Hacker News Reader

Comment by ineedasername | Hacker News Reader

I don't like Chomsky in his main area of expertise of Linguistics. My post-grad work was in NLP, embedded within a linguistics department, and requiring a heavy background in formal linguistics. A fair number of faculty, while appreciative of his contributions to the field, held him in low regard because he & some of his main "acolytes" systematically worked to block any researchers in the U.S. working on ideas that ran counter to his own. Meaning more than a generation or more of advancement has stagnated in the US. Advancement in syntactic theory (his main contribution to linguistics) has tended to come from outside the US.

It's possible this has changed somewhat since my formal education, but it was the status quo for quite some time.

View on HN

I fully agree with you. I started studying linguistics since I come from a math background and Chomskian promise of mathematical linguistics was something I found very enticing. I quickly became disillusioned. Most of the Chomsky-inspired research boils down to fitting carefully selected data into an artificial formal machine and manipulating the said machine into accepting more and more data, until the machinery ends up being turing complete therefore rendering any result trivial. Don't get me wrong, there are a lot of very smart people working in that particular field and they have produced some very interesting results and observations, but IMO the real academic value of these models if very limited. It's all about solving intellectual puzzles in a theoretical void. But does it advance our understanding of human brain, learnability or language change? After being a linguist for over 15 years, I'd say "not really".

And yes, as you say, Chomsian followers are a highly guarded, closed camp. There is a strong tradition of giving maximal support to acolyte who wants to work in this area while casting out the "heretics".

As another (ex but not disillusioned) linguist, can I ask what’s the alternative? Say I’m puzzled by the acquisition of Antecedent Contained Deletion structures. I know how to tell a broadly Chomskian story about it. What other options are on the table?

I hear lots of broad brush criticism of generative linguistics, but in almost all cases it turns out that the proposed alternative approaches are just working on different problems, and aren’t really alternatives at all.

I don't think that you will find my answer satisfactory, but does there have to be an alternative? I think that many syntactic mysteries are mysterious only because syntacticians assume that they work in particular ways. Once you start looking at these problems from the perspective of pragmatics, competing interpretations and conventionalisation, many of these mysteries disappear.

The point to make is that interpretation makes use of multiple sources of information, of which syntax is just one. You don't need an elegant, tight fitting, constrained syntactic structure to use language. Which incidentally is why we can understand things even they are not grammatical. I firmly believe that the issue of "grammaticality" simply boils down to convention. Not to mention that many constructed examples used to motivate formal grammar research (be it syntax or semantics) are borderline acceptable themselves to many native speakers.

Ok. How would you apply this approach to ACD? It is not clear to me how to make the mystery “disappear” in this case.

I studied NLP in Prague, and we tend to use "dependency grammars" more than "phrase structure grammars" of Chomsky

https://en.wikipedia.org/wiki/Dependency_grammar

I didn't work in the field at all after finishing the studies, so I cannot really tell you any actual implications of the different grammar approach anymore.

Dependency grammars are just another grammar formalism. They’re an alternative to Chomsky’s specific technical proposals but not to his overall approach.

> Most of the Chomsky-inspired research boils down to fitting carefully selected data into an artificial formal machine and manipulating the said machine into accepting more and more data, until the machinery ends up being turing complete therefore rendering any result trivial.

Ah, that's interesting. Because that describes very well what I felt when I read the early Chomskian works. Very cool theories and models with very little data to support them. At some point, I was wonderering whether the authors of those theories had any practical experience with languages apart from their own (English) mother tongue. And then the models became more and more complex, seemingly in an attempt to make them somehow work with real data, and at the end they became so complex that any data could be matched and no scientifically verifiable predictions could be made.

somehow reminds me of the epicycle model astronomy used back then

I don't think astronomers were using epicycles in the 1960s :)

>> A fair number of faculty, while appreciative of his contributions to the field, held him in low regard because he & some of his main "acolytes" systematically worked to block any researchers in the U.S. working on ideas that ran counter to his own. Meaning more than a generation or more of advancement has stagnated in the US.

I've heard this about Chomsky before and I don't get it. It sounds like he's criticised for doing what any researcher would do- defend his work and criticise competing approaches. I know that's what I do and what everybody else I know does (and has done to me too). Of course any researcher worth her salt will defend her work and of course they will criticise competing approaches. Robustly so. That is how research works and that is how researchers work.

He "systematically worked to block any researchers in the U.S. working on ideas that ran counter to his own"? Yes, that's precisely what you do when you think the other approach is rubbish, you fight to put it out like a fire that threatens to burn down your beloved field. You only ever stop criticising others' work and promoting your own ideas when you are 100% convinced -no doubts at all- that it's your work that's flawed.

Otherwise, how does science advance? Nobody has the absolute truth in any scientific question. Our best chance to know the truth is to try to formulate an answer and then have everybody's answers duke it out, like pit fighters in Conan the Barbarian. The alternative is to let your ideas go without putting up a fight - but then you run the risk of letting ideas that are worse than yours prevail. And I'm probably getting all this from Schopenhauer on the art of dialectic, which I read a while ago, but this is how debates go. And science is a debate. And if you're having a debate, it means there is disagreement- that's the point of having a debate; and the point of having research: that we can't agree about what the truth is in the first place. Otherwise, we wouldn't need research.

So I think that those complaints are best explained as sour grapes by those who failed to convince of the merit of their ideas, and blame Chomsky for doing his job as a scientist better than them. Of course you're starting at a disadvantage if you're going up against an established authority in your field, no less one of the size of Chomsky, but that's exactly what everyone else spends all their time trying to become, an established authority in their field. And they want that so that they can defend their own ideas more easily. So that, too, is not such a great piece of criticism.

Thanks it is good to know there are still those who think there is a truth to be settled by politicking like real gentleman. Still I love how his favourite question of universal grammar was utterly sidelined when it was shown that problems of parser ambiguity also show up in the vision system, a much older neural structure.

Parser ambiguities have never really been a major focus of Chomsky’s research. He’s certainly never argued for the existence of UG on the basis of parser ambiguities.

What if some of the ideas that Chomsky "put out like a fire" were actually correct, and would have been discovered to be so if more time and (collaborative) effort had gone into them?

Are you accusing his sincerity? The correctness of the ideas which failed to gain traction is irrelevant. The burden is on the authors of those works to explain and decent the merit of their ideas. If they failed to convince their peer reviewers then that is on them. Unless you are accusing the peer-reviewers of acting in bad-faith.

> Are you accusing his sincerity?

No, I am pointing out the shortcomings in this statement:

>> He "systematically worked to block any researchers in the U.S. working on ideas that ran counter to his own"?

> Yes, that's precisely what you do when you think the other approach is rubbish, you fight to put it out like a fire that threatens to burn down your beloved field. You only ever stop criticising others' work and promoting your own ideas when you are 100% convinced -no doubts at all- that it's your work that's flawed.

> The correctness of the ideas which failed to gain traction is irrelevant.

Do you perceive yourself to have knowledge of what difference an alternative outcome would have made in the world?

> The burden is on the authors of those works to explain and decent the merit of their ideas. If they failed to convince their peer reviewers then that is on them.

It is also "on" anyone else who would have been affected by good ideas not being "put out like a fire". Things that do not happen are not stored in your mind, therefore it is quite natural for them to not exert much influence on your thinking.

> Unless you are accusing the peer-reviewers of acting in bad-faith.

I have no knowledge of their actions and therefore no opinion, I am simply pointing out shortcomings in thinking.

Then everybody loses, of course. But we can't know that beforehand and our best chance to arrive at the truth is to let competing ideas actually compete- and hope that the ones that surive the ordeal are the ones closest to the truth.

There is a lot of variability available within the vague expression "let competing ideas actually compete".

I don't know what you mean by that. Could you please explain?

Ideas can compete on a level playing field, purely on their merits, or they can be "put out like a fire" by someone with power and influence, and many variations in between.

Exactly, science is vigorous debate.

When ever you see someone trying to quell debate, you should immediately recognize it as counter to the scientific method.

"Science progresses one funeral at a time"

https://en.wikipedia.org/wiki/Planck%27s_principle

I hold him in low regard because he defended Chavez for years while his dictatorship destroyed my country.

Disclaimers: It’s inevitable that his political activism will be a topic here, and I’m sensitive to the harm Chavez did, and I may very well be out of the loop.

That said I’m not aware of Chomsky defending Chavez, though I’m certainly aware of him challenging US intervention. And his general stance of non-intervention is often interpreted as support for the would-be intervened regime.

Is there a more clear position of support/defense on record? I know that I’ve been disappointed by some of his later in life positions dismissing Palestinian efforts, it wouldn’t surprise me if he’s regressed on other topics, I’m just not familiar in this case. And asking sincerely.

Chomsky has more citations to his peer-reviewed work, than your whole former department or maybe university.

I don’t agree with him on numerous topics, but that doesn’t mean that I have a credible critic of his work.

This is not a critic. This is a complaint that that dominance sucked up a lot of mental freedom and dareness to break out established regime. That's not even complaint to most people who are doing professional research, i.e., modern professional research is just like any other industrialized profession, it requires profit. Of course, with that comes conscious or unconscious desire to follow the lead. That inevitably manifests into low esteem to some prominent figure, but when you look closely, that low esteem is nothing but a sentiment towards the lack of resources; you almost never see any well regarded openly criticize people without different views, because intellectual ideas being different is just the inherent situation of any academic research, everyone knows that, otherwise one would not actually want to be in that profession in the first place.

Isn't this more or less how a paradigm shift would look from the perspective of someone who sticks to the old paradigm?

I find it very hard to believe. Could you please provide any citations.

Eric Weinstein (The Portal podcast on Youtube) has run into a similar problem with physics. He's a mathematician by trade and over the last 37 years has developed a new theory of everything he calls Geometric Unity. He wasn't able to make any headway into the physics community for a long time - and so now instead he's educating people about what he calls DISC and GIN - about the problems with these systems, and building his own audience of people while waiting for peers to get interested in learning/understanding his theory.

It's happened in health too, arguably it has in all subjects, whether due to regulatory capture of government policy and/or academic capture influencing what research gets done via funding - the system is broken, well - it works for industrial complexes who generally have ulterior motive than wanting the very best for society.

Edit to add: why the hell would anyone downvote this comment?

Can't speak for everyone else, but I personally think you are misrepresented the Eric Weinstein story. From all that I have read about him, it seems like the academic peer review system did it's job: kept a charlatan with half baked ideas out of reputable journals. Sometimes when no one takes you seriously it doesn't necessarily mean there is a conspiracy against you.

"kept a charlatan with half baked ideas out of reputable journals."

LOL. Citations needed. Please point to the peer reviewed criticism that disproved his math. You've clearly never watched him speak to people about mathematics - there hasn't been a person he hasn't been able to dance with when discussing mathematics/physics principles.

  criticism that disproved his math

has weinstein made any testable predictions with his theory/math?

https://www.youtube.com/watch?v=Z7rd04KzLcg - him releasing the video of his multi-hour long guest lecture.

He's working on publishing a paper version soon to hash out all the math for people.

As far as I know predictions wise, all of the math equations fit within his foundation, allowing a connection between string theory and relatively - which AFAIK didn't previously have math that allowed them to interconnect/co-exist.

He's been trying for 15 years to get people in physics interested in it - and realized he has run into what he calls DISC and GIN, as mechanisms that keep outsiders ideas out and status quo narratives perpetuated. And so he's built an audience to educate, to share interviews/talks with high level conversation with a wide range of people, and build momentum/attention for his Geometric Unity theory because the existing institutions seem so indoctrinated into the last 50 years that they aren't looking into other ideas (yet); and an example of this is right here on HN, the previous commenter who called him a charlatan - with what experience are they themselves basing that off of, or who did they hear that narrative from, and how much time did that person invest investigating [interesting how invest is in the word investigating, it takes time which is an investment] and did they actually engage with Eric to see if they perhaps weren't understanding themselves and so then Eric could help them understand, etc.

Fortunately science is not about being able to dance with people in a conversation or showmanship of any sort - it's about making falsifiable hypotheses and then carrying out experiments to test them. It requires a lot of tedious work where you have to convince your peers that your ideas are solid and worthy of publication. If someone fails to do that, maybe, just maybe, the reason is not a vast conspiracy against this person because his ideas are too powerful and academia too stupid to see the truth, but rather the ideas could just be wrong.

Hey, can I connect with you 1-1 to discuss your post-grad work? I’ve been playing around with the idea of getting a graduate degree in linguistics, but with a CS-focus. Would love to hear from someone who’s been through it.

Reply on news.ycombinator.com