Looking at Tractatus logico-philosophicus, I'm reminded of Ayn Rand's Objectivism - another attempt to derive a philosophy from axiomatic principles.
However, it's probably impossible to deductively derive something that lines up with reality, given how language itself is an imperfect medium for that.
It looks like Wittgenstein eventually points out the problems with the Tractatus in his later works like "The Philosophical Investigations"
IIRC doesn’t the Tractatus itself famously end with a kind of acknowledgment of its own failure? “Whereof one can’t speak, thereof one must be silent?”
While I don't know a lot about Objectivism, it feels like there is a very fundamental difference. The Tractatus bans things like ethics and perhaps even metaphysics (and therefore itself, which is the ladder-type argument) to the realm of the unspeakable, whereas Ayn Rand seems to try to draw ethical or moral conclusions?
Is this just some hyped up collection of interconnected assertions (for lack of a better description) or does it actually reveal something non-trivial if you take the effort to dive into it?
My theory is that the tractacus was a bullshit detector made by Wittgenstein.
If after reading it you think it is genius you fail the test, if you understand that it is a bunch of refined nonsense you pass.
Honestly it is obvious to anybody that actually spent time reading it, e.g even the truth tables in it are invalid. He conclude the book by saying that when you don't know what you're talking about, it would be better for you to not speak about it, this is obviously a meta-critic about the tractacus.
Ironically, most philosophers fail the test (obviously most of most philosophers admire it out of authority, without having read a single page from it nor it's subsequent book (as if the subsequent book was even needed for understanding it's true semantically-empty nature..))
There is no x nor y in 4.442. I have two guesses at what you might mean.
1. Maybe you're referring to the last sentence "(The number of places in the left-hand pair of brackets is determined by the number of terms in the right-hand pair.)" and saying: haha, if the number of terms in the right-hand pair is 3 then the number of places in the left-hand pair is 8, not 3. But Wittgenstein doesn't say "is equal to", he says "is determined by", and that's correct.
2. Maybe you're referring to the concrete truth-table, where if you label its rows and columns 1,2,3 etc., the slot with x=y=3 is left blank. That's a weird notational choice but I'm pretty sure it isn't a mistake. Look at 4.44: "The sign that results from correlating the mark 'T' with truth-possibilities is a propositional sign." Wittgenstein's being terse as usual, but I think what he means is that when reading one of these truth-tables what we need to know is where the Ts are. For a bit more explicitness, look further back at 4.43: "We can express agreement with truth-possibilities by correlating the mark 'T' (true) with them in the schema. The absence of this mark means disagreement." (Emphasis mine.) So the right-hand column of his truth tables uses T to mean that the corresponding proposition is true, and the absence of a T to mean that it's false. Which is exactly what he does in 4.442. He's clearly aware that there's a cost in clarity, which is why after writing it as "(TT-T)(p,q)" he says: "or more explicitly (TTFT)(p,q)".
Either way, I think you're wrong to characterize this as a mistake.
I would always give Wittgenstein the benefit of the doubt on such relatively trivial things given 1) this work was checked thoroughly by the likes of Bertrand Russell 2) the topic has evolved considerably over the last ~100 years, or at least the language and notation used to talk about it, which may lead to misunderstandings for today's readers.
Two objects of the same logical form are—apart from their ex-
ternal properties—only differentiated from one another in that
they are different.
differentiated in that
they are different.
It is a mediocre circular reasoning. The correct statement could have been to say that they are internally structurally the same object but have different identity like e.g in nominal typing.
HOWEVER, I skimmed the tractacus years ago and now I update my jugement:
The tractacus has many flaws but it is not pure bullshit nor a bullshit detector test despite the striking irony of Wittgenstein saying in the preface and in the conclusion
Its whole meaning could be summed up somewhat as follows: What can be said at all can be said clearly; and whereof one cannot speak thereof one must be silent.
And then spending a hundred a pages talking in a non clear, vague manner about a few conclusions that are one 1) obvious to any rationalist
And 2) that can be summed up in ~one page Cf the symbol grounding problem, defining empiricism and that metaphysical questions are by design out of the scope of science and therefore cannot be answered (well an exhaustive list of the possibles can still be given).
This book has a very low signal to noise ratio which make it easy to mistakely take for bullshit/pseudo intellectualism.
As a NLU researcher some of his thoughts are actually interesting about how do define a semantic space even if most of them are trivial.
Personal attacks will get you banned here. No more of this, please.
Rather than superciliously putting someone down when they're wrong (or you think they are), why not share some of what you know, so the rest of us can learn?
The Tractatus is basically an exploration of the limits of Tautology - anything that follows from the axioms is tautologous. And we have no grounds to speak about anything else
A perhaps useful quick reference point-by-point summary of the Tractatus but it would be even better if one could see the whole tree and branches in full or expandable/collapsible branch by branch. As is, it requires the text to be open in front of one (this may not always be convenient in the internet age).
I've actually always wanted something like this for Spinoza's Ethics. The whole book is a directed graph, and is hard to read without constantly leafing around (propositions supported by other propositions and their corollaries, etc.). Similar in structure to this book (the "geometric method") as both are inspired by Euclid's Geometry (and I'm sure Wittgenstein had read Spinoza as well).
Spinoza's Ethics is probably more deeply nested than this, and it would be handy to be able to see the original Latin forms of propositions alongside the English translation(s). The highly structured and granular form of the book means that you could see the translations on a proposition by proposition basis, so it would really lend itself to that kind of thing.
I went so far as to work the whole of the book into XML back in 1999 or 2000, but I've long since lost that. These days I'd probably put it into a relational database.
Every few years I attempt it, but UI dev isn't my forte.
A truly great tool might as well generalize to the axiomatic method in general. When I've gone through math texts (Euclid's Elements, calculus, topology, etc.) I find it annoying to flip back so much when they are constantly referencing past figures. Even if you write the often used axioms and definitions down, there will still be some that are a pain to access.
I think there's a big opportunity to figure out a way to display these systems in a way no one has done well yet. I'm surprised that if I want to go through Euclid's elements in 2021, it's probably going to be in a pdf or a really basic static webpage. How is there not a super slick app that reveals all kinds of connections in a way only possible with interactive media?
This is cool in general, but I feel like it disrupts the flow of the writing/presentation. There is something fundamentally poetic about the structure and rhythm of the Tractatus. Perhaps that is why it has held so much appeal. In any case I feel like indentation or some other nesting convention found in software formatting would do a better job of visually representing the logical structure of the book without disrupting the flow.
I personally think the structure of the book is it's biggest weakness. No matter how poetic, it really doesn't support communicating it's material very well. At least not to normal people.
That said, I actually did a redesign of the book (https://i.imgur.com/QFyuIsd.jpg) years ago using indentation and lines as guides. I reckon that's the best way to improve the experience in book form without fundamentally altering the text itself.
If I were to do the same for web, I'd also make sure the reader always sees the statements from upper layers to get the full picture of the context in which the current statement is said.
29 comments
[ 2.8 ms ] story [ 74.0 ms ] threadHowever, it's probably impossible to deductively derive something that lines up with reality, given how language itself is an imperfect medium for that.
It looks like Wittgenstein eventually points out the problems with the Tractatus in his later works like "The Philosophical Investigations"
Yes, he came to the same conclusion you did, that what he attemtped in the Tractatus was misguided at best and impossible at worst.
But it remains his most famous and most quoted (well, that one sentence specifically) book.
that would be a very specific and relevant point to make. can you back this up with sources or more than one example?
> My theory is that the tractacus was a bullshit detector
Are you serious? This seems like a fun idea but doesn't really pass my own bullshit detector.
1. Maybe you're referring to the last sentence "(The number of places in the left-hand pair of brackets is determined by the number of terms in the right-hand pair.)" and saying: haha, if the number of terms in the right-hand pair is 3 then the number of places in the left-hand pair is 8, not 3. But Wittgenstein doesn't say "is equal to", he says "is determined by", and that's correct.
2. Maybe you're referring to the concrete truth-table, where if you label its rows and columns 1,2,3 etc., the slot with x=y=3 is left blank. That's a weird notational choice but I'm pretty sure it isn't a mistake. Look at 4.44: "The sign that results from correlating the mark 'T' with truth-possibilities is a propositional sign." Wittgenstein's being terse as usual, but I think what he means is that when reading one of these truth-tables what we need to know is where the Ts are. For a bit more explicitness, look further back at 4.43: "We can express agreement with truth-possibilities by correlating the mark 'T' (true) with them in the schema. The absence of this mark means disagreement." (Emphasis mine.) So the right-hand column of his truth tables uses T to mean that the corresponding proposition is true, and the absence of a T to mean that it's false. Which is exactly what he does in 4.442. He's clearly aware that there's a cost in clarity, which is why after writing it as "(TT-T)(p,q)" he says: "or more explicitly (TTFT)(p,q)".
Either way, I think you're wrong to characterize this as a mistake.
HOWEVER, I skimmed the tractacus years ago and now I update my jugement: The tractacus has many flaws but it is not pure bullshit nor a bullshit detector test despite the striking irony of Wittgenstein saying in the preface and in the conclusion Its whole meaning could be summed up somewhat as follows: What can be said at all can be said clearly; and whereof one cannot speak thereof one must be silent. And then spending a hundred a pages talking in a non clear, vague manner about a few conclusions that are one 1) obvious to any rationalist And 2) that can be summed up in ~one page Cf the symbol grounding problem, defining empiricism and that metaphysical questions are by design out of the scope of science and therefore cannot be answered (well an exhaustive list of the possibles can still be given). This book has a very low signal to noise ratio which make it easy to mistakely take for bullshit/pseudo intellectualism. As a NLU researcher some of his thoughts are actually interesting about how do define a semantic space even if most of them are trivial.
https://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
Rather than superciliously putting someone down when they're wrong (or you think they are), why not share some of what you know, so the rest of us can learn?
https://hn.algolia.com/?dateRange=all&page=0&prefix=true&sor...
https://news.ycombinator.com/newsguidelines.html
That’s NOT what he wrote. The quote is:
“whereof we cannot speak, thereof we must remain silent.” (or “What we cannot speak about we must pass over in silence.” in this translation)
Proposition 7 is about the limits of language, not about dissing idiots.
Spinoza's Ethics is probably more deeply nested than this, and it would be handy to be able to see the original Latin forms of propositions alongside the English translation(s). The highly structured and granular form of the book means that you could see the translations on a proposition by proposition basis, so it would really lend itself to that kind of thing.
I went so far as to work the whole of the book into XML back in 1999 or 2000, but I've long since lost that. These days I'd probably put it into a relational database.
Every few years I attempt it, but UI dev isn't my forte.
I think there's a big opportunity to figure out a way to display these systems in a way no one has done well yet. I'm surprised that if I want to go through Euclid's elements in 2021, it's probably going to be in a pdf or a really basic static webpage. How is there not a super slick app that reveals all kinds of connections in a way only possible with interactive media?
That said, I actually did a redesign of the book (https://i.imgur.com/QFyuIsd.jpg) years ago using indentation and lines as guides. I reckon that's the best way to improve the experience in book form without fundamentally altering the text itself.
If I were to do the same for web, I'd also make sure the reader always sees the statements from upper layers to get the full picture of the context in which the current statement is said.
https://wittgenstein-network.gitlab.io
https://arxiv.org/abs/1703.04336
Well, clearly, that's a good enough reason why the Tractatus isn't taught in the first semester of phil. 101. :-)