X Is Not X | WTF Alfred?!
Updated: Oct 8
There comes a point in any sufficiently prolonged encounter with the internet when a perfectly ordinary word starts to look a little suspicious. For me, that word recently became content.
A piece of content. Content creator. Content strategy. Content marketing. Content consumption. Content engagement. Content performance.
Content everywhere.
Yet the more I thought about the word, the less certain I felt that everyone meant quite the same thing.
So I started making little videos: Content Is Not Content.
Then a few sub-videos, including: A View Is Not A View, Attention Is Not Attention, A Number Is Not A Number.
At first, these phrases felt like jokes. Then they started to feel less like jokes. What exactly do we mean when we say a view?
Someone watched a video. Or someone watched three seconds, or thirty seconds. They watched it twice, or even scrolled past while the video played automatically.
A system counted an event, according to rules that remain partly hidden from the person who uploaded the video. The platform processed that event and produced a number. I looked at the number and I called it a view.
Now I had a view. Except I didn’t have the thing itself. I had a number, and I had a word. Somewhere between the two, I had made a map. And that brings us, relatively neatly, to Alfred Korzybski.¹
Korzybski developed what he called general semantics during the first half of the twentieth century. His work grew partly from a question that seems almost absurdly contemporary: human beings had developed extraordinary abilities in science and engineering, yet misunderstanding, conflict and destructive behaviour continued to flourish.²
His 1933 book Science and Sanity developed the idea in considerable detail. Korzybski originally planned a title involving time-binding — his term for the human capacity to accumulate and transmit knowledge across generations.³
We now possess an enormous machine for time-binding. We call it the internet.
Someone writes something. Someone records something, photographs something, uploads something. Someone else encounters it. Someone reacts. Someone measures the reaction. Someone measures that measurement. Someone turns the measurement into a graph. Someone writes a report about the graph. Someone makes a video about the report. And then someone comments underneath the video.
And in the middle of all this, a human being tries to remember what they originally wanted to say.
Korzybski’s central concern involved abstraction: the nervous system does not deliver the world to us whole. We select, filter, organise and name. We abstract. His phrase “consciousness of abstracting” points towards awareness of that process — awareness that whatever we experience or describe has already undergone selection.⁴
A map can help us navigate a territory. But the map does not contain the territory. Neither does the word. Neither does the number. In fact, neither does the category, nor the thumbnail, nor the algorithmic profile assembled from thousands of tiny behavioural signals. And neither does this essay.
That last point probably deserves a footnote.⁵
X = X
Consider the apparently innocent equation: X = X
It looks almost embarrassingly obvious. A tree equals a tree. A person equals a person. An event equals an event. A feeling equals a feeling. A video equals a video. A view equals a view. Obvious, no?
Yes, of course. Except the moment we name something, we have already done something to it. We have selected it. Separated it from everything around it. We have placed a boundary around an otherwise continuous field of events. And we have given ourselves something we can point towards.
Useful? Absolutely. Complete? Not at all. And that’s a vital distinction.
The category tree allows me to communicate with another person. I can say “there’s a tree over there” and, assuming reasonable circumstances, another human being can probably locate the object I mean.
But the word does not contain the tree. It doesn’t contain its history, its microscopic structure, the movement of water through its tissues, the insects living on it, the precise pattern of light falling across its leaves, or whatever the tree might mean to someone standing underneath it.
The label really helps. It does leave things out, though. And the trouble really starts when the label disappears from view. That’s because we stop noticing the abstraction and start treating the abstraction as the thing itself.
We don’t just have a number representing some collection of platform-defined events; we have views. We don’t just have a fluctuating set of responses; we have engagement. We don’t just have a collection of files, posts, clips, images, essays and songs. We have … content.
And then the language starts to do something peculiar.
The strange power of a tiny word
Consider these sentences:
I am tired.*⁶
She is difficult.*⁷
He is successful.*⁸
This is art.*⁹
That is rubbish.*¹⁰
The video is popular.*¹¹
The algorithm is confused.*¹²
Each sentence performs an astonishing amount of compression.
“I am tired” takes an experience occurring now and attaches it to an apparently stable entity called I. “She is difficult” converts an enormous quantity of interactions, perceptions and interpretations into a property. “He is successful” hides a measurement system behind a noun. “This is art” turns a complex encounter into a classification. “That is rubbish” performs the same operation with considerably less diplomacy. “The video is popular” turns a collection of measurements into an apparently intrinsic quality. And “the algorithm is confused” performs perhaps the most entertaining trick of all: it takes a system whose internal operations I cannot fully observe and gives it a human psychological state.
That little verb, “is”, does a lot of work.
Forms of the verb to be can express identity, classification, attribution, location and other relationships. General semantics has paid particular attention to the way such constructions can encourage us to collapse different levels of abstraction. E-Prime later offered one practical experiment: write English without conjugations of to be.¹³
So: The algorithm is confused.
might become: I cannot determine why the algorithm produced this result.
Something changes. The second sentence contains uncertainty. The first sentence conceals it. The first gives me a thing called the algorithm and a state called confused. The second gives me an observer, a limitation and an event.
Neither sentence gives me direct access to the underlying machinery. But the second sentence leaves more room for that ignorance.
Let X = X
Laurie Anderson gave the phrase another life in “Let X = X”, from her (I think) wonderful 1982 album Big Science.¹⁴
The phrase has always appealed to me partly because of its ridiculous simplicity.
Let X equal X. Fine.
But what exactly have we allowed X to remain? Anderson has spoken openly about the relationship between her artistic practice and Buddhism, including meditation and the way those practices have shaped her understanding of perception, identity and experience.¹⁵
I don’t want to turn that connection into a grand theory. General semantics does not secretly equal Buddhism. E-Prime does not secretly equal Zen. Laurie Anderson does not provide a missing footnote to Alfred Korzybski.
But something interesting happens when these different lines of thought begin touching the same problem. We mistake our descriptions for the described. We mistake the category for the thing. We mistake the map for the territory. We mistake the measurement for the event. We mistake the story about the experience for the experience. We mistake the representation for the represented. And then, occasionally, something interrupts the mistake.
The algorithm
The phrase “the algorithm” makes an especially convenient object for this kind of thinking.
The algorithm rewarded me. The algorithm buried me. The algorithm wants shorter videos. The algorithm hates long essays. The algorithm likes faces. The algorithm prefers hooks. The algorithm knows what people want. The algorithm has decided that my video deserves 900 views.
Except … which algorithm? Which system? Which signals? Which moment? Which experiment? Which viewer behaviour? Which recommendation surface? Which measurement?
Platforms contain many interacting systems, and the details change constantly. Yet that singular noun remains wonderfully convenient.
The algorithm. One character. One antagonist. One invisible intelligence. One thing we can blame.
Once again, language compresses. And once again, compression helps. At least up to a point.
Without abstraction, we could hardly communicate at all. The point does not involve abandoning abstraction. The point involves noticing abstraction while using it. That small shift changes the game.
A view
Suppose someone watches one of my videos.
Something happens. Light reaches eyes. Sound reaches ears. Attention moves. Perhaps an emotional response occurs. Maybe boredom. Maybe curiosity. Maybe nothing particularly memorable. The person watches for five seconds and leaves. Or maybe they watch the entire thing twice. They click “like”. Or they don’t.
Some system records something. The system converts that event into data. The data enters a measurement system. The measurement becomes a number. The number appears on my screen. I experience a tiny surge of pleasure. I call the number a view.
At that point, the original event has travelled a long way. The number can now affect my behaviour.
I might make another video. I might change the thumbnail. I might shorten the introduction. I might alter the title. I might decide that people want something else. I might even start trying to make videos that the number appears to reward.
The map has started changing the territory. Or maybe the territory has always included the map. Either way, the distinction becomes difficult to maintain.¹⁶
Content
The same thing happens with content. That single word performs an extraordinary act of flattening.
A song becomes content. A poem becomes content. A tutorial, a political speech, a joke, a documentary, a dreamcast all become content. A 25 minute argument about the philosophical implications of platform metrics becomes content.
And once everything counts as content, everything can enter the same machinery.
Views. Likes. Retention. Shares. Comments. Followers. Watch time. Conversion. Performance. The language - a language of aggregation - begins to suggest interchangeability.
The artist and the advertiser can now discuss “content”. The platform and the creator can discuss “content”. The viewer can consume “content”. The algorithm can recommend “content”.
And underneath all of this, the actual thing keeps happening.
Someone makes something. Someone encounters something. Something happens between them.
I guess the word content helps us coordinate that activity. But it also makes the activity easier to forget.
Difficulty
This brings me to another word. Difficult.
The difficult word doesn’t optimise. The meme does.
A difficult idea asks the reader to remain somewhere uncomfortable. A difficult piece of music may refuse immediate recognition. A difficult image may resist interpretation. A difficult essay may take twenty minutes. A difficult thought may require several attempts.
The platform can measure whether someone stayed. It cannot easily measure what happened during that staying. And that distinction says more than the number.
I keep returning to the word difficult because language can perform another little trick here.
“It’s not difficult.” People use the phrase casually.
They might mean “I understand it”, “you’ll manage”, “don’t worry”, or “I don’t want to discuss why you find it difficult.” They might mean “I have forgotten how long it took me to learn this.”
The same sentence can conceal radically different territories. The word doesn’t settle the matter. The word opens it.
The experiment
So this essay has tried an experiment. I have attempted to remove forms of to be from the main body.
Not perfectly. Not absolutely. The title gets a pass. Quoted examples get a pass, although I have marked them with an asterisk.
The asterisk signals a deliberate breach. The superscript number sends the reader somewhere else. And that ‘somewhere else’ waits until the end.
I wanted the notes to remain invisible while the main text continued. At first, they behave like ordinary footnotes. Then they begin to question the sentences that summoned them. Then they begin questioning one another. Then they begin questioning the distinction between text and footnote. Eventually, they question you, dear reader. But you may well have already started questioning them.
This experiment therefore has another layer.
This essay talks about maps, while the footnotes create another map. This essay talks about abstraction, but the footnotes abstract the very same essay. As this essay talks about categories, the footnotes categorise the sentences.
While this essay talks about measurement, the superscript numbers measure the violations. This essay talks about hierarchy. Its own footnotes sit beneath it. Literally.
At least for a while. Before something strange happens. The notes accumulate. The main text continues. The notes grow. Until eventually the notes contain arguments that the main text never made.
The footnotes contain qualifications. Objections. Corrections. Jokes. Contradictions. And even a little rebellion.
X Is Not X
I don’t think the point involves declaring that nothing means anything. Nor does it involve throwing away categories, language, measurement or abstraction.
We need maps. We need words. We need numbers. We need categories, models, algorithms. We need the shorthand they enable. We need all the things that allow one human nervous system to communicate with another. But perhaps we also need occasional reminders that the shorthand came second.
The event came first. The territory came first. The experience came first. The relationship came first.
The label arrived afterwards. And quite often that label becomes so familiar that we forget the arrival.
So: X Is Not X.
There. I have broken the rule. I think I had to.
I think the sentence maybe demonstrates the problem more effectively than an explanation could.
It may well be that X never needed rescuing from X in the first place. Perhaps the interesting thing happens in the gap between them. That gap contains the whole essay. Or maybe it contains nothing.
Look again.⁸⁵
And then, if you want, read the footnotes.
⸻
Footnotes
1. Korzybski makes a useful starting point because his project concerned the relationship between language, abstraction and human behaviour rather than merely the accuracy of individual words. The Institute of General Semantics describes his work as an attempt to develop a general theory of human evaluation, with consciousness of abstracting forming a central part of the approach.
2. I have compressed a complicated intellectual history into a paragraph because essays have limited patience. Korzybski’s experience of the First World War contributed to his thinking, as did his observation of the contrast between rapid scientific and technological development and persistent difficulties in human affairs.
3. Korzybski’s concept of time-binding described the human capacity to pass knowledge across generations, allowing later generations to build upon previous achievements. His 1933 Science and Sanity developed the wider system; contemporary Institute material also notes his original intention to use Time-Binding: The General Theory as the book’s title.
4. The phrase consciousness of abstracting carries more weight than I have given it here. Korzybski’s point involves recognising that our nervous systems select characteristics from events rather than somehow receiving total reality without mediation. The Institute’s explanation uses the useful formulation that awareness of an event does not equal the event itself.
5. The essay has now performed its first small sleight of hand. I have written “this essay” as though the words on the screen could contain a stable object called the essay. They cannot. You have an arrangement of marks, rendered by software, presented through a device, interpreted by a nervous system, and currently called an essay. The map has already begun.
6. I could have written “I feel tired.” That construction still contains abstraction, obviously. E-Prime doesn’t magically produce unmediated experience. It merely removes one particularly convenient grammatical mechanism for turning experience into identity.
7. Who supplied the difficulty? Her? Me? Someone else? A previous interaction? A cultural expectation? A memory? A projection? The sentence quietly encourages us to search for an essence called difficult inside the person.
8. “Successful according to which measurement?” creates an awkward pause. Financially? Socially? Artistically? Algorithmically? Personally? Spiritually? According to whom? The original sentence hides all of those questions behind one adjective.
9. “This is art” can function as a declaration, a classification, an invitation, a provocation, a marketing decision or an attempt to end an argument. Context performs a considerable amount of work.
10. “That is rubbish” can perform a similar trick with considerably less patience. It can describe a judgement while disguising the person making the judgement.
11. A platform can report popularity through numbers. The numbers can describe particular forms of activity. They cannot, by themselves, settle what popularity means to the people involved.
12. I have no evidence that an algorithm experiences confusion. I have only experienced confusion while attempting to interpret algorithmic behaviour. The sentence therefore gives my uncertainty to the algorithm.
13. D. David Bourland Jr. developed E-Prime during the 1960s. The system removes conjugations of to be from English, partly to make writers distinguish more carefully between identity, classification, observation and inference. The Institute of General Semantics describes E-Prime as an application of Korzybskian principles to ordinary English.
14. Laurie Anderson’s “Let X = X” appeared on Big Science in 1982. The phrase provides an almost perfect companion for this experiment because it looks like a proposition while refusing to tell us much about the X on either side of the equation.
15. Anderson has discussed Buddhism, meditation, art and death in several interviews, including a 2026 conversation in which she described art and Buddhism as deeply intertwined in her life. I don’t want to turn her remarks into evidence for an argument she didn’t make. The connection belongs to her practice, not to my footnote.
16. “Either way” quietly tries to rescue the distinction it just weakened. Perhaps maps alter territories. Perhaps territories produce maps. Perhaps both descriptions depend upon a model that already assumes some separation between map and territory. Footnote 16 therefore refuses to settle the matter. ¹⁷
17. The footnote has now started doing something the main text cannot easily control. ¹⁸
18. Consider like. The word can describe pressing a button, experiencing affection, expressing agreement, making a social gesture, participating in a measurement system or producing a data point. One small word crosses several territories. ¹⁹
19. And perhaps the platform’s achievement involves making these meanings useful to one another without making them identical. A button can generate a metric. A metric can influence a creator. A creator can influence a viewer. The viewer can press the button because of something entirely different from the reason the platform values the resulting data. ²⁰
20. Content can therefore function as an extremely efficient abstraction. It allows radically different objects to enter a common technical system. A poem and an advertisement can occupy the same dashboard. ²¹
21. The footnote now qualifies the footnote. This probably counts as a warning sign. ²²
22. The original essay says that language can open a question. This note wonders whether language can also close one by pretending that the question already has a stable answer. ²³
23. We have reached the point where the apparatus no longer merely explains the text. It comments upon the act of explanation. ²⁴
24. Which sounds grander than it deserves. ²⁵
25. The main text asked the reader to notice abstraction. The footnotes have created another abstraction around the main text. We now have text, footnote, commentary on footnote, commentary on commentary. ²⁶
26. This could continue indefinitely. ²⁷
27. It won’t. ²⁸
28. Probably. ²⁹
29. The previous two notes demonstrate an interesting property of footnotes: once granted permission to multiply, they can generate their own little ecology. One note refers to another. Another disputes it. A third comments upon the dispute. Eventually the original sentence looks like an excuse for the apparatus. ³⁰
30. Which perhaps makes the apparatus the territory. ³¹
31. Or the main text. ³²
32. Or neither. ³³
33. Someone reading this far may reasonably ask when the actual conclusion will arrive. ³⁴
34. That question already assumes a distinction between the essay and its conclusion. ³⁵
35. Fair enough. ³⁶
36. The footnote has now become argumentative. ³⁷
37. It also has developed a tone. ³⁸
38. I did not explicitly give it one. ³⁹
39. Or perhaps the tone emerged from the accumulated constraints. ⁴⁰
40. Or perhaps I have simply projected personality onto punctuation. ⁴¹
41. The same thing happens with algorithms. ⁴²
42. And cats. ⁴³
43. I have no useful evidence for the cats. ⁴⁴
44. We should probably return to the essay. ⁴⁵
45. We cannot. ⁴⁶
46. The essay ended several paragraphs ago. ⁴⁷
47. The footnotes continued. ⁴⁸
48. The reader continued. ⁴⁹
49. Therefore something has happened after the supposed ending. ⁵⁰
50. Or perhaps the ending never contained the authority we assigned to it. ⁵¹
51. Which gives us a problem. ⁵²
52. Who decides where the main text ends? ⁵³
53. The author? ⁵⁴
54. The page? ⁵⁵
55. The formatting system? ⁵⁶
56. The reader? ⁵⁷
57. The footnote? ⁵⁸
58. Note 58 would like to nominate itself. ⁵⁹
59. Note 59 objects. ⁶⁰
60. Note 60 has no objection but would like to register an interest. ⁶¹
61. We have now created a committee. ⁶²
62. The committee has no jurisdiction. ⁶³
63. This may represent the most accurate description of an algorithm yet produced in this essay. ⁶⁴
64. The footnotes have overtaken the text. ⁶⁵
65. Except the sentence above still belongs to the footnotes. ⁶⁶
66. So perhaps they haven’t. ⁶⁷
67. Perhaps the hierarchy has simply become harder to locate. ⁶⁸
68. Perhaps that provides a better ending than another conclusion. ⁶⁹
69. The title remains at the top of the page. ⁷⁰
70. It contains the forbidden verb. ⁷¹
71. I left it there deliberately. ⁷²
72. The title names the experiment. ⁷³
73. The essay performs the experiment. ⁷⁴
74. The footnotes complicate the experiment. ⁷⁵
75. The reader completes it. ⁷⁶
76. Or not. ⁷⁷
77. That final distinction matters. ⁷⁸
78. I almost wrote “The experiment is complete.” ⁷⁹
79. I didn’t. ⁸⁰
80. You noticed. ⁸¹
81. Perhaps. ⁸²
82. And now the footnotes have finally reached the point where they can stop explaining themselves. ⁸³
83. Unless this note has already failed to stop. ⁸⁴
84. It has. ⁸⁵
85. Look again.
See the audio-visual version of this essay ... content ...
Comments