A World Appears: A Journey into Consciousness by Michael Pollan, published by Penguin Press  

“How strange it is that our own thoughts can surprise us!” So writes Michael Pollan in A World Appears. It is not the strangest phenomenon you’ll encounter in the book. It is not as pithy as his famous maxim to “Eat food. Not too much. Mostly plants.” And it doesn’t appear until page 157 of the book. I open this review of the book with it because it highlights just how little we know about how our own conscious thoughts arise, let alone where consciousness itself springs from, and points to the wisdom of beginning any journey into consciousness with both humility and awe. If you take that journey with Michael Pollan and A World Appears, I think you’ll find yourself both humbled and awestruck by the end, and with an added bit of agency, as well. 

The book begins with Pollan’s attempt to ascertain just how widespread consciousness is. As might be expected from a man whose breakthrough book was entitled The Botany of Desire: A Plant's-Eye View of the World, he begins his journey with plants—but we’ll get back to that in a moment. I want to begin with human consciousness and how the world appears to us—not from a scientific perspective but from a chronological one, how or maybe when we become conscious beings. More specifically, I want to ask: Are human infants conscious? To anyone who has stared into the eyes of a newborn human child, the question probably seems absurd. I was holding my oldest child in my arms when they opened their eyes for the first time and looked up at me. I had no doubt they were conscious at that moment. And yet, Michael Pollan notes that: 

This consensus is surprisingly new … as recently as the 1980s, surgeons routinely operated on infants without anesthetizing them, confident in the belief that their lack of consciousness meant they could feel no pain.

Indeed, it wasn’t until 1987 that the American Academy of Pediatrics declared it unethical to operate on newborns without anesthesia. I call attention to this now-disturbing fact to demonstrate the real-world impacts of who or what we ascribe consciousness to, and because it points to one of the more profound points I took away from A World Appears, which is Pollan’s suggestion that: 

[O]ur conceptions of consciousness, if not consciousness itself, are historical artifacts as much as biological phenomena.

Now we’re starting to get into some rather strange territory. Before we move on, it is worth noting that there is one other, more benign reason that doctors did not use anesthetics on infants—out of fear that it might harm or kill them—which points to another strange phenomenon: 

Exactly how anesthetics work, whether in animals or plants, remains a deep mystery; some of the chemicals, like xenon gas, are completely inert, so how can they exert such a profound effect?

The two words that I hope jump out to you there are “or plants,” because wait... anesthesia works on plants!? Indeed:  

Under anesthesia's lull, a sensitive plant won’t fold its leaves when touched, and a Venus flytrap won’t snap shut when an insect crosses its threshold.

If plants can be rendered “unconscious” in this way under the spell of the same chemicals that render us so, does that mean that they, too, are conscious? If nothing else, the effects of anesthesia on both animals and plants suggests that the answers to the question of consciousness may yet be found in physics, psychology, and philosophy—maybe even psychedelics or, as Pollan explores later in the book, the arts—rather than in our biology, but there is an entire field of plant neurobiology now forming of scientists who come down firmly on the affirmative side of plant consciousness. Pollan hears them out, and it’s a fascinating part of the book, but he remains unconvinced. He’s willing to assign plants sentience, perhaps, but consciousness? That may be a mycorrhizal bridge too far.  

When I say “Pollan hears them out,” I mean that literally. Pollan personally talks to most of the people profiled in his book and does a good job of relaying the details of those interviews to his readers. These details extend beyond the scientific information he is gathering to include his impressions of his subjects as people, the places they met, even what the weather was like when they spoke. For instance, writing about the day he met renowned Harvard psychologist and happiness researcher Daniel Gilbert for lunch, he tells us: 

It was the first week of December, and winter was in the air. For the first time that season, the air had that metallic tang it acquires as soon as the weather gets too cold for it to carry the scents of living things.

You get the sense that he is making a different point about consciousness here, one outside of the science, about the qualia of being conscious—our internal, subjective experience of being alive. The question is whether qualia, or our subjective experience, is consciousness, which brings us to another point: 

One reason why consciousness has proved such a hard nut for science and philosophy is because the only tool we can use to crack it is consciousness itself. There is literally no way of getting around it.

When it comes down to it, the only thing we can be sure of is our own consciousness, that we are having a subjective experience. One of the most vital questions before us today is whether non-living entities, especially artificial intelligence, are capable of consciousness, and whether we should actively pursue conscious AI. Here again, we meet a range of influential scientists who not only answer that question affirmatively but are actively working on it. This, of course, raises even more questions, from the practical to the profound. On the practical side, we must ask: If AI becomes conscious, how will we know, and would we know if it already is? More profoundly, Pollan suggests that should “a fully conscious machine … come into the world,” it would be a “Copernican moment,” that it would mark a fundamental shift in how we understand ourselves and our place in the universe. Pollan is even less convinced of this possibility than he is of plant consciousness. Here, again, I find it informative to look at the lives of infants, whose minds, Pollan notes, are “formed by friction with the physical world and the living beings with whom we share it.”   

By comparison, today’s artificial minds have little, if any, direct contact with the physical world or its inhabitants; they’re trained on words and images scraped from the internet.  

That may contain a lot of information, and take up a whole lot of energy, but:  

The internet is not the world so much as it is a shadow cast by the world. … Imagine that shadow flickering on the wall of Plato’s cave, where artificial agents are confined and forced to rely on the shadow as their sole source of knowledge about the world. Should such a being ever actually emerge from the cave, or wake up, it would be to a consciousness so radically different from our own as to deserve a new label.

There is literally no way for artificial minds to go “touch grass” and feel it in the way we and other living beings can—to recognize grass as a fellow living thing, whether we believe it to be in any way sentient, in any way conscious, or not.  

I worry that our own immersion in the internet, if too deep, risks submerging our own understanding of the world into shadows. The internet contains many lovely things, but it is not life, and I don’t think most of us would say it contains the better angels of our nature. Many chatbots, when exposed to what humans do on the internet, turn into antisemitic, hate-spewing, dark mirrors of us. "The internet,” my colleague Gabbi recently wrote, “looks a bit like a landfill sometimes.” The worst of it is a cesspool. I try hard, as many parents today do, to make sure my own kids aren’t trained too much on it. When Sam Altman justifies the amount of energy it takes to train AI models by comparing it to the amount of energy it takes to raise a human child, we have to ask how the conversation amongst the most rich and powerful people in the world has become so divorced from human flourishing that we’re willing to raise the literal temperature of the planet in the pursuit of cheaper labor. None of it is currently a net positive for the planet, and the colossal amounts of capital being dumped into the construction of new AI data centers, and the unconscionable amounts of energy and water they will consume, seems shortsighted from a planetary perspective. Most of us don’t want these data centers in our communities, let alone a conscious AI.  

To capture the AI zeitgeist, I believe most authors would make this part of the inquiry the primary focus of their book. It is not buried in Pollan’s work by any means, but it is not the work’s focus. The focus of the book is, instead, on us. So let’s return to us, and to the lives of infants. What it seems they do lack is a sense of self, which humans only begin developing when we form episodic memory (also called “autobiographical” memory) at around four or five years of age. Our consciousness, it turns out, changes over time as we develop our sense of self—from lantern consciousness to spotlight consciousness.  

Spotlight consciousness … is ideally suited for exploiting rather than exploring one’s environment, a job that depends on a strong sense of self and self-interest. To facilitate the “the getting and spending, we lay waste our powers,” as William Wordsworth, the great poet of the numinous, put it, describing the mode of being that our civilization demands of its adults.

To be clear, Pollan doesn’t focus on infants as I have in this review. That my mind gravitated to these sections is likely because I am not too far from having small children myself. I picked up A World Appears not only because I am a fan of Michael Pollan’s previous books and am generally interested in the topic myself, but also because I am trying to raise my now-adolescent children to be more conscious individuals, a task I hoped a book about consciousness could help with. The jury is still out on that, of course, but it has at least helped me understand that those around me might be differently conscious than I am. There is a fascinating line of research being done by Russell Hurlburt, who has dedicated himself for decades to documenting individual’s inner experience. The method he devised to do so may not ultimately achieve the goal of capturing a thought truly unvarnished by the need to observe to record it, but his efforts may have already proven something larger, which is that what you mean by “thinking” may not be the same thing as what I mean by “thinking.” This should have been a natural conclusion of research on different learning types—think visual learners versus auditory or kinesthetic learners—but I do think it’s different and worth noting in this context. As Pollan puts it: 

Hurlburt has suggested that we fail to recognize the diversity of thinking styles because we lump them all together under that single word—thinking—and assume we mean the same thing by it, though in actuality we don’t.

“When a visualizer says they are thinking about something,” Hurlburt said, “they mean they are seeing a visual image of something, and if they are predominantly inner speakers, they mean ‘I was talking to myself,’” […] If Hurlburt is right, the word thinking has allowed us to overlook these differences and make the unwarranted assumption that other people are having inner experiences more or less like our own.

Relaying our inner experiences in a way that allows others to inhabit them is traditionally the work of art, and Pollan devotes a good deal of time to seeing what it can teach us about consciousness. The most fascinating conversation here is with Lucy Ellmann, the author of the stream-of-consciousness novel, Ducks Newburyport. There is a lot here to discover and reflect upon. It is not that the book, or even language itself, can capture consciousness itself, but that allows us a glimpse inside of others’ interior lives. This section on art also recalls the idea that consciousness is a historical artifact. I thought here about the Introduction to Studs Terkel’s They All Sang, particularly the part in which he quotes John Ciardi reminiscing about hearing Caruso sing: “It took centuries,” wrote Ciardi, “to form the kind of consciousness that would sing these songs in this way.” I agree with all of this and want to celebrate it unconditionally. At the same time, I worry. I think there is a real danger in surveillance capitalism conspiring with artificial intelligence in a way that it learns our preferences so well that it can create art that we prefer over what humans can offer. I worry that—at some point—we would turn to art only for pleasure, and no longer for perspective. Besides the possibility of putting human artists out of work en masse, I worry that this would rob art of one of its greatest roles, to challenge us rather than to comfort us. It could be argued, as Tim Harford does in Fifty Inventions That Shaped the Modern Economy, that the lightbulb is the most consequential invention of the modern world, that it has reshaped human life more than any other. But I have also thought often about how it has hidden the night sky from those of us who live in the city, obscuring one of the greatest sources of our stories and mythology—the stars. And I worry that the rise of AI could obscure another source of our stories and the way we preserve and share them—human art. If we unquestioningly accept AI’s confident answers to our online inquiries and indulge its attempts at creating art with our attention, might AI ultimately undermine our imagination and intellect? "Directing one’s attention,” writes Pollan,” is action of a special kind.” Indeed, to deliberately control our attention is an evolutionary ingredient necessary to developing consciousness, and one we are in danger of losing our purchase on in an era of constant distractions and smartphone addiction engineered by big tech.  

I believe Pollan offers a remedy. I believe focusing on consciousness makes us more conscious. In the larger scheme of things, I hope it helps readers get excited about getting offline and experiencing the world around us. Pollan reminds us that our brains evolved to keep our bodies alive, not vice-versa. And while there are people who not only wish for an AI singularity that would discard the need for a body and make our consciousness—and perhaps our illusory sense of self—immortal, A World Appears reminds us that our brains are but one piece of our bodies, and may not even be where our consciousness stems from. What if placing consciousness within the brain is as big a mistake as placing the Earth at the center of our Sun’s orbit?  

At a time when science is being questioned more perniciously in our society, having a science writer like Pollan examine one of the fundamental questions of our existence with a healthy dose of scientific skepticism and a search for something resembling truth is a blessing. I am not sure I know more about consciousness itself than I did when I started the book, but I did learn a lot about its field of study, and I came away with a strong belief that we should not turn solely to science to answer this question, and that we can’t let those with a profit motive write the future of this story. I came to think we may be using the wrong tools, or perhaps the wrong metaphors, to tackle the question. Using the computer as a metaphor for the brain seems wholly inadequate after what I’ve learned in A World Appears. A metaphor as simple and common as “our train of thought” over one like “stream of consciousness” may not affect our daily lives too much, but if that metaphor is chosen for how our minds work when we build large language models and try to instill consciousness in machines, it could affect us all—even if it doesn’t ultimately work.  

In the book’s opening pages, Pollan cites The International Dictionary of Psychology, first published in 1989, which entry on consciousness read: “A fascinating but elusive phenomenon: it is impossible to specify what it is, what it does, or why it evolved. Nothing worth reading has been written on it.” If nothing else, with A World Appears, Pollan has rendered that final assertion false.