The air is thick with bad takes. Scroll for five minutes and you’ll find someone announcing, usually with the pomp of a TEDx speaker, that “AI has no emotions” or “It’s not really reading.” These objections are less profound insights than they are linguistic face-plants. The problem isn’t AI. It’s the speakers’ near-total ignorance of how language works.
Audio: NotebookLM podcast on this topic.
Language as the Unseen Operating System
Language is not a transparent pane of glass onto the world. It is the operating system of thought: messy, recursive, historically contingent. Words do not descend like tablets from Sinai; they are cobbled together, repurposed, deconstructed, and misunderstood across generations.
If you don’t understand that basic condition, that language is slippery, mediated, and self-referential, then your critique of Large Language Models is just noise in the system. LLMs are language machines. To analyse them without first understanding language is like reviewing a symphony while stone deaf.
The Myth of “Emotions”
Critics obsess over whether LLMs “feel.” But feeling has never been the measure of writing. The point of a sentence is not how the author felt typing it, but whether the words move the reader. Emotional “authenticity” is irrelevant; resonance is everything.
Writers know this. Philosophers know this. LLM critics, apparently, do not. They confuse the phenomenology of the writer with the phenomenology of the text. And in doing so, they embarrass themselves.
The Licence Test
So here’s the proposal: a licence to comment on AI. It wouldn’t be onerous. Just a few basics:
Semiotics 101: Know that words point to other words more than they point to things.
Context 101: Know that meaning arises from use, not from divine correspondence.
Critical Theory 101: Know that language carries baggage, cultural, historical, and emotional, that doesn’t belong to the machine or the individual speaker.
Fail these, and you’re not cleared to drive your hot takes onto the information superhighway.
Meta Matters
I’ve explored some of this in more detail elsewhere (link to Ridley Park’s “Myth of Emotion”), but the higher-level point is this: debates about AI are downstream of debates about language. If you don’t grasp the latter, your pronouncements on the former are theatre, not analysis.
Philosophy has spent centuries dismantling the fantasy of words as perfect mirrors of the world. It’s perverse that so many people skip that homework and then lecture AI about “meaning” and “feeling.”
Don’t get salty with me when I tell you I asked AI to write this for me. I was thinking that “take it with a grain of salt” or “take it with a pinch of salt” in English did not share the same meaning as “mettre son grain de sel” en français, so I asked ChatGPT for other uses of salt. This is why it doesn’t follow by usual style, if one can call it that.
🧂 Salt: That Most Misunderstood Metaphor
Salt has an image problem.
Despite being one of the most ancient and revered substances in human civilisation—once used as currency, treaty-sealer, and god-bait—it somehow gets dragged through the metaphorical gutter in modern idiom. In English, to take something “with a grain of salt” is to doubt it. To “add your grain of salt,” per the French idiom mettre son grain de sel, is to interrupt uninvited. Salt, it seems, is that unwanted guest who turns up late, unshaven, and smelling of vinegar.
And yet, salt is also life. Necessary. Essential. Literal. So what gives?
Let’s do what the internet never does and look at context.
🏴☠️ English: Cynicism in a Crystal
The English expression “take it with a grain of salt” (or, in older form, a pinch) comes from Latin cum grano salis, which likely implied adding a figurative preservative to dubious claims—treat this as you would old meat. In other words, don’t fully trust it unless you like dysentery.
We also say “he’s a bit salty” to mean grumpy, caustic, or prone to verbal cutlery. “Adding your two cents” is bad enough, but adding your grain of salt implies that what you’re contributing is both unsolicited and probably irritating.
Put simply, English idioms treat salt as if it’s the person in the meeting who thinks they’re clever. There’s a faint whiff of Protestantism here—suspicious of flavour, pleasure, and expressive enthusiasm. Plain oatmeal, plain truths, no seasoning required. Salt is vice. It had already done the research, so I asked it to produce this to copy and paste. You’re welcome.
🇫🇷 French: Salty Saboteurs
The French mettre son grain de sel is more or less the same: to butt in. To lob your unwanted opinion into someone else’s stew. Not unlike “putting in your two penn’orth” in British English—but somehow meaner, as if your salt is not just annoying, but wrong.
Salt, in this idiom, doesn’t enrich—it ruins. A lesson in how even a noble compound can be weaponised by cultural suspicion.
🏺 Hindi: Loyalty Seasoned with Honour
Contrast this with Hindi: namak harām — literally “unfaithful to salt.” This is a powerful accusation. It means you’ve betrayed someone who fed you, someone who sustained you. You’ve taken their salt and spat in their dish.
Conversely, namak halāl is a compliment: someone loyal, trustworthy, faithful to the hand that seasoned them. Salt is the symbol of obligation and honour—not interference.
It is covenantal.
🗾 Japanese: Salt as Mercy
塩を送る (shio o okuru) – “to send salt” – is a Japanese idiom meaning to help your enemy in their time of need. Based on a historical moment when Uesugi Kenshin sent salt to his rival, Takeda Shingen, when the latter’s supply was blockaded.
Salt, here, transcends enmity. It’s noble. A tool of ethics.
In short: send salt, don’t throw it.
🇩🇪 German & 🇪🇸 Spanish: Flavour as Personality
The Germans say “das Salz in der Suppe sein”—to be the salt in the soup. You’re what makes life interesting. Without you, it’s just… wet nutrition.
In Spanish, “ser la sal de la vida” means to be the zest of existence. Without salt, life is dull, bland, morally beige.
In these idioms, salt is essential. A little dangerous, maybe, but necessary. Just like any compelling person.
🇹🇷 Turkish: The Dry Salt of Privilege
The Turkish idiom “tuzu kuru” (lit. “dry salt”) means you’re doing fine. Perhaps too fine. You’re unaffected, aloof, in your tower of comfort while others stew.
Dry salt is privilege: unbothered, unsalted tears. An idiom with side-eye built in.
🕊️ Christianity: Salt of the Earth
The Gospels famously commend the righteous as “the salt of the earth.” Not merely good people, but the ones who preserve and season the whole damn world. And yet, “if salt loses its savour,” says Matthew 5:13, “wherewith shall it be salted?” A warning to remain vital. Relevant. Useful.
Even Jesus had thoughts about flavour fatigue.
⚖️ So… Is Salt Praised or Pitied?
Depends who you ask.
For some, salt is civic virtue (Hindi).
For others, it’s moral generosity (Japanese).
Sometimes it’s life’s spark (German, Spanish).
Sometimes it’s trouble in a shaker (English, French).
But the ambivalence is the point. Salt is essential—but easily overdone. Too little, and life is bland. Too much, and it’s ruined.
Like language, then: salt mediates between flavour and clarity. Add carefully. Stir well.
🧂 Final Sprinkle
Before you disparage someone for being “a bit salty,” ask yourself whether they’re really interfering—or simply adding what your grey little broth lacked all along.
And for heaven’s sake, be faithful to the salt you’ve eaten.
Few philosophical thought experiments have managed to torment empiricists quite like Molyneux’s problem. First posed by William Molyneux to John Locke in 1688 (published in Locke’s An Essay Concerning Human Understanding), the question is deceptively simple:
If a person born blind, who has learned to distinguish a cube from a sphere by touch, were suddenly granted sight, could they, without touching the objects, correctly identify which is the cube and which is the sphere by sight alone?
I was inspired to write this article in reaction to Jonny Thmpson’s post on Philosophy Minis, shared below for context.
Video: Molyneux’s Problem
Locke, ever the champion of sensory experience as the foundation of knowledge, gave a confident empiricist’s answer: no. For Locke, ideas are the products of sensory impressions, and each sense provides its own stream of ideas, which must be combined and associated through experience. The newly sighted person, he argued, would have no prior visual idea of what a cube or sphere looks like, only tactile ones; they would need to learn anew how vision maps onto the world.
Audio: NotebookLM podcast on this topic.
This puzzle has persisted through centuries precisely because it forces us to confront the assumptions at the heart of empiricism: that all knowledge derives from sensory experience and that our senses, while distinct, can somehow cohere into a unified understanding of the world.
Empiricism, Epistemology, and A Priori Knowledge: The Context
Before we dismantle the cube further, let’s sweep some conceptual debris out of the way. Empiricism is the view that knowledge comes primarily (or exclusively) through sensory experience. It stands opposed to rationalism, which argues for the role of innate ideas or reason independent of sense experience.
Epistemology, the grandiloquent term for the study of knowledge, concerns itself with questions like: What is knowledge? How is it acquired? Can we know anything with certainty?
And then there is the spectre of a priori knowledge – that which is known independent of experience. A mathematical truth (e.g., 2 + 2 = 4) is often cited as a classic a priori case. Molyneux’s problem challenges empiricists because it demands an account of how ideas from one sensory modality (touch) might map onto another (vision) without prior experience of the mapping—an a priori leap, if you will.
The Language Correspondence Trap
While Molyneux and Locke framed this as an epistemological riddle, we can unmask it as something more insidious: a failure of language correspondence. The question presumes that the labels “cube” and “sphere” – tied in the blind person’s mind to tactile experiences – would, or should, carry over intact to the new visual experiences. But this presumption smuggles in a linguistic sleight of hand.
The word “cube” for the blind person means a specific configuration of tactile sensations: edges, vertices, flat planes. The word “sphere” means smoothness, unbroken curvature, no edges. These are concepts anchored entirely in touch. When vision enters the fray, we expect these words to transcend modalities – to leap from the tactile to the visual, as if their meanings were universal tokens rather than context-bound markers. The question is not merely: can the person see the cube? but rather: can the person’s tactile language map onto the visual world without translation or recalibration?
What Molyneux’s problem thus exposes is the assumption that linguistic labels transparently correspond to external reality, regardless of sensory apparatus. This is the mirage at the heart of Locke’s empiricism, the idea that once a word tags an object through experience, that tag is universally valid across sensory experiences. The cube and sphere aren’t just objects of knowledge; they are signs, semiotic constructs whose meaning depends on the sensory, social, and linguistic contexts in which they arise.
The Semiotic Shambles
Molyneux’s cube reveals the cracks in the correspondence theory of language: the naïve belief that words have stable meanings that latch onto stable objects or properties in the world. In fact, the meaning of “cube” or “sphere” is as much a product of sensory context as it is of external form. The newly sighted person isn’t merely lacking visual knowledge; they are confronted with a translation problem – a semantic chasm between tactile signification and visual signification.
If, as my Language Insufficiency Hypothesis asserts, language is inadequate to fully capture and transmit experience across contexts, then Molyneux’s problem is not an oddity but an inevitability. It exposes that our conceptual frameworks are not universal keys to reality but rickety bridges between islands of sense and meaning. The cube problem is less about empiricism’s limits in epistemology and more about its blind faith in linguistic coherence.
In short, Molyneux’s cube is not simply an empirical puzzle; it is a monument to language’s failure to correspond cleanly with the world, a reminder that what we call knowledge is often just well-worn habit dressed up in linguistic finery.
A Final Reflection
Molyneux’s problem, reframed through the lens of language insufficiency, reveals that our greatest epistemic challenges are also our greatest linguistic ones. Before we can speak of knowing a cube or sphere by sight, we must reckon with the unspoken question: do our words mean what we think they mean across the changing stage of experience?
That, dear reader, is the cube that haunts empiricism still.
“All that is necessary for evil to triumph is for a few words to go missing from the bylaws.” — not Edmund Burke, but it ought to be.
The Trump administration—America’s reigning monarch of meaningless bombast—has done it again. This time, with an executive order so linguistically cunning it deserves a Pulitzer for Subtextual Menace.
Issued on 30 January 2025, the decree known as “Defending Women from Gender Ideology Extremism” (because, of course, it couldn’t just be called Let’s Erase Legal Protections for People We Don’t Like) removed “political affiliation” and “marital status” from the list of protected classes within certain federal frameworks.
And the result? According to documents unearthed by The Guardian, VA doctors can now legally refuse treatment to patients based on their politics or marital status. You know, because being a Democrat apparently makes you too much of a pre-existing condition.
Naturally, the VA and White House are insisting this means absolutely nothing. “Don’t worry,” they coo. “No one’s actually doing it.” Ah yes, the old Schrödinger’s Protections defence—simultaneously removed and unchanged, invalid but somehow still effective.
But here’s the point—and where it ties to the Language Insufficiency Hypothesis I’ve been peddling like a raving madman at the crossroads of post-structuralism and bureaucratic despair: language isn’t just failing to communicate meaning—it’s being weaponised to obscure it.
The Erosion of Meaning Through Omission
This isn’t the blunt-force idiocy of Orwell’s Newspeak. This is something more elegant—more insidious. This is legislative lacunae. It’s what happens when not saying something says everything.
The words “political affiliation” and “marital status” weren’t replaced. They weren’t clarified. They were simply deleted. Erased like a bad tweet, like a conscience, like a veteran with the wrong bumper sticker.
This is language subtraction as a tool of governance.
We’re not criminalising dissent. We’re just making it legally ignorable.
We’re not discriminating against the unmarried. We’re just no longer required to treat them the same.
It’s the bureaucratic cousin of the dog-whistle: not quite audible in court, but perfectly clear to the base.
The Slippery Slope is Now a Slip-n-Slide
This is how you rewrite civil rights without the fuss of saying so. You just… remove the language that once held the dam in place. Then, when the flood comes, you feign surprise:
“Oh, dear. Who could have guessed that removing protections would result in people being unprotected?”
(Everyone. Everyone could have guessed.)
This is not a bug in the legal language. It’s the feature. The silence is the speech act. The absence is the argument.
This is what I mean by language insufficiency: not merely that our words fail to convey truth, but that their very structure is liable to be gamed—exploited by those who understand that ambiguity is power.
Beyond Intentionality: The Weaponised Void
In philosophy of language, we often debate intentionality—what the speaker meant to say. But here we’re in darker waters. This isn’t about intention. It’s about calculated omission.
The executive order doesn’t declare war on Democrats or single mothers. It simply pulls the thread and lets the tapestry unravel itself.
It’s an act of rhetorical cowardice disguised as administrative efficiency.
This is the Trumpian genius: use language like a stage magician uses sleeves. Distract with one hand, disappear with the other.
Final Diagnosis: Policy by Redaction
We now inhabit a political climate where what is not said carries more legal force than what is. Where bylaw gaps become policy gateways, and where civil rights die not with a bang, but with an elision.
So no, the VA hasn’t yet denied a Democrat a blood transfusion. But the table has been set. The menu revised. The waitstaff told they may now “use discretion.”
Language doesn’t merely fail us. It is being made to fail strategically.
Welcome to the new America: where rights aren’t removed—they’re left out of the memo.
Yet again, ChatGPT renders an odd image. Can’t be bothered to amend it.
Let us begin with the heresy: Truth is a rhetorical artefact. Not a revelation. Not a metaphysical essence glimmering behind the veil. Just language — persuasive, repeatable, institutionally ratified language. In other words: branding.
Audio: NotebookLM podcast on this topic.
This is not merely a postmodern tantrum thrown at the altar of Enlightenment rationalism. It is a sober, if impolite, reminder that nearly everything we call “knowledge” is stitched together with narrative glue and semantic spit. Psychology. Neuroscience. Ethics. Economics. Each presents itself as a science — or worse, a moral imperative — but their foundations are built atop a linguistic faultline. They are, at best, elegant approximations; at worst, dogma in drag.
Let’s take psychology. Here is a field that diagnoses your soul via consensus. A committee of credentialed clerics sits down and declares a cluster of behaviours to be a disorder, assigns it a code, and hands you a script. It is then canonised in the DSM, the Diagnostic Scripture Manual. Doubt its legitimacy and you are either naïve or ill — which is to say, you’ve just confirmed the diagnosis. It’s a theological trap dressed in the language of care.
Or neuroscience — the church of the glowing blob. An fMRI shows a region “lighting up” and we are meant to believe we’ve located the seat of love, the anchor of morality, or the birthplace of free will. Never mind that we’re interpreting blood-oxygen fluctuations in composite images smoothed by statistical witchcraft. It looks scientific, therefore it must be real. The map is not the territory, but in neuroscience, it’s often a mood board.
And then there is language itself, the medium through which all these illusions are transmitted. It is the stage, the scenery, and the unreliable narrator. My Language Insufficiency Hypothesis proposes that language is not simply a flawed tool — it is fundamentally unfit for the task it pretends to perform. It was forged in the furnace of survival, not truth. We are asking a fork to play the violin.
This insufficiency is not an error to be corrected by better definitions or clever metaphors. It is the architecture of the system. To speak is to abstract. To abstract is to exclude. To exclude is to falsify. Every time we speak of a thing, we lose the thing itself. Language functions best not as a window to the real but as a veil — translucent, patterned, and perpetually in the way.
So what, then, are our Truths™? They are narratives that have won. Stories that survived the epistemic hunger games. They are rendered authoritative not by accuracy, but by resonance — psychological, cultural, institutional. A “truth” is what is widely accepted, not because it is right, but because it is rhetorically unassailable — for now.
This is the dirty secret of epistemology: coherence masquerades as correspondence. If enough concepts link arms convincingly, we grant them status. Not because they touch reality, but because they echo each other convincingly in our linguistic theatre.
Libet’s experiment, Foucault’s genealogies, McGilchrist’s hemispheric metaphors — each peels back the curtain in its own way. Libet shows that agency might be a post-hoc illusion. Foucault reveals that disciplines don’t describe the subject; they produce it. McGilchrist laments that the Emissary now rules the Master, and the world is flatter for it.
But all of them — and all of us — are trapped in the same game: the tyranny of the signifier. We speak not to uncover truth, but to make truth-sounding noises. And the tragedy is, we often convince ourselves.
So no, we cannot escape the prison of language. But we can acknowledge its bars. And maybe, just maybe, we can rattle them loudly enough that others hear the clank.
Until then, we continue — philosophers, scientists, diagnosticians, rhetoricians — playing epistemology like a parlour game with rigged dice, congratulating each other on how well the rules make sense.
As I put the finishing touches on the third revision of my Language Insufficiency Hypothesis manuscript, I find myself reflecting on the role of Foucault’s concept of Power Relations in shaping the use and interpretation of language in institutional contexts.
A key aspect of my hypothesis is the notion that some abstract conceptual language is intentionally vague. I touched on this idea in my recent article on the ambiguity of the term ‘gift’, but the implications extend far beyond that specific example. The strategic use of linguistic indeterminacy is a pervasive feature of many professional domains, serving to veil and enable subtle power plays.
The strategic use of linguistic indeterminacy is a pervasive feature of many professional domains
In my manuscript, I examine the concept of ‘reasonableness’ as a prime example of this phenomenon. This term is a favourite hiding spot for legal professionals, appearing in phrases like ‘reasonable doubt’ and ‘reasonable person’.Yet, upon closer inspection, the apparent clarity and objectivity of this language dissolves into a morass of ambiguity and subjectivity. The invocation of reasonableness often serves as a rhetorical sleight of hand, masking the exercise of institutional power behind a veneer of impartiality.
While I don’t wish to venture too far into Nietzschean cynicism, there is a sense in which the legal system operates like a casino. The house always seeks to maintain its edge, and it will employ whatever means necessary to preserve its authority and legitimacy. In the case of reasonableness, this often involves a strategic manipulation of linguistic indeterminacy.
interpretation of key legal concepts is always subject to the shifting imperatives of institutional power
The court reserves for itself the power to decide what counts as reasonable on a case-by-case basis. Definitions that prove expedient in one context may be swiftly discarded in another. While skilled advocates may seek to manipulate this ambiguity to their advantage, the ultimate authority to fix meaning rests with the judge – or, in some instances, with a higher court on appeal. The result is a system in which the interpretation of key legal concepts is always subject to the shifting imperatives of institutional power.
This example highlights the broader significance of the Language Insufficiency Hypothesis. By attending to the ways in which abstract and contested terms can be strategically deployed to serve institutional ends, we can develop a more critical and reflexive understanding of the role of language in shaping social reality. In the process, we may begin to glimpse the complex interplay of power and meaning that underlies many of our most important professional and political discourses.
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 