, without sharp discontinuities. There won't have been any point in history where we could draw a line: unconscious that side, conscious this. So, some kind of consciousness must go all the way down.

Panpsychists—‘conscious everywhere’ theorists—believe that consciousness is a basic property of physical matter. Even a teacup has a smidgeon of conscious feeling. Panpsychism seems to me a really bad idea. What would a smidgeon be like? Whose experience would it be?

It seems to me that consciousness must either be fully fledged or not there at all. That's certainly my experience of it. I move abruptly in and out consciousness when I wake from sleep or fall back into it again. Why not a similarly abrupt transition in the course of evolution: a critical point when all of a sudden our ancestors woke up, the lights came on? One small step for the brain; one giant leap for the mind?

There's a moving star, slowly arcing across the sky. No, not a star, a manmade satellite.

We human beings are on our way to becoming extra-terrestrial ourselves. Soon there will be sentient beings in space. But we, in our human bodies, won't be able to go beyond our solar system. If we want to explore the stars, we'll have to send intelligent robots in our place. Could these be sentient robots—machines that value their consciousness as we do? What extra ingredient would be required in their design?

Daniel Hillis has suggested that the World Wide Web has already become conscious simply as a consequence of its complexity. Only it hasn't deigned to tell us yet. Could the World Wide Web be hurting? Do we have a duty of care towards it?

It's often claimed that, as we build more and more complex robots, it will just happen: a threshold will be reached where sentience arrives as an emergent property—in the same way it happened in evolution. But I don't think it did just happen like that in evolution. I believe circuitry had to be built into the brains of our ancestors by natural selection for the special purpose of adding sentience.

There are many who disagree with me. They point out—quite rightly—that sentience can have been selected for only if it makes a positive difference to survival. But then, they ask, where's the evidence it makes any difference at all?

Any difference at all? In my own case, I want to say it makes all the difference in the world: the difference between being me and not being me! Yes, but wanting to say it doesn't make it true. I could be kidding myself. There are a good many people who maintain that I am kidding myself.

Hmm. This is going to need some turning round.

Back home in the UK, a new ‘Animal Welfare Bill (Recognition of Sentience)’ is under consideration by parliament. Clause 1 is the ‘animal sentience’ clause. The Secretary of State says that this will ‘embed in UK statute the principle that animals are sentient beings, capable of feeling pain and pleasure’. I see that Sir Stephen Laws, former First Parliamentary Counsel, has commented at the Committee stage: ‘It is fair to say that all the concepts in Clause 1 seem to be problematic in one way or another.’

He's right. It's a philosophical, scientific, ethical, and legal mess. As of now, we lack not only direct evidence but even agreed arguments as to how far consciousness extends. Arguably, the only sure case is our own, and every other is within reasonable doubt. Yet, all the time, we are obliged to act as if we know the answers.

Mary Oliver ends her lovely poem about whether stones, trees, and clouds have conscious feelings, ‘Do Stones Feel?’ by protesting that even if the world says it's not possible, she refuses to concur. ‘Too terrible it would be, to be wrong.’

I understand her: the poet's refusal to bow to impossibility, the insistent tug of the ‘What if?’ What if stones feel? I may say I share the world's opinion about stones. I'm as certain as can be that they don't feel. But lobsters, octopuses?

Terrible to be wrong. Yes. But irresponsible not to be right—if we can only establish what right is. Let's suppose we can discover how conscious feeling is generated in the brain and how it shows up in animals’ behaviour. Then perhaps we'll even be able to have a diagnostic test.

When Archimedes, in his bath, realized how he could test whether or not the king's crown was pure gold, he leapt out and ran naked through the streets of Syracuse. I lie back in the tub in the desert and wait for that Eureka moment.

Philosopher Jerry Fodor has said, ‘We don't know, even to a first glimmer, how a brain (or anything else that is physical) could manage to be a locus of conscious experience. This is, surely, among the ultimate metaphysical mysteries; don't bet on anybody ever solving it.’

It seems it might be a long wait.

1

SENTIENCE AND CONSCIOUSNESS

I have used the terms ‘sentience’ and ‘consciousness’ liberally in the prologue without attending to their definition. Easy to do when you're lounging in a bath. Easy to do, actually, when you're sitting at your desk and writing an academic paper. I admit I sometimes find myself at cross purposes even with myself.

Given the seriousness of our topic, we can't afford to be slapdash about the language. So, from here on, I'll be more careful, at the expense of being more long-winded.

Let's start with the words ‘sentient’ and ‘sentience’. The adjective ‘sentient’ came into use in the early seventeenth century to describe any creature—human or otherwise—that responds to sensory stimuli. But the meaning subsequently narrowed to put emphasis on the inner quality of the experience: what sensations feel like to the subject. And by the time sentience, the state of being sentient, was under discussion, notably in an 1839 book about cruelty to animals,¹ what was at issue was whether animals have experiences that feel to them the way ours do to us.

Thus sentience, in the first instance at least, gets its meaning by ostension. If we ask whether another creature is sentient, our understanding of the question rests on our being able to point to personal examples of what it's like for us.

Being sentient means having experiences like this: like the sensation of redness we have when we look at a poppy or the sensation of sweetness we have when we taste a sugar lump.

As scientists, however, we must step back from our first-person involvement. We must get a fix on what sensations are objectively about. I'll return to this over and over again as we go forward. But let's say for now that sensations are basically mental states— ideas—that track what's happening at our sense organs: light at our eyes, sound at our ears, scent at our nostrils, and so on.

They provide us, as subjects, with information about the quality of the sensory stimulus, its distribution and intensity, its bodily location, and—especially—how we evaluate it: the pain is in my toe and horrible; the red light is at my eyes and stirs me up.

But ‘tracking’ this information is only half the story. For, as we can each of us attest, sensations have a qualitative dimension that sets them apart from all other mental states and attitudes. There's something that our pains, smells, sights, and so on have in common that our thoughts, beliefs, wishes, and so on don't. For want of a better word, let's say it's something uniquely ‘charming’.

We may not be sure exactly what this charming something is, but we can be sure that it is. Suppose you were to acquire a new kind of sense organ in addition to those you already have, for example, a sense organ that registers magnetic fields. Your magnetic sensations might be as different from visual sensations as visual are from tactile or auditory. But if they were to be in their own way similarly charming, you would recognize immediately that they were up there with the rest.

Now, when we ask about sentience in a non-human creature, the same applies. The creature's sensations need not correspond to ours in every way. The creature might indeed have sense organs that we don't have. But the quality of its sensations must be such that, if they were ours, we would recognize them as belonging to the same charmed class.

Philosophers do have another word than charming. They call this special quality ‘phenomenal quality’, and they call particular examples of it—such as phenomenal redness or phenomenal sweetness—‘qualia’. Moreover, they say that, when we experience qualia, ‘it's like something’ to have the experience: it's like something to feel the pain of a bee sting, and it would—or could be— like something to have the sensation of magnetic north. Though none of these terms is ideal, I'll make use of them as others do. A creature is sentient if but only if it consciously experiences qualia—by virtue of which it becomes like something to be itself.

*

Let's turn to the term ‘consciousness’. When we are aware of having experiences with phenomenal qualities, we can be said to be ‘phenomenally conscious’. For many philosophers, phenomenal consciousness is the only kind of consciousness that really matters. David Chalmers, for example, says, ‘I use experience, conscious experience, and subjective experience more or less interchangeably as synonyms for phenomenal consciousness.’ ²

However, here we do need to exercise some care. The word ‘consciousness’ has been in use a good deal longer than sentience or phenomenal consciousness, and along the way it has acquired a considerably wider reach, both in everyday speech and in the science of psychology. The oldest meaning, going back to classical times, has to do with self-knowledge: we say a person is conscious of a mental state when they know they have it. A modern meaning in cognitive science has to do with information processing: we say a state is conscious when its contents are available to a global workspace in the brain. Neither of these meanings restricts consciousness to states that have phenomenal quality.

Phenomenal consciousness is unquestionably a variety of consciousness. We know about our sensations. They influence our judgements and decisions. But, compared to other mental states of which we're conscious, sensations are clearly in a class of their own. To understand what's so special about them, we're going to have to tell the story of phenomenal consciousness separately from the story of consciousness in its entirety.

So, here at the start, let's take a closer look at the landscape, to see where phenomenal consciousness fits in.

*

We can begin with a simple definition, going back to the original meaning. Consciousness means having knowledge of what's in your mind. Your conscious mental states comprise just those states to which at any one time you have introspective access and of which you are the subject.

These can include all sorts of mental states: memories, emotions, wishes, thoughts, feelings, and so on. When you introspect, you observe these various states, as it were, with an inner eye. Thus, it comes naturally—and people everywhere do this— for you to think of consciousness as some kind of window on the mind, a private view of the stage where your mental life is being played out.

A view from whose standpoint? Well, from the standpoint of whom else but ‘You’, your conscious self. Wherever your self focuses its gaze, it takes over as the singular subject of these states. And this speaks to one of the most striking features of consciousness: its unity. Across different states and across time, the conscious subject remains one and the same. There's only one ‘You’ at the window, only one self. When you find yourself feeling pain, or wanting breakfast, or remembering your mother's face, it's the same you in each case.

We might think it obvious it has to be so. But actually this unity is by no means a logical necessity. It's quite conceivable— and indeed psychologically plausible—that your brain could house several independent selves, each representing a different segment of the mind. In fact, this fragmented state may have been the way you started out at birth. Thankfully, however, it was never going to stay that way. As your life got going and your body— your one body—began interacting with the outside world, these separate selves were destined to come into register, orchestrated, as it were, by the single line of music that made up your life.

The unity of the self underwrites the most obvious cognitive function of consciousness, which is to create what Marvin Minsky has called the ‘society of mind’. Just as—in fact just because— there is only one ‘You’ at the window looking in, there comes to be only one mind on the other side. Anything that is in consciousness becomes shareable with whatever else is. Information from different agencies is being brought to the same table, and it's here that your sub-selves can meet up, shake hands, and engage in fertile cross-talk. This means you now have a mind-wide forum for planning and decision making—a conscious workspace wherein you can recognize patterns, marry past and future, assign priorities, and so on. A computer engineer might recognize this as an ‘expert system’, designed to anticipate your environment and make intelligent choices. You, of course, recognize it as ‘You’.

Then, alongside this, a different kind of opportunity emerges. Once you can observe the parts of the mind interacting on a single stage, you are in a position to make sense of the interaction and track its history. Observing, for example, how ‘beliefs’ and ‘desires’ generate ‘wishes’ that lead to