Learning Through Wonder

The hummingbird looked dead. It was hanging upside down from a thin twig at the edge of a cloud forest clearing, just after first light, feet locked around the branch, eyes closed, body angled at nearly ninety degrees to the ground in a way no living bird holds itself. A visitor on the walk touched my arm and pointed, worried, and I understood the worry immediately: nothing about the posture read as sleep. I told her to wait and watch the chest instead of the head. After a while, a faint tremor started along the flank, so slow at first that it could have been wind. The tremor built, the bird's whole body began to shiver visibly, and perhaps eight minutes later it opened its eyes, released the twig, and was gone into the mist as though the last ten minutes hadn't happened.

What she had watched, and what I had shown her without yet explaining, was a hummingbird coming out of nightly torpor, a controlled, drastic shutdown of its own metabolism that it enters every single cold night of its life and that it must, each morning, actively reverse before it can fly. I could have left it there. The moment had its own power without a single fact attached to it: a small thing that looked dead, that was not dead, that came back. Plenty of naturalist tradition holds that this is where a guide should stop, that adding physiology to a moment like this is a kind of vandalism, trading a small miracle for a mechanism and coming out poorer for the exchange. This essay is going to argue the opposite, but it owes that tradition a fair hearing first, because the concern behind it is neither foolish nor new.

The worry has a long pedigree. In the early nineteenth century, the poet John Keats complained that Isaac Newton had ruined the rainbow by explaining it, reducing a marvel of the sky to a mere fact about the refraction and dispersion of light through water droplets, and stripping it, in the process, of the poetry that had made it worth looking at. The complaint has outlived the specific target. Versions of it surface every time someone learns that a sunset's colors come from atmospheric scattering, that a firefly's glow is a chemical reaction with a name, that the hummingbird's dead-looking stillness is a scheduled physiological event rather than a small daily resurrection. The underlying claim is always the same: understanding a mechanism replaces the felt experience of a thing with an account of it, and the account, however accurate, is thinner than what it replaces.

There is a reason this complaint keeps resurfacing across two centuries rather than being settled once and for all, and it has to do with what each side takes the word understanding to mean. Keats and his heirs tend to picture understanding as a kind of translation: the rainbow's color becomes a wavelength, the wavelength becomes a number, and the number, however precise, seems to have quietly replaced the experience of standing under an arc of color rather than added anything to it. On this picture, every fact learned is a small subtraction from the original phenomenon, a conversion of something felt into something merely known, and a sufficiently thorough scientific education would, by this logic, eventually convert the entire visible world into a ledger of such conversions, leaving less and less to actually feel. The argument this essay is making depends on rejecting that picture of what a fact does to an experience, not on denying that the worry is coherent.

It is worth taking this claim seriously rather than dismissing it, because it identifies something real: explanation delivered badly can flatten an experience. A fact recited too quickly, before an observer has had time to feel their own surprise, can foreclose the noticing it was meant to deepen, in exactly the way an earlier chapter in this collection warned against when discussing how interpretation installs frames in someone else's attention. If I had opened that morning by announcing that the bird was in torpor and explaining metabolic rate before my visitor had a chance to worry, watch, and wonder on her own, something would genuinely have been lost, not because the fact was wrong but because it arrived at the wrong moment, doing the noticing on her behalf rather than deepening noticing she had already started to do herself. The Keatsian complaint is not baseless. It is aimed, correctly, at a real failure mode. It is simply aimed at the wrong culprit, mistaking bad timing and clumsy delivery for a problem with explanation itself.

The strongest reply to Keats came, more than a century and a half later, from a physicist rather than a poet. Richard Feynman, in a recorded conversation about a disagreement with an artist friend over the beauty of a flower, made a point worth restating in full because of how directly it bears on the hummingbird. His friend had argued that a scientist's habit of taking a flower apart, analytically, left it duller than the artist's undivided appreciation of its surface beauty. Feynman's response was that his own knowledge of the flower did not compete with that surface beauty but sat alongside it and beneath it: he could still see the same color and shape his friend saw, and could additionally imagine the cellular machinery producing that color, the evolutionary logic connecting the flower's hue to the vision of the insects it needed to attract, and the further, still-open questions that logic raised about whether something like aesthetic sensitivity might exist in those insects too. Nothing about the deeper layer replaced the surface one. It sat underneath it, and on top of it, and around it, multiplying the places from which the flower could be found beautiful rather than substituting one vantage point for another. Carl Sagan later made a related point more compactly, noting that a little astronomical knowledge does nothing to spoil the emotional pull of a sunset. Richard Dawkins built an entire book around defending the same claim against its harsher critics, arguing that the poetic imagination and the scientific imagination are not natural enemies competing for the same territory in a person's mind, but two different instruments capable of being played at once.

The hummingbird gives this abstract disagreement a concrete test case, because the physiology behind its apparent death is unusually extreme even by the standards of vertebrate biology. Hovering flight is among the most energetically expensive forms of locomotion known, and the muscle driving it demands a metabolic rate that, scaled to body mass, is roughly seventy times that of a resting human. A hummingbird's heart, sustaining this, can beat over a thousand times a minute in flight, occasionally approaching twelve hundred, a rate that pushes against what a muscle of that size can physically achieve. That metabolic furnace has to be fed almost continuously during the day, because a bird this small cannot carry enough fuel reserve to sustain it for long. The problem this creates at night is not decorative. A hummingbird holding a normal body temperature through a cold cloud-forest night would, in many cases, simply burn through its entire fat reserve before dawn and starve in its sleep. Torpor exists to prevent exactly that outcome: an actively regulated collapse of body temperature, sometimes by ten to thirty degrees Celsius, and of metabolic rate, by as much as ninety-five percent, cutting the night's energy bill down to something the bird's tiny reserves can actually cover.

The word regulated matters more than it might first appear to, and it is worth pausing on why. Early physiological studies of torpid hummingbirds found that a bird in torpor does not simply go cold and passive, tracking whatever the surrounding air happens to be; it defends a specific, actively maintained body temperature, generally somewhere between eighteen and twenty degrees Celsius regardless of how much colder the air around it gets, and its oxygen consumption rises again if the ambient temperature drops far enough to threaten even that lowered target. This is not a bird that has simply failed to keep itself warm. It is a bird running a second, deliberately lower thermostat setting, one it can enter and exit on a schedule tied to its own fat reserves, arousing itself each morning through a burst of shivering thermogenesis that can carry its core temperature from near-ambient back up to flight-ready in well under an hour. More recent field studies tracking wild hummingbirds across a range of climates have found that species and even individuals vary considerably in how they deploy this tool, some entering only shallow torpor on mild nights and reserving the deepest, most dramatic version for the coldest ones, suggesting the whole system is tuned with a precision current research is still working out rather than fixed and identical across the group. None of this was visible to my visitor watching the bird's chest for the first tremor of the morning. All of it was already true of the bird she was watching, whether or not either of us knew it.

Put in terms a human body offers some rough comparison for, a resting adult's heart beats somewhere around seventy times a minute, and a person in a medically induced hypothermic state, kept alive under close monitoring for surgery, might have that rate slowed to a third or a quarter of normal at most, with body temperature lowered by a handful of degrees under tightly controlled conditions built around elaborate equipment. The hummingbird does something considerably more extreme than that, unassisted, every single cold night, with no equipment beyond its own nervous system, and reverses the whole procedure again by breakfast. The comparison is not offered to suggest the two processes are biologically equivalent; the mechanisms differ in nearly every particular. It is offered because human physiology is the frame most people already carry into an encounter with a hummingbird, and the scale of the difference between what a human body can be made to do under intensive medical care and what this bird's body does on its own, nightly, without assistance, is closer to what the research on awe means by vastness than the bare fact of torpor's existence ever manages to convey by itself.

This is where the naive version of the morning's wonder and the informed version genuinely diverge, and the divergence favors the informed version by a wide margin. The naive wonder available to an observer who stops at what looked dead, then wasn't is real, but it is also thin: surprise at a single, isolated event, quickly available and about as quickly exhausted, since a bird waking up is not, on its own, an unusual enough occurrence to sustain much reflection once the initial startle passes. It is the kind of wonder that produces a good photograph and a story told once at dinner, and then, having nowhere further to go, quietly recedes into the general category of that time I saw a weird bird. The informed wonder available to an observer who goes on to learn what torpor actually costs and enables is built from a different kind of material entirely. It includes the fact that this bird runs, every single night of its life, one of the most extreme controlled physiological shutdowns known in the vertebrate world, on a fixed energy budget that can be calculated in fractions of a calorie, and that a shortfall of even one evening's nectar intake can be the difference between an arousal at dawn and a death overnight. The bird did not do anything unusual on the morning I watched it with my visitor. It did what it does on ordinary nights, and the ordinariness, once understood, is more remarkable than the apparent miracle it briefly resembled. Explanation did not replace the wonder with a mechanism. It replaced a small, momentary surprise with a much larger, sturdier one, because the actual scale of what the bird's body accomplishes nightly turns out to exceed what the naive impression of it ever suggested.

There is a psychological account of why this happens, and it is worth bringing in because it moves the argument from intuition to evidence. Research on the emotion of awe has proposed that it reliably involves two components: a perceived vastness in whatever is being encountered, and what researchers, borrowing a term from developmental psychology, call a need for accommodation, meaning an experience large or strange enough that a person's existing mental categories cannot simply absorb it without being restructured to fit. A bird that looks dead and then isn't supplies a small amount of both, enough for a moment of genuine surprise. Learning the actual numbers, the heart rate collapsing from over a thousand beats a minute to fifty, the body temperature dropping by a third of its normal value and being actively held there rather than merely allowed to fall, supplies considerably more of both, because those figures are harder to fit comfortably into an ordinary, pre-existing sense of what a small, warm-blooded animal's body can do. On this account, explanation is not awe's opponent. It is frequently awe's actual delivery mechanism, since the felt sense of vastness this research describes tends to come from encountering a scale or a strangeness that untrained observation, on its own, usually cannot supply.

A separate body of research on curiosity adds a second piece to the same argument, from an unexpected angle. One influential account holds that curiosity arises specifically from the perception of a gap between what a person already knows and what they sense they could know, a gap that requires some existing knowledge to be perceived as a gap at all; pure ignorance of a subject, on this account, does not reliably generate the sharpest curiosity, because a person with no framework yet has no way to notice that anything specific is missing. A visitor who has just learned that hummingbird torpor involves an actively defended body temperature is, by this logic, positioned to feel a sharper curiosity about further specific unknowns, how this particular species' torpor threshold compares to a relative living at a different elevation, or what actually triggers a given individual's decision to enter shallow rather than deep torpor on a given night, than a visitor who knows nothing yet about torpor at all and therefore has no place from which to ask a targeted question. Informed wonder, on this reading, does not use itself up on delivery. It tends to generate the conditions for the next round of curiosity, which is closer to the opposite of the exhaustion the disenchantment argument predicts.

None of this licenses explaining everything, immediately, regardless of context, and the earlier caution about timing deserves to be restated rather than quietly dropped now that the main argument has been made. The two research threads just described both depend on a person already having encountered the phenomenon directly, with enough unhurried attention to register genuine surprise, before any mechanism gets layered on top of it. Awe requires an experience large enough to strain existing categories, and a fact delivered before anyone has had the chance to form an initial, naive impression has nothing yet to strain against, no felt gap for the explanation to fill. This suggests a specific, practical discipline rather than a blanket license to explain on sight: let the observation happen first, in something close to silence, and only once genuine noticing and surprise have had room to form should the mechanism be offered, at which point it tends to compound the wonder already underway rather than substitute for it. Getting this sequence backward, leading with the physiology before the visitor has been allowed her own moment of worry and watching, would have produced something closer to what Keats feared. Getting it in the right order produced, by her own report walking back down the trail, something considerably larger than what either the naive sighting or the explanation could have supplied alone.

This sequencing requirement connects the argument here to a discipline argued for elsewhere in this collection under a different name. Treating identification as an entry point rather than an endpoint, and treating an interpreter's task as installing a perceptual frame rather than delivering a fact, both presuppose that there is a right order in which information should arrive, and both warn against the same failure: information supplied before the observer has done any of their own noticing, which forecloses the noticing rather than rewarding it. What this essay adds to that earlier argument is a claim about what the noticing is eventually for. It is not merely a courtesy extended to the visitor's own perceptual work, valuable for its own sake and then handed off to explanation once complete. It is the raw material awe requires in order to register a subsequent fact as vast rather than merely additional, the gap curiosity requires in order to register a subsequent fact as answering something rather than arriving unprompted. Skip the noticing, and the explanation that follows has nothing left to compound. It becomes trivia, correctly stated and largely inert, precisely the flattened outcome Keats was afraid of, produced not by the presence of explanation but by its arrival before anything was there yet to deepen.

The failure mode is not hypothetical, and it is worth describing concretely rather than leaving it as an abstract warning, because the difference between the two deliveries is almost entirely a matter of pacing rather than content. On a different morning, with a different group, I made the mistake this essay has been warning against. A visitor spotted the same species of hummingbird perched motionless on a similar cold morning and asked, lightly, whether it was sick. I answered immediately and completely, torpor, metabolic rate, the seventy-fold multiple, the whole explanation delivered in under a minute before he had done anything but glance at the bird for a few seconds. He nodded, said that was interesting, and moved on down the trail without breaking stride. Nothing in what I told him was wrong, and he had, technically, learned exactly what my other visitor had learned watching the chest tremor build for eight minutes. But he had learned it the way a person learns a fact from a placard, arriving with no felt gap for it to fill and leaving with nothing to compound, because there had been no time for any noticing of his own to happen first. The information was identical in both cases. The experience was not, and the difference had nothing to do with how much was explained.

I do not know whether this compounding effect has a ceiling. Every naturalist who spends decades with the same handful of extraordinary phenomena, the same torpid hummingbirds, the same swarm-following antbirds, the same quetzals swallowing fruit whole, eventually reaches a point where very few new facts about them still strain against existing categories the way the first ones did, because the categories have themselves grown large enough to absorb almost anything that arrives. A newcomer's first torpid hummingbird can produce genuine vastness with a comparatively modest fact, since almost anything about the bird's physiology will exceed a beginner's prior sense of what small warm-blooded animals do at night. A researcher who has spent a career measuring exactly this phenomenon, publishing the numbers this essay has borrowed secondhand, presumably experiences something quieter by now watching the same bird wake up, not because the physiology has stopped being extraordinary but because their own mental categories have grown large enough to hold it without strain. If awe genuinely depends on a gap between what is encountered and what a mind can already accommodate, and if a career of informed attention steadily closes that gap phenomenon by phenomenon, something has to give as the years accumulate, either a narrowing of what still counts as vast, or a compensating deepening in some other register this essay has not yet named.

What is left, decades in, for the naturalist who has genuinely come to understand almost everything they still choose to look at closely?

Recommended scientific references

Keltner, D., & Haidt, J. (2003). Approaching awe, a moral, spiritual, and aesthetic emotion. Cognition and Emotion, 17(2), 297-314.
Loewenstein, G. (1994). The psychology of curiosity: A review and reinterpretation. Psychological Bulletin, 116(1), 75-98.
Hainsworth, F. R., & Wolf, L. L. (1970). Regulation of oxygen consumption and body temperature during torpor in a hummingbird, Eulampis jugularis. Science, 168(3929), 368-369.
Lasiewski, R. C. (1963). Oxygen consumption of torpid, resting, active, and flying hummingbirds. Physiological Zoology, 36(2), 122-140.
Shankar, A., Schroeder, R. J., Wethington, S. M., Graham, C. H., & Powers, D. R. (2020). Hummingbird torpor in context: duration, more than temperature, is the key to nighttime energy savings. Journal of Avian Biology, 51(6).

Author's note

The scientific content of this essay, including the extreme mass-specific metabolic demands of hummingbird flight, the physiological details of regulated torpor and its documented energy savings, and the psychological research on awe and curiosity, reflects established, cited findings. Where current research is still actively refining a finding, as with species and individual variation in torpor use, this essay has tried to say so rather than present a settled picture.

What is not an established finding, and belongs to The Neotropic Philosophy of Interpretation as this collection's own proposal, is the essay's central claim: that scientific understanding deepens rather than diminishes the experience of wonder, and that the highest form of wonder available to a naturalist is specifically an informed one. The psychological research cited above shows that awe depends on encountering something that exceeds existing mental categories, which is compatible with this essay's argument without strictly requiring it; a person could accept every finding cited here and still reasonably hold that some experiences are better left unexplained, a position this essay has tried to take seriously rather than dismiss, even while arguing against it as a general rule.

Learning Through Wonder

​

Recommended scientific references

Suggested further reading

Author's note