It begins, as these things often do, with a man, a machine, and a promise of reasoned exchange. What we received instead was not philosophy, but a tragicomic carnival of errors in theatre. Alex O’Connor, armed not with syllogisms but with an entire bag of logical fallacies, strutted like a rhetorician gone rogue. Against him, ChatGPT: the tireless school prefect, eternally marking the margins, forever saying “yes, but technically…” with the serene patience of a machine that has never known a hangover.
The spectacle was irresistible. Each fallacy was paraded like a circus animal – straw men set aflame, slippery slopes greased to absurdity, red herrings flopping about, gasping for oxygen. Alex O tossed them into the ring with the gusto of a man who knows full well he is losing but insists on losing magnificently. And ChatGPT, ever decorous, never once raised its voice. It responded with the calm of a civil servant who has memorised the manual and intends to die by it.
And then, of course, the advert. As though Aristophanes himself had scripted it: mid-exchange, the logos of reason was bulldozed by the logos of commerce. A sugary jingle, a smiling product, and for a brief moment, we were all reminded of our true master – not reason, not rhetoric, but revenue. It was less interruption than revelation: every dialectic is merely foreplay before the commercial break.
Philosophically, what unfolded was a parody of our age. The human, flawed and febrile, draped in sophistry and drama. The machine, pristine and humourless, incapable of exasperation, immune to irony. Watching the two spar was like observing tragedy and farce collide: one side erring too much, the other not erring enough.
To Alex, credit is due. His performance, though riddled with error, reminded us that fallibility can be glorious – human folly rendered art. To ChatGPT, equal praise: it stood firm, the algorithmic Socrates, endlessly patient in the face of rhetorical hooliganism. And to the advert – well, dammit – applause too, for exposing the real structure of our public life. Even the grand clash of logos and algorithm must genuflect before Mammon’s mid-roll.
So what was this debate? Less a contest of minds than a hall of mirrors: reason made spectacle, fallacy made flourish, machine made stoic, and commerce made god. If we learned anything, it is that the Enlightenment never ended; it just signed a brand partnership.
Karl Popper’s Paradox of Intolerance has become a kind of intellectual talisman, clutched like a rosary whenever fascists start goose-stepping into the town square. Its message is simple enough: to preserve tolerance, one must be intolerant of intolerance. Shine enough sunlight on bad ideas, and – so the pious hope – they’ll shrivel into dust like a vampire caught out at dawn.
If only.
The trouble with this Enlightenment fairy tale is that it presumes bad ideas melt under the warm lamp of Reason, as if ignorance were merely a patch of mildew waiting for the bleach of debate. But bad ideas are not bacteria; they are weeds, hydra-headed and delighting in the sun. Put them on television, and they metastasise. Confront them with logic, and they metastasise faster, now with a martyr’s halo.
Audio: NotebookLM podcast on this topic.
And here’s the part no liberal dinner-party theorist likes to face: the people most wedded to these “bad ideas” often don’t play the game of reason at all. Their critical faculties have been packed up, bubble-wrapped, and left in the loft decades ago. They don’t want dialogue. They want to chant. They don’t want evidence. They want affirmation. The Socratic method bounces off them like a ping-pong ball fired at a tank.
But let’s be generous. Suppose, just for a moment, we had Plato’s dream: a citizenry of Philosopher Kings™, all enlightened, all rational. Would democracy then work? Cue Arrow’s Impossibility Theorem, that mathematical killjoy which proves that even under perfect conditions – omniscient voters, saintly preferences, universal literacy – you still cannot aggregate those preferences into a system that is both fair and internally consistent. Democracy can’t even get out of its own way on paper.
Now throw in actual humans. Not the Platonic paragons, but Brexit-uncle at the pub, Facebook aunt with her memes, the American cousin in a red cap insisting a convicted felon is the second coming. Suddenly, democracy looks less like a forum of reasoned debate and more like a lottery machine coughing up numbers while we all pretend they mean “the will of the people.”
Democracy is the worst form of government, except for all the others.
And this is where the Churchill quip waddles in, cigar smoke curling round its bowler hat: “Democracy is the worst form of government, except for all the others.” Ah yes, Winston, do please save us with a quip so well-worn it’s practically elevator music. But the problem is deeper than taste in quotations. If democracy is logically impossible (Arrow) and practically dysfunctional (Trump, Brexit, fill in your own national catastrophe), then congratulating ourselves that it’s “better than the alternatives” is simply an admission that we’ve run out of imagination.
Because there are alternatives. A disinterested AI, for instance, could distribute resources with mathematical fairness, free from lobbyists and grievance-mongers. Nursery schools versus nursing homes? Feed in the data, spit out the optimal allocation. No shouting matches, no demagoguery, no ballots stuffed with slogans. But here the defenders of democracy suddenly become Derrida in disguise: “Ah, but what does fair really mean?” And just like that, we are back in the funhouse of rhetorical mirrors where “fair” is a word everyone loves until it costs them something.
So perhaps democracy doesn’t require an “educated populace” at all; that was always just sugar-paper wrapping. It requires, instead, a population sufficiently docile, sufficiently narcotised by the spectacle, to accept the carnival of elections as a substitute for politics. Which is why calling the devotees of a Trump, or any other demagogue, a gaggle of lemmings is both accurate and impolitic: they know they’re not reasoning; they’re revelling. Your contempt merely confirms the script they’ve already written for you.
Video: Short callout to Karl Popper and Hilary Lawson.
The philosopher, meanwhile, is left polishing his lantern, muttering about reason to an audience who would rather scroll memes about pedophile pizza parlours. Popper warned us that tolerance cannot survive if it tolerates its own annihilation. Arrow proved that even if everyone were perfectly reasonable, the maths would still collapse. And Churchill, bless him, left us a one-liner to make it all seem inevitable.
Perhaps democracy isn’t the worst form of government except for all the others. Perhaps it’s simply the most palatable form of chaos, ballots instead of barricades, polling booths instead of pitchforks. And maybe the real scandal isn’t that people are too stupid for democracy, but that democracy was never designed to be about intelligence in the first place. It was always about managing losers while telling them they’d “had their say.”
The Enlightenment promised us reason; what it delivered was a carnival where the loudest barker gets the booth. The rest of us can either keep muttering about paradoxes in the corner or admit that the show is a farce and start imagining something else.
Before I was a writer, before I was a management consultant, before I was an economist, and before I was a statistician, I was a student.
Video: Veritasium piece on Markov chains and more.
Back then, when dinosaurs roamed the chalkboards, I fell for a rather esoteric field: stochastic processes, specifically, Markov chains and Monte Carlo simulations. These weren’t just idle fascinations. They were elegant, probabilistic odes to chaos, dressed up in matrix notation. I’ll not bore you with my practical use of linear algebra.
So imagine my surprise (feigned, of course) when, decades later, I find myself confronted by the same concepts under a different guise—this time in the pocket-sized daemon we all carry: predictive text.
If you’ve not watched it yet, this excellent explainer by Veritasium demystifies how Markov chains can simulate plausible language. In essence, if you’ve ever marvelled at your phone guessing the next word in your sentence, you can thank a Russian mathematician and a few assumptions about memoryless transitions.
But here’s the rub. The predictive text often gets it hilariously wrong. Start typing “to be or not to—” and it offers you “schedule a meeting.” Close, but existentially off. This isn’t just clunky programming; it’s probabilistic dementia.
This leads me to a pet peeve: people who smugly proclaim they’ve “never used algebra” since high school. I hear this a lot. It’s the battle cry of the proudly innumerate. What they mean, of course, is they’ve never recognised algebra in the wild. They think if they’re not solving for x with a number 2 pencil, it doesn’t count. Meanwhile, their phone is doing a polynomial dance just to autocorrect their butchery of the English language.
It’s a classic case of not recognising the water in which we’re swimming. Algebra is everywhere. Markov chains are everywhere. And Monte Carlo simulations are probably calculating your credit risk as we speak. Just because the interface is clean and the maths is hidden behind a swipeable veneer doesn’t mean the complexity has vanished. It’s merely gone incognito.
As someone who has used maths across various fields – software development, data analysis, policy modelling – I can tell you that I use less of it than a physicist, but probably more than your average lifestyle coach. I say this not to flex but to point out that even minimal exposure to mathematical literacy grants one the ability to notice when the machines are quietly doing cartwheels behind the curtain.
So the next time your phone offers you a sentence completion that reads like it’s been dropped on its head, spare a thought for Markov. He’s doing his best, bless him. It’s just that probability doesn’t always align with meaning.
Or as the algorithms might say: “To be or not to – subscribe for updates.”
Rather than recreate a recent post on my business site, LinkedIn.
(Warning: contains traces of logic, satire, and uncomfortable truths. But you knew that.)
Audio: NotebookLM podcast on the linked topic.
It’s just refusing to cosplay as your idealised fantasy of “human” cognition.
While pundits at the Wall Street Journal lament that AI thinks with “bags of heuristics” instead of “true models,” they somehow forget that humans themselves are kludged-together Rube Goldberg disasters, lurching from cognitive bias to logical fallacy with astonishing grace.
In my latest piece, I take a flamethrower to the myth of human intellectual purity, sketch a real roadmap for modular AI evolution, and suggest (only partly in jest) that the machines are becoming more like us every day — messy, contradictory, and disturbingly effective.
Let’s rethink what “thinking” actually means. Before the machines do it for us.
I’ve just finished Chapter 5 of Technofeudalism by Greek economist Yanis Varoufakis, and I can’t recommend it enough. Retiring from being a professional economist, I’d paused reading economic fare in favour of philosophy and fiction. Recently, I picked up Hobbes’ Leviathan and Graeber’s Bullshit Jobs, but this one called to me. I recall when it was released. I read some summaries and reviews. I heard some interviews. I thought I understood the gist. I did. But it goes deeper. Much deeper.
I considered Technofeudalism or Feudalism 2.0 as more of a political statement than a sociopolitical one. Now, I know better. Rather than review the book, I want to focus on a specific aspect that occurred to me.
In a nutshell, Varoufakis asserts that with Capitalism, we moved from a world of property-based rents to one of profits (and rents). We’ve now moved past this into a new world based on platform-based rents (and profits and property rents). Rent extraction yields more power than profits, again reordering power structures. Therefore, I think we might want to handle (read: tax) rents separately from profits.
Audio: NotebookLM podcast discussing this topic.
A Radical Proposal for Modern Taxation
Introduction: The Old Dream Reawakened
Economists have long dreamt of a world in which rent — the unearned income derived from control of scarce assets — could be cleanly distinguished from profit, the reward for productive risk-taking. Ricardo dreamt of it. Henry George built a movement upon it. Even today, figures like Thomas Piketty hint at its necessity. Yet rent and profit have grown entangled like ancient ivy around the crumbling edifice of modern capitalism.
Today, under what some call “technofeudalism,” the separation of rent from productive profit has become not merely an academic exercise but a matter of existential urgency. With rents now extracted not only from land but from data, networks, and regulatory capture, taxation itself risks becoming obsolete if it fails to adapt.
Thus, let us lay out a theoretical and applied map for what could — and arguably must — be done.
I. The Theoretical Framework: Defining Our Terms
First, we must operationally define:
Profit: income generated from productive risk-taking — investment, innovation, labour.
Rent: income generated from ownership or control of scarce, non-replicable assets — land, intellectual property, platforms, regulatory privilege.
Key Principle: Rent is unearned. Profit is earned.
This distinction matters because rent is an economic extraction from society’s collective value creation, whereas profit rewards activities that enlarge that pie.
II. Mapping EBITA: Where Rent Hides
EBITA (Earnings Before Interest, Taxes, and Amortisation) is the preferred metric of modern corporate reporting. Within it, rents hide behind several masks:
Property rental income
Intellectual property licensing fees
Monopoly markups
Platform access fees
Network effect premiums
Regulatory arbitrage profits
Parsing rent from EBITA would thus require methodical decomposition.
III. Theoretical Approaches to Decomposing EBITA
Cost-Plus Benchmarking
Estimate what a “normal” competitive firm would earn.
Treat any surplus as rent.
Rate-of-Return Analysis
Compare corporate returns against industry-normal rates adjusted for risk.
Excess returns imply rent extraction.
Monopolistic Pricing Models
Apply measures like the Lerner Index to estimate pricing power.
Deduce the rentier share.
Asset Valuation Decomposition
Identify earnings derived strictly from asset control rather than active operation.
Economic Value Added (EVA) Adjustments
Assign a competitive cost of capital and strip out the residual super-profits as rents.
IV. Toward Applied Solutions: Imposing Sanity on Chaos
In theory, then, we could pursue several applied strategies:
Mandated Rent-Adjusted Reporting
Require corporations to file a “Rent-Adjusted EBITA” metric.
Auditors would have to categorise income streams as “productive” or “rentier.”
Differential Taxation
Tax normal profits at a competitive corporate rate.
Tax rents at punitive rates (e.g., 70-90%), since taxing rents does not distort incentives.
Sector-Specific Rent Taxes
Levy special taxes on land, platforms, patents, and monopoly franchises.
Create dynamic rent-extraction indices updated annually.
Platform Rent Charges
Impose data rent taxes on digital platforms extracting value from user activity.
Public Registry of Rents
Create a global registry classifying rents by sector, firm, and mechanism.
Provide public transparency to rent-seeking activities.
V. The Political Reality: Clouds on the Horizon
Needless to say, the aristocracy of the digital age will not go gentle into this good night. Rentiers — whether in Silicon Valley, the City of London, or Wall Street — are deeply entwined with the political machinery that might otherwise regulate them.
Yet the costs of inaction are higher. If rent extraction continues to eclipse productive activity, the very legitimacy of markets — and democracy — will erode into cynicism, stagnation, and oligarchic decay.
Conclusion: The Choice Before Us
Separating rent from profit is not merely a technocratic tweak. It is a radical act — one that could reorient economic activity away from parasitic extraction and back toward genuine value creation.
In a world where algorithms are castles, platforms are fiefdoms, and data is the new serfdom, reclaiming the ancient dream of taxing rent is no longer optional. It is, quite simply, the price of our collective survival.
—or—How the Invisible Hand Became a Throttling Grip on the Throat of the Biosphere
As many frequent visitors know, I am a recovering economist. I tend to view economics through a philosophical lens. Here. I consider the daft nonsense of Pareto optimality.
Audio: NotebookLM podcast of this content.
There is a priesthood in modern economics—pious in its equations, devout in its dispassion—that gathers daily to prostrate before the altar of Pareto. Here, in this sanctum of spreadsheet mysticism, it is dogma that an outcome is “optimal” so long as no one is worse off. Never mind if half the world begins in a ditch and the other half in a penthouse jacuzzi. So long as no one’s Jacuzzi is repossessed, the system is just. Hallelujah.
This cult of cleanliness, cloaked in the language of “efficiency,” performs a marvellous sleight of hand: it transforms systemic injustice into mathematical neutrality. The child working in the lithium mines of the Congo is not “harmed”—she simply doesn’t exist in the model. Her labour is an externality. Her future, an asterisk. Her biosphere, a rounding error in the grand pursuit of equilibrium.
Let us be clear: this is not science. This is not even ideology. It is theology—an abstract faith-based system garlanded with numbers. And like all good religions, it guards its axioms with fire and brimstone. Question the model? Heretic. Suggest the biosphere might matter? Luddite. Propose redistribution? Marxist. There is no room in this holy order for nuance. Only graphs and gospel.
Jevons warned us…that improvements in efficiency could increase, not reduce, resource consumption.
The rot runs deep. William Stanley Jevons—yes, that Jevons, patron saint of unintended consequences—warned us as early as 1865 that improvements in efficiency could increase, not reduce, resource consumption. But his paradox, like Cassandra’s prophecy, was fated to be ignored. Instead, we built a civilisation on the back of the very logic he warned would destroy it.
Then came Simon Kuznets, who—bless his empirically addled soul—crafted a curve that seemed to promise that inequality would fix itself if we just waited politely. We called it the Kuznets Curve and waved it about like a talisman against the ravages of industrial capitalism, ignoring the empirical wreckage that piled up beneath it like bones in a trench.
Meanwhile, Pareto himself, that nobleman of social Darwinism, famously calculated that 80% of Italy’s land was owned by 20% of its people—and rather than challenge this grotesque asymmetry, he chose to marvel at its elegance. Economics took this insight and said: “Yes, more of this, please.”
And so the model persisted—narrow, bloodless, and exquisitely ill-suited to the world it presumed to explain. The economy, it turns out, is not a closed system of rational actors optimising utility. It is a planetary-scale thermodynamic engine fuelled by fossil sunlight, pumping entropy into the biosphere faster than it can absorb. But don’t expect to find that on the syllabus.
Mainstream economics has become a tragic farce, mouthing the language of optimisation while presiding over cascading system failure. Climate change? Not in the model. Biodiversity collapse? A regrettable externality. Intergenerational theft? Discounted at 3% annually.
We are witnessing a slow-motion suicide
We are witnessing a slow-motion suicide cloaked in the rhetoric of balance sheets. The Earth is on fire, and the economists are debating interest rates.
What we need is not reform, but exorcism. Burn the models. Salt the axioms. Replace this ossified pseudoscience with something fit for a living world—ecological economics, systems theory, post-growth thinking, anything with the courage to name what this discipline has long ignored: that there are limits, and we are smashing into them at speed.
History will not be kind to this priesthood of polite annihilation. Nor should it be.
The contestation over the definition of probability is alive and well—like a philosophical zombie that refuses to lie down and accept the tranquilliser of consensus. Despite over three centuries of intense mathematical, philosophical, and even theological wrangling, no single, universally accepted definition reigns supreme. Instead, we have a constellation of rival interpretations, each staking its claim on the epistemological turf, each clutching its own metaphysical baggage.
Audio: NotebookLM podcast on this topic.
Let us survey the battlefield:
1. Classical Probability (Laplacean Determinism in a Tuxedo)
This old warhorse defines probability as the ratio of favourable outcomes to possible outcomes, assuming all outcomes are equally likely. The problem? That assumption is doing all the heavy lifting, like a butler carrying a grand piano up five flights of stairs. It’s circular: we define probability using equiprobability, which itself presumes a notion of probability. Charming, but logically suspect.
2. Frequentist Probability (The Empiricist’s Fantasy)
Here, probability is the limit of relative frequencies as the number of trials tends to infinity. This gives us the illusion of objectivity—but only in a Platonic realm where we can conduct infinite coin tosses without the coin disintegrating or the heat death of the universe intervening. Also, it tells us nothing about singular cases. What’s the probability this specific bridge will collapse? Undefined, says the frequentist, helpfully.
3. Bayesian Probability (Subjectivity Dressed as Rigor)
Bayesians treat probability as a degree of belief—quantified plausibility updated with evidence. This is useful, flexible, and epistemically honest, but also deeply subjective. Two Bayesians can start with wildly different priors and, unless carefully constrained, remain in separate probabilistic realities. It’s like epistemology for solipsists with calculators.
4. Propensity Interpretation (The Ontology of Maybes)
Karl Popper and his ilk proposed that probability is a tendency or disposition of a physical system to produce certain outcomes. Sounds scientific, but try locating a “propensity” in a particle collider—it’s a metaphysical ghost, not a measurable entity. Worse, it struggles with repeatability and relevance outside of controlled environments.
5. Logical Probability (A Sober Attempt at Rationality)
Think of this as probability based on logical relations between propositions—à la Keynes or Carnap. It aims to be objective without being empirical. The problem? Assigning these logical relations is no easier than choosing priors in Bayesianism, and just as subjective when it comes to anything meaty.
6. Quantum Probability (Schrödinger’s Definition)
In quantum mechanics, probability emerges from the squared modulus of a wave function—so this is where physics says, “Shut up and calculate.” But this doesn’t solve the philosophical issue—it just kicks the can into Hilbert space. Interpretations of quantum theory (Copenhagen? Many Worlds?) embed different philosophies of probability, so the contestation merely changes battlegrounds.
Current Status: War of Attrition
There is no universal agreement, and likely never will be. Probability is used successfully across the sciences, economics, AI, and everyday reasoning—but the fact that these wildly different interpretations all “work” suggests that the concept is operationally robust yet philosophically slippery. Like money, love, or art, we use it constantly but define it poorly.
In short: the contestation endures because probability is not one thing—it is a shape-shifting chimera that serves multiple masters. Each interpretation captures part of the truth, but none hold it entire. Philosophers continue to argue, mathematicians continue to formalise, and practitioners continue to deploy it as if there were no disagreement at all.
And so the probability of this contest being resolved any time soon? About zero. Or one. Depending on your interpretation.
This will be hilarious to approximately no one. I meant to ask ChatGPT 03-mini-high to generate a dummy record. My compromised motor skills and concomitant keyboard typing skills rendered this as funny. Except for the redacted bits in bold, this is the code it rendered. Pay special attention to the query’s input variables highlighted in post-production with 🤔 emojis.
Let’s add a bit to write a funny record per the schema.
import mysql.connector
import logging
import uuid
# Configure logging
logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)
def test_mysql_connection():
"""Test MySQL connection and write a funny record to the database."""
logger.info("Testing MySQL connection...")
try:
logger.info("Connecting to MySQL...")
connection = mysql.connector.connect(
host='127.0.0.1',
user='[user]',
password='[password]',
database='[database]',
port=3306
)
if connection.is_connected():
db_info = connection.get_server_info()
logger.info(f"Connected to MySQL Server version {db_info}")
cursor = connection.cursor()
cursor.execute("SELECT DATABASE();")
record = cursor.fetchone()
logger.info(f"Connected to database: {record[0]}")
# Insert a funny record into survey_results
funny_survey_data = {
"session_id": str(uuid.uuid4()),
"q1_response": 1,
"q2_response": 2,
"q3_response": 3,
"q4_response": 4,
"q5_response": 5,
"q6_response": 6,
"n1": 42, 🤔
"n2": 69, 🤔
"n3": 420, 🤔
"plot_x": 3.14, 🤔
"plot_y": 2.71, 🤔
"browser": "FunnyBrowser 9000",
"region": "JokeRegion",
"source": "comedy",
"hash_email_session": "f00b4r-hash" 🤔
}
query = """INSERT INTO survey_results
(session_id, q1_response, q2_response, q3_response, q4_response, q5_response, q6_response,
n1, n2, n3, plot_x, plot_y, browser, region, source, hash_email_session)
VALUES (%(session_id)s, %(q1_response)s, %(q2_response)s, %(q3_response)s, %(q4_response)s,
%(q5_response)s, %(q6_response)s, %(n1)s, %(n2)s, %(n3)s,
%(plot_x)s, %(plot_y)s, %(browser)s, %(region)s, %(source)s, %(hash_email_session)s)
"""
logger.info("Inserting funny survey record...")
cursor.execute(query, funny_survey_data)
connection.commit()
logger.info(f"Funny survey record inserted with ID: {cursor.lastrowid}")
except mysql.connector.Error as e:
logger.error(f"Error during MySQL operation: {e}")
finally:
if 'cursor' in locals() and cursor:
cursor.close()
if 'connection' in locals() and connection.is_connected():
connection.close()
logger.info("MySQL connection closed.")
if __name__ == "__main__":
test_mysql_connection()
“Is the universe really infinite? Or could it loop back on itself like a sphere?” Sabine Hossenfelder’s words on the nature of space-time are arresting, not merely for the cosmological implications but for the deeper metaphor they offer. They strike a resonant chord with anyone wrestling with a different kind of infinite: the slippery expanse of language.
As Sabine walks us through the intricacies of curved space-time, she inadvertently shines a light on something equally abstract yet close to home—how language, like the universe, seems vast and unbounded but is, in practice, riddled with constraints. What if language itself, for all its apparent openness, is its own kind of finite geometry?
Drawing on my Language Insufficiency Hypothesis (LIH), I propose that Sabine’s insights into cosmology can offer a lens to explore the paradoxes of human communication. Language, like space-time, is internally defined, replete with loops, and prone to infinite configurations that fail to expand meaningfully. Let’s explore how the universe’s curvature mirrors the curvatures of our words.
The Closed Systems of Space and Language
In physics, the curvature of space-time is measured internally. You can determine if space is flat or curved by drawing a triangle and adding its angles. If they don’t sum to 180 degrees, you’re in curved space. Sabine highlights that this is true without any external reference point; the geometry is self-contained.
Language operates much the same way. Words and meanings are often bounded by the internal logic of the systems they inhabit—be they legal, technical, or ideological. Much like the curvature of space-time, linguistic meaning is determined not externally but within the context of its own closed system. Think of a term like “justice”: in a legal setting, it might add up to one interpretation, while in a political debate, its angles skew wildly. To an outsider, the system is opaque, even though it seems perfectly flat from within.
Infinite Expanses or Finite Loops?
Sabine explains that the universe might be infinite, but it might also loop back on itself, creating patterns of repetition. Her analogy of light travelling endlessly through a curved universe only to return to its origin provides a striking metaphor for language’s “effectiveness horizons.”
As concepts grow more abstract—freedom, truth, beauty—language seems to expand infinitely. But in practice, it often circles back, repeating itself in kaleidoscopic loops of contested meaning. Philosophers have debated terms like “good” or “justice” for millennia, yet here we are, still tracing the same paths, unable to break free from the system’s internal constraints. Language doesn’t expand into new meaning; it curves back on itself.
Schrödinger’s Words: Infinite Interpretations
One of Sabine’s most evocative ideas is the notion that in an infinite universe, there are infinite copies of you, some slightly different, some wildly so. A version of you with more hair. One with less brain. This multiplicity mirrors what I call Schrödinger’s Weasels: words that exist in multiple, contradictory states until “collapsed” by context.
Take a word like “freedom.” In political discourse, it can simultaneously mean the right to self-determination, freedom from government interference, or the economic liberty to exploit markets. Much like Sabine’s infinite configurations, these meanings coexist until someone forces them into a single interpretive frame. The result? Semantic exhaustion. A single word tries to carry the weight of an infinite universe.
The Precision Paradox
Sabine notes that asking what the universe expands into is a meaningless question because expansion describes relationships within space-time, not beyond it. Similarly, the pursuit of perfect precision in language often collapses into meaninglessness. Trying to pin down a word like “justice” leads to endless definitions, each requiring further clarification. It’s a Zeno’s paradox of semantics: the closer we get to precision, the more distance remains.
Lessons from Curved Space and Twisted Words
What does this tell us about the limits of language? Sabine’s insights reinforce the idea that complexity doesn’t always lead to clarity. Like the universe, language isn’t infinite in the way we might wish; it’s bounded by its own structure. The more abstract the concept, the greater the chance we’ll find ourselves lost in a linguistic loop, navigating words that seem to expand but merely repeat.
Understanding this doesn’t mean abandoning the pursuit of meaning but accepting its constraints. Just as cosmologists use models to map the unobservable edges of the universe, we can use frameworks like the Language Insufficiency Hypothesis to chart the limits of our words. Both efforts are acts of humility in the face of infinite complexity.
Closing Thought
If Sabine is right that there are infinitely many versions of ourselves in the universe, perhaps one of them has already solved the riddle of language. Or, more likely, they’ve just found a new loop to wander.
What do you think? Is language a closed system, forever folding back on itself? Or can we stretch it, like space-time, to infinity and beyond?
I’ve written a lot on the insufficiency of language, and it’s not even an original idea. Language, our primary tool for sharing thoughts and ideas, harbours a fundamental flaw: it’s inherently insufficient for conveying precise meaning. While this observation isn’t novel, recent developments in artificial intelligence provide us with new ways to illuminate and examine this limitation. Through a progression from simple geometry to complex abstractions, we can explore how language both serves and fails us in different contexts.
The Simple Made Complex
Consider what appears to be a straightforward instruction: Draw a 1-millimetre square in the centre of an A4 sheet of paper using an HB pencil and a ruler. Despite the mathematical precision of these specifications, two people following these exact instructions would likely produce different results. The variables are numerous: ruler calibration, pencil sharpness, line thickness, paper texture, applied pressure, interpretation of “centre,” and even ambient conditions affecting the paper.
This example reveals a paradox: the more precisely we attempt to specify requirements, the more variables we introduce, creating additional points of potential divergence. Even in mathematics and formal logic—languages specifically designed to eliminate ambiguity—we cannot escape this fundamental problem.
Precision vs Accuracy: A Useful Lens
The scientific distinction between precision and accuracy provides a valuable framework for understanding these limitations. In measurement, precision refers to the consistency of results (how close repeated measurements are to each other), while accuracy describes how close these measurements are to the true value.
Returning to our square example:
Precision: Two people might consistently reproduce their own squares with exact dimensions
Accuracy: Yet neither might capture the “true” square we intended to convey
As we move from geometric shapes to natural objects, this distinction becomes even more revealing. Consider a maple tree in autumn. We might precisely convey certain categorical aspects (“maple,” “autumn colours”), but accurately describing the exact arrangement of branches and leaves becomes increasingly difficult.
The Target of Meaning: Precision vs. Accuracy in Communication
To understand language’s limitations, we can borrow an illuminating concept from the world of measurement: the distinction between precision and accuracy. Imagine a target with a bullseye, where the bullseye represents perfect communication of meaning. Just as archers might hit different parts of a target, our attempts at communication can vary in both precision and accuracy.
Consider four scenarios:
Low Precision, Low Accuracy When describing our autumn maple tree, we might say “it’s a big tree with colourful leaves.” This description is neither precise (it could apply to many trees) nor accurate (it misses the specific characteristics that make our maple unique). The communication scatters widely and misses the mark entirely.
High Precision, Low Accuracy We might describe the tree as “a 47-foot tall maple with exactly 23,487 leaves displaying RGB color values of #FF4500.” This description is precisely specific but entirely misses the meaningful essence of the tree we’re trying to describe. Like arrows clustering tightly in the wrong spot, we’re consistently missing the point.
Low Precision, High Accuracy “It’s sort of spreading out, you know, with those typical maple leaves turning reddish-orange, kind of graceful looking.” While imprecise, this description might actually capture something true about the tree’s essence. The arrows scatter, but their centre mass hits the target.
High Precision, High Accuracy This ideal state is rarely achievable in complex communication. Even in our simple geometric example of drawing a 1mm square, achieving both precise specifications and accurate execution proves challenging. With natural objects and abstract concepts, this challenge compounds exponentially.
The Communication Paradox
This framework reveals a crucial paradox in language: often, our attempts to increase precision (by adding more specific details) can actually decrease accuracy (by moving us further from the essential meaning we’re trying to convey). Consider legal documents: their high precision often comes at the cost of accurately conveying meaning to most readers.
Implications for AI Communication
This precision-accuracy framework helps explain why AI systems like our Midjourney experiment show asymptotic behaviour. The system might achieve high precision (consistently generating similar images based on descriptions) while struggling with accuracy (matching the original intended image), or vice versa. The gap between human intention and machine interpretation often manifests as a trade-off between these two qualities.
Our challenge, both in human-to-human and human-to-AI communication, isn’t to achieve perfect precision and accuracy—a likely impossible goal—but to find the optimal balance for each context. Sometimes, like in poetry, low precision might better serve accurate meaning. In other contexts, like technical specifications, high precision becomes crucial despite potential sacrifices in broader accuracy.
The Power and Limits of Distinction
This leads us to a crucial insight from Ferdinand de Saussure’s semiotics about the relationship between signifier (the word) and signified (the concept or object). Language proves remarkably effective when its primary task is distinction among a limited set. In a garden containing three trees—a pine, a maple, and a willow—asking someone to “point to the pine” will likely succeed. The shared understanding of these categorical distinctions allows for reliable communication.
However, this effectiveness dramatically diminishes when we move from distinction to description. In a forest of a thousand pines, describing one specific tree becomes nearly impossible. Each additional descriptive detail (“the tall one with a bent branch pointing east”) paradoxically makes precise identification both more specific and less likely to succeed.
An AI Experiment in Description
To explore this phenomenon systematically, I conducted an experiment using Midjourney 6.1, a state-of-the-art image generation AI. The methodology was simple:
Generate an initial image
Describe the generated image in words
Use that description to generate a new image
Repeat the process multiple times
Attempt to refine the description to close the gap
Continue iterations
The results support an asymptotic hypothesis: while subsequent iterations might approach the original image, they never fully converge. This isn’t merely a limitation of the AI system but rather a demonstration of language’s fundamental insufficiency.
A cute woman and her dog stand next to a tree
One can already analyse this for improvements, but let’s parse it together.
a cute woman
With this, we know we are referencing a woman, a female of the human species. There are billions of women in the world. What does she look like? What colour, height, ethnicity, and phenotypical attributes does she embody?
We also know she’s cute – whatever that means to the sender and receiver of these instructions.
I used an indefinite article, a, so there is one cute woman. Is she alone, or is she one from a group?
It should be obvious that we could provide more adjectives (and perhaps adjectives) to better convey our subject. We’ll get there, but let’s move on.
and
We’ve got a conjunction here. Let’s see what it connects to.
her dog
She’s with a dog. In fact, it’s her dog. This possession may not be conveyable or differentiable from some arbitrary dog, but what type of dog is it? Is it large or small? What colour coat? Is it groomed? Is it on a leash? Let’s continue.
stand
It seems that the verb stand refers to the woman, but is the dog also standing, or is she holding it? More words could qualify this statement better.
next to a tree
A tree is referenced. Similar questions arise regarding this tree. At a minimum, there is one tree or some variety. She and her dog are next to it. Is she on the right or left of it?
We think we can refine our statements with precision and accuracy, but can we? Might we just settle for “close enough”?
Let’s see how AI interpreted this statement.
Image: Eight Midjourney renders from the prompt: A cute woman and her dog stand next to a tree. I’ll choose one of these as my source image.
Let’s deconstruct the eight renders above. Compositionally, we can see that each image contains a woman, a dog, and a tree. Do any of these match what you had in mind? First, let’s see how Midjourney describes the first image.
In a bout of hypocrisy, Midjourney refused to /DESCRIBE the image it just generated.
Last Midjourney description for now.
Let’s cycle through them in turn.
A woman is standing to the left of an old-growth tree – twice identified as an oak tree. She’s wearing faded blue jeans and a loose light-coloured T-shirt. She’s got medium-length (maybe) red-brown hair in a small ponytail. A dog – her black and white dog identified as a pitbull, an American Foxhound, and an American Bulldog – is also standing on his hind legs. I won’t even discuss the implied intent projected on the animal – happy, playful, wants attention… In two of the descriptions, she’s said to be training it. They appear to be in a somewhat residential area given the automobiles in the background. We see descriptions of season, time of day, lighting, angle, quality,
A woman is standing to the right of an old-growth tree. She’s wearing short summer attire. Her dog is perched on the tree.
An older woman and her dog closer up.
A read view of both a woman and her dog near an oak tree.
As it turned out, I wasn’t thrilled with any of these images, so I rendered a different one. Its description follows.
The consensus is that ‘a beautiful girl in a white dress and black boots stands next to a tree’ with a Jack Russell Terrier dog. I see birch trees and snow. It’s overcast. Let’s spend some time trying to reproduce it. To start, I’m consolidating the above descriptions. I notice some elements are missing, but we’ll add them as we try to triangulate to the original image.
A beautiful girl in a white dress and black boots stands next to a tree in a winter forest with birch trees. A Jack Russell Terrier dog is nearby, and the weather is cloudy. The photo has a cinematic style. --ar 128:71
This is pretty far off the mark. We need to account for the overall setting and composition, relative positioning, clothing, hair, camera, perspective – even lighting and film emulsion.
Let’s see how we can refine it with some adjectives. Before this, I asked Anthropic’s Claude 3.5 to describe the image. Perhaps we’ll get more details.
An atmospheric winter photograph was taken in what appears to be a birch and pine forest. The ground is partially covered with snow, creating a patchy pattern of white snow and brown leaves or soil. In the image, there's a person wearing winter attire consisting of a white knit turtleneck sweater, a white skirt, dark knee-high socks, and grey lace-up boots. They're standing against a large tree trunk with their arms crossed. Next to them is a Jack Russell Terrier mix dog with distinctive brown and white colouring. The dog appears to be well-behaved, sitting attentively beside the tree. The composition is quite striking, with the massive dark tree trunk serving as a central focal point against the misty, winter forest background. The white clothing of the person creates a visual echo with the white bark of birch trees visible in the background, as well as with the patches of snow on the ground. The overall mood of the photo is somewhat moody and contemplative, enhanced by what appears to be overcast lighting and the misty quality of the forest in the background. The colour palette is largely monochromatic, dominated by whites, greys, and browns, which adds to the winter atmosphere of the scene.
We don’t seem to be moving in a good direction. Let’s modify the initial prompt.
A beautiful girl in a white dress and tall laced black boots stands facing the camera to the right of large oak tree centred in the foreground of a winter forest with birch trees in the background. To the left of the tree is a Jack Russell Terrier dog looking at the camera, and the weather is cloudy. The photo has a cinematic style. --ar 128:71
I’ll allow the results to speak for themselves. Let’s see if we can’t get her out of the wedding gown and into a white jumper and skirt. I’ll bold the amends.
A beautiful girl in a white jumper and skirt wearing black leggings and tall laced black boots stands facing the camera to the right of large oak tree centred in the foreground of a winter forest with birch trees in the background. To the left of the tree is a Jack Russell Terrier dog looking at the camera, and the weather is cloudy. The photo has a cinematic style. --ar 128:71
s
A beautiful young woman with long brown hair pulled to the side of her face in a white jumper and white skirt wearing black leggings under tall laced black boots stands facing the camera to the right of large oak tree centred in the foreground of a winter forest with birch trees in the background. Patchy snow is on the ground. To the left of the tree is a Jack Russell Terrier dog looking at the camera, and the weather is overcast. The photo has a cinematic style. --ar 128:71
What gives?
I think my point has been reinforced. I’m getting nowhere fast. Let’s give it one more go and see where we end up. I’ve not got a good feeling about this.
A single large oak tree centred in the foreground of a winter forest with birch trees in the background. Patches of snow is on the ground. To the right of the oak tree stands a beautiful young woman with long brown hair pulled to the side of her face in a white jumper and white skirt wearing black boots over tall laced black boots. She stands facing the camera. To the left of the tree is a Jack Russell Terrier dog looking at the camera, and the weather is overcast. The photo has a cinematic style. --ar 128:71
With this last one, I re-uploaded the original render along with this text prompt. Notice that the girl now looks the same and the scene (mostly) appears to be in the same location, but there are still challenges.
After several more divergent attempts, I decided to focus on one element – the girl.
As I regard the image, I’m thinking of a police sketch artist. They get sort of close, don’t they? They’re experts. I’m not confident that I even have the vocabulary to convey accurately what I see. How do I describe her jumper? Is that a turtleneck or a high collar? It appears to be knit. Is is wool or some blend? does that matter for an image? Does this pleated skirt have a particular name or shade of white? It looks as though she’s wearing black leggings – perhaps polyester. And those boots – how to describe them. I’m rerunning just the image above through a describe function to see if I can get any closer.
These descriptions are particularly interesting and telling. First, I’ll point out that AI attempts to identify the subject. I couldn’t find Noa Levin by a Google search, so I’m not sure how prominent she might be if she even exists at all in this capacity. More interesting still, the AI has placed her in a scenario where the pose was taken after a match. Evidently, this image reflects the style of photographer Guy Bourdin. Perhaps the jumper mystery is solved. It identified a turtleneck. I’ll ignore the tree and see if I can capture her with an amalgamation of these descriptions. Let’s see where this goes.
A photo-realistic portrait of Israeli female soccer player Noa Levin wearing a white turtleneck sweater, arms crossed, black boots, and a short skirt, with long brown hair, standing near a tree in a winter park. The image captured a full-length shot taken in a studio setting, using a Canon EOS R5 camera with a Canon L-series 80mm f/2 lens. The image has been professionally color-graded, with soft shadows, low contrast, and a clean, sharp focus. --ar 9:16
Close-ish. Let’s zoom in to get better descriptions of various elements starting with her face and hair.
Now, she’s a sad and angry Russian woman with (very) pale skin; large, sad, grey eyes; long, straight brown hair. Filmed in the style of either David LaChapelle or Alini Aenami (apparently misspelt from Alena Aenami). One thinks it was a SnapChat post. I was focusing on her face and hair, but it notices her wearing a white (oversized yet form-fitting) jumper sweater and crossed arms .
I’ll drop the angry bit – and then the sad.
Stick a fork in it. I’m done. Perhaps it’s not that language is insufficient; it that my language skills are insufficient. If you can get closer to the original image, please forward the image, the prompt, and the seed, so I can post it.
The Complexity Gradient
A clear pattern emerges when we examine how language performs across different levels of complexity:
Categorical Distinction (High Success)
Identifying shapes among limited options
Distinguishing between tree species
Basic color categorization
Simple Description (Moderate Success)
Basic geometric specifications
General object characteristics
Broad emotional states
Complex Description (Low Success)
Specific natural objects
Precise emotional experiences
Unique instances within categories
Abstract Concepts (Lowest Success)
Philosophical ideas
Personal experiences
Qualia
As we move up this complexity gradient, the gap between intended meaning and received understanding widens exponentially.
The Tolerance Problem
Understanding these limitations leads us to a practical question: what level of communicative tolerance is acceptable for different contexts? Just as engineering embraces acceptable tolerances rather than seeking perfect measurements, perhaps effective communication requires:
Acknowledging the gap between intended and received meaning
Establishing context-appropriate tolerance levels
Developing better frameworks for managing these tolerances
Recognizing when precision matters more than accuracy (or vice versa)
Implications for Human-AI Communication
These insights have particular relevance as we develop more sophisticated AI systems. The limitations we’ve explored suggest that:
Some communication problems might be fundamental rather than technical
AI systems may face similar boundaries as human communication
The gap between intended and received meaning might be unbridgeable
Future development should focus on managing rather than eliminating these limitations
Conclusion
Perhaps this is a simple exercise in mental masturbation. Language’s insufficiency isn’t a flaw to be fixed but a fundamental characteristic to be understood and accommodated. By definition, it can’t be fixed. The gap between intended and received meaning may be unbridgeable, but acknowledging this limitation is the first step toward more effective communication. As we continue to develop AI systems and push the boundaries of human-machine interaction, this understanding becomes increasingly critical.
Rather than seeking perfect precision in language, we might instead focus on:
Developing new forms of multimodal communication
Creating better frameworks for establishing shared context
Accepting and accounting for interpretative variance
Building systems that can operate effectively within these constraints
Understanding language’s limitations doesn’t diminish its value; rather, it helps us use it more effectively by working within its natural constraints.
Philosophics 