5 ms·

I'd argue that every programmer interested in language design needs to read Wittgenstein. His later work in ethics and epistemology (that the article recommends

I'd argue that every programmer interested in language design needs to read Wittgenstein. His later work in ethics and epistemology (that the article recommends, Philosophical Investigations) is hit or miss, but his Tractatus Logico-Philosophicus, which deals with the nature of language as it relates to our ability to think and formulate ideas, is undeniable. He essentially builds a framework for breaking reality down into logically composable components which can be reasoned over in their own right, regardless of semantics. See this article which was posted here a few years ago: Wittgenstien for Programmers http://www.hxa.name/notes/note-hxa7241-20110219T1113Z.html http://www.hxa.name/notes/note-hxa7241-20110219T1113Z.html

kiliantics 9y ago
As I understand it, the Investigations arrives at opposing conclusions to Tractatus and Wittgenstein believed he was mistaken in his premises for the earlier work.
aschampion 9y ago
> which deals with the nature of language as it relates to our ability to think and formulate ideas, is undeniable Except by Wittgenstein himself, who largely renounced this philosophical project in the second half of his career. Still essential reading, though.
- philwelch 9y ago
  Wittgenstein is like a rock band that abruptly changes style at some point in their career, alienating many of their fans but gaining new ones. One of my philosophy professors used Radiohead as an example, with (I believe) OK Computer as the turning point, but I don't know Radiohead well enough to judge that particular analogy.
  - carlob 9y ago
    Probably Kid A https://en.wikipedia.org/wiki/Kid_A https://en.wikipedia.org/wiki/Kid_A
    - mercer 9y ago
      I'd agree that the turning point was OK Computer. While that record shows that the turning point wasn't as abrupt as it's often made out to be, OK Computer was the first time a lot of Radiohead listeners went 'huh?!' and either accepted or rejected this new direction. Kid A just affirmed this new direction as more than just some 'influence'.
- carlob 9y ago
  This is really interesting in the context of that link: if I remember correctly Wittgenstein renounced his previous ideas on language after working for a few years in an elementary school. Some think that this change was prompted by direct observation of how language is actually learnt. However it might be the case that what he wrote in the Tractatus is still relevant for formal languages, such as programming.
pmoriarty 9y ago
Wittgenstein was more of a mystic than a logician, though he tried to dress up his mysticism in what may at first glance appear to be logical garb in the Tractatus.[1] This is something the early Logical Positivist analytics of the Vienna Circle, perhaps willfully, did not understand. He was championing the very thing that they abhorred: religious mysticism. When Wittgenstein said "Whereof one cannot speak thereof one must be silent." He meant it in a respectful way, as a defense of religious mysticism. The Logical Positivists assumed that attitude in the exact opposite direction: to attack religious mysticism and religion in general. They labeled it as meaningless nonsense. No wonder Wittgenstein wanted nothing to do with them, despite their (misplaced and misunderstood) hero-worship of him. I'm really not sure what any programmer would get out of reading the Tractatus... at least not for anything programming-related. [1] - The full title being the "Tractatus Logico-Philosophicus", just in case you missed that it was supposed to be logical and philosophical. Wittgenstein starts by labeling some of his statements as "axioms" and constructing what might appear as derivations.. which don't actually follow. On superficial inspection it might look like some kind of logical treatise, but that impression vanishes if you actually try to read and make sense of it.
- B1FF_PSUVM 9y ago
  Well, the later work has a caution that applies to programmers ... https://en.wikipedia.org/wiki/Private_language_argument https://en.wikipedia.org/wiki/Private_language_argument
  - paganel 9y ago
    Thanks for the link, didn't know this was a "concept" per-se, so to speak, even less so that it had its own wikipedia page and that Wittgenstein had written about it. I've started thinking about this concept ever since I read Wilhelm von Humboldt's "The Heterogeneity of Language and its Influence on the Intellectual Development of Mankind", where at some point he mentions that, taken at the limit, each of us speak our own, very private language (the book was written in the early 1800s). That idea has stuck with me ever since and I've tried ways of both disproving and affirming it, now I'm mostly in the "affirming it" phase. As a quick example, one can just think of a sentence like "I love you", which means different things for different people, even if spoken in the same language, and, most important of all, it might mean (and in many cases it does mean) different things to different people who have lived together for decades (as husband and wife, for example, or as parents and children). On the other hand, more "neutral" sentences like "turn off the lights" seem to mean the same thing for a larger proportion of speakers.
igravious 9y ago
> His later work in ethics and epistemology (that the article recommends, Philosophical Investigations) is hit or miss A couple of things about this. Wittgenstein famously did not write much on ethics. A couple of lines in the Tractatus, no passages that I know of in Philosophical Investigations (not saying there are none, I just don't know of them). He did give a lecture on ethics to the Heretic Society in Cambridge in 1929. It is short, here is the transcript: http://www.geocities.jp/mickindex/wittgenstein/witt_lec_et_en.html http://www.geocities.jp/mickindex/wittgenstein/witt_lec_et_e... Well worth the read, I don't think his position evolved from this. I would rather say that the subject matter of Philosophical Investigations is Philosophy of Language and also Epistemology. To call it "hit or miss" is nowhere near the consensus view among philosophers, in fact one is compelled to say that it is the direct opposite of the view that contemporary philosophers hold. Given that the list in the article is rank-ordered and Philosophical Investigations is #1 should clue you into its status. Easily in the top three most important books in philosophy in the last 100 years.
codeulike 9y ago
Not that its relevant, but this is an opportunity to drop in my favourite Wittgenstein quote: A philosopher who is not taking part in discussions is like a boxer who never goes into the ring.
lgessler 9y ago
You are noting that the logical atomism in the Tractatus reminds us of a large software system, which it does. The problem is that logical atomism isn't actually tenable, as many (not least W himself) have pointed out. But you've actually read the Tractatus and PI, so I'm surprised to hear you still supporting the Tractatus. A full discussion of why logical atomism is wrong is, of course, out of scope here, but essentially, the problem is that there are no logical atoms. As inspiration for programmers, the Tractatus is probably fine. But as the only take on philosophy of language/epistemology that a person might ever read, it is at least very far from where modern thinking is at.

kiliantics 9y ago

As I understand it, the Investigations arrives at opposing conclusions to Tractatus and Wittgenstein believed he was mistaken in his premises for the earlier work.

aschampion 9y ago

> which deals with the nature of language as it relates to our ability to think and formulate ideas, is undeniable Except by Wittgenstein himself, who largely renounced this philosophical project in the second half of his career. Still essential reading, though.

philwelch 9y ago
Wittgenstein is like a rock band that abruptly changes style at some point in their career, alienating many of their fans but gaining new ones. One of my philosophy professors used Radiohead as an example, with (I believe) OK Computer as the turning point, but I don't know Radiohead well enough to judge that particular analogy.
- carlob 9y ago
  Probably Kid A https://en.wikipedia.org/wiki/Kid_A https://en.wikipedia.org/wiki/Kid_A
  - mercer 9y ago
    I'd agree that the turning point was OK Computer. While that record shows that the turning point wasn't as abrupt as it's often made out to be, OK Computer was the first time a lot of Radiohead listeners went 'huh?!' and either accepted or rejected this new direction. Kid A just affirmed this new direction as more than just some 'influence'.
carlob 9y ago
This is really interesting in the context of that link: if I remember correctly Wittgenstein renounced his previous ideas on language after working for a few years in an elementary school. Some think that this change was prompted by direct observation of how language is actually learnt. However it might be the case that what he wrote in the Tractatus is still relevant for formal languages, such as programming.

pmoriarty 9y ago

Wittgenstein was more of a mystic than a logician, though he tried to dress up his mysticism in what may at first glance appear to be logical garb in the Tractatus.[1] This is something the early Logical Positivist analytics of the Vienna Circle, perhaps willfully, did not understand. He was championing the very thing that they abhorred: religious mysticism. When Wittgenstein said "Whereof one cannot speak thereof one must be silent." He meant it in a respectful way, as a defense of religious mysticism. The Logical Positivists assumed that attitude in the exact opposite direction: to attack religious mysticism and religion in general. They labeled it as meaningless nonsense. No wonder Wittgenstein wanted nothing to do with them, despite their (misplaced and misunderstood) hero-worship of him. I'm really not sure what any programmer would get out of reading the Tractatus... at least not for anything programming-related. [1] - The full title being the "Tractatus Logico-Philosophicus", just in case you missed that it was supposed to be logical and philosophical. Wittgenstein starts by labeling some of his statements as "axioms" and constructing what might appear as derivations.. which don't actually follow. On superficial inspection it might look like some kind of logical treatise, but that impression vanishes if you actually try to read and make sense of it.

B1FF_PSUVM 9y ago
Well, the later work has a caution that applies to programmers ... https://en.wikipedia.org/wiki/Private_language_argument https://en.wikipedia.org/wiki/Private_language_argument
- paganel 9y ago
  Thanks for the link, didn't know this was a "concept" per-se, so to speak, even less so that it had its own wikipedia page and that Wittgenstein had written about it. I've started thinking about this concept ever since I read Wilhelm von Humboldt's "The Heterogeneity of Language and its Influence on the Intellectual Development of Mankind", where at some point he mentions that, taken at the limit, each of us speak our own, very private language (the book was written in the early 1800s). That idea has stuck with me ever since and I've tried ways of both disproving and affirming it, now I'm mostly in the "affirming it" phase. As a quick example, one can just think of a sentence like "I love you", which means different things for different people, even if spoken in the same language, and, most important of all, it might mean (and in many cases it does mean) different things to different people who have lived together for decades (as husband and wife, for example, or as parents and children). On the other hand, more "neutral" sentences like "turn off the lights" seem to mean the same thing for a larger proportion of speakers.

igravious 9y ago

> His later work in ethics and epistemology (that the article recommends, Philosophical Investigations) is hit or miss A couple of things about this. Wittgenstein famously did not write much on ethics. A couple of lines in the Tractatus, no passages that I know of in Philosophical Investigations (not saying there are none, I just don't know of them). He did give a lecture on ethics to the Heretic Society in Cambridge in 1929. It is short, here is the transcript: http://www.geocities.jp/mickindex/wittgenstein/witt_lec_et_en.html http://www.geocities.jp/mickindex/wittgenstein/witt_lec_et_e... Well worth the read, I don't think his position evolved from this. I would rather say that the subject matter of Philosophical Investigations is Philosophy of Language and also Epistemology. To call it "hit or miss" is nowhere near the consensus view among philosophers, in fact one is compelled to say that it is the direct opposite of the view that contemporary philosophers hold. Given that the list in the article is rank-ordered and Philosophical Investigations is #1 should clue you into its status. Easily in the top three most important books in philosophy in the last 100 years.

codeulike 9y ago

Not that its relevant, but this is an opportunity to drop in my favourite Wittgenstein quote: A philosopher who is not taking part in discussions is like a boxer who never goes into the ring.

lgessler 9y ago

You are noting that the logical atomism in the Tractatus reminds us of a large software system, which it does. The problem is that logical atomism isn't actually tenable, as many (not least W himself) have pointed out. But you've actually read the Tractatus and PI, so I'm surprised to hear you still supporting the Tractatus. A full discussion of why logical atomism is wrong is, of course, out of scope here, but essentially, the problem is that there are no logical atoms. As inspiration for programmers, the Tractatus is probably fine. But as the only take on philosophy of language/epistemology that a person might ever read, it is at least very far from where modern thinking is at.