What do we talk about when we talk about dreams? Martin Luther King had one of the most famous — a waking vision of justice, projected onto a future he wouldn’t live to see. We say “I dreamt of you” and the ambiguous sentence trembles between confession and accusation. We call someone “a dream” and collapse an entire person into an idealisation. And then there are the other dreams — the ones that visit us without permission while we sleep, the ones we can’t control, the ones we forget by breakfast and occasionally remember decades later in the middle of a meeting, struck by a feeling we can’t name.

The word carries all of this: aspiration and irrationality, prophecy and nonsense, desire and dread. When we share a dream, we rarely specify which kind. Maybe we don’t need to. Maybe all dreams — sleeping and waking — share the same root: a restless attempt to give shape to what hasn’t happened yet, to find meaning in what resists it.

i. a meaning-seeking animal in the dark

We are meaning-seeking creatures. If that sounds obvious, consider how deep it runs: we search for significance in everything — in patterns, in coincidences, in the arrangement of stars, in the colour of a stranger’s shirt, in the silence after someone says something terrible. As I’ve written before, “‘nothing’ usually means ‘something,’(…). It’s also easy to see ‘nothing’ as the opposite of ‘everything’, and that adds little to our understanding. After all, no-thing can be anything. Another use for ‘nothing’ is when we want to categorize effort or an event as having little importance: ‘It was nothing,’ we say. In any case, it’s wonderful how we can generally understand each other regarding ‘nothing,’ with little effort.” We are beings that cannot stop searching for sense, even in experiences that might have none. Especially in those.

And then we sleep. Every night, we surrender consciousness — the very faculty that allows us to seek meaning — and enter a state that, without dreams, would approximate the one experience where meaning becomes permanently unreachable: death. Sleep without dreams would be too close to dying for a meaning-seeking animal to bear. Not because the body would fail, but because the search would stop.

This is the thought I keep circling. If we are built for meaning, then dreams might be what keeps that search alive when we’re in no condition to pursue it. A way to avoid the proximity of an experience too close to the one we fear most. Not entertainment. Something closer to survival.

Death is frightening not because we know what it is, but because we can’t experience it. It is, paradoxically, an experience of the living — we think about death, fear death, prepare for death, but death itself remains the one thing we cannot encounter and report back. Dreams, in this reading, are our nightly proof that we haven’t crossed that threshold. The search for meaning continues. We’re still here.

If dreams function as a training ground — a kind of gymnasium for our capacity to find meaning while consciousness is away — then they also serve as the unconscious, and therefore purer, less controlled expression of our intuitive and inventive faculties. Dreams are imagination uncensored. They are where we give shape to what has no shape yet, including the future, which is the largest meaningless territory we face every day.

Intuition, imagination, forecasting, projecting, planning, and dreaming: these are all ways we deal with the ultimate existential threat — not death itself, but the absence of meaning. The unknown. The future that hasn’t been given a story yet.

ii. how we’ve tamed the wild

Every major intellectual tradition has taken a crack at dreams, and every one has domesticated them in its own way.

Freud treated dreams as encrypted telegrams from the unconscious. There’s a manifest content — what you see — and a latent content — what it “really” means. The dreamer is hiding something from themselves, usually involving desire or fear, and the analyst’s task is to decode the cipher. Dreams, in this framework, are problems to be solved. A woman dreams of being chased by a snake, and we are told this represents her fear of sexual intimacy. Jung broadened the vocabulary — collective archetypes, universal symbols — but kept the same architecture: dreams carry a hidden meaning that requires a trained interpreter to extract.

Psychoanalysis, at its best, takes dreams seriously as expressions of unconscious desires and fears, as a way to approach what remains unsolved while we’re awake. Its limitation is that it assumes the meaning is already there, encrypted, waiting to be found. The dreamer is an obscure text; the analyst, a supposedly experienced and capable reader. The possibility that the dream might resist reading — might be valuable precisely because it resists reading — is hard for this tradition to accommodate. Wilfred Bion came closest, with his discipline of approaching experience “without memory, without desire, without understanding.” But that’s a stance most psychoanalytic practice struggles to sustain. ¹

The existential-phenomenological tradition offers a different temperament. Medard Boss, who studied dreams more systematically than any other existentialist, offers a clear provocation for the psychoanalysis enthusiast: a dream about pigs is about pigs. Not about repressed desires, not about random neurons firing, not about evolutionary rehearsal. The dream reveals the dreamer’s existential possibilities as they are, without the need for translation. Boss didn’t believe in the unconscious as a storage room for repressed material. He saw dreams as another way existence discloses itself — different from waking life in texture but not in ontological status. Dreams, in this view, are a “bearing” toward the world, toward others, toward life — unlived possibilities of being oneself that haven’t yet found their form in daylight.

This tradition comes closest to tolerating mystery. But it can also become so respectful of the phenomenon that it has little to say about it beyond describing its contours.

And then the neuroscientists, arriving with different tools and a different temperament entirely. Hobson and McCarley’s activation-synthesis hypothesis, published in 1977, proposed the bluntest possible answer: dreams don’t mean anything. They’re the cortex trying to make narrative sense of random electrical signals generated by the brainstem during REM sleep. We dream, and then we wake up and invent the story afterward. The signals come from the brainstem; the meaning — if you can call it that — is retrofitted by a waking mind that can’t tolerate the absence of narrative.

Antti Revonsuo’s threat simulation theory offered a more generous reading: dreams evolved as a rehearsal space for dangerous scenarios — an ancient biological defence mechanism that enhances our capacity for threat perception and avoidance. The memory consolidation theory treats dreams as a byproduct of the brain sorting and filing the day’s experiences. Erik Hoel’s overfitted brain hypothesis argues that dreams introduce noise into our neural networks to prevent cognitive rigidity. ² And Robert Stickgold and Antonio Zadra’s NEXTUP model (Network Exploration to Understand Possibilities) suggests dreams help the brain explore weak associations across its memory network, enhancing flexibility and creative integration.

David Eagleman and Don Vaughn proposed what might be the most materialist theory of all: the Defensive Activation Theory. ³ Their argument runs as follows. Brain regions maintain their territory through continuous activity. If activity slows or stops — as happens with blindness — neighbouring regions invade. And the speed of this takeover is alarming: measurable within an hour. Now, of all our senses, only vision shuts down cyclically, because of the rotation of the planet into darkness. Touch, hearing, smell — they keep working while we sleep. A bug crawls on your skin, your baby cries, smoke fills the room: you’ll notice. But there’s nothing to see. So the visual cortex, which occupies substantial neural real estate, goes dark every night and becomes vulnerable to annexation. Eagleman and Vaughn propose that REM sleep exists to pump activity into the occipital cortex — to keep the lights on, so to speak — and that dreams are the experiential byproduct of this defensive operation. Dreams are primarily visual, they argue, precisely because vision is the only sense that needs defending against the dark.

It’s an ingenious theory. It’s also met with scepticism. Critics have noted that the evidence is largely circumstantial — correlations between REM sleep percentages and developmental speed across primate species, rather than direct proof that harmful neuroplasticity occurs within a single night’s sleep. No one has demonstrated that the rapid cortical changes Eagleman cites are strong enough, in the span of a few hours, to genuinely threaten the visual system of a healthy brain. The theory remains, for now, a compelling hypothesis rather than established science. Which, given the subject, feels appropriate: even neuroscientists can’t reach consensus on what dreams are for.

But here’s what interests me about Eagleman’s theory, regardless of whether the mechanism holds. Even in the most reductive account — dreams as a cortical screensaver, a territorial defence strategy — the brain’s method of self-protection is imagination. The organ that supposedly just needs to keep some neurons firing ends up producing narrative, emotion, absurdity, beauty, terror. If the goal were merely to maintain neural territory, random visual noise would suffice. Static. Flickering patterns. Instead, the brain generates rich, filmic, emotionally charged sequences. It creates Michel standing in a thunderstorm, explaining why lightning can’t talk to thunder (more on this later). The mechanism may be maintenance; the output is something else entirely.

Neuroscience, taken as a whole, keeps oscillating between “dreams mean nothing” and “dreams are essential for survival” — which is itself a version of the tension I want to explore.

Each tradition reveals a particular anxiety about the unknowable. Freud can’t bear the idea that the unconscious isn’t speaking in code — there must be a hidden meaning. Neuroscience can’t bear the idea that something without measurable function might matter — dreams must do something useful, or else they’re noise. The existentialists come closest to tolerating mystery but still need to frame it philosophically.

Each school performs its own version of the compulsion I want to diagnose: the need to explain, to domesticate, to make dreams safe for rational consumption and sharing. The inherent challenge with dreams does not stem from their subjective nature. It stems from our relentless desire to strip away that subjectivity — our eagerness to find the theory or the method that will crack them open and hand us a clean, translatable meaning.

iii. before freud, there were the wendat

But there are older traditions than any of these, and they unsettle all our categories.

In David Graeber and David Wengrow’s The Dawn of Everything, they describe a concept from the Wendat (Huron) people of North America: Ondinnonk — a secret desire of the soul, revealed through dreams. The Wendat believed that dreams were the soul’s native language, and that when these desires went unacknowledged, the soul became angry, causing illness and even death. ⁴

What makes this remarkable is not the parallel with psychoanalysis — though it exists, and was documented by Jesuit missionaries some 250 years before Freud published The Interpretation of Dreams. Father Ragueneau, writing in the late 1640s, described dream practices that the missionaries considered demonic and tried to refute in order to bring their interlocutors to the truth of Holy Scripture. What’s remarkable is the difference.

The Wendat practice was collective, not clinical. “Dream-guessing” was carried out by groups, and realising the desires of the dreamer — either literally or symbolically — could mobilise an entire community. Ragueneau reported that winter months in a Wendat town were largely devoted to organising collective feasts and dramas to make important dreams come true. Dreams belonged to the community, not to the individual on a couch.

There were significant differences between Freudian psychoanalysis and Wendat practice, as Graeber and Wengrow note. But the point isn’t the parallel. The point is that an intellectual tradition — sophisticated, internally consistent, and based on generations of accumulated observation — existed in full operation centuries before ours, and we are taught to treat it as folklore. We dismiss such traditions as primitive or mythic while our own relationship with dreams lurches between clinical reduction and scientific dismissal.

Consider what the Wendat had that we’ve lost: a way to take dreams seriously without needing to explain them away. They didn’t need a theory of the unconscious or a model of REM sleep. They needed a community willing to listen and to act.

iv. the kitchen in sintra

In June 2023, at The___Dream — the House of Beautiful Business festival in Sintra — I facilitated a session called “The Conversations of Dreams: an inquiry into the irrational.” ⁵

We were in the kitchen of the estate. Not a stage, not a conference room, but a kitchen — which is where the best conversations happen. ⁶ We were exploring dreams, irrationality, the porous border between what we call sanity and what we call madness. The conversation flowed. People were generous with their vulnerability. Nobody was defending positions — they were discovering them, together, in real time. It was a conversation onírica — a dream-like conversation where traditional logic could be suspended, where the categories of “madness” and “sanity” could be questioned without anyone feeling the need to resolve the questioning into an answer.

Then I made an unscripted, improvised joke about performance reviews. Can’t remember why or what the context was. But I was swept away by the immediate shift of the quality of the conversation. Two participants started arguing. Not exploring — arguing. Defending positions, marshalling past experience, speaking past each other. The room contracted. Something had changed, and everyone felt it, especially after I made that shift manifest.

What had entered the room wasn’t just a different topic. It was a different language. Dreams don’t have a technical vocabulary; performance reviews do. And that vocabulary brought with it an entire architecture of certainty: evaluation, measurement, right answers, hierarchical judgment. The register shifted from imaginary to virtual — from the generative, unpredictable space of genuine exploration to the predetermined, rule-based territory of corporate speak.

That kitchen moment was a collective insight. We didn’t need to theorise about it. Everyone in the room had just lived the argument: that we can actually converse when we’re in the territory of the irrational, the ambiguous, the unexplained — and that conversation collapses the moment we enter the language of certainty and control. We stop turning toward each other and start turning against each other. ⁷

This, incidentally, is why business abhors the unknown. And why dreams are not welcome in meeting rooms. The corporate world runs on the premise that uncertainty is a problem to be solved, that ambiguity signals incompetence, that not knowing is a failure rather than a starting condition. Dreams — messy, illogical, resistant to KPIs — represent everything that world works to eliminate. ⁸

v. love, death, and curiosity: a different approach

If traditional logic collapses when applied to dreams, what do we use instead?

I’ve been working with a triadic framework drawn from psychoanalytic thinking — specifically from Wilfred Bion’s reformulation of the drives and Paul Stokoe’s development of curiosity as a third fundamental force alongside love and hate. ⁹

Instead of asking “what does this dream mean?” — which already presupposes a hidden code — try:

• What do you secretly want to create? (love/life)
• What do you secretly want to destroy? (hate/death)
• What explanations are you searching for, and haven’t found yet? (curiosity)

These don’t pretend to be interpretation techniques. They’re orientations. They respect the emotional core of dreams rather than subjecting them to logical dissection. They also acknowledge something that the neuroscientific tradition tends to overlook and that the psychoanalytic tradition often buries in theory: that dreams are felt before they’re thought. They arrive in the register of love, hate, and wonder — not in the register of analysis.

This framework isn’t necessarily contradictory to any of the traditions I’ve sketched. If dreams keep our meaning-seeking capacity alive during unconsciousness, then of course they draw on desires and fears — Freud was right about the raw material. Of course they involve neural activation and memory processing — the neuroscientists are right about the mechanism. Of course they reveal our mode of being-in-the-world — the existentialists are right about the phenomenology. But each captures a dimension. Each misses the whole. And the whole might be simpler and wilder than any of them suspects.

What follows is an attempt to sit with a dream rather than solve one.

vi. lightning and thunder

During a severe thunderstorm in Tulum, on my honeymoon, sleeping in a cabin while the sky fell apart outside, I dreamed that Michel was saying: “Lightning doesn’t have time to talk to thunder; they don’t have time to exchange words.”

Michel was one of my father’s closest friends. My brother’s godfather. One of the most intelligent people I’ve known. He was the person who, when I was in my third year at university, when I had to choose between clinical, social/organizational or educational psychology, told me: “Go into clinical psychology — besides being what you enjoy most, even if you end up working in another area, understanding how people function will always be an advantage.” He shaped the trajectory that led me to the consulting room, and from there to everything else.

Let me do what each tradition would do with this dream.

A Freudian might ask: what does Michel represent? What desire is being disguised by this image of impossible communication? The lightning and thunder as symbols of some repressed conflict — sensation and meaning, forever out of sync.

A neuroscientist would say: there was a real thunderstorm. Your brain was incorporating external stimuli during REM sleep. The auditory cortex was processing the thunder; the visual cortex, the lightning. The prefrontal cortex, being offline, couldn’t flag the absurdity of a deceased family friend narrating atmospheric physics. It means nothing beyond that.

An existentialist would ask: what mode of being-in-the-world does this dream reveal? What does the impossibility of conversation between lightning and thunder illuminate about the dreamer’s existence?

My approach: sit with it.

The dream speaks about temporal gaps between experience and expression. Lightning arrives and is already gone by the time the thunder follows. Two manifestations of the same event that can never meet, never exchange words. One is light, the other sound. They travel at different speeds. By the time one arrives, the other has already left.

It’s a dream about what I keep writing about: the difficulty of certain conversations, the weight of what goes unsaid, the physics of interrupted exchange. The man who sent me toward a life of listening to people is telling me, in a storm, on the first nights of a marriage, that some things that belong together will never have time to speak to each other. I think of Nietzsche while writing this: when marrying, he wrote, one should ask whether one will enjoy conversing with this person into old age, because everything else in marriage is transitory. ¹⁰

I don’t need to solve it. The dream isn’t a problem. Its meaning lives in its resonance, not in its translation.

And there’s something else: the dream emerged during a real storm. Reality and dream were having the same conversation simultaneously — the cabin shook with thunder while the dreaming mind heard Michel’s voice explaining why lightning and thunder can’t converse. Which is itself a demonstration of the porosity between waking and dreaming life that Boss insisted on, and that our neat theoretical categories work so hard to deny.

vii. who interprets the interpreter?

There’s another dimension I want to name. Whoever helps interpret a dream needs the same self-knowledge as the dreamer — perhaps more. The interpreter must know themselves well enough not to contaminate the dreams dreamed by someone else.

This sounds like a clinical principle, and it is — the psychoanalytic tradition formalised it as countertransference. But it extends well beyond the consulting room. Every time we listen to someone’s dream, we bring our own desires, fears, and frameworks to the encounter. We hear our own meanings before we hear theirs. Without awareness of this, interpretation becomes projection — we decode the dream into our own language and hand it back as though it were the dreamer’s truth.

This is why the Wendat practice of collective dream-guessing is so interesting. It distributed the interpretive burden across a community, reducing the risk that any single interpreter’s biases would dominate. Our clinical model, by contrast, concentrates that power in one person — the analyst, the therapist, the expert — and trusts their self-awareness to manage the contamination. Sometimes it works. Often it doesn’t.

The question is not whether we should interpret dreams. We will, because that’s what we do when faced with experiences that resist explanation. The question is whether we’re honest about the limits of our interpretation, and humble about what we bring to it. Bion’s discipline of approaching experience “without memory, without desire, without understanding” is precisely an attempt to address this contamination — to create conditions where the interpreter doesn’t impose their pre-existing frameworks on the material. It’s extraordinarily difficult to sustain.

viii. madness, health, and the dreams that keep us sane

Our notions of sanity and madness need revision before we can approach dreams without violence.

Georges Canguilhem argued that illness is not the opposite of health but part of it — being sick is a normal state. Émile Durkheim argued that crime is not only normal but necessary, useful, and central to societal evolution. Both were making the same structural move: refusing to treat what makes us uncomfortable as a deviation from some pure norm, and recognising it instead as integral to how the system works.

Dreams belong in this company. They are our nightly dose of madness — and that madness is central to our health. Dreams keep our irrational parts alive and functional. They keep us processing and expanding both individual and collective realms. To pathologise dreams, to treat them as symptoms requiring decoding, is to misunderstand their function as badly as treating fever as the disease rather than the healing response.

Fernando Pessoa captured the inversion with characteristic precision: “Whenever I arrived at a certainty, I remembered that those with the greatest certainties are lunatics.” If the sane live in constant mental instability — doubting, revising, tolerating ambiguity — and the mad are certain, then the relationship between certainty and health is the reverse of what we assume. ¹¹

If Pessoa is right, then dreams — which are uncertain, unstable, illogical, and ungovernable — might be our healthiest hours. Not in spite of their madness, but because of it. The dream doesn’t distinguish between good and bad, logical and absurd, possible and impossible. It operates with a freedom that our waking mind — so busy being sane, so busy being right, so busy performing certainty — can rarely access.

Living in a dream is madness. But what if we separated madness from its negative connotations? What if we saw it as imagination, as fantasy that hasn’t yet found its meaning — loucura que ainda não encontrou sentido? The wise, then, wouldn’t be those who never dream wildly, but those who manage to articulate rationally and clearly what began as something formless and uncontrolled. We could apply the same to dreams. Consciousness, in this view, wouldn’t be the executioner of dreams — the faculty that wakes us up and kills the vision — but an intermediary, a bridge toward a rationality that is deeper, wider, and freer than the one we settle for between nine and five.

ix. is a controlled dream still a dream?

This leads to a question I can’t stop asking: is the virtual substituting the imaginary?

A controlled dream, if such a thing exists, would be a contradiction in terms — a domesticated wildness, an administered irrationality. And yet this is increasingly what our technological environment offers: simulated experiences that mimic the surprise and freedom of genuine imagination while being entirely predetermined. Virtual reality, generative AI, algorithmically curated “discovery” — all promise the experience of the new while delivering the already-known in a different wrapper. ¹²

Companies want virtual innovation — safe, controlled, with measurable outcomes. They don’t want imaginary revolution — dangerous, unpredictable, impossible to capture in a slide deck. This is the same dynamic that played out in that kitchen in Sintra: the moment performance review language entered, the conversation shifted from imaginary to virtual. From genuine exploration to a simulation of dialogue.

Dreams are stubbornly imaginary. They can’t be programmed, predicted, or performance-managed. This is their value, and it’s why they make us nervous. We live in a culture that has become profoundly uncomfortable with any experience that can’t be immediately converted into something useful. Dreams resist this conversion. They are — in the most generous sense — useless. Which, if you’ve been reading this newsletter for any length of time, you know is not an insult. ¹³

x. toward a dreamlike rationality

Dreams are not problems awaiting solutions. They are not codes to crack, noise to filter, or data to process. They are madness — in the generous, non-pathological sense of that word: imagination that hasn’t yet found its explanation.

What would it mean to approach dreams — and, by extension, all that is irrational, ambiguous, and uncertain in our experience — without the compulsion to immediately explain it? Not as abandonment of reason, but as expansion of it. A deeper rationality, as I suggested in Sintra — one that includes rather than excludes what it can’t control.

This is, I suspect, what happened in that kitchen for a brief stretch of time. In the company of strangers who’d chosen to be honest, we achieved something like a dreamlike rationality — an exploration unconstrained by the need to be right, productive, or certain. And then performance reviews arrived, and the room woke up. What we lost in that moment wasn’t just a topic of conversation — it was a way of being together that most of us rarely experience outside of sleep.

The invitation is to stay asleep a little longer. Not as escape. As practice.

Because dreams, in the end, teach us something our waking lives keep trying to forget: that the search for meaning doesn’t require certainty. That understanding can emerge from confusion without ever fully resolving it. That the conversation between lightning and thunder — the one that can never quite happen, because they travel at different speeds — might be the truest kind of conversation there is.

This essay draws on notes and ideas first developed for “The Conversations of Dreams,” a session I facilitated at The___Dream, the House of Beautiful Business festival, in Sintra, Portugal, in June 2023. Some of those raw thoughts — written in Portuguese and English, in fragments, across notebooks and voice memos — have found their way here in reworked form. The piece by Manfred Kets de Vries that initially provoked this reflection treats dreams as “built-in counselling sessions” and “cryptic postcards from the unconscious.” I wanted to write something that treated them with more respect and less packaging.

1. I explored Bion’s radical discipline of suspension — and what happens when understanding arrives too early — in “Interlude: before understanding.” ↩︎
2. See Hoel, E. (2021). “The overfitted brain: Dreams evolved to assist generalization.” Patterns, 2(5), 100244. ↩︎
3. Eagleman, D.M. & Vaughn, D.A. (2021). “The Defensive Activation Theory: REM Sleep as a Mechanism to Prevent Takeover of the Visual Cortex.” Frontiers in Neuroscience, 15, 632853. ↩︎
4. Graeber, D. & Wengrow, D. (2021). The Dawn of Everything: A New History of Humanity. Allen Lane. ↩︎
5. The House of Beautiful Business is a global network that gathers leaders, thinkers, and artists. The___Dream took place at Quinta da Bella Vista, in the Sintra hills north of Lisbon. More at houseofbeautifulbusiness.com/thedream. ↩︎
6. This connects to what I’ve written about how the space of conversation shapes its quality. See “The Physics of Interrupted Conversations” and “On the Value of Conversation.” ↩︎
7. The Latin conversari means “to turn toward, together.” I’ve explored this etymology in several earlier pieces — see “The Physics of Interrupted Conversations” and “Silence, Conversation, and Intimacy.” ↩︎
8. I’ve explored how organisations perform certainty while drowning in doubt across several pieces, including “Corporate Hierarchy as the New Feudalism” and “Good Enough: Challenging the Tyranny of Excellence.” ↩︎
9. Bion’s conceptualisation of L (love), H (hate), and K (knowledge) as fundamental links in human experience appears throughout his work, particularly Learning from Experience (1962). Stokoe later developed curiosity as a distinct drive. I explored this framework and its practical implications in “The Curious Middle.” ↩︎
10. From Nietzsche’s Human, All Too Human (1878). The original is gendered — “this woman” — a limitation of its era that doesn’t diminish the observation about conversation as the substance of a shared life. ↩︎
11. As I developed more fully in “The Certainty Syndrome,” drawing on León Grinberg’s insight that psychopathology manifests as intolerance of ambiguity while mental health relates to a greater capacity for adaptation to complexity and uncertainty. ↩︎
12. I first sketched this distinction in notes for the Sintra session. The virtual is controlled simulation: safe, bounded, predictable. The imaginary is genuine creation: risky, open, ungovernable. Much of what passes for “innovation” in organisations is virtual — imagination with the edges sanded off. See also “The Revolution Will Not Be Technological.” ↩︎
13. See the inaugural essay that gave this project its name: “The Utility of Uselessness.” ↩︎