Pinker Sentence 7
Elliot’s Tree:
Mine:
Differences: I wasn’t sure about where to attach the section after the dash. Other than that choice, our trees appear identical.
Rereading, I think “the notion” as a modifier restating the previous “the notion” is correct, not my version.
PS I think some of your errors in previous trees were important to seek more clarity about instead of dismissing as minor. (That’s assuming your goal is to be really good at this, rather than adequate at it to e.g. participate in debates better.)
1 Like
I thought about it for a few minutes and still like my approach to the first sentence more.
One of the ways I think about where to attach stuff is to imagine dropping out other stuff and see if what you have left makes some sense.
So e.g. (my version of Elliot’s approach): Artificial intelligence is an existence proof of one […]: that the abstract realm of knowledge, reason, and purpose does not consist of an élan vital or immaterial soul or miraculous powers of neural tissue.
Versus my approach: Artificial intelligence is an existence proof of one of the great ideas […]: that the abstract realm of knowledge, reason, and purpose does not consist of an élan vital or immaterial soul or miraculous powers of neural tissue.
A different way of approaching this: “that the abstract realm of knowledge, reason, and purpose does not consist of an élan vital or immaterial soul or miraculous powers of neural tissue” reads to me more like an idea than a one. One in this context doesn’t tell us much. It’s a noun but kind of conceptually incomplete until you get to the “ideas”. “of the great ideas” is technically a modifier but it’s basically necessary to clarify which thing from a gigantic range of stuff we could be talking about and isn’t an optional detail.
I think my representation here was fine (and Elliot’s was fine too). The “that” in “that stand” is both relating the “that stand” clause and serving as a relative pronoun within the clause (standing for “patterns”).
For sentence 4:
I think Elliot’s right here. Following Elliot’s approach, the “by physical operations” prepositional clause can be seen as providing information on how the transformations are occurring. It’s answering the question “by what means are the transformations occurring?” If you follow my approach, the question comes out something like “by what means are the transformations of knowledge occurring?” So the transformations is still key and the prepositional phrase has to kind of “reach through” the “of that knowledge” phrase to connect to what it’s actually modifying.
Still think this is fine.
I’m actually a bit fuzzy on the basic structure there. Intuitively it seems like “are designed” is the verb, “to preserve” is an infinitive, “relations” is serving as the object of something.
Something I didn’t catch before from my tree: “those” should be a child of “relations”. That’s just an error.
I found your response to my post about controversial causes difficult to respond to. I could easily be misinterpreting your argument. But I think you’re bringing up some things I think of as important but separate issues from controversy. I’ll try to pull them out, name them, and discuss why I think they’re separate.
I’ll call this the breakpoint problem. Breakpoints exist in lots of causes (maybe all?), and unless your efforts or lack thereof change which breakpoint the world is at your decision has little practical effect. Some of my thoughts about that:
- The breakpoint problem can apply whether a cause is controversial or not. Suppose have a family member with cancer and you’re considering donating to the effort to find a cure fast enough that they don’t die. If you don’t donate to the effort to cure cancer (a non-controversial cause) & cancer doesn’t get cured before your relative dies should that whole failure be ascribed to your failure to donate? Just as with the political donation the answer is no. On the other hand, consider offering sex-ed books instead of philosophy books to the kids in your own example. That’d be a controversial cause. Some people would actively oppose it. But your $100 would still buy $100 worth of books that could go to kids because the breakpoint in a small(see below) book cause is 1 book, whether the book is a controversial one or not.
- The main difficulty of the breakpoint problem for a contributor is that you usually don’t know how close to a breakpoint the world actually is while you’re considering donating. And often the world is quite far from a breakpoint. This is similar to voting - you don’t know how close the vote will actually be until after you’ve already decided whether or not to vote, but it’s super rare for a vote to be close enough that your individual vote mattered. Definitely something to consider but doesn’t depend on there being a controversy.
- The breakpoint problem may be good reason to choose causes with smaller breakpoints, or clearer breakpoints, though I haven’t thought about it much. But I don’t see it as affecting the controversial vs. non-controversial cause decision because it’s not about that.
I’ll call this the control problem. You do have lots more control over the outcome of your efforts with some causes than others. But as with the breakpoint problem I think it’s mostly independent of controversy when isolated from the effects of intentional opposition.
What I mean is: Of course the existence of intentional opposition to a particular cause reduces one’s control over the outcome of that cause because of the people actively trying to thwart your efforts. I think that’s a main point of Elliot’s article, and I agree with it.
But the extra stuff you mentioned (events that are interpreted as reasons to pass new legislation, general political dynamics and cultural trends in the country, the ideologies taught to people at school) exist and can affect outcomes whether there’s intentional opposition to your specific cause or not.
Again consider the non-controversial effort to cure cancer. Maybe someone famous and beloved gives an impassioned speech in favor of cancer research prior to dying of cancer. Or maybe a Theranos of the cancer world commits massive fraud, sours lots of people on the effort and causes super restrictive legislation to be passed. You can’t control any of that, but it’s going to have a huge impact on how quickly cancer gets cured.
I’ll call this the impact problem. If there’s a million people donating $100 to something, being the million and first probably won’t have much impact. Whereas if there’s only 10 people donating $100, being the eleventh is much more likely to matter.
I think this is related to the breakpoint problem and why I specified a small book cause above. If there’s a million people donating $100 to a book cause (controversial or not), maybe the million and first donor has no positive impact at all. Maybe every kid who exists and wants a book of that type and isn’t prevented by something (like their parents) can get one, so buying even more books to give away won’t help.
Maybe nearly everyone with the skills to help cure cancer is already fully employed trying to cure cancer, and more donations will only bring in lower skilled people which actually slow down the effort.
Maybe nearly everyone who could be persuaded by additional political spending on an issue has already been persuaded. Etc.
You can probably see where I’m going…I think marginal impact depends on things like the size of the current effort vs. other factors that determine whether additional effort would have much, or any, or even negative impact. These (mostly?) don’t depend on whether or not the cause is controversial.
For Sentence 5:
I’m still kinda torn on this one. One argument for Elliot’s view: we can delete “of operations” and that part of the sentence is still pretty coherent (though the definite article before “control” reads awkwardly: “Purpose can be explained as the control […] to effect changes”). That is compatible with Elliot’s tree, since “of operations” is a separate branch and so you can prune it and it’s no big deal. In mine, “operations” is the parent of “to effect”. Deleting the parent of something seems like it should kinda break the sentence as it relates to that parent and its children.
I momentarily considered something like this btw, giving the infinitive the full dignity of a verb:
I vaguely remembered discussing some structure like that on the FI list but couldn’t find it in quick searching. Regardless, I don’t think that’s getting the relationship of the infinitive and “control” correct, but it was a thought 
.
Regarding Pinker Sentence 5:
That’s just fighting with/rewriting the existing grammar. “guided” is just a participle and is fully capable of modifying “changes” without implying additional structure.
Regarding Pinker sentence 6:
I think that representing the “that” twice is fine but I think I got confused or something in how to represent stuff. This is the right approach and much closer to Elliot’s for that part:
I think I’m correct. Articles modify their nouns directly. You could knock out “most familiar” completely and the sentence is still sensible and meaningful, so “the” doesn’t depend on anything in the adjective phrase “most familiar”.
Agreed.
@Lebowski let’s focus on the text analysis discussion for now (if you’re okay with that) as I think that’s more straightforward and likely to get somewhere. I know I opened the subthread (erm subtopic) on the controversial causes stuff here. I had some energy to comment on the controversial causes stuff and wanted to respect/use that energy, but don’t want to get bogged down in it with the prior discussion still outstanding.
I made a tree of that discussion so far
I’m experimenting with an application called RemNote for taking notes on Pinker video and on the posts in this topic about it. Here is a link to my notes so far
More Pinker Video.
I tried making my own paragraph tree before looking at Elliot’s. Will analyze later (well, likely tomorrow)… Just posting trees for now.
Mine:
Elliot’s:
I numbered the Pinker paragraph sentences from 1 through 7 for ease of reference
Analysis regarding Pinker Paragraph trees:
- I wrote lengthy descriptions regarding the connections between the parts of the paragraph and Elliot wrote short ones. I intentionally wrote long stuff as a strategy, as I was having some trouble/doubts in making my own tree, and I thought writing more stuff would help. So I think my approach was fine for my context.
- For the connection between Sentences 1 and 2
- Elliot wrote "restating thesis positively."
- I wrote "The child tells us what the abstract realm *is* instead of the parent, which tells us what it is not (or at least, what it does not consist of)."
- This sentence is unclear and was trying to do much in one sentence. A better rewrite is:
- The child tells us what the abstract realm is/consists of. The parent merely tells us what the abstract realm does not consist of.
- This sentence is unclear and was trying to do much in one sentence. A better rewrite is:
- Analysis: Both descriptions are getting at the idea that the child frames Pinker's claims positively (which doesn't happen in the parent, which is negative). I would be hesitant to call Sentence 2 a restatement of the thesis because I don't see that Pinker's quite gotten his thesis out until Sentence 2. It'd be possible to get it out in Sentence 1, but I don't think he succeeded in that. So I guess I have some mild disagreement with that description.
- If I were rewriting Pinker's sentences in order to put what I think is the main point in the first sentence, I'd do it something like this:
- Artificial intelligence proves that knowledge, reason and purpose, rather than being miraculous or supernatural, are actually connected to physical reality via the concepts of information, computation, and control.
- If I were rewriting Pinker's sentences in order to put what I think is the main point in the first sentence, I'd do it something like this:
- For the connection between Sentence 2 and Sentences 3, 4, and 5
- Elliot wrote "explanation for knowledge", "explanation for "reason", and "explanation for purpose:
- I wrote: "The children give examples of how knowledge, reason, and purpose can be linked to the physical realm via information, computation, and control"
- I think "explanation" (ET's description) is a better word choice than "example". I think that, for each case of knowledge, reason, and purpose, Pinker was trying to explain how it could be linked to the physical realm via information, computation, and control, respectively. "Example" suggests that he's talking about some specific concrete examples, which isn't the case.
- For the connection between Sentence 1 and Sentence 6
- Elliot wrote "elaboration: intelligence comes from computation not special tissue"
- I wrote: "Pinker previously discussed information/computation/control. The child wants to cash in on those concepts in order to claim that the naturally evolved brain isn't special. This connects to the parent's rejection of "miraculous powers of neural tissue" as an explanation."
- I think Elliot's description is better as a high level summary of what Pinker's point is, despite me using many more words.
- For the connection between Sentence 1 and Sentence 7
- Elliot wrote "evidence/proof"
- I wrote: " Pinker started out, in the parent, claiming that AI proves something about the abstract realm, and then discussed a connection between the abstract realm and information, computation, and control. The child further develops or elaborates the point about what Pinker thinks AI proves. "
- My description is too complex and confusing. It'd be better rewritten as:
- In the parent, Pinker claimed that AI proves something about the abstract realm. He then (in other nodes) discussed a connection between the abstract realm and information, computation, and control. The child further develops or elaborates the point about what Pinker thinks AI proves.
- My description is too complex and confusing. It'd be better rewritten as:
- I don't see any significant disagreement on this point.
I discovered how to paste nested lists from RemNote to the forum: RemNote has an html export mode and I can just paste from that to the forum and it keeps the formatting 
Yes except that I already have substantial text analysis skill. Outside of CF I think I am the best person at it of anyone I regularly interact with. LOL I just remembered last week I was on a zoom call with a bunch of other people, one of whom is an attorney, and I was the one answering most of the questions about the meaning of a particular area of contract law.
Rewriting the last bit, the truth is more like:
So it’s best to not work towards developing additional text analysis skill, and to thus not become even more aware of the literal meaning of what other people are saying.
If someone says something where it’s easy to guess what they intended, then it’s easy to go with it and not even think about what they literally said. Like if someone says “I could care less if it’s cold outside” they mean they don’t care that it’s cold outside even though they literally said they could care less, which means they do care some. That’s common enough that unconsciously guessing it doesn’t stop the flow of thought or seem to contradict the rest of the context.
But if they something where it’s hard to guess what they intended, that stops the flow. I start consciously wondering about the possibilities and using literal analysis skills to generate possibilities. If the literal analysis fits the rest of the context better than more socially acceptable possibilities I’ll tend to want to go with that.
In practice literal analysis sets up situations where I’m internally conflicted between for example, calling people out on their bullshit or contradictions vs. maintaining social harmony by pretending I didn’t notice. I don’t know about ‘inherently’ but I do find it super hard to resolve conflicts like that in ways I feel good about.
You’re right, it’s not super safe. However in most situations I think literal analysis is an additional cause of blow ups, or makes other blow ups worse. It’s plausible I’m wrong about that; it’s an intuition not an area I’ve consciously studied.
I know & take it as given that most people outside CF talk nonsense and bullshit on a regular basis. Just to get started, a majority of them literally believe in some version of an invisible man in the sky who grants favors if you ask nicely. And most of the ones who don’t believe in an interventionist God literally believe in stuff like the power of healing crystals, that we’ve been visited by aliens, or that we’re all gonna die soon because of gun owners, car exhaust & cow farts.
In most situations, literal analysis of what people like that say is less useful than figuring out what script they’re playing and whether that script and my knowledge of how to play along with it permits a mutually beneficial interaction. Which it often does, and is kinda the point.
If I bailed on everyone who talks nonsense or bullshit I’d be worse off and they would also. Interacting with those people are the areas of life where I say literal analysis has negative value. Because literally analyzing what for example God believers say will reliably cause me to want to do less mutually beneficial interactions or say things that piss them off so they want to do less interactions with me.
More Pinker, comparing analyses of Sentence 1
Edit: made this a reply to Lebowski by accident. My bad.
- Analyzing Sentence 1: Artificial intelligence is an existence proof of one of the great ideas in human history: that the abstract realm of knowledge, reason, and purpose does not consist of an élan vital or immaterial soul or miraculous powers of neural tissue.
- Justin's criticisms (note that I've watched Elliot's video before, so by "Justin's criticisms" I just mean "criticisms that occurred to Justin before watching Elliot's discussion of this sentence a second time.")
- The sentence is pretty long. It could easily be two sentences.
- "Existence proof" seems to possibly be some sort of specialized math term. I think Pinker's using it in a more standard way (like "proof of existence") but there's a clash between his use and the normal use of the term.
- "abstract realm" seems unnecessary. Ideally you could just talk about knowledge, reason, and purpose directly (as I did in my rewrite below). Otherwise you could just group them under something simpler like "ideas".
- He says something is "one of the great in human history" but then he says that the great idea is that something isn't in certain categories. Anticlimactic! Doesn't really seem like that great of an idea. He doesn't give us a positive statement until the next sentence.
- Élan vital seems like a fancy and unnecessary term/concept to introduce.
- The primary definition for "soul" includes the fact that it's immaterial, so that's unnecessary to specify.
- The last bit ("of an élan vital or immaterial soul or miraculous powers of neural tissue.") could use the clear structure provided from being separated by commas, especially given that the sentence is pretty long as it is.
- Elliot's Analysis:
- Elliot's rewrite: AI is proof of idea: abstract realm does not consist of soul.
- Elliot's points/criticisms:
- So: the abstract realm could be made of 90% soul. ET doesn't think that's what Pinker's means (he thinks Pinker means soul isn't involved at all, and I agree) and so there's a problem here.
- How AI proves what Pinker thinks it does is underexplained.
- Pinker wants to advocate atheism so he brings up rejecting souls early, but then goes and talks about other stuff later.
- The really great ideas in human history are typically positive and not negative claims. (This is in agreement with my 4th criticism above)
- The idea that souls don't exist isn't that great. People advocating for souls must be familiar with it and have encountered e.g. children doubting souls.
- Existence proofs
- It's a math concept that basically means proving something exists without providing an example
- Pinker doesn't mean that, though
- Seems like he wanted a fancy term but used it in a non-standard way to mean "the existence of X proves Y" (which you have to guess)
- Readers often blame themselves for being less educated than Pinker, but some of what he writes isn't correct or expert jargon
- Justin's criticisms (note that I've watched Elliot's video before, so by "Justin's criticisms" I just mean "criticisms that occurred to Justin before watching Elliot's discussion of this sentence a second time.")