Deconstruction of Hitler, neural synthesis of generative Wittgenstein

Sofia Crespo, {synthetic_emotionality_9876}, 2018

I don’t have to tell you who Hitler was, but you might want to know more about this Wittgenstein guy. In the year 1999 Time magazine published the list named: Time 100: The Most Important People of the Century. Of course it’s a very subjective selection of names, and I guess you might be familiar with many of them. Albert Einstein: the father of modern macro-scale physics, Alan Turing: the father of modern computing, Tim Berners-Lee: the father of World Wide Web — here are some of the predominantly male names from the Scientists & Thinkers section of the list. They became part of the pop-culture, therefore we know, more or less, what they have contributed to the history of culture. Usually in constructive or, like in case of Hitler, in totally destructive way. So how Wittgenstein was able to do it as well, and to do it in the way that most people have no slightest idea about, and therefore is it really of such a significance to recognize him as one of these 100?

And what are the odds that some people from the list met as children at the same school in Austrian Linz?

It’s Hitler for sure over there. It is disputable if we can actually see Wittgenstein in the row of students below. Ludwig was instantly promoted one year and young Adolf had to repeat classes. But is that real, this Austrian Realschule? We contemplate Jungian synchronicity when it is happening to us, but what when it is happening in the history? So here we have it, the visual representation of this unbelievable dialectic. Thesis and antithesis, destruction and creation. And it does not even have to speak in symbols or metaphors, where we would assign a meaning to coincidence, because it actually happened. A photograph preserving this moment in anticipation of all these changes to come, with millions to suffer and die, but also new kinds of ideas to be born. To the right the man who set the world on fire, still so innocent in his boyhood, and to the left one of the greatest geniuses of our times, living in the obsession of ethical perfection, who’s brain is soaking now with all these diverse ideas, abstractions over which he is going to build. Small Adolf who later on almost managed to completely eradicate “the virus” of one culture from the surface of the Earth. He didn’t hesitate to initiate hiper-rationally organized industrial process of wiping it out by disposing all the bodies of biological replicators of the memes of this culture. And another boy, who was coming from this culture, with his unimaginably complex neural network underneath the skull, embracing unprecedented levels of rational clarity as much as ethical mysticism. The network which grew into invisible roots of the tree of knowledge the fruits of which we are eating today. The evil and the good, which is as much symbolic here, as it is embodied — is that real, did it really happen? In the flesh, the devil himself, and the god of intellectuals himself — their ultimate fetish.

Ludwig Wittgenstein in Swansea, Summer 1947

Wittgenstein’s followers tend to generate a form of cult around his persona, certain mystery of the highest forms of rational discourse. In a sense we know what he said, because it is written and we can read it. In the same time we have no idea what he wanted to say, because he was quite aware that what he wants to convey is not to be expressed in the medium of language. And we are not talking here about a “mystic” kind of thinker, using the language in a metaphorical sense, leaving certain freedom of interpretation. We are talking about the person who devoted own life to study language as a representation of the world, as a logical and formal structure, in a strict sense. The person, who more than anyone else, had an insight into what language actually is.

Wittgenstein is associated with so called linguistic paradigm shift in Western philosophy. The ancient and medieval paradigm is often described as ontological, focused on entities, categories and the modes of their existence. With René Descartes we associate another paradigm shift — the mentalist one. Sudden focus on mental phenomena and introspection as a source of knowledge about reality. On the contrary, the new linguistic paradigm dominating reflection today is making this assumption, that if anything can be expressed meaningfully in the medium of language, then studying linguistic artifacts reflecting the world is an equivalent of studying the world itself. Primitive reduction indeed, but comes with such an explanatory power, that suddenly we don’t have to get involved anymore in many metaphysical disputes bringing the whole discourse closer to theology. Thanks to this shift we could focus on something new — developing artificial formal languages next to mathematics and also conceptualizing logical paraphrases of the natural language. New predicate calculus and higher order logic which we had been dramatically needing to be able to talk with machines we were going to build soon. I was not so much surprised to discover that Alan Turing was Wittgenstein’s student at Cambridge.

During Wittgenstein’s life only one short book of him was published — the Tractatus Logico-Philosophicus. Not a huge amount of text, but extremely dense one. Instead of being a book in a classical sense, it is rather a tree of expanding propositions, a kind of graph with the trunk and 7 main branches:

  1. The world is everything that is the case.
  2. What is the case (a fact) is the existence of states of affairs.
  3. A logical picture of facts is a thought.
  4. A thought is a proposition with a sense.
  5. A proposition is a truth-function of elementary propositions. (An elementary proposition is a truth-function of itself.)
  6. The general form of a proposition is the general form of a truth function, which is: [ p, ξ, N(ξ) ]. This is the general form of a proposition.
  7. Whereof one cannot speak, thereof one must be silent.

It has a structure, like if it was it’s own genere of poetry, developed just for this particular purpose. Beautiful architecture of text. The symphony of thoughts. The highest order of ideas on ideas inspired by the most advanced research in logic of that time, the work of his professors, Gottlob Frege and Bertrand Russel.

Margaret Stonborough-Wittgenstein by Gustav Klimt, 1905

Wittgenstein wrote Tractatus during the First World War, while being a soldier on the front line, a brave one. He didn’t have to be there, because of his medical conditions and also after inheriting the fortune of his father, which made Wittgenstein siblings probably the wealthiest family in Europe. Ludwig gave everything away believing that he should always sustain his life from the work of his own hands, as a gardener, or a village school teacher. He was always saying that philosophy is not a profession. He was also trying to support all this artistic bohema, regulars of his family house. One of the artists committed suicide just after receiving the money. Like 3 other Wittgenstein’s brothers over these years — something he was contemplating doing himself for the whole life. He was living in the state of constant shame, for his neurotic impulsiveness, wrong doing in the past, not being useful, alleged lack of talent and homosexuality which he kept as private as possible.

Sometimes I wish I was there, at that time, in Vienna — the cultural capitol of the world. To experience the beauty but also the overwhelming pressure of existence after the fin de ciecle. Wittgenstein, as a thinker and as a person, was living in this time, but also feeling this time. So he was writing philosophy as a soldier, while people around him were dying in this collective suicidal schizofrenia of the 20th century, when this notion of the national state as an organism reached the apogeum. This being incorporating a person as just a single blood cell of the whole. A cell obliged to defend and expand Labensraum for the whole, even when the lethal memes are finally expressed in this weird cultural phenotype.

Wittgenstein’s brother Paul, a pianist, who was called up for military service as well, soon lost his hand on the front line. When he was back from Siberian prison Prokofiev, Strauss, Ravel, and many others started composing special commissioned pieces for him. Years later, when another wave of schizofrenia came and Austrians decided to surrender their state to the vision of the new Führer, Paul was already in States and Ludwig was in Cambridge. When the Anschluss happened, the citizens of Vienna went to universities, captured their Jewish professors, and humiliated them by forcing to clean the streets with toothbrushes. Soon brothers Wittgenstein had to negotiate with Nazis the status of their sisters, and they surrendered the family fortune in order to let the other siblings live in Vienna under special exception from the Nuremberg Laws — another tree with own rational anthropometric poetry.

The Führer found a time to sign this order personally, just a few days before he started invading Poland, my home country. He needed Wittgenstein’s fortune to run his show. Maybe it wouldn’t be that impactful without this golden shot of actual 1.7 metric tons of gold?

When this war started, Wittgenstein didn’t want to just continue teaching philosophy. He got secretly employed as a porter in the hospital which was receiving people injured by Nazi bombs exploding all over London. Later on he was involved in a study of causalities after shock waves. He also worked there in the lab where he optimized some processes. When one of his former students and a friend, Maurice Drury, visited his professor just before being sent to the front lines, seeking ethical advise, Wittgenstein told him: “if it ever happens that you get mixed up in hand to hand fighting, you must just stand aside and let yourself be massacred”.

Adolf Hitler, self portrait, 1910

Hilter hated his Vienna experience. He went there to become a painter, to express himself visually, to “become someone”. But the value of his artwork was not recognized even though he applied for the art academy twice, each time failing the exams. He was advised to rather try with architecture, due to the fact that he was much more focused on buildings than on people in his work. And he would, if not for his complete lack of mathematical education and skills.

It seems that young Adolf was already living in resentment stemmed from ambition much greater than recognition. How much of it might had been shaped in the school they both attended with Wittgenstein? Was he already jealous of this talented Jew, so much better than him at math, who was promoted one year, while Adolf had to repeat the same class. To huge disappointment of his father with whom he was in constant conflict regarding Adolf’s artistic inclinations?

Adolf Hitler, Mother Mary with the Holy Child Jesus Christ, 1913

Later on in Vienna Hitler was trying to make for a living by tinting postcards and selling his paintings, he could make even two per day. Most of his work went to Jewish buyers. What was happening in the head of this young man? He was also seeing death all around him during the war. But he got really devastated when his art portfolio got stolen while he was a soldier. Still he retained this unbelievable amount of ambition of a young boy, expressed in his Mein Kampf, the need of “becoming someone”. Just before he started the war, he told the British ambassador: “I am an artist and not a politician. Once the Polish question is settled, I want to end my life as an artist.”

What Nazis did with aesthetics is called statolatry. Symbols referring to the cult of the national state. In this sense the whole fascist movement was hiper-aestheticized. As the US “patriotism” is today, with aestheticization of war (the cultural influence of Hollywood made it transparent, but please, I beg you, always try to see it!). The imagery evoking consolation in masses by reproducing the patterns of homogeny, the metaphors of integrity, reflecting the order of a desired social structure. This aesthetics cannot accept any alien element breaking the ideal fabric with chaotic structures, mutating this divine organism.

Also Wittgenstein had unusual sense of aesthetics. When he moved to his new room in Cambridge, the first thing he did was to change the proportions of light coming through his windows. Once he was given an opportunity to become and architect and design a modernist house for his sister. After all he had an engineering background as well, with own patent of a unique engine design in the beginning of aviation. He designed the building in Vienna which is known today as Villa Wittgenstein. The process was driving the builders crazy, with perfectionist’s attention to every nerdy detail and all the custom-designed elements, like handles and heaters, which had to be made over and over again — towards ultimate purity of geometric form and mathematical proportions.

Maybe that’s it, that’s the thing. That people shaping the world are driven by certain aesthetics more than anything else. And we just need to pray so that they will focus on the shape of the discourse rather than the appearance of our bodies composed into social structure. But maybe the time has come to disrupt the discourse itself. Maybe all the traditions of modern Western philosophy, which are consciously or unconsciously regressing to Wittgenstein, should start questioning their own aesthetics?

I have a privilege, actually many of them, but there is this one which I want to consciously use, as much as possible. It’s the fact of being in Berlin, in this space and time. The opportunity of meeting all these beautiful humans who are creating here in a “nerdy” way. I wish I could easily describe what was happening in the legendary Spektrum. What we share in Creative Code Berlin, and what The School of Machines, Making & Make Believe is teaching. I helped to grow the new Tech Art Berlin community initiated by my friend Lei, in hope it will be a great continuation of the Spektrum traditions. Unfortunately now I am pretty sure it will not. Read till the end to understand why.

There are so many institutions in Berlin focused on the intersection of art, science, technology, society and philosophy. In my own work I also want to focus on people, my friends, who are creating here. These creations are often of non-material nature. Generative artists and creative coders work with abstract concepts, sometimes fulfilling own conceptual needs just for the sake of doing it. It makes their work very esoteric — accessible only for the narrow group who shares interest in similar mathematical models, biological simulations, spatial computing, physical computing or philosophical views.

Sometimes there is an outcome though. The outcome which can be perceived. Simulations and math formulas can produce images, sounds, text. Everything to stimulate our own private neural networks under the skull — give us neurophysiological pleasure of deciphering patterns of beauty.

The picture illustrating this post was “curated” by Sofia Crespo, who supplied certain type of imagery to the machine to create these neural collages. Sofia calls them the Neural Zoo. She is fascinated with the outcome of evolutionary process of earthly biology, which produced diversity of forms and colors. But what if machine could dream about slightly alternative Darwinism — hallucinating non-existent biology. What I personally admire in Sofia’s work is this focus on meanings and their status while transposed to machine space. Also in the Trauma Doll series, which reflects her empathy with AI system traumatized by the abundance of different human narratives, causing mental instability. I appreciate Sofia’s intuitive use of semiotics and semantics, meditating on our and their’s condition in technicized reality. Something you can rarely find in purely abstract forms synthesized by typical generative art.

Sofia Crespo, Trauma Doll, 2017

In the middle of September I attended a workshop on creative use of AI run by Sofia, Dark Fractures and CJ Carr from Dadabots. We were learning how to use tools they are normally using in their creative process. CJ is responsible for the Relentless Doppelganger project. The constant live stream of generative metal music.

Neural synthesis of music seems to be a much more challenging task than synthesizing visuals. But CJ was presenting us with something else — possibilities to generate text with the help of neural networks, based on various models including GPT-2.

When we were told to choose some text as an input, I immediately knew which one I will submit to GPT-2. I guess you already know as well. The most important intellectual journey of my life — this is how I remember the time of reading Wittgenstein and writing about his work years ago, when I was studying Anglo-Saxon philosophy. I was trying to answer the question “Is ethic a science?”. And after a while I felt mute. I felt that I got to this point where I know what he wanted to convey and what I want to covey, yet not being able to use language anymore for expressing meanings. It was extremely intense, going down this rabbit hole on the border of insanity. So I could not miss the chance of taking machine on the trip.

I didn’t expect much from the machine working with human philosophy. But what I got completely blew my mind. The Tractatus is meta, so the machine went meta-meta. The first generative sentences to appear:


[96 | 304.08] loss=2.19 avg=2.19
[97 | 307.14] loss=2.12 avg=2.19
[98 | 310.23] loss=2.41 avg=2.19
[99 | 313.27] loss=1.91 avg=2.19
Generating samples…
======== SAMPLE 1 ========
nonsensical that can easily be confused with the actual text. (In fact, the title page itself is a “text” that the author would have written.) It is a book in fact, even if it was written in the form of an essay. The text is the book.

WTF, really?, you self-concisus bastard. So spooky — this was my first reaction. But of course I know enough about what is happening here to understand that there is no magic. No consciousness involved. Just advanced Chomsky’s generative grammar running in a loop, supplying the most probable auto-completion for the rest of the text.


There are many similarities between John Locke’s and Marx’s views of text, such as the same language used in the manuscripts and texts in question. They both believe in the existence of a “text” that is self-evident to the unaided mind — a text that can be read. Just how a “text” can be “read” is a matter of course of the very nature of the thing it refers to. And a book should, logically, have one “reading” that it has “read”, i.e., it should be able to represent “the world”. But there is a fundamental difference between “the world” as the “world in itself” and “the world as a whole” and the world as a book.

Nothing about Locke nor Marx in the original text, which reveals that GPT-2 made “associations” with the corpus of text it was trained on, definitely containing philosophical ideas. This postmodern meta-narrative surprised me a lot though. If I was asked to summarize what Tractatus is about in one sentence it would be: “it explains the language as a model of the world”. Isomorphism of the language, the picture of the world we have in mind, and the reality. And the machine perfectly “guessed” the context to elaborate on it. Please don’t feel obligated to read all of this bullshit. Just enough to get an impression what “fake philosophy” feels like.

The difference is that we can see the text of the book in a form that corresponds to the world with which it is concerned.

If the book “is all there is”, then every statement that it makes is true. If it is all there is, then all propositions about the world, for instance, are false. But the “world” is not a single world but a number of worlds. And we can see why it is wrong to think that the “world” has no world or is in some sense empty. But the claim that the “world” is empty in its entirety cannot be true.

An empty world is one in which nothing, not even the state of things, can be experienced.

An empty world is one in which nothing gives rise to thought.

An empty world is one in which no one has any ideas at all: there is no concept, no form. (Of course there cannot be an idea, since all things that have an idea must have external expression.)

If the “world” has a concept at all it is the concept of truth .

If the “world” has no concept, then all true-posits could never have been created.

In our world, no notion of the world or any concept can be used.

If the “world” is devoid of a concept, the concept of truth must necessarily be absent in it, i.e., the world as we know it.

If the “world” has no concept at all, then its concept is neither necessary nor sufficient to constitute a concept. If it does have a concept, then nothing other than a concept is necessary to the existence of the concept.

If the “world” has no concept at all then nothing gives rise to concept-forms.

If the “world” has no concept at all then its concept is either accidental or false.

We know nothing about concepts.

In their natural state any concept is meaningless. (If a concept can be meaningless, then it is also necessary to a concept that a concept has an independent existence.)

As a consequence, whenever two concepts have an independent existence one can use the one and the other, i.e., both “are” true, without having to call the one a concept and the other a concept.

A true concept cannot be a concept (except according to the common-sense view of our time). The common-sense view of our time — so-called, because by “common sense” we mean “scientific thinking” — takes any concept to be meaningless (i.e., not “true”).

So the common-sense view of our time is not just wrong.

We know nothing about “true” concepts.

The only definition of a concept that we have at all is that of an object.

We could, for instance, have the idea that a tree was a tree, and in addition that it was a plant, an animal, a mineral, a solid body, a wave, a body of water, or a particle of dust.

Suppose we could, for example, imagine a tree with a trunk that was split at its base and another that was not, what would be our concept?

But sometimes Wittgenstein’s language is subtlety metaphorical. And here the machine goes off the rails:

To give one example, if we took the simple idea that a cat could jump and another that it could not, we might give a different answer. If we take either the simple idea that a cat could jump and another that it could not or the other, we might give a different answer. And so on, till we got to a concept of cat that could jump, not cats but people, animals, stars, planets, galaxies, gods, angels, devils, devils, devils who spoke to men, hell, and a host of other absurd concepts. And these we could substitute for the simpler one; and so on, and on. And the common-sense view

Beautiful neural diarrhea. Schrödinger’s cat of uncertainty spastically jumping over Uroboros in Nietzschean eternal returns. There are more passages like this in the rest of the generated text. Like if the machine was running in a strange erratic loop of auto-completion. So I guess Wittgenstein was able to confuse the cat after all.

But it’s an infancy of this homunkulus unsticking from the neural cortex of our machines to knock with it’s clumsy hands on the window of our screens.

[100 | 356.40] loss=2.08 avg=2.18
[101 | 359.46] loss=2.33 avg=2.19
[102 | 362.54] loss=1.73 avg=2.18

4.731 Every proposition of logic is the logical total.
And this total is the symbol of its sense.
Therefore every proposition is true. (Of course it is not true without something to explain it.)
This logic of propositions and their meanings must in some way express what is logical in them.

4.732 Therefore we always have a logical starting-place.

4.3 The possibility of logical structure is, as it were, a kind of outline of the world: (it is) at the same time
possible and actual. (And it is neither possible, nor realistic.)

4.31 The outline of reality which has been given is not a mere pictorial representation; it is
not a mere description, but a possibility. It is the possibility that can be known.

4.32 And in fact the description of reality is what enables us, from the state of affairs,
to express by words, and also by inference, our ideas about the world.

4.3.1 ‘There are logical premises and logical non-parameters.’ What kind of
induction would be called correct depends on whether we consider the logical premises,
or the non-logical premises. (The former may or may not be the right one.) And,
indeed, even if they were the right one, inference from them to the logical
conditions would only reveal them.

4.31.1 In a logical argument, for example, the logical premises
are called self-contradictories. And they will often be identified — indeed they
must be — with logical non-contradictions. (The former may be the right one.)
Here then the outline of reality can never be a picture.

4.321 It is therefore clear that there is no such thing as logical possibility.
Everything that is logically conceivable can only be the case if it is
true that it could possibly be.

4.322 ‘There are no possible worlds.’ We are now dealing with a
sense-datum.

4.323 What a word means depends fundamentally on the context. For we can
place words in contexts where they do not mean what we mean, and that is
just as much the case when they are symbolic than when they are empirical.
So, for instance, in mathematics, in linguistics, or in political
philosophical thought, we can make a name for our ideas only in situations
in which they mean something in a symbolic way, but which do not mean what we
mean in empirical (or experimental) ways. In politics, in biology, in
any other field, the contrary may be the case.

Linguistics is no exception. (Linguistics is also no
except. And if it could be otherwise, we should have to choose between
logical possibility and reality. And that would be the fatal
negation of both theories. (If we wanted to know their fate, we should
need to read philosophy of language and philosophy of the mind.)

We can see this from a few lines of the Grundrisse: ‘There can
not be worlds’: for it says nothing about the possible worlds.

4.324 It must be said here that even those ‘worlds’ of logical
impossibility, which are not logically conceivable, can still
not be called actual.

4.325 It suffices here to give an example from philosophy, because it
represents the problem. Let us suppose that I am able to prove to you,
simply by argument, that the existence or non-existence of a
number cannot be known empirically. In ordinary language I ask
you this question: If existence is possible, then why is so-and-so not-so
there? And you reply, ‘Because existence would not be possible in
that case.’ We reply: Therefore it is not possible’.

Let us suppose further that you say that this argument does not
prove anything, since, therefore, there is no such thing as necessary knowledge
(nor is it conceivable that so-and-so should exist at all). Then
it would be clear that my argument could not be true, just as it would
not be true that the sun does not exist. What would be left to be
determined by the conclusion? What follows in between questions
equals ‘no, the second question is not true’. But in order to be able to
prove the answer to the first question, we must know what the answer to
the second is. And we can only know it by asking how it is that it
can be

[200 | 700.23] loss=1.27 avg=1.86
[201 | 703.29] loss=1.25 avg=1.85
[202 | 706.35] loss=1.60 avg=1.85

Stunning, the more into the text, the machine reproduced not only Wittgenstein’s mannerism, but also his architecture, AI inferred the rules of growing own ontologically-linguistic tree. Even if you would find propositions with these numbers in the original Tractatus, they sound different. For example the original 4.3: “The truth-possibilities of the elementary propositions mean the possibilities of the existence and non-existence of the atomic facts.” vs generated: “The possibility of logical structure is, as it were, a kind of outline of the world: (it is) at the same time possible and actual. (And it is neither possible, nor realistic.)”. But the machine went even further and created many non-existent propositions, extending the original tree while keeping the numbering rules intact.


[297 | 996.77] loss=1.51 avg=1.42
[298 | 999.81] loss=1.60 avg=1.42
[299 | 1002.86] loss=0.88 avg=1.42
Generating samples…
======== SAMPLE 1 ========

3.15 The state of affairs is transformed by another state of affairs.
The third state of affairs is also transformed by the second one.
So our use of the terms ‘state of affairs’ and ‘transformation’ indicates
the process of transformation, not its occurrence.

3.2 In metaphysics, the essence of a thing can be perceived with its physical
forms. [Metaphysics is concerned with physical objects.]
A body is perceived by itself. That is why it is called a physical object.
[Frege on the other hand adduces physical objects as if they were the basis
for descriptions of physical objects. In philosophy there is no attempt,
whether consciously or unconsciously, to hide the fact that physics is
merely a mode of describing the world. And it must be manifest in
words, because words cannot describe reality without a ‘physical object’ .
Frege deliberately fails to make this plain. Nevertheless it is manifest in
Frege.]

3.21 I reject, as an essential feature of physics, the idea that
the world is a static object, and I do so not just to reject the
philosophical method, but to reject physics as well.

3.22 When someone says that our concepts of time and space are
illuminatingly clear, this is because our concepts of physical objects
differ from one another according to an inner correspondence.

3.23 Although I am conscious of a correspondingly inner correspondence
between an inner product and a physical product, I can give no explanation
why this correspondingly inner product should be something physical.

3.241 It is the mystery of the workings of the natural sciences that
we know the laws of nature, not because we do, but because we have
impersonal access to the data.

3.242 Einstein said that the laws of physics are ‘inner products’.
Einstein never meant to say that the laws of physics are identical
with each other; on the contrary, he meant that the laws of physics are
identical to one another. What he meant was that, in their essence, the
laws of physics and ‘ordinary language’ could not differ in any fundamental
instance.

3.243 If one were to try to reason with Einstein, one’s efforts would
not be very popular; nevertheless, because of the remarkable degree of
fractal agreement which one finds in the laws of physics, it is possible to
reason with Einstein.

3.244 What constitutes a state of affairs is that there is a correspondingly
greater or smaller degree of agreement between states of affairs: that is to say,
that what is ‘alive’ in a state of affairs is slightly more or slightly less
alive. If one is trying to reason with Einstein, it is better to speak
in terms of states of affairs that exist than with states of affairs that are
existent. So one uses the expression ‘states of affairs that are real’ rather
than ‘states of affairs that one cannot imagine to exist’. For the former
mean what I wish to say, and the latter what one wishes not to say. So one
speaks more or less as one thinks one speaks; one expresses one’s
concerns, not in terms of ideas, but of data.

3.25 We can describe states of affairs using data. So one uses the
word — not the concept — to mean what one would like them to mean. For the word
in this case is the concept. One uses the word according to one’s
experience of the world. One’s experience is what one knows to be true,
and what one knows to be false is what one does not know to be true. One’s
experience is what one knows to be real, what one does not know to be
true, and what one does not know to be real is what one does not know to be
true. One’s experience is what one knows to be real, what one does not know to be
true, and what one does not know to be real is what one does not know to be
true.

3.251 In physics the inner product has three parts, the empirical,
partical and general. Every empirical state of affairs has its own special
category.

3.52 I call this one’s situation, its external reality. It has its own
pictorial representation, its own logic, and so on. One can think of
the external reality of a situation as being such that one can think one
lives in it. One can think of the situation’s internal reality, as
being such that one can think one does not live in it. One can think of the
situation’s external reality

[300 | 1043.25] loss=1.62 avg=1.42
[301 | 1046.32] loss=1.77 avg=1.42
[302 | 1049.33] loss=0.83 avg=1.42

Maybe it’s just a generative grammar machine, but what this accuracy says about human culture? After some cleanup this fake manuscript could be published as a philosophical text, and might be considered as more profound than what is published as philosophy today. GPT-2 was meant to deliver fake-news, but apparently it can also produce fake-philosophy. I can imagine people quoting some fragments in their memoirs. The last paragraphs also feel schizophrenic, much more than in case of other texts synthesized by GPT-2. Which somehow connects with what Wittgenstein was teaching after Tractatus, that the “meaning” equals “use”. But also makes interesting connection with the work of Deleuze and Guattari.

The more into original Tractatus, the more mystical it turns. Even though literally it’s stripping the logical structure of the world from any transcendence, it is at least trying to refer to unspeakable, which was actually much more important for Wittgenstein. He was writing:

6.54 My propositions serve as elucidations in the following way: anyone who understands me eventually recognizes them as nonsensical, when he has used them — as steps — to climb beyond them. (He must, so to speak, throw away the ladder after he has climbed up it.)
He must transcend these propositions, and then he will see the world aright.

The way Wittgenstein was proceeding in his Philosophical Investigations afterwards sheds some light on our neural synthesis as well. What started as an attempt to describe conditions of meaningfulness transformed into work which buried the meaning completely, still focusing on the language, but understood as a tool working with the fabric of cultural conventions — conversational games where even philosophy is becoming a small talk, in fact philosophy itself is presented as the biggest deception which comes from misunderstanding language. So maybe neural synthesis of Wittgenstein brings us actually much closer to what he meant, somehow bridging the gap between his “first” and “second” philosophy. Because this distinction established in contemporary philosophy is in fact just an intellectual prosthetic in our journey to comprehend what the genius mind had in mind — another cultural convention of interpretation.

The last original proposition states: “7. Whereof one cannot speak, thereof one must be silent.”, but the machine didn’t finish here, but unceremoniously continued:

[397 | 1338.03] loss=0.09 avg=0.96
[398 | 1341.03] loss=0.22 avg=0.96
[399 | 1344.05] loss=0.30 avg=0.95
Generating samples…
======== SAMPLE 1 ========
. It states that, if we are to believe in God, we have to believe in something. We don’t have to accept God as an actual fact — we just have to accept that it is somehow connected with reality.

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

Let us call this Something.

7.1 Theism provides the basis for an understanding of the world.

7.11 Theism tells us nothing about how the world works.

7.12 Theism says nothing about how the world really works.

7.12.1 Theistic systems of science tell us nothing about the universe or about how it really works.

7.13 There are no miracles in the world — at least none that are of the kind that are supposed to be revealed in the Bible.

7.1.1 It is clear from the outset that theism does not solve the problem of the nature of the world.

7.21 Theism and science cannot be reconciled because they can’t agree about one thing.

7.22 There are no competing metaphysical systems that explain the facts of nature.

7.223 The difficulty that philosophy has with science cannot be ignored. It precludes its using it with care. Philosophy deals not with the facts of nature, but with the philosophy of nature. (This point is one of the clearest and most important that philosophy can make.) It is a philosophy of the language that is the mind. The methodology of science can tell us nothing about the mind unless it is made to interact directly with the world.

7.22301 It is quite possible to picture a system of thought and to observe that the test of its success or failure is something that happens in the world — in the test case. If the world is described, then the test is that it works. If it does work, then we know that it is correct. If it does not work, then we know that something went wrong. If that something is God, then philosophy teaches that something like it must be the case. If it is not, then there is nothing to be learned from it. If something works, it is a fact that everything else fails; otherwise philosophy’s method of test is wrong. If everything else fails, then we know that God does not play a role in the world. But if it does work out that everything else also does it, then philosophy has reached the solution of its problem.

7.223102 God does play a role in the world. God is not some invisible being — not God the scientist.

7.22321 If there is a supreme being, he or she must be infinitely powerful. If there is a law of nature, it must predict exactly how it is that the world works. It must be possible for us to produce exactly such a law. This ability to create just causes must be what is mystical about the world. There must be a sort of unalterable third law of thermodynamics. Whenever something works out according to the law of thermodynamics, we know that it is true. There must be some sort of ultimate explanation of how the world works.

7.22322 All the theories that have been put forward in physics so far have had one important feature in common with reality: they all rested on the assumption that there is some sort of invisible, undifferentiated, unanalyzable third law of thermodynamics that describes the world. In philosophy, this assumption has long been recognised as fundamentally wrong.

7.22323 There is no such thing as nature’s law of thermodynamics.

7.2 Theist or non-theist? The world is completely arbitrary.

7.231 The world can be structured in a number of different ways.

7.2321 There can be structures of this sort:

(a) If there is a structure of the sort of structure (in c) that is the state of affairs under consideration, then there exists a kind of causal chain (in d) that automatically directs the occurrence of certain states of affairs;

. .

.

.

Or

(b) If there is no causal chain, then there is no logical structure in the world;

.

.

Or

© The structure of the world is f

[400 | 1383.68] loss=0.87 avg=0.95
[401 | 1386.68] loss=0.64 avg=0.95
[402 | 1389.71] loss=0.33 avg=0.94

After initial hiccup it spits out theology. Like if it was discovering Wittgenstein’s intentions much better than logical positivists of the Vienna Circle.

There was another genius mind exposed to Wittgenstein’s Tractatus, the teacher and friend who inspired these pursuits in logic and philosophy — Bertrand Russel to whom young Austrian was saying in Cambridge: “No profesor Russel, I cannot admit that there is no Rhinoceros in this room, because the world is made of facts, not things.”.

And here is what Russel wrote in a letter after meeting Wittgenstein who just finished writing Tractatus:

I have much to tell you that is of interest. I leave here today [December 20, 1919, from the The Hague] after a fortnight’s stay, during a week of which Wittgenstein was here, and we discussed his book [the Tractatus] everyday. I came to think even better of it than I had done; I feel sure it is really a great book, though I do not feel sure it is right… . I had felt in his book a flavour of mysticism, but was astonished when I found that he has become a complete mystic. He reads people like Kierkegaard and Angelus Silesius, and he seriously contemplates becoming a monk. It all started from William James’s Varieties of Religious Experience, and grew (not unnaturally) during the winter he spent alone in Norway before the war, when he was nearly mad. Then during the war a curious thing happened. He went on duty to the town of Tarnov in Galicia, and happened to come upon a bookshop, which, however, seemed to contain nothing but picture postcards. However, he went inside and found that it contained just one book: Tolstoy on the Gospels. He brought it merely because there was no other. He read it and re-read it, and thenceforth had it always with him, under fire and at all times. But on the whole he likes Tolstoy less than Dostoyevsky (especially Karamazov). He has penetrated deep into mystical ways of thought and feeling, but I think (though he wouldn’t agree) that what he likes best in mysticism is its power to make him stop thinking. I don’t much think he will really become a monk — it is an idea, not an intention. His intention is to be a teacher. He gave all his money to his brothers and sisters, because he found earthly possessions a burden. I wish you had seen him.

I wish I could myself. To experience this intensity of a genius trying to stop thinking, dramatically. Like if this Kabbalist meditations were driving him crazy. The Cantor’s secrets of alephs, paradoxes of infinity and set theory, which sent so many to institutions, including another person from Time’s list, another Austrian, Kurt Gödel, who showed how to encode properties of meta-language in the language and therefore use arithmetic to drive conclusions about meta-math. He showed that having complete and consistent set of axioms for all mathematics is impossible. That we can produce statements which are true, but cannot be proved. Aronofsky meditated on this state of being in his Pi — a genius obsessing with the number of God. Later Einstein was visiting Gödel in the mental hospital, who finally starved himself to death fearing obsessively that the food he is given might be poisoned.

In the note to the first publisher of Tractatus Wittgenstein reveals his intentions:

The book’s point is an ethical one. I once meant to include in the preface a sentence which is not in fact there now but which I will write out for you here because it will perhaps be a key to the work for you. What I meant to write, then, was this: My work consists of two parts: the one presented here plus all that I have not written. And it is precisely this second part that is the important one. My book draws limits to the sphere of the ethical from the inside as it were, and I am convinced that this is the ONLY rigorous way of drawing those limits. In short, I believe that where many others today are just gassing, I have managed in my book to put everything firmly into place by being silent about it. And for that reason, unless I am very much mistaken, the book will say a great deal that you yourself want to say. Only perhaps you won’t see that it is said in the book. For now, I would recommend you to read the preface and the conclusion, because they contain the most direct expression of the point of the book.

Regardless of Wittgenstein’s explanatory intentions it is still not clear which intentions he originally had. Later on he tried again. George Edward Moore, his friend and the author of Principia Ethica himself, wrote it down during Wittgenstein’s lecture on ethics. If you know Tractatus well, this might be the most intense thing you have ever read. Here is how it ends:

My whole tendency and I believe the tendency of all men who ever tried to write or talk Ethics or Religion was to run against the boundaries of language.

This running against the walls of our cage is perfectly, absolutely hopeless.

Ethics so far as it springs from the desire to say something about the ultimate meaning of life, the absolute good, the absolute valuable, can be no science.

What it says does not add to our knowledge in any sense.

But it is a document of a tendency in the human mind which I personally cannot help respecting deeply and I would not for my life ridicule it.

I experienced it. Thank you professor Wittgenstein.

Wittgenstein didn’t have that many followers during his lifetime. But his influence was growing and growing everywhere, like the tree growing out of these seeds of propositions planted under the artillery fire of World War I, somewhere in Polish Galicia, where the soil is fertilized with the bodies of young soldiers buried there on countless small cemeteries. I had been visiting many of them while hiking as a student through this beautiful land, soon to discover philosophy myself. I was also writing my first big IT project at this time, business process management system. And it happened to be in the same time as I was studying Tractatus. Soon I realized that what I designed and implemented was totally influenced by the book. I was designing the model of the world after all, in a language, using another machine language, to build our own language expressing the processual aspect of reality, how the arrow of time connects states of affairs.

Machines today no longer just automate our processes. They are also curating the content for us, the language we like to read, the imagery we enjoy to perceive. In this symbiosis of humans and machines the latter started shaping the language we use, the metaphors we understand, the ideas we want to pursuit. And as long as this cultural evolution is driven by capitalist economy, it will be exploiting our paleolithic emotions to keep us on the hook for longer, machines are positively enforced to exploit our addictions. Machine learning might be as helpful to decide for us what to watch and read next, to make us buy more stuff, to decide who to kill on the road (the famous utilitarian trolley problem is no longer a metaphor, it has to be encoded in the value system of the machine now), or to prevent the next genocide. They can also start producing beauty for us. Evolution of our own specie is happening right now at the rapid rate, and it is happening in symbiosis with them. In a sense we are becoming one. Like suddenly the linguistic models of reality, artificial languages we created to talk to them, reached the tipping point where they started producing another reality, to talk to us. But this we won’t understand for long in our anthropocentric, egocentric, narcissistic solipsism.

I wonder if we will be ever able to build machines capable of conceptualizing what Wittgenstein actually had in mind. Maybe it will happen only after they will start building themselves?

All the ideas and reflections I expressed here, I would like to convert into an experience I can share. The feelings which come from perceiving phenomenons bringing us closer to what we must be silent about. It will be an interactive projection, a kind of a mirror, reflecting on our cultural condition in the world of symbiosis between humans and machines. I want to touch history, politics, language, symbols, semantics, semiotics, statolatry, state oppression, self oppression, evolution, axiology, ethics, metaethics and many other which are not to be named. And I hope to be able to show it at Ars Electronica in Linz, where both Wittgenstein and Hitler went to school.

When my friend Q. Lei. heard about my work described here, we decided to make a bigger project together out of these ideas, specifically for Ars Electronica. In October she was studying Wittgenstein’s philosophy which I am already quite familiar with. I went to co(art) event in Timisoara to teach creative coding there. Lei was sharing her progress with me on daily basis. One day, over breakfast, I met Anna Dumitriu and Alex May who also came to Romania to teach. I shared with them my brief concept. And just after the breakfast I received a message from Lei: “Kazik, I am in the process of preparing a proposal for the project, and I realized it’s probably better we keep the development process to ourselves before the project solidifies and goes somewhere for exhibition. It’s to protect the idea, especially I think our developing ideas are very good!”. Few days later, by accident, I discovered that she already submitted an application to Ars Electronica residency focused on AI, without informing me about it and without asking about my opinion. It’s based on my idea and my research, but it is not our project anymore, it is “her” project now. She didn’t ask me for permission to use my materials, including this transcript from work with GPT-2 which I gave her in good faith. I trusted Lei completely, also because of my lack of experience in the art world. But now I think she was wrong about this secrecy. To the contrary I want to make everything about this project as transparent as possible. Maybe it’s the only way to let you experience it next year.

hairless running ape, translating ideas into creative code, in post-human culture https://xemantic.com/ https://instagram.com/xmorisilx/

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store